{"paper_id":"8b2f64f0-5da1-4a8f-b6d8-c9a683d140f5","body_text":"An investigation of basic science and clinical research \nmethodologies to benefit clinical practice \n \n \n \nRAJESH VARMA  \nMA(Cantab.), MBBS (Hons.) MRCOG \n \n \n \nA  thesis  submitted  to \nThe University  of  Birmingham \nfor  the  degree  of \nDOCTOR  OF  PHILOSOPHY \n \n      \n \n \n \n \nStudent id 573790 \n \nAcademic Department of Obstetrics and Gynaecology \nDivision of Reproductive and Child Health \nUniversity of  Birmingham \n1st submitted 8th February 2008 \n(Amended after PhD Viva 23rd September 2008 and re-submitted 19th December 2008)   \n\n \n \n \n \n \n \n \n \n \nUniversity of Birmingham Research Archive \n \ne-theses repository \n \n \nThis unpublished thesis/dissertation is copyright of the author and/or third \nparties. The intellectual property rights of the author or third parties in respect \nof this work are as defined by The Copyright Designs and Patents Act 1988 or \nas modified by any successor legislation.   \n \nAny use made of information contained in this thesis/dissertation must be in \naccordance with that legislation and must be properly acknowledged.  Further \ndistribution or reproduction in any format is prohibited without the permission \nof the copyright holder.  \n \n \n \n\nContents \nii \n \nContents of Thesis \nCONTENTS OF THESIS II \nList of tables included in Thesis v \nList of figures included in Thesis viii \nPRELIMINARIES X \nAbstract x \nAcknowledgements xi \nINTRODUCTION TO THESIS 1 \nPublications and presentations arising from Thesis 5 \nCHAPTER 1: INVESTIGATING THE GENETIC AND MOLECULAR AETIOLOGY OF \nENDOMETRIOSIS 7 \nIntroduction 7 \n1.1: Exploring the hypothesis that endometriosis is a neoplastic precursor to ovarian cancer 8 \nA) Clinico-pathological AND epidemiological data 15 \nB) Genetic and molecular data 18 \nC)  Testing causality of endometriosis and ovarian cancer using Bradford Hill criteria 27 \nD. Summarising the published evidence whether endometriosis is a neoplastic precursor to ovarian cancer\n 35 \nDiscussion of methodology used in testing hypothesis 38 \n1.2. Experimental investigation of endometriosis and EAOC 39 \nIntroduction 39 \nHypothesis 39 \nPlan of investigation 41 \nExperimental Methods 42 \n1.3 Investigation of epidemiological factors associated with EAOC and SOC 49 \n1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 using multiple \nmicrosatellite genetic markers and their prognostic significance. 56 \nRationale for selecting Glycodelin and Progesterone Receptor as candidate disease-modifying  genes 63 \n1.5. Laser Capture Microdissection (LCM) of endometriosis and selected LOH mapping 64 \n1.6 Immunohistochemical investigation of EAOC using Glycodelin (9q34.3) and Progesterone receptor \n(11q22) 65 \n\nContents \niii \n \n1.7. Preliminary nuclear morphometric analysis of endometriosis adjacent to ovarian cancer 69 \n1.8.  Affymetrix SNP DNA microarray genotyping of ovarian endometriosis 70 \nCHAPTER 2. ANALYTICAL OBSERVATIONAL STUDIES 76 \nIntroduction 77 \n2.1. Predicting negligence in female sterilization failure using time interval to sterilization failure: analysis of \n131 cases 79 \n2.2. The effectiveness of a levonorgestrel-releasing intrauterine system (LNG-IUS) in the treatment of \nendometrial hyperplasia – a long-term follow-up study. 96 \n2. 3. Hospital recovery following Thermachoice ablation is not dependent on setting (outpatient or daycase) \nor rescue analgesia: unexpected result 111 \n2.4. Outpatient Thermachoice endometrial balloon ablation: long-term, prognostic and quality of life \nmeasures 125 \n2.5. Long term outcomes following hysteroscopic myomectomy for abnormal uterine bleeding  148 \nCHAPTER 3. SYSTEMATIC REVIEWS AND SYSTEMATIC LITERATURE APPRAISAL\n 162 \nIntroduction 163 \n3.1. Do screening-preventative interventions in asymptomatic pregnancies reduce the risk of preterm \ndelivery: a systematic review and meta-analysis 168 \nEvidence for the value of screening-preventative interventions on routine antenatal population screening\n 173 \nEvidence for screening-preventative interventions based on routine antenatal care plus specialist \ninvestigations 176 \nEvidence for the value of screening-preventative strategies in specific high risk groups 183 \nEvidence for the value of population-wide preventative strategies in high and low risk groups 186 \nAntenatal management plan and role of specialist antenatal prematurity clinics 190 \nDiscussion 193 \nDiscussion on bacterial vaginosis meta-analyses 197 \n3.2. Non-contraceptive uses of levonorgestrel releasing hormone system (LNG-IUS)- a systematic enquiry \nand overview 198 \nDiscussion 229 \nCHAPTER 4: CLINICAL GUIDELINE DEVELOPMENT 233 \nIntroduction 234 \nSummary of evidence for each clinical guideline according to RCOG  and GRADE guideline development tools\n 240 \n\nContents \niv \n \n4.1. Birth after previous caesarean section 243 \n4. 2. What treatments improve outcomes in women with unruptured tubal ectopic pregnancy?  276 \n4. 3. Laparoscopic entry techniques: clinical guideline, national survey and medicolegal ramifications  304 \nA. Evidence based criteria for safe laparoscopic entry 309 \nB. Questionnaire survey 313 \nC. Medico-legal ramifications 319 \nDiscussion 320 \n4.4.  Minimising the Risk of Sterilisation Failure-an evidence-based approach  322 \nClinical Guideline: Minimising the risks of sterilisation failure 335 \nCHAPTER 5. THESIS CONCLUSION 343 \n5.1. Experimental investigation of endometriosis 344 \n5.2. Observational Analytical Studies 350 \n5.3. Systematic reviews 355 \n5.4. Clinical guideline development 358 \n5.5. Future research themes arising from Thesis 363 \nIntegrating genomic, transcriptomic and proteomic high throughput technology 363 \nAnatomical and molecular re-classification of endometriosis 363 \nTissue banks for endometriosis (and other important diseases) 364 \nDeveloping and utilising animal models of disease 364 \nImproved basic science and clinical science collaboration 364 \nThe need to increase translational potential of basic science research 365 \nCo-ordinated research programmes and commitment from Government funded research bodies 365 \nUtilising and developing high quality clinical datasets 366 \nCaution with over-reliance on meta-analyses: value according to the quality and methodology of the RCTs \nincluded 367 \nEnsure that clinical guidelines and their utilisation adds value to clinical practice 368 \nREFERENCES 369 \nChapter 1 372 \nChapter 2 395 \nChapter 3.1 406 \nChapter 3.2 423 \nChapter 4 437 \nChapter 5 481 \n\nList of Tables included in Thesis \nv \n \nList of tables included in Thesis \n \nTable Title \n \nPage  \nA * Classification of evidence used by RCOG guideline development 3 \n1.1 Hanahan‘s criteria of properties exhibited by a cancer cell ‗the hallmarks of \ncancer‘  \n12 \n1.2 How endometriosis displays the ‗Hallmarks  of cancer‘ 13 \n1.3 Criteria and fulfillment of Bradford Hill criteria of causality for \nendometriosis and ovarian cancer \n14 \n1.4 Risk of ovarian cancer and other types of cancer in women with \nendometriosis \n30 \n1.5 Prevalence of ovarian cancer in women with and without endometriosis 31-32 \n1.6 Summarising the published evidence that supports or refutes the hypothesis \nthat endometriosis is a neoplastic precursor to the development of ovarian \ncancer \n35 \n1.7 Characteristics of endometriosis associated ovarian cancer (EAOC) and \nsporadic ovarian cancer (SOC), matched for endometrioid and clear cell \nhistologies,  used in epidemiological analysis (N=62) \n50 \n1.8 Characteristics of endometriosis associated ovarian cancer (EAOC) and \nsporadic ovarian cancer (SOC), matched for endometrioid and clear cell \nhistologies,  used in genetic analysis(N=50) \n51 \n1.9 Multivariate survival regression analysis of EAOC and SOC 52 \n1.10 Allelic Loss at chromosome 9 57-58 \n1.11 Allelic loss at chromosome 11 59-60 \n1.12 Genome wide microsatellite analysis of  endometriosis adjacent to ovarian \ncancer \n64 \n1.13 Summary of immunohistochemistry findings 65 \n1.14 Summarising genome-wide LOH regions identified in ovarian endometriosis \nthrough SNP Affymetrix microarray analysis \n70 \n2A Advantages and Disadvantages as displayed by Centre for Evidence-Based \nMedicine (Oxford, UK; www.cebm.net ) \n78 \n2.1 Filshie Clip sterilisation failure rates 81-82 \n2.2 Databases used to acquire failed sterilisation records 85 \n2.3 Sterilisation method and time interval to pregnancy 87 \n2.4 Negligent and Non-negligent failure group compositions and intervals to \npregnancy \n88 \n2.5 Empirical probabilities and likelihood ratios at incremental time intervals. 90-91 \n2.6 Baseline characteristics (n=105) of LNG-IUS treatment of endometrial \nhyperplasia \n102 \n2.7 Outcome of the study according to histological data derived from outpatient \nendometrial Pipelle and hysterectomy histologies \n103 \n2.8 Correlation between endometrial Pipelle histology and hysterectomy \nhistology (n=23 hysterectomies) \n107 \n2.9 Baseline and procedural characteristics of LA vs GA TBEA \n \n119 \n\nList of Tables included in Thesis \nvi \n \nTable Title \n \nPage  \n2.10 Outcomes of LA vs. GA TBEA 120 \n2.11 Regression analysis 121 \n2.12 Baseline demographic data for outpatient TBEA 130-131 \n2.13 Peri-procedure outcomes of outpatient TBEA 132 \n2.14 Long-term outcomes of outpatient TBEA 139-140 \n2.15 Patient satisfaction and its relationship to quality of life and other treatment \noutcomes following endometrial ablation \n141 \n2.16 Prognostic outcomes for endometrial ablation (using multivariate regression \nanalysis) \n144 \n2.17 Baseline characteristics for 92 women undergoing hysteroscopic \nmyomectomy \n155 \n2.18 Characteristics associated with hysteroscopic myomectomy procedure 156 \n2.19 Outcomes after hysteroscopic myomectomy 157 \n2.20 Women (n=10) undergoing hysterectomy following hysteroscopic \nmyomectomy \n158 \n2.21 Multivariate analysis of prognostic factors 159 \n3i * Classification of evidence used by RCOG Guideline development (originate \nfrom US Agency for Health Care Research and Quality) \n165 \n3ii * GRADE approach- The Grading of Recommendations Assessment, \nDevelopment and Evaluation (GRADE) \n166-167 \n3.1 Gestation-specific perinatal mortality 169 \n3.2 Risk factors associated with increased risk of preterm delivery.    172 \n3.3 Defining elective and indicated types of cervical cerclage 175 \n3.4 Suggested antenatal strategy to prevent preterm delivery 192 \n3.5 Summary of screening and preventative strategies that may reduce the risk of \npreterm delivery \n194 \n3.6 Summary of studies that assess LNG-IUS use in various non-contraceptive \ntherapeutic indications as primary study outcome measures \n202-203 \n3.7 LNG-IUS studies assessing therapeutic effect in women with menorrhagia 206-208 \n3.8 LNG-IUS studies directly or indirectly assessing therapeutic effect on \nfibroids or fibroid related menorrhagia \n210-211 \n3.9 LNG-IUS studies assessing therapeutic effect in women with endometriosis 213-214 \n3.10 LNG-IUS studies assessing therapeutic effect in women with adenomyosis 215 \n3.11 LNG-IUS studies assessing use to provide uterine protection during \noestrogen replacement or tamoxifen therapy \n217-223 \n3.12 LNG-IUS studies assessing therapeutic effect in women with endometrial \nhyperplasia \n225 \n4.1 Guideline publications arising from chapter 4 233 \n4i Assessment criteria for selecting topics for clinical guideline development 235 \n4ii * Classification of evidence used by RCOG Guideline development (originate \nfrom US Agency for Health Care Research and Quality) 7 \n236 \n4iii  Classification of evidence used by Scottish Intercollegiate Guidelines \nNetwork (SIGN) Grading System  \n237 \n4iv * GRADE approach. The Grading of Recommendations Assessment, \nDevelopment and Evaluation (GRADE) \n238-239 \n\nList of Tables included in Thesis \nvii \n \nTable Title \n \nPage  \n4v Summary of evidence for each clinical guideline according to RCOG  and \nGRADE guideline development tools \n240-242 \n4.2 Definition of obstetric terms 245 \n4.3 Definition of perinatal terms 245 \n4.4 Items to be discussed when determining mode of delivery 250 \n4.5 Risks and Benefits of opting for VBAC or ERCS 251-252 \n4.6 Clinical features associated with uterine scar rupture 267 \n4.7 Risks of planned VBAC labours from NICHD study (N=17,898 planned \nVBACs) 15;29 \n271 \n4.8 Management of augmentation in established VBAC labour 271 \n4.9 Glossary of terms used in ectopic pregnancy guideline 278-279 \n4.10 Comparison of fertility outcomes of salpingotomy versus salpingectomy 285-288 \n4.11 RCTs and meta-analyses of surgical and surgical versus medical treatments \nin the management of ectopic pregnancy \n293 \n4.12 Evidence-based criteria for safe laparoscopic entry: 10 steps 307 \n4.13 Laparoscopic entry technique in uncomplicated vs. high-risk women 314 \n4.14 Frequency of angle of entry for Veress and Primary Trocar 315 \n4.15 Safety checks performed to ensure correct Veress placement 316 \n4.16 Safety checks performed prior to primary trocar insertion 317 \n4.17 Awareness of evidence-based guidance and previous experience of \nlaparoscopic injury \n318 \n4.18 Female surgical sterilisation techniques 323 \n4.19 Filshie Clip: reported sterilisation failure rates 326 \n4.20 Classification system for mechanism of sterilisation failure 332 \n5.1 Individualise therapeutic approach to endometriosis according to cancer cell \nhallmarks \n346-347 \n5.2 Summary of studies comparing genomic, transcriptomic and proteomic \nprofiling of endometriosis using high-through put microarray technology \n348 \n5.3 Implications of thesis findings and future research directions for \nendometriosis \n349 \n5.4 Improved health care resulting from analytical observational studies. 351 \n5.5* Classification of evidence used by RCOG Guideline development (originate \nfrom US Agency for Health Care Research and Quality)  \n352 \n5.6* GRADE approach. The Grading of Recommendations Assessment, \nDevelopment and Evaluation (GRADE) \n353-354 \n5.7* Screening and preventative strategies that may reduce the risk of preterm \ndelivery \n356 \n5.8* Suggested antenatal strategy to prevent preterm delivery 357 \n5.9* Ectopic pregnancy evidence appraised using RCOG and GRADE criteria 361 \n \n \n\nFigures included in Thesis  \nviii \n \nList of figures included in Thesis \n \nFigure Title \n \nPage  \nA* Ascension of research pyramid 4 \n1.1 Image of an endometriotic lesion surrounded by adhesions 9 \n1.2 Acquired stepwise genetic somatic mutations that predispose to \ndevelopment of cancer (Fearon and Vogelstein 1990) \n23 \n1.3  Proposed genetic and molecular aeitopathogenesis of endometriosis 37 \n1.4 Genetic allelic products images observed following microsatellite \namplification of target DNA and analysis on ABI Prism analyser \n44 \n1.5 Importance of Laser capture microdissection of target disease (such as \nendometriosis epithelium glandular lining) from surrounding tissue (such \nas endometriosis stroma \n46 \n1.6 Increased genetic resolution of Affymetrix Single Nucleotide \nPolymoprhism DNA microarray compared to ‗traditional‘ multiple \nmicrosatellite marker genome wide mapping \n48 \n1.7 Survival differences between subtypes of ovarian cancer 53 \n1.8 Survival analysis according to stage of ovarian cancer. 54 \n1.9 Survival analysis according to presence of endometriosis 55 \n1.10 Contribution to allelic loss at chromosome 9 by each cancer subtype  61 \n1.11 Contribution to allelic loss at chromosome 11  by each cancer subtype 61 \n1.12 Significant association of LOH at 9q34.3 and 11q 23.3 on survival of all \novarian cancers according to Cox Proportional Hazards survival analysis \n62 \n1.13 Immunohistochemistry images of endometriosis and ovarian cancer using \nGlycodelin and CD10 \n66-67 \n1.14 Immunohistochemistry images of endometriosis and ovarian cancer using \nProgesterone receptor subtypes A and B  (individually labeled) \n67 \n1.15 Immunohistochemistry: patchy positive staining of PR-B in endometrioid \ncancer \n68 \n1.16 Nuclear morphometric analysis of endometriosis, atypical endometriosis \nand ovarian cancer that appear as one continuum on the histology slide \n69 \n1.17 Selected images of chromosomal abnormality (chrom 11) in ovarian \nendometriosis (patient 3) compared to their matched normal ovarian \nsurface epithelium (patient 2) \n71 \n1.18 Selected images of chromosomal abnormality (chrom 15) in ovarian \nendometriosis (patient 5) compared to their matched normal ovarian \nsurface epithelium (patient 3) \n72 \n1.19 Selected images of chromosomal abnormality (chrom 21) in ovarian \nendometriosis (patient 6) compared to their matched normal ovarian \nsurface epithelium (patient 9) \n73 \n1.20 Selected images of chromosomal abnormality (chrom 6 and chrom 11 and \nchrom X) in ovarian endometriosis for patient 2 and patent 3 and patient \n6,  respectively. \n74-75 \n2.1 The probability of sterilisation failure for negligent and non-negligent \ncases against time interval to failure (Cox Regression model) \n \n92 \n\nFigures included in Thesis  \nix \n \nFigure Title \n \nPage \n2.2 Outcome of study according to outpatient endometrial Pipelle histology at \npre-treatment and 2-years following LNG-IUS insertion \n104 \n2.3 Correlation of duration of stay with strength of analgesia for combined \nLA and GA TBEA cohort \n121 \n2.4 Correlation of morphine usage to post ablation VAS Score and duration of \nhospital stay \n138 \n2.5 Survival analysis for likelihood of surgical re-intervention post TBEA 142 \n3.1 Meta-analyses for treatment of bacterial vaginosis and reduction in \npreterm delivery \n177 \n4.1 Plan of care for singleton uncomplicated pregnancy with previous LSCS 246 \n4.2 Clinico-pathological mechanisms proposed in sterilisation failure based \non Canadian dataset 389 \n328 \n4.3 Cumulative risk of pregnancy by method from US CREST study 388 and \nFilshie clip references \n329 \n4.4 Filshie clip under-closure due to operator fault 338 \n5.1* Ascension of a research pyramid of research methodologies to benefit \nclinical practice \n343 \n5.2 Evaluating, in parallel, differences between genomic, transcriptomic and \nproteomic array platforms to identify candidate molecular pathways \n347 \n5.3 Derivation of clinical guidelines 358 \n \n \nFootnote to List of Tables & Figures included in Thesis\n \n* There has been repetition in the listing of these tables and figures in the Thesis text. This \nhas been adopted in order to help the reader when reading the particular section, rather than \nthe reader having to cross-reference to other chapters of the Thesis in order to locate the \nrelevant item.\n\nPreliminaries \nx \n \nPRELIMINARIES  \nAbstract  \n \nThe aim of this PhD thesis was to produce research that could inform and benefit clinical \npractice by exploring the application of basic science and clinical research methodologies to \ndisorders in obstetrics and gynaecology. Chapter 1’s investigation of endometriosis is the \nfirst to 1) report detailed genetic mapping of endometriosis-associated ovarian cancer, 2) \nreport the existence of micro-LOH (loss of heterozygosity)  in ovarian endometriosis through \na SNP 100K DNA array. Chapter 2 explores the efficacy of interventions to treat menstrual \nabnormalities using clinical cohort studies.  Furthermore, Chapter 2 highlights how \nnegligence in female sterilization failure may be mathematically (Bayesian) modelled. \nChapter 3 explores the value of systematic reviews for preventing preterm delivery and use \nof LNG-IUS (Mirena coil). The clinical guidelines published in Chapter 4 include: vaginal \nbirth after previous caesarean, ectopic pregnancy, safe laparoscopic entry and minimising risk \nof sterilisation failure. The thesis concludes (Chapter 5) by suggesting strategies to augment \nthe research methodological approaches evaluated in this thesis in order fulfill the aim of \nbenefitting clinical practice. Work included in this PhD thesis has been orally presented at \ninternational conferences, published in peer-reviewed journals, and published as a national \nclinical guideline by the Royal College of Obstetricians and Gynaecologists, UK (RCOG). \n \n\nPreliminaries \nxi \n \nAcknowledgements \n \nBASIC SCIENCE    Prof  ER Maher \nDr N Morgan, Dr S Sahota, Dr J Arrand \nHISTOPATHOLOGICAL   Dr R Ganesan, Dr T Rollason \nCLINICAL EPIDEMIOLOGICAL  Prof JK Gupta, Mr J Clark \n      Dr H Soneja \nSTATISTICAL    Prof P Patil \nMENTORSHIP    Prof JK Gupta & Prof ER Maher \nPhD EXAMINERS    Prof T Barrett & Prof J Konje \nSECRETARIAL    A.Intennimeo, D.Leake, H.Khan \nFUNDING     Birmingham Springboard Fellowship \n      MRC/RCOG  Clinical PhD Fellowship \n      RCOG Endometriosis Millennium Fund \nAND \nFRIENDS, FAMILY, EMPLOYERS \n \n \n \n\nIntroduction to Thesis \n1 \n \nIntroduction to Thesis \n \nClinical research, although a commonly used term, is actually difficult to achieve a \nconsensus definition for. A definition stated by Department of Health (United Kingdom) is \nthat research is ―the attempt to derive generalisable new knowledge by addressing clearly \ndefined questions with systematic and rigorous methods‖. This definition includes studies that \naim to generate hypotheses as well as studies that aim to test them 1 . The Medical Research \nCouncil (United Kingdom) aims to support research that is aimed at ― maintaining and \nimproving human health” 2; a commitment endorsed by all other research funding bodies, \nprofessional medical colleges and National Health Service (UK).  \nThere are numerous basic science and clinical research methodologies employed in clinical \nresearch. I suggest that these may be depicted as components of a ‗research pyramid‘ (Figure \nA).  For research to ultimately translate to clinical benefit, there needs to be ascension of the \npyramid to its peak through appropriate selection of the ‗next level‖ research methodology. \nThe graphical depiction is useful as it highlights methodologies existing within the context  of \na particular level that corresponds to the level of evidence that is considered during guideline \ndevelopment (Table A).  Furthermore, the pyramid shape mirrors the typical frequency of \npublications on disease, with several existing at the base, and fewer identified as the pyramid \nis ascended. \n \nMultiple components are necessary to ensure that research is relevant, effective, efficient, \nethical, and will ultimately translate to health gain.  The aim of this PhD thesis was to \nproduce research that could inform and benefit clinical practice. The chapters have been \nordered to follow a stepwise ascension of the research methodology pyramid (Figure A). \n\nIntroduction to Thesis \n2 \n \nEach chapter illustrates the use of a specific research methodology by considering selected \ndisorders in obstetrics and gynaecology.  In Chapter 1, the thesis explores the molecular \naetiology of endometriosis and tests whether it behaves as a neoplastic precursor to ovarian \ncancer. Maintaining a gynaecological theme, Chapter 2 explores the efficacy and \neffectiveness of interventions to treat menstrual abnormalities using clinical cohort studies; \nthis work also led to the production of a RCOG educational module for specialist trainees in \nabnormal uterine bleeding 3.  Furthermore, Chapter 2 highlights how rare outcome measures, \nsuch as failed sterilisation, may be adequately explored using cohort study design and \nBayesian mathematical modelling. Chapter 3 explores the clinical value and potential \ndrawbacks of systematic review by assessing screening-preventative interventions to reduce \nthe risk of preterm delivery.  In addition, the chapter includes a systematic review of the non-\ncontraceptive uses of Levonorgestrel-releasing hormone system. The publications of the \nclinical guidelines in Chapter 4 are likely to have immediate and maximal benefit on clinical \npractice. The production of clinical guidelines adopted a structured approach and considered \nall levels of research evidence, not just systematic reviews or RCTs, to generate \nrecommendations for best medical practice. The guidelines included: vaginal birth after \nprevious caesarean (RCOG national guideline), ectopic pregnancy (BMJ Clinical evidence), \nsafe gynaecological laparoscopic entry, and minimising the risk of sterilisation failure.  The \nthesis concludes (Chapter 5) by summarising the benefits to clinical practice for each \nresearch methodology. The chapter also suggests future research themes that may augment \nthe research methodological approaches evaluated in this thesis in order to benefit clinical \npractice. \n\nIntroduction to Thesis \n3 \n \nTable A Classification of evidence used by RCOG guideline development \nClassification of Evidence Levels  \nIa Evidence obtained from meta-analysis of randomised controlled trials.  \nIb Evidence obtained from at least one randomised controlled trial.  \nIIa Evidence obtained from at least one well-designed controlled study without \nrandomisation.  \nIIb Evidence obtained from at least one other type of well-designed quasi-experimental study.  \nIII Evidence obtained from well-designed non-experimental descriptive studies, such as \ncomparative studies, correlation studies and case studies.  \nIV Evidence obtained from expert committee reports or opinions and/or clinical experience of \nrespected authorities.  \nGrades of Recommendations  \n \nRequires at least one randomised controlled trial as part of a body of literature of \noverall good quality and consistency addressing the specific recommendation. \n(Evidence levels Ia, Ib) \n \nRequires the availability of well controlled clinical studies but no randomised \nclinical trials on the topic of recommendations. (Evidence levels IIa, IIb, III) \n \nRequires evidence obtained from expert committee reports or opinions and/or \nclinical experiences of respected authorities. Indicates an absence of directly \napplicable clinical studies of good quality. (Evidence level IV) \nGood Practice Point    \n \nRecommended best practice based on the clinical experience of the guideline  \ndevelopment group \n \n \n\nIntroduction to Thesis \n4 \n \nFigure A. Ascension of research pyramid \nAudit\nClinical \nGuidelines\nSystematic \nreviews & RCTs\nCohort studies\nDescriptive studies \nElucidating aetiopathogenesis\nmolecular  in vitro &  in vivo models\nMolecular & epidemiological associations\n \nFootnotes \nRCT randomised controlled clinical trials \nAudit refers to clinical audit to assess impact of clinical guidelines \n\nIntroduction to Thesis \n5 \n \nPublications and presentations arising from Thesis  \n \nCHAPTER ONE \nProwse AH, Manek S, Varma R, Liu J, Godwin AK, Maher ER et al. Molecular \ngenetic evidence that endometriosis is a precursor of ovarian cancer. Int J Cancer \n2006; 119(3):556-562. \n4 \nVarma R, Rollason T, Gupta JK, Maher ER. Endometriosis and the neoplastic \nprocess. Reproduction 2004; 127(3):293-304. \n5 \nGenome wide SNP 100K analysis of endometriosis: demonstration of genomic \nimbalance, Oral Presentation, British Congress of Obstetrics and Gynaecology \nRCOG, London, July 2007. \n6 \nGenetic evidence for malignant transformation of endometriosis (1st July 2007, \nOral Presentation, European Society for Human Reproduction & Embryology \n(ESHRE), Lyon, Paris. \n7 \nCHAPTER TWO  \nVarma R, Gupta JK. Predicting negligence in female sterilization failure using \ntime interval to sterilization failure: analysis of 131 cases. Hum Reprod 2007; \n22(9):2437-2443. \n8 \nVarma R, Soneja H, Bhatia K, Ganesan R, Rollason T, Clark TJ et al. The \neffectiveness of a levonorgestrel-releasing intrauterine system (LNG-IUS) in the \ntreatment of endometrial hyperplasia--a long-term follow-up study. Eur J Obstet \nGynecol Reprod Biol 2008; 139(2):169-175. \n9 \nVarma R, Soneja H, Samuel N, Sangha E, Clark TJ, Gupta JK. Hospital recovery \nfollowing Thermachoice ablation is not dependent on setting (outpatient or \ndaycase) or rescue analgesia: unexpected result. Eur J Obstet Gynecol Reprod \nBiol 2008; 140(1):76-81. \n10 \nVarma R, Soneja H, Clark TJ, Gupta JK. Outpatient Thermachoice endometrial \nballoon ablation: long term, prognostic and quality of life measures. European \nJournal of Obstetrics & Gynaecology and Reproductive Biology. In submission, \n2008. \n11 \nVarma R, Soneja H, Clark TJ, Gupta JK. Hysteroscopic myomectomy for \nmenorrhagia using Versascope trade mark bipolar system: Efficacy and \nprognostic factors at a minimum of one year follow up. Eur J Obstet Gynecol \nReprod Biol 2008. In Press \n12 \nVarma R, Gupta JK. Royal College of Obstetricians and Gynaecologists. \nAbnornal Uterine Bleeding. Module 13. StratOG.net. Stuctured Training \nResource to Assist Trainees in Obstetrics and Gynaecology. June 2007.  \n3 \n\nIntroduction to Thesis \n6 \n \nCHAPTER THREE  \nVarma R, Gupta JK, James DK, Kilby MD. Do screening-preventative \ninterventions in asymptomatic pregnancies reduce the risk of preterm delivery--a \ncritical appraisal of the literature. Eur J Obstet Gynecol Reprod Biol 2006; \n127(2):145-159. \n13 \nVarma R, Gupta JK. Antibiotic treatment of bacterial vaginosis in pregnancy: \nmultiple meta-analyses and dilemmas in interpretation. Eur J Obstet Gynecol \nReprod Biol 2006; 124(1):10-14. \n14 \nVarma R, Sinha D, Gupta JK. Non-contraceptive uses of levonorgestrel-releasing \nhormone system (LNG-IUS)--a systematic enquiry and overview. Eur J Obstet \nGynecol Reprod Biol 2006; 125(1):9-28. \n15 \nCHAPTER FOUR  \nVarma, R, Gupta, J.K., Smith, G.C. Birth after previous caesarean section. Royal \nCollege of Obstetricians and Gynaecologists Clinical Green top guideline No.45. \nFebruary 2007. http://www.rcog.org.uk/index.asp?PageID=1913  \nVarma R, Smith GC. Management of women with previous caesarean section. In \nPress. In: Warren R, Arulkumaran S, editors. Best Practice in Labour and \nDelivery. Cambridge University Press, Cambridge, UK.; 2008. \n16 \n \n17 \nVarma R, Gupta JK. Ectopic Pregnancy. \nhttp://clinicalevidence.bmj.com/ceweb/conditions/pac/1406/1406_background.jsp \n. BMJ Clinical Evidence . 2006. \n18 \nVarma R, Gupta JK. Laparoscopic entry techniques: clinical guideline, national \nsurvey, and medicolegal ramifications. Surg Endosc 2008; 22(12):2686-2697. \n19 \nVarma R, Gupta JK. Failed sterilisation: evidence-based review and medico-legal \nramifications. BJOG 2004; 111(12):1322-1332. \nVarma R, Gupta JK.  Minimizing the risk of sterilization faliure: An evidence \nbased approach.  In: Complications in Gynecological Surgery.  Editor: \nO'Donovan P.  Spinger-Verlag, London 2008.  Chapter 12; pages 106-126 \n20 \n \n21 \n\nChapter 1. Endometriosis: Basic Science  \n \n 7 \nCHAPTER 1: INVESTIGATING THE GENETIC AND MOLECULAR \nAETIOLOGY OF ENDOMETRIOSIS \n \nIntroduction \n \nEndometriosis is a gynaecological disorder affecting 10-15% of women of reproductive age. \nThe condition often presents with infertility and pelvic pain, causing significant impairment \nof quality of life.  The precise aetiology of endometriosis is unclear, but is considered to \ninvolve multiple genetic, environmental, immunological, angiogenic and endocrine \nprocesses. Although endometriosis is a benign disorder, recent studies suggest endometriosis \ncould be viewed as a neoplastic process. This chapter presents a basic science investigation of \nthe genetic and molecular aetiology of endometriosis. The similarities between endometriosis \nand neoplasia have been used to investigate endometriosis using techniques normally applied \nin cancer biology. Initially, the chapter presents the epidemiological, genetic and molecular \nevidence that justifies the rationale for using a cancer biology model to study endometriosis. \nThereafter, the chapter discusses various genetic and immunohistochemistry techniques used \nin the investigation. Traditional approaches (such as microsatellite genetic marker genetic \nmapping) are contrasted with newer technologies (laser capture microdissection and \nAffymetrix SNP 100K DNA microarray). The aim was to identify the key genes involved in \nthe initiation, proliferation and malignant transformation of endometriosis to enable \ndevelopment of improved screening-preventative therapies for both endometriosis and \novarian cancer.\n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 8 \n1.1: Exploring the hypothesis that endometriosis is a neoplastic precursor \nto ovarian cancer \n \nIntroduction \nEndometriosis is defined as the implantation of endometrium-like glandular and stromal cells \noutside their normal location in the uterus. Endometriotic lesions are usually identified at \nlaparoscopy localised to ovaries and the Pouch of Douglas (Figure 1.1). Endometriosis is \ndiagnosed in 30% of cases referred for infertility investigation 1 and in 10%-70% of women \nwith pelvic pain 1 . Overall, studies estimate that endometriosis may affect around 7-15% of \nwomen of reproductive age, thus making this a common condition. \nEndometriosis has been considered a ‗disease‘ because it is often identified when \ninvestigating women with infertility, pelvic pain, dyspareunia (pain on intercourse) and \ndysmenorrhoea (painful periods). Traditionally the classification of endometriosis has been \nmade by anatomical (surgical staging by revised American Fertility Society score) and \nhistopathological (atypical and non-atypical endometriosis) criteria 2. However, this \ncombined approach of classification does not correlate closely with pelvic pain or \nreproductive outcome, and is prone to inter-observer error. Furthermore, the emphasis of \ntargeting the endometriotic lesion, by surgical removal or hypo-oestrogenic inactivation, does \nnot necessarily correct the aberrant underlying molecular mechanism(s). This explains why \ncurrent endometriosis treatment does not alleviate clinical symptoms in all cases, and \nrecurrence is common 3.  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 9 \nFigure 1.1. Image of an endometriotic lesion surrounded by adhesions  \n \nThese disparities suggest that endometriosis may not be a true ‗disease‘ but a heterogeneous \nentity with differing subtypes. One subtype may be capable of causing symptomatic disease \ndirectly consequent to endometriotic pathology (e.g. ovarian endometriomas, pelvic \nadhesions), whereas, another subtype, may be associated with symptoms without obvious \nendometriotic-lesion basis.  Another subtype may be clinically asymptomatic and its presence \nbe considered a normal ‗non-pathogenic‘ phenomena. Consequently the current focus on \ntreating the endometriotic lesion should be reconsidered, and efforts to understand the \npathogenesis of endometriosis, and its temporo-spatial relationship to symptomology, should \nbe increased. \n \n \n \n \n \n \n \n \n\n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 10 \nTraditionally, endometriosis research has focused on the lesion itself and comparing \nmolecular processes between ectopic and eutopic endometrium4. This has identified multiple \nanomalies in genetic, environment, angiogenic, endocrine, metabolic, and immunological \nmechanisms. Some of these correlate with the severity of endometriosis and/or associated \nclinical sequelae implying a causative rather than simply associative role. However, the major \nobstacle has been the difficulty in discriminating between processes fundamental to \nendometriosis aetiopathogenesis and epiphenomena. Importantly, these physiological \ndifferences are multi-compartment (endometrium, peritoneal fluid, follicular fluid and blood) \nand not just localised to the site of the endometriotic lesion, implying a fundamental \nwidespread alteration in reproductive tract function. This multifactorial multi-compartment \npathogenesis, coupled with the clinical heterogeneity, has created a confusion of data, with \nlittle consensus on a unifying mechanism. Nevertheless, since Sampson first reported in \n19255 that endometriosis may give rise to malignant change, and proposed criteria for \ndiagnosis of malignancy arising in endometriosis, extensive evidence for an association \nbetween endometriosis and cancer (especially ovarian) has now accumulated.  \n \nAim To evaluate the hypothesis that endometriosis is a neoplastic precursor to the \ndevelopment of ovarian cancer based on systematic literature search and critical appraisal of \nclinical and basic science data  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 11 \nMethods All observational and experimental studies examining the relationship between \nendometriosis and ovarian cancer were retrieved from MEDLINE (1966-2004) and EMBASE \n(1980-2004) medical databases using combination of specific keywords and MeSH terms. \nThe following search terms and word variants were used: ‗endometriosis‘, ‗endometriotic‘, or \n‗endometrio$‘ combined with ―AND‖ to ‗ovarian neoplasms‘, ‗neoplasms‘, ‗carcinoma‘ \n‗genital neoplasms, female‘, ‗carcinogens‘, ‗carcinogen$‘, ‗carcinogens, enivronmental‘, \n‗tumo$‘, ‗malignan$‘, ‗cancer$‘, or ‗neoplas$‘. In addition, bibliographies of retrieved \narticles were examined to identify further relevant studies. The search was completed in April \n2004. At the time of submission of this PhD thesis, a further search of the medical databases \nwas performed and specific key articles have been included where they substantially alter the \nevidence-base. The hypothesis was examined by examining by considering the following \nmethodological approaches: \nA. Clinico-pathological epidemiological data \nB. Genetic and molecular data of endometriosis and cancer. In particular, considering \nhow endometriosis demonstrates a molecular cancer phenotype according to \nHanahan’s  ‘Hallmarks of Cancer6‘ criteria for a cancer cell [defined as seven \ncritical features of the cancer phenotype (Tables 1.1 and 1.2)] \nC. Testing association vs. causality of endometriosis and cancer using Bradford-Hill \n(1965) epidemiological causality criteria7 (Table 1.3). \nD. Summary of the published evidence that supports or refutes the hypothesis that \nendometriosis is a neoplastic precursor to the development of ovarian cancer. \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 12 \nTable 1.1 Hanahan’s criteria of properties exhibited by a cancer cell: ‘the hallmarks of \ncancer’ 6 \n1 Self-sufficiency in growth \nsignals \nMitotic growth signals are needed for cells to move from a quiescent \nstate into active proliferative state. These signals are transmitted into \nthe cell by transmembrane cell-surface receptors that bind to: \ndiffusible growth factors, ECM components, cell-to-cell adhesion \ninteraction molecules \n2 Insensitivity to \nantiproliferative signals \nGrowth inhibitory signals (soluble or immobilized in ECM and on \nsurfaces of nearby cells) are received by transmembrane cell-surface \nreceptors coupled to intracellular signaling circuits \n3 Resistance to Apoptosis Evasion mechanisms of programmed cell death \n4 Limitless replicative \npotential \nDisruption of intrinsic cell-autonomous program that limits their \nmultiplication. This program operates independently of the cell-to-cell \nsignaling pathways described above \n5 Sustained angiogenesis Virtually all cells in a tissue are obligated to reside within 100um of a \ncapillary blood vessel to allow adequate permeation of oxygen and \nnutrients crucial for cell survival The cells within aberrant \nproliferative lesions initially lack angiogenic ability, but in order to \nprogress, incipient neoplasias must develop angiogenic ability \n6 Tissue invasion and \nmetastasis \nThis enables cancer cells to escape the primary tumour mass and \ncolonise new sites where, at least initially, nutrients and space are not \nlimited \n7 Genomic instability Mutations or inactivation/activation of tumour suppressor genes, \noncogenes, DNA monitoring and repair enzymes, checkpoint systems \nat mitosis. These are carried out by intragene (e.g. mutation, deletion) \nand epigenetic (e.g. promoter hypermethylation) mechanisms \nFootnotes \nThe listed capabilities are mostly acquired directly, or indirectly, through changes in the \ngenomes of cancer cells. \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 13 \nTable 1.2 How endometriosis displays the ‘Hallmarks  of cancer’6 \nHallmarks of cancer How endometriosis demonstrates the signified hallmark \n1 Self-sufficiency in \ngrowth signals \nIncreased local production of estrogen and responsiveness to estrogen8 \nInherited genetic polymorphisms in drug-metabolizing enzymes (CYP1A1, \nCYP19, and GSTM1) which predispose to endometriosis9 and ovarian \nendometrioid and clear cell cancer10 \n2 Insensitivity to \nantiproliferative \nsignals \nExpression of the inhibitory progesterone receptor isoform PR-A instead of the \nstimulatory isoform PR-B 11 \nAltered expression of p27Kip1 protein (cdk inhibitor) in active and inactive \nendometriotic lesion, and increased p21 expression in endometriomas compared \nwith benign and malignant ovarian tumours 12;13 \n3 Resistance to \nApoptosis \nElevated soluble Fas ligand and IL-8 in endometriotic peritoneal fluid.14 \nGermline and somatically acquired inactivating mutations of p53 gene 15;16 \nUp-regulation of survivin, matrix metalloproteinases, and bcl-2, and decreased \nBAX 17-19 \n4 Limitless \nreplicative \npotential \nNo studies examining telomerase function in endometriosis, but it is noted \nestrogen and progesterone stimulate, whilst tamoxifen and wild-type (normal \nvariant) p53 inhibit, telomerase activity in estrogen dependent neoplasms (breast \nand endometrial cancer cells) 20;21 which endometriosis represents. \n \n5 Sustained \nangiogenesis \nPathological angiogenesis, immune cell suppression and immune cell activation \nco-exist in endometriosis22 and cancer processes23. Mediators of angiogenesis \nexhibit genetic polymorphisms that either predispose to endometriosis (e.g. \nICAM-1, IL-6 and IL-10 gene promoters)24-26 or ovarian cancer (e.g. IL-6, MMP-\n1, integrin beta3, TGFBR1 ,IL-1R antagonist)27-31 \n6 Tissue invasion \nand metastasis \nEndometriosis exhibits invasiveness that is mediated through de-regulation of \nsimilar cell adherence signaling (such as integrins, beta-catenin, cadherins and \nmatrix metalloproteinases17-19;32-34 to cancer. Beta catenin mutations occur in \nendometrial and ovarian endometrioid cancers 35;36 but have not been investigated \nfor in endometriosis. \n7 Genomic instability Like cancer, endometriosis is monoclonal37 and shows allelic imbalance38. \nMutations of tumour suppressor genes occur in endometriosis39;40, which are in \nsome cases similar to those ovarian cancers arising directly from the \nendometriosis41. \n  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 14 \nTable 1.3  Criteria and fulfillment of Bradford Hill criteria7 of causality for \nendometriosis and ovarian cancer \n \nListe\nd \nfactor \n \nCausality criteria \nComments \n \nStrength of supporting \nevidence identified by \nthis review \n1 Temporal sequence Did exposure precede outcome? \n \nWeak \n2 Strength of association  How strong is the effect, measured as relative \nrisk or odds ratio? \n \nModerate \n3 Consistency of \nassociation \nHas the effect been seen by others? \n \nStrong \n4 Biological gradient  \n \nDoes increased exposure result in more of the \noutcome (dose-dependency)? \n \nNone \n5 Specificity of \nassociation \nDoes exposure lead only to outcome? \n \nWeak \n6 Biological plausibility Does the association make pathophysiological \nsense? \n \nModerate \n7 Coherence with \nexisting knowledge \n \nIs the association consistent with available \nevidence? \nWeak \n8 Experimental evidence Has a randomized controlled trial been done? \n \nHuman-Weak \nAnimal-Strong \n \n9 Analogy Is the association similar to others? \n \nWeak \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 15 \n A) Clinico-pathological AND epidemiological data \n \n1.Histopathology. Like malignancy, endometriosis displays features of atypia, adherence, \ninvasion and metastases. Atypical endometriosis is characterized histologically by \nendometrial glands with cytological or architectural atypia 42, and has been observed in 12%- \n35% of ovarian endometriosis 43-45. Around 60%-80% of cases of endometriosis-associated \novarian cancer (EAOC) occur in the presence of atypical ovarian endometriosis46-48. Of these \ncases, 25% showed direct continuity of the atypical ovarian endometriosis with ovarian \ncancer49 , underlying a potential ‗premalignant‘ transition spectrum of non-atypical to \natypical and malignant variants. \n \n2. Nuclear morphometry This involves a structured histological approach to grading mitotic \nactivity using nuclear size and pleomorphism.  Morphometric analysis of cancer has been \nshown to correlate to clinical prognosis 50;51. There is published data on nuclear \nmorphometric analysis of endometriosis (and related adenomyosis), albeit most limited to \nmainly non-prognostic correlations52-61.  Morphometric analysis of non-atypical \nendometriosis showed no difference between active (red lesions) and inactive (black or white \nlesions) lesions; it is yet to be studied in atypical endometriosis62 . Nonetheless, mild \ncytological atypia in the glandular epithelium of endometriotic cysts has been associated with \nnormal DNA diploid patterns, whereas severe atypia may be associated with aneuploidy63 . \nFurthermore, the existence of morphometric differences between  peritoneal, ovarian and \nrectovaginal endometriosis supports the earlier stated hypothesis that endometriosis at \ndifferent anatomical locations are likely to be molecularly diverse entities53;55;56. \n \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 16 \n3.Ovarian malignancy may arise directly from ovarian endometriosis Around 60% of \nEAOC occurs with the cancer adjacent to endometriosis or arising directly from ovarian \nendometriosis, with the remaining 40% occurring with distant endometriotic disease 64;65. \nClear cell and endometrioid carcinomas are the commonest EAOCs with ovarian \nendometriosis, while clear cell adenocarcinoma and adenosarcoma the commonest EAOCs in \nextraovarian endometriosis66-70. The risk of direct malignant transformation of ovarian \nendometriosis has been estimated as 0.7% to 1.6% over an average of eight years 43;44. \nInterestingly, there is a common unexplained left-sided predominance for endometriotic \ncysts, and ovarian endometrioid and clear-cell cancers71.72 \n \n4. Increased risk of ovarian cancer in women with endometriosis, irrespective if \nendometriosis is distant or adjacent to ovarian tumour. The age standardised incidence of \novarian cancer in women in the UK is 21.9 per 100,000 (0.02%), with around 75% of cases \ndiagnosed in postmenopausal women 73. If there were no association between cancer and \nendometriosis then the incidence of endometriosis in women with ovarian cancer would be \nsimilar to that in the general population. However, the incidence of endometriosis in women \nwith ovarian cancer is 8%-30% 46;74;75. This compares to a background incidence of \nendometriosis of 7-15% in women of reproductive age, and less than 2% in postmenopausal \nwomen 1. This data correlates with the finding from a Swedish population study, where the \nrisk of ovarian cancer was increased 4.2-fold (95% confidence interval 2.0 to 7.7) in the \npresence of endometriosis 76.  Furthermore, the histology of EAOC (40-55% clear cell , 20-\n40% endometrioid and <10% serous and mucinous subtypes)77-79 differs considerably from \nthat seen in all ovarian cancers (FIGO 1998 annual report 55% serous, 13% mucinous, 14% \nendometrioid, 6% clear cell) 80.  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 17 \n5.Increased risk of synchronous endometrial and ovarian cancers, especially \nendometrioid type, in presence of endometriosis. Simultaneously detected endometrial and \novarian carcinomas are most often associated with endometrioid subtypes, and ovarian \nendometriosis was identified in around 30% of these cases68;70;81. \n \n6.Clinical behavior and prognosis of endometriosis associated ovarian cancer (EAOC) \ndiffers from matched ovarian cancer subtypes not associated with endometriosis. EAOC \ncompared to ovarian cancer without endometriosis presents at a less advanced stage, lower \ngrade, predominantly endometrioid and clear cell type, and has a better overall survival 82;83. \n \n7.Increased risk of extra-ovarian cancers. Around 80% of intraperitoneal cancers \nassociated with endometriosis relate to ovarian cancer, with the remainder extra-ovarian84.  A \nseparate study showed an increased risk of extra-pelvic cancers (breast and non-Hodgkin‘s \nlymphoma) in women with endometriosis 85. \n \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 18 \nB) Genetic and molecular data  \n \nThese have been considered according the SEVEN listed criteria that a cancer cell possess \naccording to Hanahan and Weinberg (2000)6 (Tables 1.1 and 1.2). \n1.Self-sufficiency in growth signals:  Like uterine and breast cancer, endometriosis behaves \nas an oestrogen dependent neoplasm. Endometriosis has specifically adapted to oestrogen-\ninduced signaling by: \n Increased local production of oestrogen through increased expression of aromatase \ncytochrome P450 expression but deficient 17B-hydroxysteroid dehydrogenase (17B-\nHSD) type 2 expression (which impairs inactivation of potent oestradiol E2 to less potent \noestrone E1) 86. \n Increased responsiveness to oestrogen. There is increased oestrogen receptor (ER-alpha) \nexpression in active (red lesions) than inactive (black lesions) endometriosis 87.  \n Inherited genetic polymorphisms in oestrogen and progesterone receptors, which \npredispose to endometriosis88;89 \n Inherited genetic polymorphisms in drug-metabolizing enzymes (CYP1A1, CYP19, and \nGSTM1) which predispose to endometriosis9;90-92 and ovarian endometrioid and clear cell \ncancers 93. These alterations may induce endometriosis or cancer by altering a dioxin-\ninduced oestrogen growth signal. Dioxins have been shown to induce endometriosis-like \nand oestrogen-dependent tumours in animal models 94. Of importance, there is a doubled \nrisk of developing endometriosis amongst women with high serum dioxin levels 95 . \nActivation of oestrogen receptors in endometriosis may occur indirectly through \nupregulated CYP1A1 activity, which causes increased aromatase P450 and oestrogen \nproduction 96, or directly by dioxin activated aryl hydrocarbon receptor 97.  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 19 \nOther growth factors, such as transforming growth factor alpha (TGF-alpha) and insulin-like \ngrowth factor-1 (IGF-1) have also been implicated in endometriosis and cancer \ndevelopment98. IGF-1 signaling is required for cell cycle progression and appears to be a pre-\nrequisite for malignant transformation and implantation. A higher risk for cervical, ovarian \nand endometrial cancer is related to high IGF-1 levels in post- and premenopausal women. \nPlasma IGF-1 levels are higher in cases of severe endometriosis; however, in endometriosis \nIGF-1 levels locally in the endometrium are reduced 99. \n \n2.Insensitivity to antiproliferative signals \nCell division relies on the activation of Cyclins (e.g. Cyclin D1), which bind to cyclin-\ndependent kinases (cdk) to induce cell-cycle progression towards S phase and later to initiate \nmitosis. Since uncontrolled cdk activity is often the cause of human cancer, their function is \ntightly regulated by cdk inhibitors (e.g. p21 and p27 Cip/Kip proteins). For example, \nincreased expression of Cyclin D1 and cdk occurs in breast cancer and is associated with \npoor outcome.  At the cellular level, differences in expression of p27Kip1 protein (cdk \ninhibitor) in active and inactive endometriotic lesions13, coupled with increased p21 \nexpression in endometriomas compared with benign and malignant ovarian tumours 12, \nsuggests a role for increased cyclin-dependent kinase activity through reduced cell-cycle \ninhibitor activity; which is an imbalance frequently seen in cancer.  At the tissue level, \nendometriosis may resist the anti-proliferative effect of progesterone by the predominant \nexpression of the inhibitory progesterone receptor isoform PR-A instead of the stimulatory \nisoform PR-B 11.  \n3.Resistance to Apoptosis \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 20 \nMalignancy commonly displays overexpression of anti-apoptotic (Bcl-2), under-expression \nof pro-apoptotic (BAX) factors, and inactivation of p53 gene (p53 is a tumour suppressor \ngene whose protein is pro-apoptotic) through mutation. Similarly, endometriotic lesions have \nalso evolved strategies to evade apoptosis by: \n Increased bcl-2, and decreased BAX100. \n Up-regulation of survivin and matrix metalloproteinases (MMPs)17-19. \n Elevated soluble Fas ligand and IL-8 in endometriotic peritoneal fluid. The increased \nFasL expression by IL-8 may induce apoptosis of T lymphocytes and thus enable \nendometriosis to evade immune mediated cell death14. \n Germline 16 and somatically acquired 101 inactivating mutations of p53 gene. \n \n4.Limitless replicative potential \nWith each replicative cycle, telomeres (repetitive DNA sequences capping each \nchromosome) become progressively shorter, eventually resulting in cell senescence and cell \ndeath. Tumours commonly express the enzyme telomerase, which protects the telomeres \nfrom shortening and thus preventing ‗cell ageing‘. Oestrogen and progesterone stimulate, \nwhilst tamoxifen and wild-type (normal variant) p53 inhibit, telomerase activity in breast and \nendometrial cancer cells20;21. Although there are no published studies examining telomerase \nfunction in endometriosis, it is notable that oestrogen dependent neoplasms are potentially \nsusceptible to telomerase control. \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 21 \n5.Sustained angiogenesis \nPathological angiogenesis, immune cell suppression and immune cell activation co-exist in \nendometriosis and cancer processes 22;23. Genetically transmitted or environmentally induced \n(e.g. exposure to dioxins) alterations in the angiogenic and/or immune response may \npredispose women to the ectopic implantation of endometrial cells, transported into the \nperitoneal cavity by retrograde menstruation which thereby lead to endometriosis. \nSignificantly, both cancer and endometriosis share some of the mediators implicated in this \n‗inflammatory angiogenesis‘ model. Furthermore, the genes of these mediators exhibit \ngenetic polymorphisms that either predispose to endometriosis (e.g. ICAM-1, IL-6 and IL-10 \ngene promoters) 102-104 or cancer (e.g. IL6, IL8, TNF-alpha, NFKB1, and PPAR-gamma \ngenes) 105-109. \n \nAnti-angiogenic therapy involves the inhibition of pro-angiogenic factors (e.g. anti-vascular \nendothelial growth factor VEGF monoclonal antibodies) or activation of endogenous \ninhibitors of angiogenesis (e.g. endostatin and angiostatin). Pre-clinical studies have shown \nthat endostatin effectively inhibits tumor growth and shrinks existing tumor blood vessels. \nPhase 1 clinical cancer trials of endostatin and angiostatin are ongoing, and preliminary \nresults show minimal toxicities. Similarly, anti-angiogenic strategies for treating \nendometriosis exist, but are still at the experimental phase110.  Soluble truncated receptor (flt-\n1) and an affinity-purified antibody to human VEGF-A, significantly inhibited the growth of \nendometrial explants in a mouse in vivo model of endometriosis by disrupting the vascular \nsupply. Gene transfection (using a replication-deficient adenovirus vector Ad-Angiostatin) of \nthe endogenous angiogenesis inhibitor angiostatin to the peritoneum of a mouse was \nsuccessful in treating a mouse in vivo model of endometriosis111. \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 22 \n6.Tissue invasion and metastasis  \nThe ability to invade through the basement membranes characterizes the transition from non-\ninvasive to invasive cancer. Tumours secret proteases (e.g. matrix metalloproteinases MMPs) \nto degrade the basement membrane and surrounding stroma. Expression of MMP-2 and \nMMP-9 is correlated to grade and stage of many cancers. Likewise, MMP activity is \nupregulated in endometriotic lesions 112. De-regulation of cell adherence signaling involving \nintegrins, beta-catenin, E-Cadherin and P-Cadherin has been demonstrated in the genesis of a \nnumber of malignancies113, and has also been implicated in endometriosis aetiopathogenesis \n32;33;114. Beta-catenin mutations have been identified in endometrial and ovarian endometrioid \ncancers35;36but have not been invesitgated in endometriosis. Cytokeratin-positive and E-\nCadherin-negative endometriotic cells have an invasive phenotype in an in vitro collagen \ninvasion assay similar to metastatic carcinoma cells 115. \n7.Genomic Instability  \nThe classical model of malignant transformation of the cell involves the stepwise acquisition \nof multiple genetic alterations, which confers a clonal selective advantage at each step \npredisposing to the next step (Fearon and Vogelstein 116;117, Figure 1.2). This is often \naccompanied by activation of proto-oncogenes to oncogenes (transformation of normal \ncellular growth, proliferation and differentiation genes) and inactivation of tumour suppressor \ngenes (TSG) (genes that encode for proteins which inhibit excess cellular proliferation and \nmalignant transformation). The genetic alterations can occur at different levels and include \nsingle nucleotides, small stretches of DNA [microsatellites], whole genes, chromosomal \ncomponents or whole chromosomes. The genetic alterations can be intragene or epigenetic \n(e.g. gene silencing by promoter hypermethylation).  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 23 \nFigure 1.2, Acquired stepwise genetic somatic mutations that predispose to development \nof cancer (Fearon and Vogelstein 1990116;117) \nStepwise genetic alterations create cancer\nGenetic model of colorectal carcinogenesis \n[Fearon and Vogelstein (1990)] \n \n \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 24 \nSix principle genetic mechanisms have been identified to contribute to genomic instability \nin cancer, but only the first three have been examined for in endometriosis: \n Gain in oncogenic activity.  \n Inactivation of TSG (loss of both gene copies of allele confers functional loss), or \ninactivation of haploinsufficient TSG (loss of only a single gene copy of allele confers \nfunctional loss) \n Anomalies in DNA mismatch repair enzymes, identified by microsatellite instability \n Inactivation of genes that monitor genomic instability at cell cycling (e.g. mitotic \nspindle assembly checkpoint genes) \n Telomere dysfunction (provokes chromosomal aberrations initiating carcinogenesis) and \ntelomerase-mediated telomere maintenance  (enables cells to achieve a fully malignant \nendpoint and metastasis).  \n Hypermethylation. These mechanisms often act in synergy to promote genomic \ninstability and tumour cell proliferation. For example, deficiency of the TSG p53  alters \nthe cellular response to DNA damage, in that it leaves cells with attenuated DNA damage \ncheckpoint controls and a reduced propensity for apoptotic cell death. Thus, although the \nDNA repair capacity of these cells is reduced, survival is increased. This promotes \ngenomic instability and contributes to the resistance of p53-deficient cells to cytotoxic \nagents.  \nImportantly, pre-malignant lesions display similar genetic aberrations to established cancer. \nLoss of mismatch repair enzyme activity, and loss of PTEN (phosphatase and tensin homolog \ngene) and p53 TSGs frequently occurs in premalignant and malignant stages of breast, \nendometrial and ovarian carcinomas 118;119. Furthermore, epithelial-stromal interactions are \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 25 \nimportant in tumour microenvironment and tumour development. In a similar manner, \nendometriosis demonstrates somatically acquired genetic alterations analogous to those found \nin cancer, resulting in the clonal expansion of genetically abnormal cells. The genetic \nevidence supporting the ‗pre-neoplastic‘ state of endometriosis involves:  \n Monoclonality. Most neoplasms are monoclonal in origin and evidence for \nmonoclonality of endometriosis has been demonstrated in several studies 120-122, although \nthese findings have been challenged recently 123. \n Comparative genomic hybridization (CGH) has shown over-representation (increased \ncopy-number) of chromosomes 1, 2, 3, 5, 6p, 7, 16, 17q, 20, 21q and 22q in an \nendometriosis cell culture line FbEM-1, while chromosomes 5p, 6q, 9q, 11p, 12, 13q, 18 \nand X were under-represented. CGH repeated in endometriotic tissue revealed loss of \nDNA copy number on 1p, 22q and chromosome X, while gain on 6p and 17q. FISH \nanalysis confirmed that the gain at 17q includes amplification of the proto-oncogene \nHER-2/neu124;125. \n Fluorescent in situ hybridization (FISH) analysis of late stage endometriotic lesions \nshowed monosomy of chromosome 17, and loss of TP53 (17p13.1) locus. Because not all \nendometriotic cells displayed this genetic alteration it was suggested that this was a \nsomatically acquired mutation, perhaps occurring in mainly advanced endometriosis \nstates126;127.  \n Loss of heterozygosity (LOH) commonly indicates regions of TSG inactivation, and has \nbeen identified in endometriosis and endometriosis derived cell lines at 5q, 6q, 9p, 10q, \n11q, 22q, p16 (Ink4), GALT, p53, APOA2 128-133. Importantly, cases with ovarian cancer \nadjacent to endometriosis or arising from endometriosis showed common genetic LOH \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 26 \nalterations in endometriosis and cancer indicating a possible malignant genetic transition \nspectrum between endometriosis and cancer 134;135. \n Microsatellite Instability (MSI) Hypermethylation of hMLH1 (gene product is a \ncomponent of the DNA mismatch repair pathway), with concurrent absence of hMLH1 \nprotein expression, is noted in 8.6% of endometriotic lesions 136.  \n Somatic mutations in TSGs. Mutations of PTEN, a TSG, were identified in 20% of \novarian endometrioid carcinomas (EAOC and sporadic) and 20% of solitary endometrial \ncysts, suggesting that inactivation of the PTEN is an early event in the malignant \ntransformation of endometriotic implants137.   A separate study identified reduced PTEN \nprotein expression in 15% of endometriosis cases 136.  \n Germline mutations in Tumour suppressor genes (TSGs). As stated earlier, germline \nand somatically acquired 138 inactivating mutations of p53 gene. \n Activation of oncogene. Both human 139;140  and mouse model141 studies of endometrosis \nhave shown that activation of the K-ras oncogene promotes the development of ovarian \ncancer, even though the mutation appears not be present in the adjacent endometriosis. \n Evidence from endometriosis associated ovarian cancer (EAOC) arising from \nendometriosis. Endometrioid EAOC arising from endometriosis shows higher expression \nof p53 and c-erB-2 oncoproteins than similar ovarian endometrioid cancers without \nendometriosis 142. The different pattern of expression in the two groups suggests different \nmolecular pathways and could explain variations in cancer subtype and prognosis \nbetween the two groups 66;143. \n \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 27 \nC)  Testing causality of endometriosis and ovarian cancer using Bradford Hill criteria \n \nCausality for aetiological factors is normally assessed using the following study designs: \nrandomized controlled clinical trials (RCTs), cohort, case-control, cross-sectional analyses \nand biological models (in vivo, ex vivo, in vitro).  Studies would normally be subjected to \ncritical analysis according to established causal inference methods, the most widely used \nbeing the criteria suggested by Austin Bradford Hill (1965) 7 and listed in Table 1.3. The \nstrength of the causal relationship between endometriosis and ovarian cancer is assessed \nusing such epidemiological causality criteria.  \n \n1. Temporal sequence The natural history of the development and progression of \nendometriosis and ovarian carcinoma is not known. No studies have examined women with \ninitially normal pelvices, who then develop endometriosis, and prospectively followed them \nwith a control cohort to establish the relative risk of developing ovarian cancer; or the need \nfor endometriosis as a pre-requisite that precedes the onset of ovarian cancer. However, \nindirect evidence exists that supports this concept. \nCross sectional studies indicate that the peak age range for endometriosis diagnosis is 25-30 \nyears 144 and for sporadic ovarian cancer, the age range is 50-55 years80, thus fulfilling the \ncriteria for temporal sequence. However, studies mainly reported estimates of incidence of \nsymptomatic endometriosis and ovarian cancer diagnosis rather than their actual incidence of \nonset. There is little evidence to support that endometriosis onset necessarily coincides with \nsymptom onset. Furthermore, delays in diagnosis may also exist. The time elapsed from onset \nof symptoms to diagnosis of endometriosis varies from 3-13 years for women mainly \ncomplaining of pelvic pain, and 2-6 years for infertility 145;146 .The time from symptom onset \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 28 \nto diagnosis in ovarian cancer is four to six weeks147, although the cancer is often at a \nsignificantly advanced stage at diagnosis. Thus, even taking into account the symptom free \nintervals before the diagnosis of endometriosis and ovarian cancer, the criteria for temporal \nsequence remains valid. One study retrospectively assessed ovarian cancer cases (n=573)  to \ninvestigate whether ovarian pathology had been identified 12 months previously148. This \nstudy showed that within this limited period approximately half of ovarian carcinomas \ndeveloped secondarily from preexisting benign-appearing cysts or endometriotic cysts, and \nthe remainder appeared to develop from an ovary of normal appearance. A case report has \ndescribed the continuous transition from benign endometrioid epithelium through epithelial \natypia to invasive ovarian carcinoma within a three year period 149, again suggesting \ncausality. \n \n2. Strength of Association Strong associations imply causality, whereas weak \nassociations are more likely to have arisen or been influenced by unsuspected bias. It has \nbeen suggested that relative risks more than 3 in cohort studies, or odds ratios greater than 4 \nin case-control studies, provide strong support for causation150. Strong evidence to support \nthis component of causality testing was identified by demonstrating:- \n Increased prevalence of ovarian cancer in women with endometriosis  \nSeveral studies have found an increased ovarian cancer incidence in women with \nendometriosis: the odds ratios range from 0.8 to 4.2 (studies are listed in Table 1.4). \n Increased prevalence of endometriosis in women with ovarian cancer \nThe age standardised incidence of ovarian cancer in women in the UK is 21.9 per 100,000 \n(0.02%), with around 75% of cases being diagnosed in postmenopausal women 73. If there \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 29 \nwere no association between cancer and endometriosis then the prevalence of endometriosis \nin women with ovarian cancer would be similar to that in the age-matched general \npopulation, and would be similar across all ovarian cancer subtypes. However, the prevalence \nof endometriosis is increased in women with ovarian cancer (7.7%-29%)46;47;67;74;151-158, in \ncomparison to a background prevalence of endometriosis of 5%-15% in women of \nreproductive age and 3%-5% in postmenopausal women144 .  Furthermore, endometrioid and \nclear cell ovarian cancer subtypes are more likely in the presence of endometriosis than those \novarian cancers occurring in the absence of endometriosis: odds ratios range from 1.87 to \n5.36 for endometrioid, and range from 1.05 to 7.30 for clear cell subtypes, and these are \nshown in Table 1.5. This table also shows increased odds ratios for ‗mixed epithelial‘ and \n‗other types‘ of ovarian cancer, but these tumours are generally uncommon and contain \nmixed varieties of endometrioid, clear cell and adenosquamous cells; the significance of this \nassociation is unclear. Nonetheless, summarizing comparative and non-comparative studies \n46;152;153;155;156;158-161, the prevalence of endometriosis for each ovarian cancer subtype is: 0-\n8% of serous, 0-6% of mucinous, 8%-74% of clear cell, and 9%-43% of endometrioid \nsubtypes.  \n \n3. Consistency of association Since Sampson‘s first report in 1925 5, numerous reports \nhave described cases of ovarian cancer arising from pre-existing endometriosis or associated \nwith ovarian cancer. This observation is consistently repeated in different populations. \nFurthermore, all the studies depicted in Tables 1.4 and 1.5, apart from one, found consistent \nand similar increases in risk of ovarian cancer and distribution of histological subtypes, and \nthus further emphasizing the validity of this association.  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 30 \nTable 1.4. Risk of ovarian cancer and other types of cancer in women with \nendometriosis \nType of study Risk of ovarian \ncancer \nin women with \nendometriosis \nOther cancers in women \nwith endometriosis \n \nReference and \nsource of data \nCase control study \nExamining 28,163 \nwomen with \nendometriosis \nOR 1.34  \n(95% CL  1.03-1.75) \nNot reported 162 \nSwedish Hospital \nDischarge Register \nPooled analysis of \neight case-control \nstudies \nOR 1.73  \n(95% CL 1.10-2.71) \nNot reported 163 \nStudies from US, \nCanada, Australia, \nDenmark \nCohort study of \nwomen with \nSelf-reported \nendometriosis \nUp to 13 year \nfollow up \nRR 0.8 \n(95% CL 0.2-2.4) \n \n \nNon Hodgkin‘s \nlymphoma \nRR  1.8 (95% CL 1.0-3.0) \n \n \n164 Iowa Women‘s \nHealth Study \n \n \nCase control study \nexamining 20,686 \nwomen with \nendometriosis \nOR 1.9 \n(95% CL 1.3-2.8) \n \nOR 4.2 \n(95% CL 2.0-7.7) \nfor long-standing \nendometriosis \n \nNon-Hodgkin‘s \nlymphoma  \nOR 1.8 (95% CL 1.2-2.6) \n \nBreast \nOR 1.3 (95% CL 1.1-1.4) \n \n165 \nSwedish Inpatient \nRegister and \nNational Swedish \nCancer Registry \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 31 \nTable 1.5. Prevalence of ovarian cancer in women with and without endometriosis  \nSubtype of \nepithelial ovarian \ncancer \nPrevalence \nof ovarian cancers \nin association with \nendometriosis \nPrevalence of \novarian cancers \nin absence of \nendometriosis  \nOdds Ratio \nWith  \n(95% confidence \ninterval) \nReference Pooled \nepidemiological \nstudies of ovarian \ncancer80 \n(prevalence of \nendometriosis not \nstated) \nSerous 32% (8/25) \n15% (8/52) \n11% (4/37) \n13% (6/48) \n7% (4/58) \n21% (3/14) \n57% (84/147) \n51% (212/414) \n62% (56/90) \n44% (57/131) \n52% (121/232) \n38% (56/146) \n0.35 (0.28-0.42) \n0.17 (0.14-0.21) \n0.07 (0.03-0.12) \n0.19 (0.13-0.24) \n0.07 (0.04-0.10) \n0.44 (0.36-0.52) \n166 \n156 \n46 \n74 \n66* \n47 \n55% \nMucinous 4% (1/25) \n11% (6/52) \n0% \n4% (2/48) \n2% (1/58) \n14% (2/14) \n23% (34/147) \n21% (88/414) \n19% (17/90) \n25% (33/131) \n11% (25/232) \n14% (21/146) \n0.14 (0.09-0.19) \n0.48 (0.44-0.53) \n- \n0.13 (0.08-0.18) \n0.15 (0.10-0.19) \n0.99 (0.98-1.01) \n166 \n156 \n46 \n74  \n66* \n47 \n13% \nMixed epithelial 0% \n22% (13/58) \n28% (4/14) \n0% \n5% (11/232) \n23% (33/146) \n- \n5.80 (5.2-6.41) \n1.37 (1.26-1.48) \n166 \n66* \n47 \n3% \nEndometrioid 12% (3/25) \n58% (30/52) \n41% (9/22) \n8% (3/37) \n27% (13/48) \n57% (33/58) \n28% (4/14) \n7% (10/147) \n20% (84/414) \n24% (14/57) \n4% (4/90) \n13% (18/131) \n27% (62/232) \n10% (14/146) \n1.87 (1.68-2.06) \n5.36 (4.92-5.80) \n2.13 (1.79-2.47) \n1.90 (1.67-2.12) \n2.33 (2.07-2.59) \n3.62 (3.26-3.97) \n3.77 (3.27-4.27) \n166 \n156 \n154 \n46 \n74 \n66* \n47 \n14% \nClear Cell 52% (13/25) \n15% (8/52) \n13% (19/147) \n7% (30/414) \n7.30 (6.29-8.31) \n2.33 (2.17-2.49) \n166 \n156 \n6% \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 32 \n32% (7/22) \n81% (30/37) \n56% (27/48) \n10% (6/58) \n7% (1/14) \n18% (10/57) \n14% (13/90) \n18% (23/131) \n5% (11/232) \n7% (10/146) \n2.19 (1.84-2.55) \n25.38 (21-29) \n6.04 (5.23-6.85) \n2.32 (2.12-2.52) \n1.05 (1.01-1.08) \n154 \n46 \n74 \n66* \n47 \nOther types 0% \n2% (1/58) \n0% \n0% \n1% (2/232) \n8% (12/146) \n- \n- \n2.02 (1.85-2.18) \n166 \n66* \n47 \n9% \n \nFootnotes  * Age matched nested case control study \n4. Biological gradient (dose-response relationship) No studies were identified that \ncorrelated volume and extent of endometriosis with acquisition of ovarian cancer. \nInterestingly, there is a common unexplained left-sided predominance for endometriotic \ncysts, and ovarian endometrioid and clear-cell cancers 167, which may suggest a ‗spatial‘ \nbiological gradient.  \n5. Specificity This criterion relates to a specific cause producing a specific effect. \nImportantly, the occurrence of endometriosis need not cause ovarian cancer, pelvic pain or \ninfertility. Similarly ovarian cancer, as well as pelvic pain and infertility, may occur without \nendometriosis. \n6. Biological plausibility There is extensive histopathological, molecular and genetic \nevidence showing that endometriosis may be considered a neoplastic process with potential \nfor malignant transformation (discussed earlier 168). \n7. Coherence with existing knowledge There is strong evidence to support this \ncausality component, as several of the risk factors known to increase or decrease \nsusceptibility to endometriosis are also common to those of epithelial ovarian cancer. These \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 33 \nfactors may indicate a common aetiological mechanism for endometriosis and ovarian cancer. \nAlternatively, these factors could act as confounders in the association between endometriosis \nand ovarian cancer. However, data of their presence in endometriosis associated ovarian \ncancers compared to matched (by age, histological subtype, grade and FIGO stage) ovarian \ncancers is unobtainable in most studies, which precludes any analysis of their confounding \ninfluence. The risk factors currently identified are: \n Infertility and nulliparity- both of multifactorial aetiology and positively associated \nwith endometriosis 144 and ovarian cancer169. \n Unopposed estrogen replacement therapy (ERT)-  this is associated with malignant \ntransformation of endometriosis 170;171 and increased the risk of endometrioid or clear cell \nepithelial ovarian tumours (OR 2.56; 95% CL 1.32-4.94)172;173. Importantly, a \nconfounding effect is unlikely with ERT as most studies reporting prevalence of \nendometriosis associated ovarian cancer were based on women not taking ERT. \n Multiple lifetime ovulations- this increases the risk of epithelial ovarian cancer174. \nThe combined oral contraceptive pill, which is known to reduce ovulations, has been shown \nto reduce the risk of ovarian cancer175;176 and endometriosis. \n Tubal ligation- this reduces the risk of ovarian cancer177, particularly endometrioid \nand clear cell types178;179. No prospective trials exist showing tubal ligation to reduce \nendometriosis occurrence, progression or recurrence, However, assuming retrograde \nmenstruation to be a main mechanism for endometriosis, it is plausible that tubal ligation \nsuppresses retrograde menstruation and endometriosis which consequently suppresses \nendometriosis associated ovarian cancer development.  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 34 \n \n8. Experimental evidence   Animal models (mice, rat or baboon) of endometriosis may \nbe created by surgically implanting host (or human) endometrial cells in to the peritoneal \ncavity and promoting proliferation of the cells by administration of supraphysiological \nestrogens141;180-186.  Of these animal models, a sentinel paper by Dinulescu (2005) 141 induced \novarian lesions with an endometrioid glandular pre-neoplastic morphology by activating an \noncogenic K-ras allele and deletion of the PTEN tumour suppressor gene; hence, fulfilling \nFearon and Vogelstein‘s classic stepwise model of cancer development (Figure 1.2)116;117 .  \nHuman studies demonstrating induction of endometriosis or its malignant transformation are \nhighly unlikely as such research would be deemed unethical. \n \n9. Analogy Malignant transformation of endometriosis is not restricted to the ovary. \nSeveral studies have reported analogous malignant transformation at extra-ovarian locations, \nsuch as the rectovaginal septum, vulva, and colon187. Principle malignancies include \nendometrial stromal sarcoma, endometrioid adenocarcinoma, clear cell carcinoma, with \nhistological confirmation of tumour and adjacent endometriosis in all cases. Furthermore, \nmalignant transformation of adenomyosis, considered the ‗uterine‘ variant of endometriosis, \nhas been observed and results in similar histological subtypes to that found for endometriosis \nrelated malignancies188. \n \n \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 35 \nD. Summarising the published evidence whether endometriosis is a neoplastic \nprecursor to ovarian cancer \n \nBased on the methodological approaches discussed earlier (A: clinicopathological, B: Genetic \nand molecular hallmarks of cancer, C: Bradford-Hill causality criteria) there is inadequate \nevidence to support the hypothesis that ‗ENDOMETRIOSIS IS A NEOPLASTIC \nPRECURSOR TO OVARIAN CANCER‘ (Table 1.6). \nTable 1.6. Summarising the published evidence that supports or refutes the hypothesis \nthat endometriosis is a neoplastic precursor to the development of ovarian cancer \n Supporting evidence for \nendometriosis \nRefuting evidence for \nendometriosis \nOverall strength \nthat hypothesis is \ntrue \nA. Clinico-\npathological \nepidemiological \ndata \n \nCancer arises directly from \nendometriosis \nIncreased risk of certain \novarian cancer subtypes \nInconsistency of histological \nobservations \n \nWeak association \nWeak \nB. Genetic and \nmolecular \n‘Hallmarks of \nCancer6’ \ncriteria \nSelf-sufficiency \nInsensitivity to anti-\nproliferative signals \nResistance to apoptosis \nAngiogenesis \nGenomic instability \n \nLimitless replication \npotential \nTissue invasion and \nmetastasis \nModerate \nC. Association \nvs. causality \nusing \nBradford-Hill \n(1965) \nepidemiological \ncriteria7 \nExperimental evidence \n(animal model) \nBiological plausibility \nExperimental analogy \nStrength of association  \nSpecificity of association \nInconsistency of association \nTemporal sequence \nBiological gradient (dose \nresponse relationship) \nExperimental evidence \n(humans) \nWeak \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 36 \nNevertheless, the identification of an association between endometriosis and ovarian cancer \nmay suggest alternative hypotheses: \n Only specific endometriotic implants may directly undergo malignant transformation, \nperhaps through environmental exposure via an atypical endometriosis transition phase,  \nanalogous to the genetic cancer model of colon cancer where colonic epithelium acquires \nstepwise somatic genetic mutations to transform to colonic polyp, adenoma and finally to \ncolonic carcinoma (Figure 1.2)116;117. Therefore, like most types of sporadic cancer 189, \nendometriosis  may be exposed to complex interactions between inherited germline \npolygenic low-penetrance alleles (polymorphisms) 190;191, somatically acquired genetic \nalterations 192 and  environmental factors 94 .  A visual summary of the main pathways of \nthis hypothesis is shown in Figure 1.3. \n Both endometriosis and cancer share common antecedent mechanisms and/or \npredisposing factors (e.g. genetic susceptibility, immune/angiogenic dysregulation, \nenvironmental toxin exposure), with obvious divergence in molecular pathways \ndownstream.  \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 37 \nFigure 1.3 Proposed genetic and molecular aeitopathogenesis of endometriosis \n              CANCER HALLMARK \nCLONALITY   GENETIC     MECHANISMS     \n \ncell population   \n \n     \n \nRETROGRADE MENSTRUATION     \n (? also Coelomic metaplasia, lymphovascular spread) \n      ADEHSION (Cadherin,  \nB-catenin, Protein Kinase C) \n      \n  PROLIFERATION (limited) \n   \n          ANGIOGENESIS (limited) \n \n \n          EVASION OF APOPTOSIS \n    \n \n    STEPWISE \nMixed cell     ACQUISITION \n populations                     OF   GENETIC \nALTERATIONS \n    (e.g. TSG, oncogenes) \n \n \n \n          SELF-SUFFICIENCY IN \n           GROWTH SIGNALS \n           (Cylcin, cdk, p14, p16) \n         \nINSENSITIITY TO  \nGROWTH INHIBITION \n \nAPOPTOSIS EVASION \n          (Fas, Bax, p21, p53, p14) \nPredominant cell population        LIMITLESS  \nREPLICATION \n \n      PATHOLOGICAL  \n          ANGIOGENESIS \n       \nPRE-MALIGNANT     PROLIFERATION OF \nTRANSITION PHASE/ZONE  CHROMOSOMALLY \n    (? further LOH 6q,5q,9p,11q,22q,   ABNORMAL \nPTEN, TP53, beta-catenin, P-cadherin) CELLS \n      \nEntire cell population       INVASION & METASTASIS \n    \nPOLYGENIC SUSCEPTIBILITY LIKELY \nINVOLVING: \n METABOLIC/ENDCORINE/IMMUNOLOGY/ \n(e.g. POLYMORPHISMS IN GSTM1, ER, PR,IL-6) \n ENVIRONMENTAL TRIGGER (e.g. dioxin) \nSomatically acquired \nGENOMIC \nINSTABILITY \n \n \nMore \nGENOMIC \nINSTABILITY \nPOLYCLONCAL \nPOLYCLONCAL \nMONCLONAL  \nMONCLONAL  \nATYPICAL  \nENDOMETRIOSIS \nENDOMETRIOSIS \n(LOH 9p,11q,22q) \nreduced PTEN,hMLH1 \nprotein \nOVARIAN ENDOMETRIOID  \nAND CLEAR CELL CARCINOMA \nENDOMETRIUM \nMONCLONAL  \n \n\n Chapter 1.1. Hypothesis that endometriosis is a neoplastic precursor to ovarian cancer \n 38 \nDiscussion of methodology used in testing hypothesis \n \nA strength of this work has been the utilization of a systematic literature search and \ncombining this with established research methodological approaches. However, it is accepted \nthere may still be grounds to challenge our conclusion.  \nTo some extent, my conclusions may be less certain, as most included studies were of small \nsample size, retrospective design, and suffered from selection bias (incomplete case \nascertainment and unmatched populations), information bias (varying histological criteria for \ncancer arising from endometriosis and atypical endometriosis) and confounding to varying \ndegrees. Such problems of interpreting epidemiological studies involving endometriosis have \nalso been observed by others193;194. \nSignificantly, my research aim of using Bradford-Hill criteria to test causality was adopted by \nanother group (Vigano 2007195)  investigating the causal link of endometriosis and cancer. \nTheir work195, which partly included and referenced my work168,concluded that there was \nonly a weak causal link. However, Vigano‘s group did not perform a systematic literature \nsearch (and omitted key references that we have included above) and utilised modified \ncausality criteria. I therefore believe my conclusion is more likely to be accurate, and in fact, \nhas been further validated through the experimental work discussed below and orally \npresented at international conferences196;197.\n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 39 \n1.2. Experimental investigation of endometriosis and EAOC \n \nIntroduction  \n \nExperimental studies on primary endometriotic tissue and endometriotic-cell lines has shown \nendometriosis and cancer to share similar molecular (limitless replicative potential, self-\nsufficiency in growth signals, insensitivity to growth-inhibitory signals, sustained \nangiogenesis) and genetic(monoclonality, genetic instability) characteristics168;198;199. Allelic \nloss in endometriosis and sporadic ovarian cancer has been demonstrated in similar \nchromosomal regions 1p, 1q21, 5p, 5q, 6q, 7p, 9p , 9q, 11q, 17p13.1, 17q and 22q 200;201. \nFurthermore, a recent in vivo mouse model study demonstrated induction of endometriosis-\nlike and ovarian cancer tissue through introduction of oncogenic K-ras and conditional \ndeletion of PTEN 202. Endometriosis-associated ovarian cancer (EAOC) (25-35% of all \novarian cancers) appears to be a separate entity from sporadic ovarian cancer without \nendometriosis (SOC). Epidemiological studies have shown EAOC tends to present in \nyounger aged women, has better survival, and more likely to be a low-grade endometrioid or \nclear cell cancer subtype 82;203-205. However, there is only limited data on the genetic \nalterations in EAOC, which to date is mainly confined to the roles of PTEN and K-ras 139;140 \nand a limited genome-wide (n=14 cases)206 LOH screen and CGH analysis (n=4 cases)207;208.  \nHypothesis  \n \nAlthough the supporting evidence is weak (Table 1.6), we could assume that, in some cases, \nit is possible that endometriosis behaves as a neoplastic precursor to the development of \novarian cancer (Figure 1.3). If this is valid, then we could better understand the genetic \naetiopathology of both endometriosis by deliberately selecting endometriosis-associated \novarian cancer (EAOC) as a model and testing the ovarian cancer and adjacent endometriosis \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 40 \nfor common or dissimilar genetic abnormalities. Therefore, in this chapter, I will explore \nwhether normal ovarian surface epithelium and adjacent endometriosis and adjacent ovarian \ncancer display a stepwise accumulation of LOH events analogous to the stepwise \naccumulation of LOH (due to inactivation of  tumour suppressor genes TSG) observed in \nFearon and Vogelstein‘s model for colon cancer (Figure 1.2).  The following patterns of \nresults may be generated by adopting this approach: \n LOH is only demonstrated in the ovarian cancer, and not in endometriosis. This would \nsuggest acquired somatic genetic events occur, perhaps due to the presence of \nendometriosis, that cause malignant transformation to ovarian cancer.  \n LOH is demonstrated in endometriosis when compared against matched normal ovarian \nsurface epithelium. This would suggest that inactivation of particular TSGs were \nresponsible for either the initiation or progression of endometriosis. \n Similar chromosomal regions of LOH occur in endometriosis and ovarian cancer. This \nwould suggest that a particular set of TSGs are in involved in both the initiation, \nprogression and malignant transformation of endometriosis and ovarian cancer, and that \nendometriosis and ovarian cancer share common antecedent genetic events.  \n Additional LOH events are identified in ovarian cancer compared to LOH events \nidentified  in adjacent endometriosis. This would confirm a stepwise accumulation of \nspecific inactivating TSG(s) (equating to the additional LOH events) are responsible for \nthe malignant transformation of endometriosis. \nFine mapping the LOH regions would therefore allow us to select candidate TSGs that were \neither responsible for the initiation and progression of endometriosis (pattern B), malignant \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 41 \ntransformation of endometriosis to ovarian cancer (pattern D), or common to development of \nboth endometriosis and ovarian cancer (pattern C). \nPlan of investigation \n \n Investigate epidemiological and prognostic factors associated with endometriosis \nassociated ovarian cancer (EAOC) and sporadic ovarian cancer (SOC), matched for \nendometrioid and clear cell ovarian cancer subtypes. (Tables 1.7, 1.8, 1.9; Figures 1.7, \n1.8,  1.9) \n Loss of heterozygosity (LOH) mapping of EAOC and SOC-fine mapping of \nchromosomes 9 and 11 using multiple microsatellite genetic markers to identify candidate \ntumour suppressor genetic loci. (Tables 1.10, 1.11; Figure 1.4-LOH mapping output; \nFigures 1.10 and 1.11).  \n Analyse survival prognostic significance of LOH at chromosomes 9 and 11. Combine \ninformation narrowed fine-mapped genetic region of LOH, frequency of LOH and \nprognostic significance of loci, to select candidate tumour suppressor genes for further \ninvestigation (Figure 1.12). \n Laser Capture Microdissection of endometriosis adjacent to ovarian cancer and perform \nLOH using candidate genetic microsatellite markers.  Compare findings to similar study \nby collaborators (we have donated our samples to their unit). (Figure 1.5-importance of \nLCM; Table 1.12). \n Immunohistochemical investigation of candidate disease-modifying genes in \nendometriosis adjacent to and distant from  EAOC. Selected gene products are \nGlycodelin (9q34.3) and Progesterone receptor (11q22). (Table 1.13; Figures 1.13, \n1.14, 1.15). Correlation of immunohistochemical expression to disease development . \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 42 \n Preliminary nuclear morphometry analysis of transition zone between endometriosis, \natypical endometriosis and ovarian cancer that exist in direct continuum. (Figure 1.16). \n Single nucleotide polymorphism genome wide analysis of endometriosis using \nAffymetrix 100K SNP microarray (Figure 1.6-SNP microarray advantages; Figures 1.17, \n1.18, 1.19, 1.20). \n \nExperimental Methods \n \nEthics: South Birmingham Local Research Ethics Committee has given full approval to all \nwork included in this thesis chapter (LREC reference No: 2002/057, August 2002). \n \nClinical material   Cases of EAOC and SOC of endometrioid and clear cell subtype were \nidentified by interrogation of a computerized histopathological database at Birmingham \nWomen‘s Hospital. All cases were gynaecological cancers operated on from 1995-2001 at \nBirmingham Women‘s Hospital. Five micron thick paraffin embedded slides were used for \nDNA extraction and three micron thick slides were cut from selected cases for \nimmunohistochemical analysis. Realising that molecular genetic alterations of ovarian cancer \nvary according to histological subtype 209, we ensured our comparative analysis of allelic loss \nbetween EAOC and SOC were matched for endometrioid and clear cell subtypes of ovarian \ncancer. Chromosomal regions showing greatest frequency of LOH in EAOC and SOC and \nthat appeared to reside within a consistent minimal region of LOH loss were prioritized for \nfurther study. \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 43 \nDNA Extraction \nCancer and matched normal DNA were extracted from five micron glass slides using one of \ntwo methods depending on slide content and composition. Needle microdissection was used \nto collect histologically labeled endometriosis and cancer. DNA was extracted from micro-\ndissected material held in an eppendorf  using a microwave-based method as previously \ndescribed 210 . Briefly, retrieved tissue material was placed in an eppendorf containing 400μl \nTrisT-EDTA buffer and heated in a 600W microwave for one minute in 15 second bursts. \nFollowing centrifugation the upper paraffin layer was discarded and the supernatant \nincubated for 48 hours with 4μl of proteinase K 20mg/ml (Sigma-Genosys) with continuous \ngentle agitation. Proteinase K was inactivated by heating to 95ºC for ten minutes and the \nsupernatant aliquoted for DNA studies.  \n \nLOH Analysis \nHighly polymorphic microsatellite markers spanning the full length of chromosomes 9 and 11 \nat approximately 20cM intervals (Wellcome Trust) were kindly provided by Oxford Group, \nDr. Stephen Kennedy. Detailed genetic fine mapping was performed using customised \nmicrosatellite markers (Sigma) spaced approximately 10cM apart and in between the \nprevious Wellcome markers. The forward primers were 5' end-labeled with FAM. A  25-µl \nPCR reaction volume containing 1xAB Gene Buffer (ABGene), Magnesium Chloride \n(ABGene),  100 µM each of dATP, dCTP, dGTP, and dTTP; 0.5 unit of DNA Taq \npolymerase (AB Gene); sterile DNAse and RNAse free water (Sigma), and 2 pmol of reverse \nprimer, 2ul (approximately 100 ng) of genomic DNA.  PCR cycling conditions were as \nfollows: (a) 5 min at 94°C; (b) 30 cycles of 30 s at 94°C, 30 s at the appropriate annealing \ntemperature (usually 55°C), and 30 s at 72°C; and a final step of 72°C for 10 min. The \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 44 \nreaction products were then diluted 1:15 with sterile water. One microlitre of the diluted PCR \nproduct was added to 10ul of 95% formamide, and 0.02ul of LIZ dye and subsequently \ndenatured for 5 minutes at 95°C and snap frozen with ice. PCR products were run on ABI \n377 gel electrophoresis analyser and fragment sizes were recorded using GeneScan software \nanalysis. LOH was scored based on the absence of alleles in tumour-derived DNA compared \nto normal DNA or a loss of at least 70%  in the relative size of alleles in the tumour-derived \nDNA compared to normal tissue; examples of GeneScan  images are shown in Figure 1.4  \n \nFigure 1.4. Gene\ntic allelic products images observed following microsatellite \namplification of target DNA and analysis on ABI Prism analyser \nMicrosatellite marker mapping of LOH \nregions\nNormal\nEAOC\ncancer\nEndometriosis\n \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 45 \nImmunohistochemistry and nuclear morphometry \nMonoclonal antibodies were obtained for p53, CD10 (Santa Cruz), Progesterone receptor \n(isoforms A and B) (AbCAM) and Glycodelin (AbCAM). Immunohistochemistry was \nperformed according to standardised protocols using the Dako Chem Mate antigen detection \nkit. Briefly, 3 micron slides were deparaffinised through stepwise Xylene, ethanol, water and \nmethanol washes; endogenous peroxidase was subsequently blocked by 20 minutes \nincubation with 0.5% hydrogen peroxide/methanol mixture. Antigen exposure was achieved \nby pressure cooker boiling for 5 to 7 minutes with pH 6 citric acid buffer.  Primary antibodies \nwere diluted to concentrations of 1 in 200 to 1 in 1000 in TBS Tris buffered saline (pH 7.6) \nand 200 microlitres were applied to each slide. The Dako Chem Mate protocol (yellow and \nred antibody washes) followed by DAB chromagen/substrate then copper sulphate solution \nstaining was performed. Brief dips in Haemotoxylin, acid-alcohol dip and Scott‘s Media \nfollowed by tap water wash allowed final ascending alcohol/xylene and coverslip slide \ncreation. \n \nLaser Capture microdissection\n \nA PALM microlaser was used. EAOC paraffin 3micron thick cut slides were de-waxed and \nsuspended in aqueous buffer. Endometriotic epithelium was separated ‗purely‘ using laser \nblot and line cutting according to the manufacturer‘s guidance (Figure 1.5). Particles were \ncatapaulted on to the inside lid surface of a single PALM 1cm3 opaque lid eppendorf. \nQUIAGEN mini-DNA prep kit buffer was placed in the conical base of the eppendorf and the \nlid closed after particle deposition and eppendorf was then inverted. DNA was extracted and \ncleaned according to the QUIAGEN mini-columns and centrifuge protocol. \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 46 \nFigure 1.5. Importance of Laser capture microdissection of target disease (such as \nendometriosis epithelium glandular lining) from surrounding tissue (such as \nendometriosis stroma) \n \nHeterogeneity in tissue sample\nThe importance of laser capture microdissection to \nobtain “pure cells”\n \n \nAffymetrix SNP 100K Microarray \nOvarian endometriosis and matched normal ovarian surface epithelium were needle micro-\ndissected immediately at the time of surgical extraction from the patients with their \ndocumented informed consent, and then promptly snap frozen in liquid nitrogen and held at -\n77°C. DNA was extracted by crushing the tissue in PureGene extraction buffer and following \nthe PureGene centrifugation and incubation protocol. The quality and concentration of \nextracted DNA was determined by spectrophotometry at A260/280. The Affymetrix \nGeneChip® Mapping Assay, in conjunction with the GeneChip Human Mapping 100K Set, \nis designed to detect > 100,000 Single Nucleotide Polymorphisms (SNPs) in samples of \ngenomic DNA. The Mapping 100K Set is comprised of two arrays (Mapping 50K Array Xba \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 47 \n240 and Mapping 50K Array Hind 240) and two assay kits (containing either Hind III, XbA 1 \nrestriction enzymes). Each array and its corresponding assay kit are processed independently \nfrom the second enzyme. The protocol starts with 250 ng of genomic DNA per array and will \ngenerate SNP genotype calls for more than 50,000 SNPs for each array of a two array set. \nThe assay first digests the genomic DNA with the Xba I or Hind III restriction; an overview \nof the remainder of the assay protocol is shown in Figure 1.6. The final PCR products \n(amplicons) are fragmented, end-labelled, and hybridized to either the Xba 1 or Hind III \nGeneChip array. Scanned images obtained from the GeneChip Mapping 50K Array Xba 240 \nand the GeneChip Mapping 50K Array Hind 240 are digitally combined and displayed by \nGeneChip Operating Software (GCOS). \n \nStatistical analysis \nStatistical data were analysed with the use of SPSS version 13 (SPSS Inc, USA). Continuous \nvariables were analysed by T-test, Mann-Whitney U and ANOVA tests . Categorical \nvariables were analysed by Chi-square. Survival regression was analysed using either the \nKaplan-Meier or Cox proportional Hazards model, depending on the parameters employed. A \np-value less than 0.05 was considered statistically significant, although a Bonferonni \ncorrection was considered (p value<0.001) when multiple testing (>10) may have led to \nincreased risk of type 1 error. \n\n Chapter 1.2. Introduction to experimental investigation of endometriosis and EAOC \n 48 \nFigure 1.6. Increased genetic resolution of Affymetrix Single Nucleotide Polymoprhism \nDNA microarray compared to ‘traditional’ multiple microsatellite marker genome wide \nmapping \nGenome Wide Screening:\nSNP Array vs. Microsatellite markers \n \n \n\n Chapter 1.3 Investigation of epidemiological factors associated with EAOC and SOC \n 49 \n1.3 Investigation of epidemiological factors associated with EAOC and SOC \n \nAt total of 62 cases were identified from the histopathological database; their epidemiological \ncharacteristics are shown in Table 1.7. Of these, paraffin tissue blocks were retrieved for 50 \ncases, and these were subjected to genetic investigation; the epidemiological characteristics \nare shown in Table 1.8.  \nOvarian cancer survival was statistically significantly associated with clear cell or \nendometrioid subtype, cancer stage and the presence of synchronous endometrial and ovarian \ncancer [regression model Chi-sq 14.1, p=0.003).  Clear cell compared to endometrioid \nsubtype of cancer increases the odds of dying earlier by 2.2 (i.e. the probability of dying \nearlier is 69%).  An advanced cancer stage increases the odds of dying earlier by 1.6 (i.e. the \nprobability of dying earlier is 62%). Synchronous cancers compared to solitary ovarian \ncancers decreases the odds of dying earlier by 0.13 (i.e. the probability of dying earlier is \n12%). The presence or absence of endometriosis did not influence survival, as did other \nfactors as listed in Table 1.9. These observations are graphically depicted by the survival \ncurves (Figures 1.7, 1.8, 1.9). \nOdds=Prob/1-Prob \nProb= Odds/1+Odds \n\n Chapter 1.3 Investigation of epidemiological factors associated with EAOC and SOC \n 50 \n Table 1.7. Characteristics of endometriosis associated ovarian cancer (EAOC) and \nsporadic ovarian cancer (SOC), matched for endometrioid and clear cell histologies,  \nused in epidemiological analysis (N=62) \n EAOC  N=34  SOC  N=28  P value  \nClear cell  18  9   \nEndometrioid  16  19   \nAge  mean [range]  57.7 [32-79]  60.4 [32-84]  n.s.  \nCancer stage:                   1  \n                  2  \n                  3  \n                  4  \n                         Mean  \n21  \n7  \n5  \n1  \n1.59  \n13  \n5  \n9  \n1  \n1.93  \n0.150  \nSidedness:       Left  \n                         Right  \n                         Bilateral  \n14  \n18  \n2  \n8  \n8  \n12  \n0.002  \nEndometriosis proximity to tumour:                                        \n   Distant  \n   Adjacent  \n   Tumour arising from endometriosis  \n \n11 (32%)  \n12 (35%)  \n11 (32%)  \nNot relevant   \nSynchronous uterine & ovary cancer  5  0  0.034  \nUterine hyperplasia  11  8  n.s  \nLeiomyoma  25  17  n.s  \nAdenomyosis  14  11  n.s  \nTumour in lymph nodes  2  3  n.s  \nTumour in omentum  4  6   n.s.  \nAscites  12  10   n.s.  \n\n Chapter 1.3 Investigation of epidemiological factors associated with EAOC and SOC \n 51 \nTable 1.8. Characteristics of endometriosis associated ovarian cancer (EAOC) and \nsporadic ovarian cancer (SOC), matched for endometrioid and clear cell histologies,  \nused in genetic analysis(N=50) \n EAOC \nClear cell \nEAOC \nEndometrioid \nSOC \nClear \ncell \nSOC \nEndometrioid \nStatistical \nTesting \nP value \nNumber of cases 15 12 7 16  \nMean Age 57.2 59.9 61.6 59.2 0.865** \nAge:lower to \nupper quartile \n51-65 56-66 45-72 51-67  \nStaging of  \novarian cancer \nStage 1 \nStage 2  \nStage 3  \nStage 4 \n \n9 \n3 \n3 \n0 \n \n8 \n3 \n0 \n1 \n \n3 \n0 \n3 \n1 \n \n9 \n3 \n3 \n1 \n \n0.486 \nAscites 5 4 4 5 0.660 \nSynchronous \nendometrial and \novarian cancer \n0 5 \n(cases \n17,19,36,37,38) \n0 1 0.222 \nPresence of \nendometriosis  \ndirectly adjacent \nto ovarian cancer \n7/15 \n(cases 2, 3, \n4,5,10,11,13) \n5/12 \n(cases \n17,18,19,24, 26) \nN/A N/A  \nSurviving  \n>48 months \n>36 months \n>24 months \n>12 months \n<12 months \n \n3 \n3 \n6 \n10 \n5 \n \n4 \n7 \n9 \n10 \n2 \n \n0 \n0 \n3 \n5 \n2 \n \n5 \n8 \n12 \n13 \n3 \n \n0.337 \n0.023 \n0.112 \n0.707 \n0.707 \n\n Chapter 1.3 Investigation of epidemiological factors associated with EAOC and SOC \n 52 \nTable 1.9. Multivariate survival regression analysis (N=62 combined cases of EAOC and \nSOC) \nVariables included in multivariable Cox \nregression analysis  \nP-value \nof  \nvariable  \nOdds of dying earlier  \nExpressed as  \nHazard Ratio  \nClear cell vs. Endometrioid subtype  0.019  2.16 (95% CI 1.14 - \n4.10)  \nAdvancing cancer stage  0.016  1.56 (95% CI 1.09 - \n2.24)  \nSynchronous cancer vs. ovarian cancer  0.012  0.13 (95% CI 0.03 - \n0.64)  \nEndometriosis presence  0.80  not significant  \nAge  0.72  not significant  \nCA125  0.43  not significant  \nTumour in Lymph nodes  0.65  not significant  \nTumour in omentum  0.92  not significant  \nAscites  0.70  not significant  \nSidedness of tumour  0.56  not significant  \nProximity of endometriosis to tumour  **   \nFootnotes \nProbability corresponds to HR/1+HR \nInterpreting Hazard Ratio results: when all variables are combined in a survival regression analysis, only \nhistological subtype, cancer stage and presence of synchronous uterine and ovarian cancer  statistically \nsignificantly impact on cancer survival [ Chi-sq 14.1, p=0.003) : \nClear cell compared to endometrioid subtype of cancer increases the odds of dying earlier by 2.2:1 (i.e. \nprobability of dying earlier is 69%) \nAn advanced cancer stage increases the odds of dying earlier by 1.6:1 (i.e. probabilit y of dying earlier is 62%). \nSynchronous cancers compared to solitary ovarian cancers increases the odds of dying earlier by 0.13:1 (i.e. \nprobability of dying earlier is 12%) \n\n Chapter 1.3 Investigation of epidemiological factors associated with EAOC and SOC \n 53 \nFigure 1.7    Survival differences between subtypes of ovarian cancer \nThere is no statistically significant association between the four individual cancer subtypes \nand survival. However, there is a statistically significant association for clear cell vs. \nendometroid types of ovarian cancer (Hazard Ratio 2.16 (95% CI 1.14-4.10)), as depicted in \nthe figure i.e.  \nClear cell compared to endometrioid subtype of cancer increases the odds of \ndying earlier by 2.2:1 (i.e. probability of dying earlier is 69%).(Hazard analysis results are \ndepicted in Table 1.9). \n100806040200\nsurvival\n1.0\n0.8\n0.6\n0.4\n0.2\n0.0\nCum Survival\nSOC endometrioid\nSOC clear cell\nEAOC endometrioid\nEAOC clear cell\n \nSurvival Analysis according to cancer subtype\n \n \n\n Chapter 1.3 Investigation of epidemiological factors associated with EAOC and SOC \n 54 \nFigure 1.8.  Survival analysis according to stage of ovarian cancer. \nThere is a statistical significant association between ovarian cancer staging and cancer \nsurvival (Hazard Ratio 1.56 (95% CI 1.09-2.24)) i.e. an advanced cancer stage increases the \nodds of dying earlier by 1.6:1 (i.e. probability of dying earlier is 62%).(Hazard analysis \nresults are depicted in Table 1.9). \n100806040200\nsurvival\n1.0\n0.8\n0.6\n0.4\n0.2\n0.0\nCum Survival\nstage 4\nstage 3\nstage 2\nstage 1\n \nSurvival Analysis according to cancer stage\n \n\n Chapter 1.3 Investigation of epidemiological factors associated with EAOC and SOC \n 55 \nFigure 1.9. Survival analysis according to presence of endometriosis \nThere is no statistical association between the presence of endometriosis and survival for all \ncancers (Log Rank Mantel Cox p=0.80). (Hazard analysis results are depicted in Table 1.9). \n100806040200\nsurvival\n1.0\n0.8\n0.6\n0.4\n0.2\n0.0\nCum Survival\nno endometriosis\nendometriosis\n \nSurvival Analysis according to presence of absence of endometriosis\n \n \n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 56 \n1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at \nchromosomes 9 and 11 using multiple microsatellite genetic markers and \ntheir prognostic significance. \n \nBased on previous published research, chromosomes 9 and 11 were selected for LOH \nmapping as chromosomes most likely to harbor tumour suppressor genes (TSG) for either \nendometriosis or ovarian cancer201;211.  \nPreliminary microsatellite markers demonstrated LOH at chromosomes 9 and 11 for both \nEAOC and SOC. Microsatellite markers that mapped to genetic loci no greater than 10cM \napart,  were selected and used to create a fine map of LOH at chromosomes 9 (Table 1.10) \nand chromosome 11 (Table 1.11).  The background frequency of genome-wide LOH \nobserved was 30-40% for chromosome 9 (Figure 1.10) and 20-40% for chromosome \n11(Figure 1.11). High frequency LOH was observed at 9q32 (65%), 9q34.3 (78%), 11q22.1 \n(57%), 11q24.1 (60%), and 11q25 (64%). There were no significant differences in the \npatterns of LOH between EAOC and SOC (Figures 1.10 and 1.11). \n \nSurvival analysis showed LOH at 9q34.3 correlated to poorer survival , suggesting that this \nregion of high frequency LOH may harbor a candidate TSG (Figure 1.12). Conversely, \nsurvival analysis showed LOH at 11q 23.3 correlated to improved survival, suggesting that \nthis region of high frequency LOH may harbor a candidate oncogene (Figure 1.12).  \n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 57 \nTable 1.10. Allelic Loss at chromosome 9  \nGene\ntic \nMar\nker \nDeC\nODE \nCM \nCytoge\nnetic \nEAOC \nCLEAR CELL \nEAOC \nENDOMETRIOI\nD \nSOC \nCLEAR \nCELL \nSOC \nENDOMETRIOID \nOVERA\nLL \nLOH \nFREQU\nENCY \n   0\n1 \n0\n2 \n0\n3 \n0\n4 \n0\n5 \n0\n6 \n0\n7 \n0\n8 \n0\n9 \n1\n0 \n1\n1 \n1\n2 \n1\n3 \n1\n4 \n1\n5 \n1\n6 \n1\n7 \n1\n8 \n1\n9 \n2\n0 \n2\n1 \n2\n2 \n2\n3 \n2\n4 \n2\n5 \n2\n6 \n2\n7 \n2\n8 \n2\n9 \n3\n0 \n3\n1 \n3\n2 \n3\n3 \n3\n4 \n3\n5 \n3\n6 \n3\n7 \n3\n8 \n3\n9 \n4\n0 \n4\n1 \n4\n2 \n4\n3 \n4\n4 \n4\n5 \n4\n6 \n4\n7 \n4\n8 \n4\n9 \n5\n0 \n \nD9S1\n71 \n45.57 9p21.3 U U U I U I U U I U U I I U I I I I I U I I █ U I █ █ I U █ U U █ I U █ I U I U █ U █ U U I U █ U I 32% \nD9S2\n73 \n66.75 9q21.1\n1 \nU I I U U I I █ U U I I █ U I I I I I I I █ U U U U █ I I █ U I █ █ █ U █ U I I █ U I █ I I U I U I 32% \nD9S9\n33 \n78.26 9q21.3\n1 \nU I I U I I I I U █ █ I I I U I █ U I █ U I U U U U █ I I █ I I █ █ U U I █ █ I █ I I I █ U U U I █ 37% \nD9S2\n83 \n94.85 9q22.2 U U U █ U U U U U U █ █ U I █ U I I U U U I U U U U █ U █ U U U U ▓ U I U █ U I █ U I I █ I I █ I U 50% \nD9S1\n816 \n101.8 9q22.3\n2 \nU ▓ U U U ▓ I █ U U U U I U U U U U I I █ U U ▓ U U U I U I █ I I ▓ ▓ U U I I U █ █ U U U U U U U I 48% \nD9S2\n87 \n98.7 9q22.3\n2 \nI I I U I U U U U U █ █ I █ U I U I I █ U ▓ U U U U U █ I █ █ I █ U U U U █ I I U ▓ I U U U U U █ U 48% \nD9S1\n690 \n104.0\n8 \n9q31.1 I █ U U U U U U U U █ █ I █ █ I U I U █ U █ U U U U U U I █ █ I U █ U U U █ I I U █ I U U █ U █ █ █ 64% \nD9S1\n677 \n112.8\n5 \n9q31.3 I █ I U U U █ █ ▓ U █ █ U █ I U U I I I U ▓ U U U U I █ U █ U U U █ █ U █ █ I U U U U U █ █ U █ U U 68% \nD9S9\n30 \n116.7\n5 \n9q32 I █ U U I I U U U U █ █ I U I I I I I █ I U U U U U █ I U █ █ █ U █ █ U █ U I U █ I I I U I U U █ I 43% \nD9S1\n776 \n121.6\n2 \n9q32 I U I U U I U U U █ █ █ I █ █ U ▓ █ U U U U █ █ I █ █ █ U █ █ I █ █ █ U I █ U I █ I I █ █ U █ █ █ █ 71% \nD9S9\n34 \n126 9q33.1 █ I I I I U I I I U I █ I I I I I █ I U I U U U I U █ I U █ █ U █ U █ U I █ I U █ I I █ U I U U █ U 35% \nD9S1\n685 \n132 9q33.2 █ U I U U I U U U I █ U U U U U U U U U █ U U ▓ I █ U █ U U U U █ █ █ I I █ I U U I I U I U U U U I 48% \nD9S1\n682 \n128.7\n7 \n9q33.2 U I I U U U U U U U █ U █ I I I ▓ I U █ U U █ █ I █ █ █ █ █ █ █ █ █ █ I I █ U █ █ █ U █ U I I █ █ I 66% \nD9S2\n90 \n136.4 9q34.1\n1 \nU U █ U U U I U U U I U I I U I U I U I U U U U U U U U U █ █ I █ █ █ U I █ I █ █ I I U █ U U I █ U 46% \nD9S2\n60 \n141 9q34.1\n1 \n█ I I U I █ █ U U U █ U U I I U I I █ U U U U U U U █ U U U █ U U █ U U U U U U █ I U U U I U █ U U 53% \nD9S1\n830 \n145.6\n5 \n9q34.1\n3 \nI I I U I █ U U █ █ █ █ U I I I ▓ I █ █ I ▓ █ ▓ I █ U U █ U █ I █ U █ U U █ I I U I I U I I U █ █ I 51% \n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 58 \nD9S2\n157 \n146.5\n4 \n9q34.2 I █ █ U █ █ U █ █ U █ I U █ I I ▓ I I █ █ I U U U █ █ U U █ █ I █ █ █ U I █ I U U I I I I I U █ █ I 57% \nD9S1\n826 \n157.7\n3 \n9q34.3 U █ I I █ U U I U U █ U I █ I I ▓ U U U U U U U U █ U I U U █ I U █ █ U U █ U I U I ▓ U U I U █ █ I 52% \nD9S1\n58 \n161.7\n1 \n9q34.3 U █ U U I █ I U U U U █ U U U U █ █ ▓ U U ▓ U U I █ U U I U █ U U ▓ U U U █ I U █ U █ █ I U █ U █ █ 74% \nD9S1\n838 \n164 9q34.3 I I █ U I █ I █ U I I U I U I I █ U U U U ▓ █ U I █ U U U █ █ I U U U U I █ I I █ I U I I I U U U I 37% \n \nFootnotes  \n█ indicates informative loci that showed LOH \n▓ indicates MSI \nI indicates informative loci, but no LOH \nU indicates uninformative loci, therefore unable to determine absence or presenc e of LOH \nLast column refers to overall frequency of LOH and MSI combined at informative loci.  \n \n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 59 \nTable 1.11  Allelic loss at chromosome 11 \nGen\netic \nMar\nker \nDe\nCO\nDE \ncM \nCyto\ngenet\nic \nEAOC  \nCLEAR CELL \nEAOC  \nENDOMETRIOID \nSOC  \nCLEAR \nCELL \nSOC  \nENDOMETRIOID \nOVER\nALL \nLOH \nFREQ\nUENC\nY \n   0\n1 \n0\n2 \n0\n3 \n0\n4 \n0\n5 \n0\n6 \n0\n7 \n0\n8 \n0\n9 \n1\n0 \n1\n1 \n1\n2 \n1\n3 \n1\n4 \n1\n5 \n1\n6 \n1\n7 \n1\n8 \n1\n9 \n2\n0 \n2\n1 \n2\n2 \n2\n3 \n2\n4 \n2\n5 \n2\n6 \n2\n7 \n2\n8 \n2\n9 \n3\n0 \n3\n1 \n3\n2 \n3\n3 \n3\n4 \n3\n5 \n3\n6 \n3\n7 \n3\n8 \n3\n9 \n4\n0 \n4\n1 \n4\n2 \n4\n3 \n4\n4 \n4\n5 \n4\n6 \n4\n7 \n4\n8 \n4\n9 \n5\n0 \n \nD11\nS13\n38 \n9.77 11p1\n5.4 \nI █ I I I I U U I U █ I I U I U ▓ I I █ U █ █ U █ █ █ I █ U █ █ U U U U U █ I U I U I U I I █ █ U U 47% \nD11\nS90\n2 \n25.6\n9 \n11p1\n5.1 \nI U U U █ I U U I U I U U I I U U I I U U U ▓ U U U █ I I U U █ █ U █ U U U I I U I U U █ I U U U I 32% \nD11\nS42\n04 \n43.1\n2 \n11p1\n4.2 \nI █ I █ █ I U I █ U U U I I U █ I █ U █ U I U U U U I U U █ █ █ █ █ U U I I I I █ I I I █ I I I I I 40% \nD11\nS93\n5 \n52.9\n4 \n11p1\n3 \nI U I █ █ U █ I █ U █ I █ I I I I U I U █ ▓ U U █ U █ U I █ U █ █ █ U I U I U U U U I U U U U U U U 52% \nD11\nS19\n93 \n59.2\n1 \n11p1\n1.2 \nI █ U U I U █ I I █ I ▓ U I U I ▓ I █ █ ▓ U I I █ ▓ U I ▓ U █ U █ U █ I I I I I I I I █ I I █ U █ I 44% \nD11\nS41\n91 \n64.9\n6 \n11q1\n2.1 \nI █ I U U I █ █ U I U I I U U I ▓ U I █ █ I ▓ U ▓ █ █ U I U █ I U I ▓ I █ I █ I █ I I I I I U U I I 41% \nD11\nS98\n7 \n72.1\n7 \n11q1\n3.2 \nI U U I I U U U I U U I I I I I ▓ I I U I I U U █ █ U I I █ I U █ █ U I U I I I I U U █ U U U U U I 24% \nD11\nS97\n1 \n76.7\n6 \n11q1\n3.4 \nU █ U U U I U U U U U I U U I U ▓ I U █ I U U U U █ U I I █ U U █ U U █ U I U I U U U █ U U U U I I 42% \nD11\nS93\n7 \n83.7\n3 \n11q1\n4.1 \nI █ I I I I I I U I I I U I I █ ▓ I I █ U ▓ U U █ █ █ I I █ I U █ U U U I U I I I I I █ █ I U █ U I 34% \nD11\nS20\n02 \n87.2\n5 \n11q1\n4.1 \nI █ I I U I I I █ U U U I I █ I ▓ I I █ I I U U I █ █ U I █ I █ █ I U U I █ I I I U I █ I U I ▓ I U 34% \nD11\nS91\n9 \n98.3\n1 \n11q2\n1 \nU U I U I I U U ▓ U U U U I U U U U U U U U U U U █ █ I U █ U U █ U █ █ U █ I I I I I █ U █ █ U I I 48% \nD11\nS89\n8 \n103.\n59 \n11q2\n2.1 \nI █ U I I I U I █ U █ U I █ I U █ I I █ █ █ █ ▓ █ █ █ U U █ I U █ U █ U █ █ I I I I █ ▓ █ U █ I █ I 59% \nD11\nS20\n00 \n106 11q2\n2.3 \nI █ I I U █ █ I ▓ U U U I I █ I █ I I U U █ U U I U █ █ I █ █ █ █ █ █ U I █ I █ I U I █ █ U U U I U 54% \n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 60 \nD11\nS19\n86 \n110 11q2\n3.1 \nI █ I I █ █ ▓ I █ I █ █ I I I I █ █ I █ I █ I ▓ █ █ ▓ I I I █ U █ █ █ █ I █ I I I I I █ █ █ █ I I I 51% \nD11\nS31\n79 \n112.\n7 \n11q2\n3.1 \nI █ I I U U U U U █ U U U I U U ▓ I U █ U █ █ █ I U █ I U U U U █ I U U U U I I U U U U U U U U U U 47% \nD11\nS90\n8 \n116.\n46 \n11q2\n3.3 \nI U I U U U █ U ▓ U U I I I █ U █ U U █ ▓ █ ▓ U ▓ █ U U █ █ U █ █ █ █ I U U I █ U U U █ U █ U U █ I 71% \nD11\nS19\n98 \n119.\n99 \n11q2\n3.3 \nI U I U I I U U I U I U U I I U █ ▓ U █ █ █ U U U █ U I U U █ U █ I █ U U █ U U I I I █ U █ █ U █ I 50% \nD11\nS40\n89 \n124.\n37 \n11q2\n3.3 \nI █ I U U I █ U U U I I U I I I ▓ █ █ █ █ U I U █ █ █ I I █ I U █ █ █ █ U U U I I I U U █ U █ █ █ I 54% \nD11\nS44\n64 \n130.\n43 \n11q2\n4.1 \nU █ U I U █ I I U U U █ █ I I U █ I I █ █ █ U U U ▓ █ I I U █ U U I █ I I █ I U U I I █ U █ █ U █ I 52% \nD11\nS93\n3 \n131.\n38 \n11q2\n4.2 \nI U I U I U U U I U U U U I I I █ U U I U U U U U U █ U ▓ U I U U I █ U I █ I █ I U U █ █ █ U U U I 39% \nD11\nS41\n50 \n132.\n95 \n11q2\n4.3 \nI █ U █ U I I I U U U U I I U U U U I I U I U U U U U U U █ I █ █ U U U I U I U I U U █ I I █ █ █ I 36% \nD11\nS20\n18 \n142 11q2\n4.3 \nU U U █ █ I I I I I U I I U I I U I █ I █ I █ I I I I I U █ U I I █ I I I U █ █ I I U █ I I █ U U I 29% \nD11\nS13\n20 \n146.\n94 \n11q2\n5 \nI █ I U U I U U I U U I I I I U ▓ U █ U U █ U U U █ █ I I █ U U U █ I I U U I U I U I █ U U U █ █ I 41% \nD11\nS13\n04 \n148.\n52 \n11q2\n5 \nU █ I U U I I U U U I U U I I █ ▓ I I U █ █ U U █ █ U I █ U I U U U I U U I U U I U █ █ U U U U █ █ 48% \nD11\nS96\n9 \n151.\n02 \n11q2\n5 \nI █ U U I U U U U █ I U U █ I I U I I █ █ █ █ U U █ U I ▓ █ █ █ █ U I █ U I I I U █ U █ █ █ █ █ U U 63% \nD11\nS41\n25 \n152.\n45 \n11q2\n5 \nI █ I U I U █ U U U U I █ █ I I U I I █ U ▓ U U U █ █ U █ █ █ U U █ █ U U U U I I I I █ █ I █ █ █ U 58% \nD11\nS96\n8 \n152.\n45 \n11q2\n5 \n█ █ I █ I I I I U U U I I U I █ ▓ █ I █ █ ▓ U U U █ █ I I █ █ █ █ I █ I U U U I U U I U I U U U I U 48% \nFootnotes  \n█ indicates informative loci that showed LOH \n▓ indicates MSI \nI indicates informative loci, but no LOH \nU indicates uninformative loci, therefore unable to determine absence or presence of LOH \nLast column refers to overall frequency of LOH and MSI combined at informative loci.\n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 61 \nFigure 1.10. Contribution to allelic loss at chromosome 9 by each cancer subtype \n0%\n10%\n20%\n30%\n40%\n50%\n60%\n70%\n80%\n9p21.39q21.319q22.329q31.1 9q32 9q33.1 9q33.29q34.119q34.29q34.3\nCytogenetic loci\nPercentage LOH\nSOC Endometrioid\nSOC Clear Cell\nEAOC Endometrioid\nEAOC Clear cell\nHigh frequency \nLOH regions\n9q32 \nD9S1690\nD9S1677\nD9S930\nD9S1776\n9q34.3\nD9S2157\nD9S1826\nD9S158\n \nFigure 1.11. Contribution to allelic loss at chromosome 11  by each cancer subtype \n0%\n10%\n20%\n30%\n40%\n50%\n60%\n70%\n80%\nPercentage LOH\nCytogenetic loci\nSOC Endometrioid\nSOC Clear Cell\nEAOC Endometrioid\nEAOC Clear cell\nHigh Frequency \nLOH regions\n11q22 \nD11S919\nD11S898\nD11S2000\n11q23.3 \nD11S3179\nD11S908\nD11S1998\nD11S4089\n11q25 \nD11S1304\nD11S969\nD11S4125\n \n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 62 \nFigure 1.12.  Significant association of LOH at 9q34.3 and 11q 23.3 on survival of all \novarian cancers according to Cox Proportional Hazards survival analysis \n \nCox survival analysis for LOH at 9q34.3\nP=0.003  \nCox survival analysis for LOH at 11q23.3\nP=0.034\n \n\n Chapter 1.4. Loss of heterozygosity (LOH) fine mapping of EAOC and SOC at chromosomes 9 and 11 \n 63 \nRationale for selecting Glycodelin and Progesterone Receptor as candidate \ndisease-modifying  genes \n \nA bioinformatic search was performed to examine the published data on genetic expression \nand functional taxonomy of genes at these two genetic loci to select candidate disease-\nmodifying genes. Previous work had identified glycodelin (9q34) expression to be \nsignificantly altered in endometriosis and it had also been implicated in tumourigenesis212-215. \nProgesterone had been implicated in both endometriosis proliferation and anti-proliferation \nand ovarian cancer11;211;216-218; mutations of the Progesterone receptor (PROGINS)(11q22) \nhad been associated with development of endometriosis219. \n\nChapter 1.5. Laser capture microdissection of endometriosis and selected LOH mapping \n 64 \n1.5. Laser Capture Microdissection (LCM) of endometriosis and selected \nLOH mapping \n \nEndometriosis adjacent to EAOC was extracted by LCM and its DNA subjected to LOH \nmapping using the 4 microsatellite markers at chromosome 9 and 11. LOH was identified in \n1/7 cases at 9q34 and 1/7 cases for LOH 11q23.3. Our results did not show strong evidence \nthat LOH events occurred in endometriosis. However, our research collaborators, who \nutilised our Birmingham Women‘s EAOC/SOC samples we had donated, showed LOH to \noccur more frequently when they microsatellite mapped their LCM endometriosis, \nparticularly when the endometriosis LOH corresponded to an adjacent ovarian cancer LOH \nevent (see Table 1.12). \n \nTable. 1.12. Genome wide microsatellite analysis of  endometriosis adjacent to ovarian \ncancer (Prowse, Varma 2006) 220  \n \n\nChapter 1.6. Immunohistochemical investigation using glycodelin and progesterone receptor \n 65 \n1.6 Immunohistochemical investigation of EAOC using Glycodelin (9q34.3) \nand Progesterone receptor (11q22)    \nA summary collation of the immunohistochemical staining for 6 EAOC cases (3 \nendometrioid, 3 clear cell) is depicted in table 1.13 and images are depicted (Figure 1.13, \nFigure 1.14, Figure 1.15). Glyocdelin staining was absent in the ovarian cancer and present \nin the endometriosis distant to the ovarian cancer, but not so strongly expressed in \nendometriosis adjacent to ovarian cancer; this is weak evidence that endometriosis adjacent \nmay be a differing molecular entity to distant endometriosis, and that glycodelin is possibly \ninvolved in causing this difference. No significant differences were observed for PR-A or PR-\nB staining. \nTable 1.13. Summary of immunohistochemistry findings\n \n Endometrioid  EAOC \npatient \nClear Cell EAOC \npatient \nEndometriosis distant \nfrom ovarian cancer \nModerate Glycodelin \nStrong PR-A,PR-B \nModerate glycodelin \nStrong PR-A,PR-B \nEndometriosis \nadjacent to ovarian \ncancer \nWeak Glycodelin \nStrong PR-A, PR-B \nWeak Glycodelin \nStrong PR-A,PR-B \nOvarian cancer Absent Glycodelin \nAbsent PR-A,  \nPatchy positive PR-B \nAbsent Glycodelin \nAbsent PR-A \nPR-B \nFootnotes \nCD10 used as a positive control for identification of endometriosis221-see Figure 1.13 \n\nChapter 1.6. Immunohistochemical investigation using glycodelin and progesterone receptor \n 66 \nFigure 1.13. Immunohistochemistry images of endometriosis and ovarian cancer using \nGlycodelin and CD10  \nGlycodelin (9q34.3) and endometriosis distant from cancer\nGlycodelin CD10\n  \n  \nGlycodelin (9q34.3) and endometriosis adjacent to cancer\nGlycodelin CD10\n \n\nChapter 1.6. Immunohistochemical investigation using glycodelin and progesterone receptor \n 67 \nFigure 1.13 continued. Immunohistochemistry images of endometriosis and ovarian \ncancer using Glycodelin and CD10  \nGlycodelin (9q34.3) negative in cancer\nEndometrioid Clear Cell\n \nFigure 1.14. Immunohistochemistry images of endometriosis and ovarian cancer using \nProgesterone receptor subtypes A and B  (individually labeled) \nProgesterone Receptor (11q22)\nPR-B PR-A\nNormal \nendometrium\nEndometriosis\n \n\nChapter 1.6. Immunohistochemical investigation using glycodelin and progesterone receptor \n 68 \nFigure 1.15. Immunohistochemistry: patchy positive staining of PR-B in endometrioid \ncancer  \n \nPR-B patchy positive in endometrioid cancer \nFigure 0.1 Immunohistochemistry of PR-B in endometrioid cancer \n \n \n\nChapter 1.7. Preliminary nuclear morphometric analysis of endometriosis \n 69 \n1.7. Preliminary nuclear morphometric analysis of endometriosis adjacent \nto ovarian cancer   \n In the single case identified, there was increasing nuclear diameter and pleomorphism in the \ndirect continuum transition between endometriosis, atypical endometriosis and EAOC \n(Figure 1.16). \nFigure 1.16. Nuclear morphometric analysis of endometriosis, atypical endometriosis \nand ovarian cancer that appear as one continuum on the histology slide \nEndometriosis Transition state \n(Atypical endometriosis)\nOvarian \nendometrioid cancer\n \n \n\nChapter 1.8. SNP DNA microarray genotyping of ovarian endometriosis  \n 70 \n1.8.  Affymetrix SNP DNA microarray genotyping of ovarian endometriosis    \nDNA from 10 patients,  5 matched ovarian endometriosis and ovarian surface epithelium and \n5 only ovarian endometriosis DNA,  were subjected to SNP microarray analysis. Multiple, \nextremely small genetic distance areas of LOH were observed in ovarian endometriosis when \ncompared to its matched ovarian surface epithelium control, without alteration of the DNA \ncopy number at that genetic locus. There was no genome-wide consistency of the \nchromosome or chromosomal region affected by this ‗micro-LOH‘ (summarized in Table \n1.14). However, regions on chromosome 11 (Figure 1.17), 15 (Figure 1.18), 21 (Figure \n1.19), 6 and X (Figure 1.20) showed considerable LOH prominence. These regions of LOH \nneed to be validated by confirmatory microsatellite marker analysis. \n \nTable 1.14. Summarising genome-wide LOH regions identified in ovarian endometriosis \nthrough SNP Affymetrix microarray analysis \nChromosomal region where \nLOH identified \nNumber of ovarian \nendometriosis cases (N= 5) \nProximity to regions of LOH \nidentified in ovarian cancer \n1q One case  \n2q One case  \n3q One case  \n6p One case  \n9q No cases 9q32 \n9q34.3 \n11q Two cases 11q23.3   YES \n11q222    YES \n11q25      NO \n15p One case  \n21p Two cases  \nXp Two cases  \nXq One case  \n\nChapter 1.8. SNP DNA microarray genotyping of ovarian endometriosis  \n 71 \nFigure 1.17. Selected images of chromosomal abnormality (chrom 11) in ovarian \nendometriosis (patient 3) compared to their matched normal ovarian surface epithelium \n(patient 2)  \nChrom 11 pat3\n \n \nChrom 11 pat2\n \nOvarian \nendometriosis \nLOH at 11q \nMatched \nnormal ovary \n\nChapter 1.8. SNP DNA microarray genotyping of ovarian endometriosis  \n 72 \nFigure 1.18. Selected images of chromosomal abnormality (chrom 15) in ovarian \nendometriosis (patient 5) compared to their matched normal ovarian surface epithelium \n(patient 3)  \nChrom 15 pat5\n \n \nChrom 15 pat3\n \n \nOvarian \nendometriosis \nLOH at 15p \nMatched \nnormal ovary \n\nChapter 1.8. SNP DNA microarray genotyping of ovarian endometriosis  \n 73 \nFigure 1.19. Selected images of chromosomal abnormality (chrom 21) in ovarian \nendometriosis (patient 6) compared to their matched normal ovarian surface epithelium \n(patient 9)  \nChrom 21 pat6\n \n \nChrom 21 pat9\nMatched \nnormal ovary \nOvarian \nendometriosis \nLOH at 21p \n\nChapter 1.8. SNP DNA microarray genotyping of ovarian endometriosis  \n 74 \nFigure 1.20. Selected images of chromosomal abnormality (chrom 6 and chrom 11 and \nchrom X) in ovarian endometriosis for patient 2 and patent 3 and patient 6,  \nrespectively. \nChrom 6 pat2\n \nChrom 11 pat3\nOvarian \nendometriosis \nLOH at 6p \nOvarian \nendometriosis \nLOH at 11q \n\nChapter 1.8. SNP DNA microarray genotyping of ovarian endometriosis  \n 75 \nFigure 1.20 continued. Selected images of chromosomal abnormality (chrom 6 and \nchrom 11 and chrom X) in ovarian endometriosis for patient 2 and patent 3 and patient \n6,  respectively. \n \nChrom X pat6\n \n \n \nOvarian \nendometriosis \nLOH at Xp, Xq \n\nChapter 2 Analytical observational studies \n76 \n \nChapter 2. ANALYTICAL OBSERVATIONAL STUDIES \n \nIntroduction The use of cohort and case-control studies to benefit clinical practice. \nAppraising the clinical value of cohort and case-control studies. \nResults Examination of methodology through five topics in gynaecology \nChapter  Title \n2.1 Predicting negligence in female sterilization failure using time interval to \nsterilization failure: analysis of 131 cases \n2.2 The effectiveness of a levonorgestrel-releasing intrauterine system (LNG-\nIUS) in the treatment of endometrial hyperplasia – a long-term follow-up \nstudy. \n2.3 Hospital recovery following Thermachoice ablation is not dependent on \nsetting (outpatient or daycase) or rescue analgesia: unexpected result \n2.4 Outpatient Thermachoice endometrial balloon ablation: long term, \nprognostic and quality of life measures \n2.5 Long term outcomes following hysteroscopic myomectomy for abnormal \nuterine bleeding                                                                                                                      \n \n\nChapter 2 Analytical observational studies \n77 \n \n Introduction \n \nCohort and case-control study methodologies are the main types of analytical observational \nstudy. Randomised controlled clinical trials are considered a superior methodology in the \nhierarchy of evidence, because they limit the potential for selection bias and minimise the \ninfluence of confounding due to differences between the two comparison groups (Table 2A). \nHowever, it is either impractical or unethical to perform RCTs to answer all clinical \nscenarios. Furthermore, cohort studies may provide important preliminary evidence to \nsuggest whether a RCT is actually warranted or not. Both Cohort and RCT studies are able to \ndetermine relative risk as both measure incidence. The differences between cohort and RCT \ndesign are depicted in the table below. There are many famous longstanding cohort studies in \nmedicine (e.g. Framingham in Heart Study) and obstetrics (e.g. UK Confidential enquiry into \nmaternal and perinatal mortalities coordinated by CEMACH). \nThe aim of this chapter was to assess the effectiveness of menstrual treatments (Outpatient \nThermachoice endometrial balloon ablation and Hysteroscopic myomectomy) over a long \ntime period in a pragmatic clinical setting (rather than highly selected population). It was felt \nthat the best study design would be a prospective cohort analysis. The chapter discusses the \nfindings in applying the cohort study design to this situation, how reliable data interpretation \ncan be given the study design, and the practical beneficial clinical impact the study has \nachieved. Furthermore, the cohort study design is applied to a rare outcome measure that \ntends to occur after considerable time (failed female sterilisation). The cohort study design is \nchosen to test a mathematical (Bayesian) hypothesis that time interval to sterilisation failure \nis predictive of negligence rather than non-negligence. The publication of this work has \nclarified the medico-legal probability of negligence in those cases where the failure \nmechanism is unknown, and has therefore had profound medico-legal impact.\n\nChapter 2 Analytical observational studies \n78 \n \nFurthermore, in our end of thesis conclusion (chapter 5), we suggest that the cohort design \nmay be under-utilised, and ways to address this. For example, provided the cohort design \nadopts strict case ascertainment and selection criteria (i.e.  minimises selection bias), is \nsufficiently powered to identify and correct for known confounders in comparison groups, \nand utilises sophisticated statistical techniques in the analysis, then the results of the cohort \nanalysis may be at least (if not more) as reliable as those obtained by a suitably powered \nRCT. To achieve this, robust large scale all inclusive prospective cohort databases are \nneeded-akin to the electronic Patient Medical Records database envisaged for both USA and \nUK. \nTable 2A.Advantages and Disadvantages as displayed by Centre for Evidence-Based \nMedicine (Oxford, UK; www.cebm.net ) \nCohort Study Randomised Controlled Clinical Trial \nAdvantages: \nethically safe \nsubjects can be matched \ncan establish timing and directionality of events \neligibility criteria and outcome assessments can \nbe standardised \nadministratively easier and cheaper than RCT  \nAdvantages \nunbiased distribution of confounders \nblinding more likely \nrandomisation facilitates statistical analysis. \nDisadvantages: \ncontrols may be difficult to identify \nexposure may be linked to a hidden confounder \nblinding is difficult \nrandomisation not present \nfor rare disease, large sample sizes or long \nfollow-up necessary \nDisadvantages: \nexpensive: time and money \nvolunteer bias \nethically problematic at times \n \n\nChapter 2.1 Predicting negligence in female sterilization failure \n79 \n \n2.1. Predicting negligence in female sterilization failure using time interval \nto sterilization failure: analysis of 131 cases \n \nBACKGROUND: Sterilization failure due to ‗tubal non-occlusion‘ or ‗wrong structure \nsterilization‘ is considered negligent, whereas ‗spontaneous tubal recanalization‘ or ‗fistula \nformation‘ is considered non-negligent. We examined whether interval to pregnancy failure \nwas predictive of a negligent rather non-negligent failure mechanism. We aim to test this \nhypothesis in a selected population series of known mechanisms of sterilization failure and \ntheir time interval to failure.  \nMETHODS: Analyses of 131 failed sterilizations pooled from UK (NHS Litigation \nAuthority, Medical Protection Society and our hospital), Australia and a qualitative \nsystematic review.  \nRESULTS: We identified 88 negligent and 43 non-negligent sterilization failures. Filshie \nand ring methods failed earlier than diathermy and Pomeroy methods. Sterilization failure \noccurred significantly earlier in negligent than non-negligent failure mechanisms [median \nfailure intervals 7.0 versus 12.0 months; Hazard ratio (2.35 95% CI 1.31–4.21)]. Knowing \nthat sterilization failure occurred early, increased the probability that the failure mechanism \nwas likely to be negligent rather than non-negligent.  \nCONCLUSIONS: A short interval to failure is suggestive of a negligent failure mechanism. \nThere is less certainty in the predictive value of longer time intervals on the mechanism of \nfailure due to a paucity of cases. A national register of failed sterilizations that have been \nsystematically investigated is needed to improve our understanding of negligent and non-\nnegligent failure mechanisms. \n\nChapter 2.1 Predicting negligence in female sterilization failure \n80 \n \nINTRODUCTION  \nFemale sterilization is one of the commonest procedures performed worldwide. In 1999 \naround 50,000 female sterilisations were performed in England in the NHS and charitable \nsectors 1 .  The procedure is performed on mainly healthy women at their request. Where \nresources permit, the preference is to use a laparoscopic technique that occludes tubal patency \nthrough tubal application of a mechanical device (e.g. Filshie, Hulka clip or Fallope ring) or \nelectrocautery. Tubal excision and separation and related techniques (e.g. Pomeroy \nprocedure) are preferred if sterilisation is performed at caesarean delivery. Conception that \noccurs after sterilisation is termed failed sterilisation and can occur several years after the \nprocedure. Two large population-wide studies have reported the ten-year cumulative \nprobability of pregnancy of 18.5 per 1000 procedures  (US CREST study) 2 and 8 per 1000 \nprocedures (Canada) 3 (Table 2.1). Differences in sterilisation failure rates arise due to \nvariation in: the characteristics of the women undergoing sterilisation; operator experience; \noperating centre; sterilisation method chosen, and the time interval to resuming sexual \nactivity post sterilisation and its frequency. However, neither of these studies reported on the \nprecise mechanism of sterilisation failure. In the UK, the RCOG 1 recommends laparoscopic \nsterilisation by either Filshie clip or ring. The 10-year sterilisation failure rate for Filshie clip \nhas been reported by studies as 2-3 per 1000 procedures (Table 2.1). \n \n \n\nChapter 2.1 Predicting negligence in female sterilization failure \n81 \n \nTable 2.1.  Filshie Clip sterilisation failure rates  \nStudy Period \ndata are \ncollected \nfrom \nSterilisations  \nPerformed \nSterilisation \nmethod \nOutcome Type of  \nstudy \nPeterson2 \nUS Collaborative \nreview of Sterilisation \n(CREST) \n1978-1986 \n \n10,685 \n \nVarious methods. \nHulka spring clip \n(1595) \nSilicone Rubber \nband (3329) \nOverall 18.5 per 1000 \nover 10 years \nHulka 36.5 per 1000 \nSilicone rubber band \n17.7 per 1000 \nProspective \ncohort \nmulticentre \nTrussell 3 1980-1999 311,960 Mainly \nLaparoscopic \nFilshie clip  \n8 per 1000 \n[2496 failures] \nRetrospective \nmulticentre  \nKovacs 6 1994-1998 30,000 \n(estimate) \nAll Filshie 2.4 per 1000 \n[73 failures]a \nRetrospective \nmulticentre \nFilshie 7 1982-1992 First 202 \nresponders \nfrom a series \nof 434 \nAll Filshie 2.3 per 1000  \n[1 failure at 6 months] \nCase series \nBirdsall 8 1988-1989 1094  Mainly \nLaparoscopic \nFilshie clip  \n12 per 1000 at 12 \nmonths b  \nCase series \nSokal 9 1984-1990 2746 Filshie clips vs. \nRings \n[2 in each group \nbecame pregnant] \n1.7 per 1000 for both \nRing and Filshie clip \ngroups at 12 months \nRCT \nDominik 10 1984-1990 2126 Filshie clips vs. \nHulka clips \n[11 pregnancies \noccurred:  \n9 Hulka,  \n2 Filshie] \n \nAt 12 months \n1.1 per 1000 for \nFilshie Clip  \n 6.9 per 1000 for \nHulka Clip group.  \nAt 24 months, 9.7 per \n1000 for Filshie  \nand  \n28.1 per 1000 for \nHulka  \nRCT \n\nChapter 2.1 Predicting negligence in female sterilization failure \n82 \n \nFootnotes to Table 2.1 \na   Kovacs: Of the 73 failures, 14 cases were due to operator error, 29 were properly applied \nclips and 30 cases had unknown reason for failure. \nb    Birdsall: Registrars had a 1.3% failure rate, consultants 1.9% and when both a consultant \nand registrar performed the procedure a failure rate was 0.7%. Eighty-six percent (6/7) of \nfailed sterilisations were due to operator error (wrong structure, initial non-occlusion). \n \nThe psychological and physical morbidity following failed sterilisation often leads to \nlitigation 4. Women who have undergone sterilisation performed negligently are entitled to \nrecover damages according to wrongful conception, negligence, and wrongful birth. Also, \nwomen are entitled to recover general damages for pain and suffering during pregnancy and \ndelivery, and loss of earnings during pregnancy. A recent judgment in the Australian High \nCourt 5  led the Australian government to amend the Civil Liberty Act to restrict the amount \nof damages that could be awarded in such situations.  \nDespite intense medico-legal activity, research into the prevention and causation of \nsterilisation failure is lacking. The mechanism of failure should be identified through a \nsystematic assessment of fallopian tube histology, X-ray hysterosalpingography and direct \npelvic visual inspection. If the mechanism of failure is due to ‗tubal non-occlusion‘ or ‗wrong \nstructure sterilisation ‘, these are considered negligent mechanisms, whereas ‗spontaneous \ntubal recanalisation‘ or ‗fistula formation‘ mechanisms of failure are considered non-\nnegligent. However, in the majority of failed sterilisation cases, even those in the advanced \nstages of litigation, the mechanism of failure remains unknown as there is no uniform \nrequirement for such cases to undergo systematic enquiry or to be reported to any supervisory \nnational registry. The RCOG should consider this requirement at the time of the sterilisation \nguideline review in 2006. 1 \n \n\nChapter 2.1 Predicting negligence in female sterilization failure \n83 \n \nThus, a common scenario in the legal setting is to cast judgment on the likelihood of \nnegligence or non-negligence in cases with unknown mechanisms of sterilisation failure. Our \nqualitative systematic review 4 pooled 81 cases of sterilisation failure that had documented \nboth interval to pregnancy and mechanism of failure. We showed that a greater proportion of \nearly (within 12 months from operation) than late (after 12 months from operation) \nsterilisation failures occurred by a negligent mechanism. We therefore propose that interval \nto sterilisation failure may represent a surrogate marker of negligence and non-negligence. \nOur aim was to:- \n1. Determine if sterilisation failure occurred earlier in negligent than non-negligent groups. \n2. Determine if time interval to sterilisation failure was predictive of negligence. \nWe aimed to test this hypothesis in a selected population series of known mechanisms of \nsterilisation failure and their time interval to failure.  \n \nMETHODS  \nA written application was made to NHS Litigation Authority (NHSLA), Medical Defence \nUnion (MDU) and Medical Protection Society (MPS) requesting anonymised information on \nfailed sterilisation cases.  The NHSLA provided 16 cases and the MPS provided 8 cases. \nSimilar anonymised failed sterilisation cases that had been subject to litigation proceedings \nwere retrieved from our hospital legal services department (n=12) and a series from an \nAustralian population (n=14) 11.  These cases were pooled with those identified in our \npreviously published qualitative systematic review 4 (n=81).  A total of 131 failed \nsterilisation cases were identified that reported mechanism of sterilisation failure, interval to \npregnancy and method used for each case. We have only included cases where the cause of \n\nChapter 2.1 Predicting negligence in female sterilization failure \n84 \n \nsterilisation failure has been established either by direct pelvic visualization or histology of \nthe fallopian tubes or a combination of both. Most of our data series examines Filshie clip \nsterilisation failures as our data set emanates from countries where Filshie clip predominates \nas the preferred sterilisation method (i.e. UK and Australia). The derivation of this set is \nshown in Table 2.2.  \n \nSTATISTICAL ANALYSES  \nStatistical analysis was undertaken using SPSS version 13. Geometric means were derived by \nexponentiating the means from the logarithm transformed interval to pregnancy data. \nCategorical correlations were assessed by Chi-squared analysis. Time-to-event methods \n(Kaplan-Meier and Cox regression) were used to investigate covariates impacting on time \ninterval to pregnancy. Graphs of log cumulative hazard for failure against time interval for \nnegligent and non-negligent cases were found to be parallel indicating that the proportional \nhazards assumption was true validating the use of the Cox proportional Hazard regression \nmodel. The probability that a randomly selected case was negligent given sterilisation failure \nbefore a specified time interval was calculated using Bayes‘ Theorem. \n\nChapter 2.1 Predicting negligence in female sterilization failure \n85 \n \nTable 2.2. Databases used to acquire failed sterilisation records \nSource of cases NHSLA  MPS BWH Australia\nn \nSeries \n \nQualitative \nSystematic \nreview \nUsed in \nStudy  \nDates of sterilisation \nprocedure \n1995-\n2004 \n1990-\n2004 \n1987- \n1996 \n1990- \n2000 \n1966- \n2005 \n \nFilshie     70 b 6 13 31 b 17 62+[2] \nDiathermy \nRing \nHulka b \nPomeroy \n0 \n1 \n0 \n0 \n \n4 \n0 \n1 \n0 \n \n0 \n0 \n0 \n0 \n \n0 \n0 \n0 \n0 \n \n20 \n24 \n1 \n19 \n24 \n24 \n[2] \n19 \nTotal included in study a 16 8 12 14 81 131 \n \nFootnotes:- \nNHSLA  National Health Service Litigation Authority \nMPS   Medical Protection Society, UK \nBWH   Birmingham Women‘s Hospital \nAustralian series  This was published in our qualitative systematic review 4. \na Only cases that included all three components (mechanism of failure, interval to \npregnancy and sterilisation method used) were included in the study‘s analysis. \nb          Individual separate analysis of 2 Hulka clip cases would be extremely limited, \ntherefore these were included with the Filshie clip category as both methods utilise similar \nmechanical tubal occlusive devices.  \n\nChapter 2.1 Predicting negligence in female sterilization failure \n86 \n \nRESULTS \n1. Overall interval to pregnancy The mean age for the group was 33.2 years ( SD 4.4;  95% \nCI 31.9-34.4; age range 24-42 years). The arithmetic mean interval to pregnancy was 13.0 \nmonths (SD 14.2; 95% CI 10.6-15.5; range 1 to 102 months). The greatest proportion of \nsterilisation failures occurred by 12 months (72.5%) in a markedly positively skewed \nfrequency distribution. The distribution was normalised by natural log transformation of the \ninterval to pregnancy times to give a geometric mean interval to pregnancy of 9.3 months \n(SD 2.2 months; 95% CI 8.1-10.6). Unlike the arithmetic mean, the geometric mean is not \noverly influenced by the large values in a skewed distribution, and so gives a better \nrepresentation of the average for the purposes of this study.  \n \n2. Negligent and non-negligent failure group compositions and intervals to pregnancy \nFilshie and Ring sterilisation methods failed significantly earlier than diathermy and Pomeroy \nmethods (Log Rank p=0.037); the mean and range intervals to pregnancy are shown in Table \n2.3.  Non-occlusion and wrong structure mechanisms of failure occurred significantly earlier \nthan fistula and recanalisation methods (Log Rank p=0.001); the mean intervals for negligent \nand non-negligent failure were 7.5 and 14.2 months respectively [Table 2.4]. There is a \nsignificant association between sterilisation method used and negligent and non-negligent \nmechanism of sterilisation failure (Chi-square, p= 0.001).  The Filshie clip, most often failing \ndue to non-occlusion or wrong structure, is the predominant method in negligent failures \n(71% of cases) [Tables 2.3, 2.4]. Whereas, Pomeroy, only failing by recanalisation and \nfistula, is the predominant method in non-negligent failures (44% of cases) [Tables 2.3,2.4]. \n \n\nChapter 2.1 Predicting negligence in female sterilization failure \n87 \n \nTable 2.3. Sterilisation method and time interval to pregnancy \nMethod of sterilisation Filshie Diathermy Ring Pomeroy \nor related \nsurgical \nmethod \nOverall \nall  \nGroups \nP value \nNumber in group 64 24 24 19 131  \nInterval to pregnancy (months) \nGeometric Mean \n95% confidence interval \n \n7.6 \n6.1-9.5 \n \n11.9 \n8.5-16.6 \n \n8.2 \n7.6-9.9 \n \n14.2 \n11.4-17.9 \n \n9.3 \n8.1-10.6 \n \n$ 0.037 \n \nRange of time intervals to pregnancy \n(months) for each method \n \nNegligent           Non-occlusion \n                           Wrong structure \n \nNon-negligent   Fistula \n                           Recanalisation  \n \n \n \n \n2-38 \n1-102 \n \n14* \n10* \n \n \n \n3-10 \n9* \n \n3-44 \n60* \n \n \n \n4-5 \n7-20 \n \n6-10 \n6-13 \n \n \n \nNo cases \nNo cases \n \n10-48 \n4-18 \n \n \nFootnotes \n$ Kaplan-Meier Log Rank (Mantel-Cox) test for interval to pregnancy difference \n* Single case only, therefore no range \n\nChapter 2.1 Predicting negligence in female sterilization failure \n88 \n \nTable 2.4. Negligent and Non-negligent failure group compositions and intervals to \npregnancy \nMECHANISM OF FAILURE NEGLIGENT NON-NEGLIGENT \n \nP value \nNumber in group \n \nmean interval to pregnancy and \n95% CI \n \nmedian interval to pregnancy and \n95% CI \n88 \n \n7.5 [6.4-8.8] \n \n7.0 [6.1-8.0] \n43 \n \n14.2 [11.8-17.2] \n \n12.0 [10.6-13.5] \n \n \n$ 0.001 \nComposition by method of \nsterilisation \nNumber of cases / [%] \nFilshie \nDiathermy \nRing \nPomeroy \n62 [71%] \n13 [15%] \n13[15%] \n0 [0%] \nFilshie \nDiathermy \nRing \nPomeroy \n2 [5%] \n11 [26%] \n11 [26%] \n19 [44%] \n \n*<0.001 \nComposition by mechanism of \nfailure \nMechanism \n \nmean interval to pregnancy and  \n95% CI \n \nMechanism \n \n \nmean interval to pregnancy and  \n95% CI \n \n \nNon- occlusion     \n45 [51%] \n6.4 [5.2-7.9] \n \n \nWrong structure   \n43 [49%] \n \n8.9 [6.9-11.3] \n \n \nFistula               \n19[44%] \n17.1 [12.1-24.1] \n \n \nRecanalisation   \n24[56%] \n \n12.4 [10.2-14.9] \n \n \n \n \n \n$ 0.001 \n \nFootnotes \n* Pearson Chi-Square for category composition difference \n$ Kaplan-Meier Log Rank (Mantel-Cox) test for interval to pregnancy difference \n\nChapter 2.1 Predicting negligence in female sterilization failure \n89 \n \n3. Regression analysis of interval to failure  \nGiven that the interval to sterilisation failure was associated with sterilisation method and \nmechanism of failure, and that both of these latter variables may interact with each other, a \nCox regression analysis was performed. The regression showed that negligence compared to \nnon-negligence significantly increased the hazard potential for sterilisation failure, and that \nnegligence (p=0.004) was the only statistically significant covariate when adjusting for \nsterilisation method (p=0.237). The unadjusted Hazard Ratio for negligence was 1.91 (95% \nCI 1.31-2.77), and adjusted Hazard Ratio was 2.35 (95% CI 1.31-4.21).  Therefore, interval \nto pregnancy was predictive of a negligent compared to a non-negligent failure mechanism, \nirrespective of the sterilisation method used. Specifically, the earlier the time interval to \nfailure the greater the likelihood of negligence than non-negligence. This is graphically \nillustrated in Figure 2.1. \n \n4. Probability of negligence for any case given the interval to pregnancy \nWe have assumed that sterilisation failure occurring before time t represents a test of \nnegligence. We have calculated the various test positive (failed before or at time t) and test \nnegative (failed after time t) likelihood ratios (LR) for negligence at various time intervals \nusing Bayes‘ Theorem (Table 2.5). This table shows statistically significantly increasing \nLikelihood Ratios for negligence at successive earlier time interval increments. This is \nconsistent with a mathematical trend that negligence is more likely the earlier the sterilisation \nfailure occurs. \n\nChapter 2.1 Predicting negligence in female sterilization failure \n90 \n \nTable 2.5. Empirical probabilities and likelihood ratios at incremental time intervals. \nTime interval \nthat \nsterilisation \nfailure has \noccurred  \nNegligent \n(n=88) \nNon-Negligent \n(n=43) \n \n \nProbability that \nrandomly \nselected case is \nnegligent from \nthe study series \ngiven failure with \ntime interval \nLikelihood Ratio \nof negligence \ngiven failure \nwithin time \ninterval  \n(LR test positive) \nNot stated 88 43 0.67* -n/a- \n0  6 40 4 0.91  4.89 (1.87-12.77) \n0- 9  61 7 0.90 2.48 (1.51-4.10) \n0  12 73 22 0.77 1.62 (1.19-2.20) \n0  18 81 33 0.71 1.20 (1.01-1.43 ) \n0  24 83 34 0.71 1.19 (1.01-1.40) \n0  48 86 42 0.67 1.15 (1.01-1.32) \n \n \n\nChapter 2.1 Predicting negligence in female sterilization failure \n91 \n \nFootnotes to Table 2.5 \n* The pretest probability of negligence from our case series is 0.67. This corresponds to the \nprobability of a randomly selected case of sterilisation failure being negligent when selected \nfrom our case series. However, knowledge of the time interval to sterilisation failure either \nincreases or decreases the probability of the case being negligent as shown in the table.  \nLikelihood ratios (LR) are derived by dividing the cumulative probabilities of sterilisation \nfailure occurring at or before a certain time interval (t) according to Bayes‘ Theorem. For \nexample, if we consider a test as failure at or before t=10 months then the LR for test positive \nis  \nP(Fail |Neg)\n = 0.7386 =2.65 \nP(Fail |NonNeg) 0.2791 \n \nand, the LR for test negative is \nP(NoFail<10m |Neg) = 1-0.7386  =  0.36 \nP(NoFail<10m |NonNeg) 1-0.2791 \nThus, the probability that a\n randomly selected case is negligent may be calculated by \nknowing the time interval to failure, the Likelihood Ratios at that time interval (as displayed \nin Table 2.5) and the Bayesian equation:  \nPRE TEST      X LIKELIHOOD RATIO        = POST TEST \nODDS   FOR THAT TIME INTERVAL  ODDS \n \nOdds =      Prob.        Prob. = Odds \n          1-Prob.         1+ Odds \nFrom our case series (88 negligent, 43 non-negligent), the pre-test probability of negligence \nwas 0.67 (88/88+43). However, our case series is highly selected. Therefore we suggest using \na pre-test probability of negligence of 0.5 (Odds=0.5/1-0.5= 1). This pre-test probability \nwould correspond to that used in legal proceedings in cases with unknown mechanism of \nfailure and therefore derivation of the post-test probability of negligence (using the Bayesian \nequation or Fagan‘s nomogram) would be useful within this medicolegal context.. \n \nLet us suppose that a sterilisation failure occurred at 8 months and the pre-test probability of \nnegligence is 0.5. The post-test probability of negligence for a case that fails before or at 8 \nmonths is 0.73 (pre-test odds of 1 x LR 3.70=3.70 post test odds; probability is 3.70/1+3.70= \n0.79). In contrast, the post-test probability of negligence if failure had occurred after 8 \nmonths is 0.32 (pre-test odds of 1 x LR 0.48=0.48 post test odds; probability is 0.48/1+0.48). \nThis suggests that failure at 8 months is likely to be negligent because the probability \ndistribution is greater in the negligent (0.73) than non-negligent (0.32) direction from a pre-\ntest probability of 0.5 (see Figure 2.1).  \n\nChapter 2.1 Predicting negligence in female sterilization failure \n92 \n \nFigure 2.1. The probability of sterilisation failure for negligent and non-negligent cases \nagainst time interval to failure (Cox Regression model) \nFootnotes \nThe graph depicts the 1-minus survival function plot of the adjusted Cox regression model \nfunction i.e. incorporates both sterilisation method and failure mechanism covariates. All \ncases have ultimately failed, therefore for both negligent and non-negligent cases the \ncumulatively probability is 1 at the maximum recorded time interval for each group.\nThe \nhazard ratio corresponds to the odds that a case in the negligent group fails before a case in \nthe non-negligent group. Thus, there is a 70% probability (converting Hazard odds of 2.35 to \nprobability by 2.35/ (1+2.35)) that sterilisation failure will occur earlier in a negligent case \nthan a non-negligent case, irrespective of the sterilisation method used. Furthermore, \ncomparing median times (Table 2.3), negligence reduces the time interval to failure by \napproximately 5 months (or 42%) compared to non-negligence.  \n\nChapter 2.1 Predicting negligence in female sterilization failure \n93 \n \nLet us suppose that a sterilisation failure occurred at 18 months and the pre-test probability of \nnegligence is 0.5. The post-test probability of negligence for any case that fails before or at \n18 months is 0.55 (pre-test odds 1 x  LR 1.20=1.20 post test odds; probability is 1.20/1+1.20= \n0.79). In contrast, the post-test probability of negligence if failure had occurred after 18 \nmonths is 0.25 (pre-test odds 1 x  LR 0.34=0.34 post test odds; probability is 0.34/1+0.34). \nThis suggests that failure at 18 months is likely to be non-negligent because the probability \ndistribution is greater in the non-negligent (0.25) than negligent (0.55) direction from a pre-\ntest probability of 0.5 (see Figure 2.1).  \n \nDISCUSSION Analysis of our selected series of failed sterilisations has shown that a \nshort interval to failure, and a long interval to failure are suggestive of a negligent and non-\nnegligent failure mechanism, whilst intervals between the two extremes are less reliable \nindicators of the mechanism of failure.  Negligence compared to non-negligence reduces the \ninterval to failure by 5 months. A test of negligence may be applied to any case of \nsterilisation failure having been provided the time interval to pregnancy and the pre-test \nprobability, as we have obtained likelihood ratios for the test at various time intervals. Such a \ntest may have important medico-legal ramifications in cases with unknown mechanism of \nfailure.  \n \nOur case series represents the world‘s largest number of failed female sterilisations with \nconcurrent knowledge of their mechanism of sterilisation failure and interval to pregnancy.  \nUntil this study, issues involving mechanism of failure, had not been addressed by the two \nlargest studies of sterilisation failure 2;3 or the Cochrane review 12. We had predicted this \nhypothesis in our earlier qualitative systematic review 4. Previous studies had showed \ndifferences in time interval to failure for different sterilisation methods 2and patient age 3. \n\nChapter 2.1 Predicting negligence in female sterilization failure \n94 \n \nWe agree there may be caveats when interpreting our results, particularly as our data series is \nselective. Firstly, our data series is composed of cases from 1975 onwards. Advances in \ntraining in laparoscopic procedures and laparoscopic video imaging may be under-\nrepresented in our data series leading us to overestimate the proportion of negligence \n(operator-fault) that may occur with earlier (1970-1990s) sterilisation failures. Secondly, our \nstudy sample is not derived from a repository of systematically investigated and recorded \nsterilisation failures. Thirdly, although NHSLA has systematically collected data on litigated \ncases in England since 1995, there are many exclusion criteria allowing hospitals to locally \nmanage some failed sterilisation cases thereby limiting case ascertainment. We were unable \nto examine the individual records from the NHSLA and MPS databases to verify the accuracy \nof the failure mechanism reported. Consequently, we are uncertain whether there are \ninconsistencies in the classification of failure mechanism used. Fourthly, we anticipate a \ngeneral under-reporting of non-negligent sterilisation failures in the published literature and \nin the legal databases that we used for the study. Therefore, it is likely that our overall \nestimate of the prevalence of negligence (i.e. pre-test probability of 0.67, 88/88+43) from our \ncase series is likely to exceed the upper limit of prevalence that would be obtained from the \ntrue population of systematically acquired sterilisation failures. \n \nNegligence litigation in the UK is based on the claimant producing the burden of proof \n(prove negligent action has occurred) and the standard of proof is the civil standard (balance \nof probabilities). The claimant has to show that the harm suffered (i.e. failed sterilisation) on \nthe balance of probabilities, is more likely than not to be caused by a negligent action than \nnon-negligent action. In this legal situation, an unknown mechanism of sterilisation failure \ncould be presumed to have a pre-test probability of negligence of 0.5 (legal equivalence). If a \ncase had failed at say 8 months, then applying our test of failure before or at 8 months (post \n\nChapter 2.1 Predicting negligence in female sterilization failure \n95 \n \ntest probability of 0.73) and failure after 8 months (post test probability of 0.32) indicates that \nfailure at 8 months is more likely to be negligent than non-negligent.  Furthermore, for any \ngiven interval to pregnancy, the post-test probabilities of negligence for failure before or after \na specified time interval could be derived using the Bayesian methodology discussed in this \nmanuscript. Although our test provides an overall probability of negligence >0.5 or <0.5  and \ntherefore satisfies the legal test of negligence or non-negligence, we would always endorse \nthat the actual negligent or non-negligent cause of sterilisation failure can only be established \nafter a systematic clinical, histopathological and X-ray examination process. \n \nA national register of systematically collected and investigated failed sterilisations, as \nrecommended by the RCOG 1, would quantify the exact prevalence (pre-test probability) of \nnegligent and non-negligent failure mechanisms, and show how this proportion is distributed \namongst the various sterilisation methods, enabling its use in the legal situation described \nabove. Little is known on non-negligent failure mechanisms due to poor case ascertainment, \nbut such a registry may show that the probability of a non-negligent sterilisation failure \nequated to the probability of a negligent sterilisation failure for a particular sterilisation \nmethod, which would then make any legal claim for negligent sterilisation unlikely to \nsucceed.  Furthermore, such a registry could identify areas of substandard care that could be \nused as an impetus to improve medical training and design effective clinical risk prevention \nstrategies. \n \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n96 \n \n2.2. The effectiveness of a levonorgestrel-releasing intrauterine system \n(LNG-IUS) in the treatment of endometrial hyperplasia – a long-term \nfollow-up study. \n \nOBJECTIVES: Medical treatment of non-atypical endometrial hyperplasia with oral progestogens \nhas limited efficacy and poor compliance. A levonorgestrel-releasing intrauterine system (LNG-IUS) \nhas been shown to successfully treat hyperplasia in small-sized studies.  Our aim was to examine the \neffectiveness of LNG-IUS in a larger study with long term follow up.  \nMETHODS: Prospective observational study of 105 women diagnosed with endometrial hyperplasia \nand treated with LNG-IUS between 1999-2004 at a University Teaching hospital. Baseline \ncharacteristics and outpatient endometrial Pipelle sampling was undertaken at 3 and 6 months post \nLNG-IUS insertion and 6-monthly intervals thereafter in all cases. Outcome included histological data \nderived from both Pipelle and uterine histologies at one and two years LNG-IUS therapy. \nRESULTS:  LNG-IUS achieved endometrial regression in 90% (94/105) of cases by two years, with \na significant proportion (96%, 90/94) achieving this within one year. Regression occurred in 88/96 \n(92%) of non-atypical and 6/9 (67%) of atypical hyperplasias, and in all 22 cases of endometrial \nhyperplasia associated with HRT. Regression rates did not differ between histological types of \nhyperplasia. Twenty-three (22%) underwent hysterectomy of which 13 were indicated and 10 were \nperformed at patient request despite regressed endometrium. Two cases of cancer (one uterine and one \novarian) were identified.  \nCONCLUSION:  LNG-IUS is highly effective at treating endometrial hyperplasia. Beneficial effects \nare observed by the majority within one year. Treatment can be reliably monitored through regular 6-\nmontly outpatient endometrial Pipelle surveillance. LNG-IUS treatment of non-atypical hyperplasias \nis likely to reduce the number of hysterectomies performed in this subgroup.  \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n97 \n \nINTRODUCTION Endometrial hyperplasia may be divided into three principal \nhistological categories listed in the order of ascending architectural and cytological \nabnormality: simple, complex and atypical hyperplasia 13 . Cytological atypia is the most \nimportant prognostic factor for progression to carcinoma 14. Around 1-3% of non-atypical \nhyperplasias progress to endometrial carcinoma, over a mean duration of 10 years. In \ncontrast, 8-30% of atypical hyperplasias progress to carcinoma over a mean duration of 4 \nyears 15. Pooling three observational studies 16-18 the rates of spontaneous regression after \nexpectant treatment for non-atypical (n=129) and atypical hyperplasia (n=28) are around 72% \nand 54% respectively.  \nThe objectives of treating women with endometrial hyperplasia are to reduce abnormal \nbleeding symptoms and to prevent progression to endometrial cancer18-20. In view of an \nincreased oncogenic potential with atypical endometrial hyperplasia, hysterectomy is \ngenerally recommended unless fertility issues or significant risk factors for surgery preclude \nthis. However, for non-atypical endometrial hyperplasia, there is debate as to whether \nhysterectomy is ‗over-treatment‘ given the low risk of malignant transformation, high \nprobability of possible spontaneous resolution, low risk of coexistent uterine cancer and high \ntherapeutic responsiveness to oral progestogen therapy. Nonetheless, oral progestogens are \nassociated with poor compliance and systemic side effects that may limit overall efficacy \n18;19;21.  Levonorgestrel-releasing intrauterine system (LNG-IUS) may be used to successfully \ntreat endometrial hyperplasia without incurring the disadvantages of oral progestogens. This \nfinding has been demonstrated in two recently published observational studies 22;23, together \nwith a systematic review 24 that included four limited sized studies25-28. Our objective was to \nexamine the effectiveness of LNG-IUS to treat endometrial hyperplasia in a larger \nprospective observational study with a long-term follow-up period. \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n98 \n \nMETHODS All women participating in this study had presented to our hospital \n(Birmingham Women‘s Hospital, England) for the investigation of abnormal uterine \nbleeding. Their reasons for referral included: women aged 40 years and over with heavy \nmenstrual bleeding or intermenstrual bleeding aged unresponsive to medical therapies (such \nas tranexamic acid, combined oral contraceptive or oral progestins), post-menopausal \nbleeding and unscheduled bleeding whilst on hormone replacement therapy or tamoxifen. \nNatural menopause was recognised to have occurred if there had been at least 12 consecutive \nmonths of amenorrhoea, for which there was no other obvious pathological or physiological \ncause. Clinical investigation involved transvaginal pelvic sonography, outpatient endometrial \nPipelle sampling (Laboratoire C.C.D, Paris, France) and outpatient hysteroscopy in all cases. \nIntrauterine polyps that were identified at hysteroscopy were removed using outpatient local \nanaesthetic Versapoint® (Gynecare, Ethicon Inc. USA) polyp resection or blind polypectomy \ntechniques. \nEndometrial hyperplasia was subdivided into three categories: simple, complex and atypical. \nFor the purposes of this study, we grouped simple atypical and complex atypical hyperplasias \nas one atypical hyperplasia group. The criteria for diagnosing endometrial hyperplasia and \nendometrial regression of hyperplasia following LNG-IUS use was as we29 and others 13;30-32 \nhave previously described. Typically, LNG-IUS resulted in atrophy of glands separated by \nplump, polygonal, pseudodecidualised stromal cells. These were accompanied by varying \ndegrees of secretory glandular changes and Metaplasia of the lining epithelium. These \nchanges have been collectively and loosely termed as ―regression‖ of hyperplasia in this \narticle. This is not a defined histological entity except in the context of follow up of \nendometrial hyperplasia. Similar morphology can be seen with both oral progestogens and \nintrauterine progestogen (LNG-IUS) when used for other clinical indications. \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n99 \n \nOur study included cases where hyperplasia was only present in the endometrial polyp but \nnot the background endometrium, a phenomenon also described by a previous study 33.All \nhistopathological diagnoses were undertaken by two experienced consultant histopathologists \n(TR, RG) working independently; referral to the other pathologist for a second opinion was \nmade in cases where there was diagnostic doubt, and a mutual consensus was then achieved.  \n \nThroughout the study period (January 1999-January 2004) there were 114 women diagnosed \nwith non-atypical hyperplasia. All were offered oral progestogens, LNG-IUS insertion \n(Mirena®,  Schering Health Care, Burgess Hill, UK) or hysterectomy as part of our routine \npractice; those opting for LNG-IUS (n=105) were included in our study cohort.  Women \ndiagnosed with atypical endometrial hyperplasia were recommended to undergo \nhysterectomy. Women who declined surgery or who were medically unfit to undergo surgery \nwere offered oral progestogens or LNG-IUS insertion; the latter LNG-IUS treated group \n(n=9) were included our study cohort. Women diagnosed with non-atypical endometrial \nhyperplasia whilst using hormone replacement therapy (HRT) were offered either withdrawal \nof HRT and LNG-IUS, withdrawal of HRT and oral progestagens, or HRT (either estrogen \nreplacement therapy or continuous combined preparations) and LNG-IUS; those opting for \ncombinations involving LNG-IUS (n=22) were included in our study cohort. \n \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n100 \n \nBaseline data and study design  \nInsertion of LNG-IUS took place between January 1999 and January 2004. For all women in \nthe study (n=105) anonymised baseline data was recorded on: histological subtype, \nsociodemographic characteristics [with emphasis on risk factors for endometrial hyperplasia \nsuch as parity, body mass index, diabetes, hypertension], use of exogenous hormones (e.g. \nhormone replacement therapy, tamoxifen), and presenting with abnormal bleeding symptoms.  \nStudy participants underwent regular outpatient clinic review and endometrial histological \nsurveillance by outpatient Pipelle sampling. Histological surveillance was performed at 3-\nmonths and 6-months following LNG-IUS insertion, and continued thereafter at 6-monthly \nintervals in all cases (n=105). We present the outcome for participants at 1 and 2 years post \nLNG-IUS insertion, however, in clinical practice, we are continuing to prospectively record \noutcome beyond this time, even in cases that show endometrial regression.  LNG-IUS \ntreatment was abandoned and hysterectomy recommended if:- \n1. There was no histological evidence of partial or complete regression of the hyperplasia by 12 \nmonths of LNG-IUS use. \n2. There was histological evidence of endometrial cancer or progression of endometrial hyperplasia \nto atypia. \n3. There was reversion to the original endometrial histology showing hyperplasia following a period \nof endometrial regression. \n4. The primary outcome was the proportion of women with complete regression of the endometrial \nhyperplasia according to both outpatient endometrial Pipelle and uterine histologies at \nhysterectomy. Secondary outcomes included time to disease regression, the proportion of women \nundergoing hysterectomy (histologically indicated or non-histologically indicated) and the \naccuracy of outpatient Pipelle compared to uterine histology at hysterectomy. \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n101 \n \nStatistical analysis  \nSPSS version 13 for Windows (Release 13.0, 1 Sep 2004, SPSS Inc.) was used.  The \nsignificance of different histological subtypes and other covariates on time interval to \nregression was determined by Kaplan-Meier and Cox-regression survival analysis. A P value \nless than 0.05 was considered statistically significant. Sensitivity, Specificity and Likelihood \nRatios were derived by constructing a 2 by 2 table and using standard techniques34. \nRESULTS \nBaseline characteristics \nThere were 105 women with endometrial hyperplasia (simple 16, complex 80, atypical 9) \nincluded in the 5-year study period. A summary of the baseline characteristics and presenting \nsymptoms are shown in Table 2.6.  The mean age was 54.5 ± SD 10.1 years (range 37-88). \nThe study comprised of 37 premenopausal and 68 postmenopausal women. Most women \npresented with postmenopausal bleeding (n=68). Endometrial polyps were visualised in \n36/105 (34%) cases at hysteroscopy. Hyperplasia in the endometrial polyp, but not in the \nbackground endometrium, occurred in 16% (17/105) of cases; all remaining cases had \nendometrial hyperplasia identified within the endometrium. \nEndometrial regression at 2 years post LNG-IUS insertion \nFigure 2.2 summarises the outcome of the 105 hyperplasias that received LNG-IUS \naccording to pre-treatment and 2-year outpatient endometrial Pipelle histologies. In contrast, \nTable 2.7 summarises the outcome of the study according to histological data derived from \nboth outpatient endometrial Pipelle and hysterectomy histologies at 1 and 2 years post LNG-\nIUS insertion. The derivation for the data are explained in the footnotes to Figure 2.2 and \nTable 2.7. \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n102 \n \nTable 2.6. Baseline characteristics (n=105) of LNG-IUS treatment of endometrial \nhyperplasia \nCharacteristic Size of parameter \n \nAge (years) Mean 54.5   (St Dev 10.1, Range 37-88) \nWeight (kg) Mean 86.0    (St Dev 28.0, Range 50-168 \nBMI kg/m2 Mean 32.0    (St Dev 8.8. Range 18-67) \nCharacteristic  \nPercentage of cases in study group (equals number of cases) \n \nParity a 21% (22)   Parity 0 \n43% (45)   Parity 1 or 2 \n23%  (24)  Parity 3 or higher \nMean 1.87; St Dev 1.34, Range 0-5 \nMenopausal status 35% (37) Premenopausal; 65% (68) Postmenopausal \nDiabetes 18% (19) \nHypertension 30%  (31) \nExogenous HRT   \nExogenous \ntamoxifen \n21% (22)  \n1%   (1)  \nAbnormal bleeding \nsymptoms on \npresentation \n27% (28)    Premenopausal, abnormal uterine bleeding \n9%   (9)      Premenopausal, unscheduled bleeding with HRT \n51% (54)    Postmenopausal bleeding \n13% (14)    Postmenopausal, unscheduled bleeding with HRT or tamoxifen \nFootnotes \na Missing parity data in 14 cases \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n103 \n \nTable 2.7. Outcome of the study according to histological data derived from outpatient \nendometrial Pipelle and hysterectomy histologies \n \nEndometrial  \nHyperplasia \n(number of \ncases at study \ncommencement) \nTotal \nnumber \nof cases \nregressing \nwith \nLNG-IUS \na Mean time \nfor regression \n(months) and  \n95% \nConfidence \nlimits \nProportion \nachieving \nregression b  by \n12 months of \nLNG-IUS \nProportion \nachieving \nregression b  by \n24 months of \nLNG-IUS \n \nSimple (n=16) 15 (94%) 6.2 (4.4-8.0) 15/16 15/16 \nComplex (n=80) 73 (92%) 9.4 (7.0-11.7) 69/80  73/80  \n \nAtypical  (n=9) 6 (67%)   8.2 (5.2-11.3) 6/9  6/9  \nOverall group \n(n=105) \n94 (90%) 9.0 (7.0-11.1 ) c 90/105 c 94/105  \n2. 1 Outcome of study according to histology from Pipelle or hysterectomy \nFootnotes \na  There are no statistically significant differences in probabilities of regression \n over time between simple, complex and atypical hyperplasias [Kaplan-Meier \n Log Rank Mantel-Cox (p=0.20)). \nb     Data on histological regression is derived from combined use of outpatient \n endometrial Pipelle and hysterectomy histologies.  \nc  Two of the 94 cases that shown regression on Pipelle, were subsequently identified to \nhave atypical hyperplasia (one case, formerly simple hyperplasia) and ovarian cancer (one \ncase, formerly complex hyperplasia). The former case underwent hysterectomy at patient \nrequest due to troublesome abnormal bleeding side effects with LNG-IUS despite Pipelle \nregression. The latter case underwent hysterectomy as this was indicated through ongoing \nsonographic surveillance for a postmenopausal cyst concurrent with the regressed Pipelle. \nBoth cases were identified within one year of LNG-IUS treatment.  \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n104 \n \nFigure 2.2.  Outcome of study according to outpatient endometrial Pipelle histology at \npre-treatment and 2-years following LNG-IUS insertion \nFootnotes \na Of the 10 hysterectomies at patient request from the stayed regressed group, \nhistologies from the uteri showed nine regressed uteri and one atypical endometrial \nhyperplasia. This is further explained in Table 2.7, footnote c.  \nb          Of the 7 reverted hyperplasias, all were non-atypical hyperplasias on Pipelle, all were \noffered hysterectomy, five declined hysterectomy in favour of continuing with LNG-IUS. Of \nthe 2 indicated hysterectomies performed, histological analysis showed one had regressed and \none had complex hyperplasia.  \nc          Of the 11 persisting hyperplasias, all were offered hysterectomy, one declined \nhysterectomy in favour of continuing with LNG-IUS. Of the 10 indicated hysterectomies \nperformed, histological analysis showed two had regressed, one simple, four complex, two \natypical hyperplasias persisted and one case of Stage 1A endometrial cancer.  \nLNG-IUS (n=105)\n(simple 16, complex 80, atypical 9)\nRegressed (n=94)\n(simple 15, complex 73, atypical 6)\nPersisting hyperplasia (n=11)\n(simple 1, complex 7, atypical 3) \nStayed regressed (n=87)\n(simple 14, complex 69, atypical 4)\nReversion of hyperplasia (n=7)\n(simple 1, complex 4, atypical 2)\nPersisting hyperplasia (n=11)\n(simple 1, complex 7, atypical 3)\naStayed regressed (n=87)\n68 continue with Mirena\n8 continue with Mirena and ERT\n1 indicated hysterectomy [ovarian ca]\n10 hysterectomy at patient request\nbReversion of hyperplasia (n=7)\n5 continue with Mirena\n2 indicated hysterectomy\ncPersisting hyperplasia (n=11)\n1 continue with Mirena\n10 indicated hysterectomy [of which, \none case endometrial ca.)\n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n105 \n \nOutpatient endometrial Pipelle regression was observed in 94/105 cases, and of these, 87/94 \ncontinued to maintain endometrial regression at 2 years follow up (Figure 2.2). Failed \ntreatment, indicated by persisting Pipelle hyperplasia or hyperplasia that regressed then \nreverted to hyperplasia, occurred in 18/105 cases (Figure 2.2).  \nOverall, 90% (94/105) of the study participants achieved endometrial regression according to \ncombined outpatient Pipelle and hysterectomy histologies (Table 2.7). A significant \nproportion (96%, 90/94) had achieved this by one year of LNG-IUS use.  \nSurvival analysis methods (Kaplan-Meier, Cox proportional hazard) showed there was no \nstatistically significant difference between the types of hyperplasia in terms of the time \ninterval to regression (Table 2.7). The overall mean interval to regression was 9 months \n(95% CI 7.0-11.1) for the overall group (Table 2.7). Furthermore, survival analysis showed \nno statistically significant association of baseline covariates (age, parity, menopausal status, \nBMI, diabetes, hypertension, exogenous estrogen or tamoxifen use) on the rate of regression. \nEndometrial hyperplasia associated with Hormone Replacement Therapy (HRT) \nOf the 22 cases of HRT associated endometrial hyperplasia and treated subsequently with \nLNG-IUS, 2 stopped HRT, 17 continued with cyclical combined HRT and 3 opted for \nestrogen only HRT.  All were non-atypical hyperplasias (19 complex and 3 simple), and all, \napart from one case, showed endometrial regression with LNG-IUS therapy. The non-\nregressed complex hyperplasia underwent hysterectomy and uterine histology subsequent \nconfirmed endometrial regression had in fact occurred. There was a single case of tamoxifen \nassociated complex hyperplasia which initially regressed with LNG-IUS then reverted back \nto complex hyperplasia; uterine histology at hysterectomy confirmed complex hyperplasia.  \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n106 \n \nTwo cases of cancer \nTwo cases of cancer were identified. One case was Stage 1B ovarian cancer, which had been \nidentified in a complex hyperplasia that had regressed at 3 months with LNG-IUS but had \nbeen under ultrasonographic surveillance for a persistent postmenopausal ovarian cyst. The \nother case was Stage 1A endometrial cancer, which had been identified in a case of complex \nhyperplasia that had shown non-regression at 12 months with LNG-IUS and therefore \nunderwent indicated hysterectomy (Figure 2.2). \n \nHysterectomy and correlation with endometrial Pipelle \nHysterectomy occurred in 23/105 women, and a summary of the origin and indication for \nhysterectomy is shown in Figure 2.2.  Most hysterectomies (12/23) were performed for \npersisting hyperplasia and reversion to hyperplasia following initial regression to normal \nhistology. However, 10/23 hysterectomies were performed in women with endometrial \nregression on Pipelle histology. The reasons cited included: worsening or persistence of \nabnormal bleeding symptoms (3), patient request (4), patient fear of progression to cancer (1), \nuterine prolapse (1) and concurrent cervical intraepithelial neoplasia (1). In all these cases the \nendometrium was extensively sampled, including the cornual aspects, and showed changes \nsecondary to the local progestogen therapy without any evidence of hyperplasia. Using \nhistology of the uterus at hysterectomy as the ―gold standard‖ and the preceding endometrial \nPipelle biopsy as a diagnostic test, then Pipelle had a sensitivity of 83% and specificity of \n73% for identifying endometrial regression (Table 2.8).  \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n107 \n \nTable 2.8 Correlation between endometrial Pipelle histology and hysterectomy histology \n(n=23 hysterectomies) \n Uterine Histology at \n Hysterectomy \nRegressed  \nendometrium \nNot regressed \nendometrium \nEndometrial \nPipelle \nbiopsy \nTest positive: showing  \nregression \n10 3 \nTest negative: \nshowing  \nnon-regression \n2 8 \nSensitivity 83% \nSpecificity 73% \nLikelihood ratio (95% confidence interval) \nLR (positive test) 3.06 (1.23-8.74) \nLR (negative test) 0.23 (0.06-0.70) \n \nDISCUSSION \nLNG-IUS is highly effective at treating endometrial hyperplasia, irrespective of whether non-\natypical or atypical hyperplasia is being treated. Beneficial effects are observed by the \nmajority within one year of treatment. Treatment success can be reliably monitored through \nregular 6-monthly outpatient endometrial Pipelle surveillance. Future widespread use of \nLNG-IUS to treat non-atypical hyperplasias is likely to reduce the number of hysterectomies \nperformed for this condition, and thereby avoid exposing women to unwarranted surgical \nrisks. \n \nThis is the largest published series of the use of LNG-IUS to treat endometrial hyperplasia \n24;26;28;35-38. Furthermore, we believe this is the first study to examine the use of LNG-IUS to \ntreat endometrial hyperplasia occurring in HRT users. The prospective design and strict data \ncollection proforma used in this study ensured uniform inclusion/exclusion criteria and \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n108 \n \nreliable collection of all outcome measures. The study was designed as a pragmatic measure \nof the effectiveness of LNG-IUS at one and two-years, therefore our results are applicable to \ncurrent clinical practice. \n \nOur study could be criticised for not incorporating a control (expectant management) or \ncohort (e.g. oral progestogens) comparison group. Furthermore, our study is under- powered \nto detect genuine differences in subtypes of endometrial hyperplasia, as well as investigate \ntheir significance along with other covariates (e.g. diabetes, hypertension, HRT) on the \nlikelihood of regression with LNG-IUS treatment.  \n \nIt has been established that outpatient endometrial biopsy is accurate in diagnosing \nendometrial hyperplasia 39. However, we accept there may be uncertainty in our estimations \nof sensitivity and specificity of endometrial Pipelle in correlating to uterine histology. This is \nbecause we only performed hysterectomy and obtained ‗gold standard‘ uterine histology in \naround a quarter of study participants, and there may be differences in histological criteria \nused by others and our own group. Nonetheless, by finding similar degrees of test accuracy as \nprevious authors 18;40-43we believe our results are at least consistent with the published \nliterature. Furthermore, we minimised the histopathological bias by utilising strict predefined \nhistological criteria and limiting the histological interpretation to two experienced \nHistopathologists. \n \n \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n109 \n \nOverall, our study‘s 90% (94/105) endometrial regression rate incorporates regression rates \nof 92% (88/96) and 67 % (6/9) for non-atypical and atypical hyperplasias, respectively.  A \nhigher regression rate of 95% (19/20) with regression rates of 100% (12/12) and 88%(7/8) for \nnon-atypical and atypical hyperplasias had been observed in a recently published long-term \nstudy 44. This difference could be explained by the longer duration of follow up in the \npublished study 45. Nevertheless, our study‘s non-atypical regression rate is similar to the oral \nprogestogen treatment regression rate (93%, n=134) 46and exceeds the expectantly managed \nregression rate of 72% (93/129) identified by pooling studies 16-18. This study‘s atypical \nregression rate does not significantly differ from the expectant regression rate of 54% (15/28) \nidentified from the same pooled studies. Importantly, this study suggests a trend for \nintrauterine progestogen therapy to regress non-atypical rather than atypical hyperplasia, \nwhich is a finding that has also been suggested by other groups 46-51. \n \nWe would have expected LNG-IUS use in our study to have led to a greater reduction in \nhysterectomy treatment for hyperplasia. However, for a variety of unexpected reasons (e.g. \npersonal choice, fear of progression) in addition to those due to failed medical treatment or \nunwanted side-effects with LNG-IUS, women opted for hysterectomy. We were unable to \nfurther explore how such patient preferences could impact on patient satisfaction, compliance \nand cost-effectiveness of LNG-IUS compared to hysterectomy treatment alternatives.   \nFurthermore, as we were dealing with a pre-malignant condition, in an age group not \nrequiring to conserve the uterus for fertility, this would lead to an increased risk of favouring \na hysterectomy decision, irrespective of whether endometrial regression had been successful \nor unsuccessful. \n \n\nChapter  2.2 Efficacy of Mirena in treating endometrial hyperplasia \n110 \n \nBoth cases of cancer identified in the study were Stage I tumours, and were readily identified \nwithin one year of insertion of LNG-IUS. It could be argued that earlier hysterectomy, \ninstead of LNG-IUS medical treatment, would have prevented cancer development or \nimproved prognosis if cancer was identified earlier. In this context, our study suggests around \n50 hysterectomies would be needed to prevent (NNT) one case of gynaecological cancer in \nwomen with endometrial hyperplasia.  \n \nOral progestagens and hysterectomy are widely accepted treatment options for endometrial \nhyperplasia 18;19;52. Newer therapies under evaluation include endometrial ablation 53 and \naromatase inhibitors54. Nonetheless, we believe that the success of this study, utilising LNG-\nIUS therapy, should provide an impetus for future robust randomised controlled trials to \nevaluate the effectiveness of medical and surgical treatments in treating endometrial \nhyperplasia. Successful validation of the treatment potential of LNG-IUS for endometrial \nhyperplasia will undoubtedly reduce the number of women undergoing hysterectomies for \nthis condition and avoid exposing them to unwarranted surgical risks.\n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n111 \n \n2. 3. Hospital recovery following Thermachoice ablation is not dependent \non setting (outpatient or daycase) or rescue analgesia: unexpected result \n \nBackground: Thermal balloon endometrial ablation (TBEA) is increasingly being performed \nin the outpatient setting under local anaesthesia (LA) rather than in a daycase setting under \ngeneral anaesthesia (GA).  Our aim was to compare the post operative rescue analgesia \nrequirements and duration of hospital say in women undergoing outpatient (LA) and daycase \n(GA) TBEA.   \nMethods: Prospective observational study of consecutively recruited women who underwent \noutpatient (LA) TBEA (n=51) and daycase (GA) TBEA (n=50) over the same time period.  \nAnalgesia that was provided additional to the standard administered analgesic regimen was \nconsidered rescue analgesia. The main outcome measures were requirement for rescue \nanalgesia and duration of hospital stay in both cohorts. \nResult(s): LA compared to GA cohorts had shorter hospital stays (11 hours [95% CI 9-13] \nvs. 17 hours [95% CI 14-20]) and lower analgesia requirements. However, multivariate \nregression, correcting for all known confounders, showed that duration of stay was \nindependent of setting for ablation or amount of rescue analgesia. \nConclusion(s): Duration of hospital stay is not entirely dependent on whether outpatient or \ndaycase endometrial ablation is considered. This unexpected preliminary finding deserves to \nbe validated in future confirmatory trials that compare outpatient and daycase treatments. We \nalso discuss the confounding factors that should be considered when designing such trials. \nAbbreviation(s): TBEA: Thermal Balloon Endometrial Ablation; LA: Local anaesthesia; \nGA: General Anaesthesia; CI: Confidence Interval. \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n112 \n \nINTRODUCTION Menorrhagia has a considerable impact on many women's lives 55.  \nEndometrial ablation is being increasingly used as a treatment option 56 and is endorsed by \nNational Institute for Health and Clinical Excellence, NICE, UK 55. There is wide variation in \nthe preferred endometrial ablation device and whether treatment should be performed in the \noutpatient local anaesthetic (LA) or daycase general anaesthesia (GA) setting 57-61. \n \nOutpatient therapy has obvious advantages in terms of safety, convenience and short \ndischarge time for the woman, and may be preferred over GA for women with high risk \nmedical conditions62;63. We 64, along with other groups 65-68, have had considerable \nexperience and success in performing outpatient thermal balloon endometrial ablation \n(TBEA).  We perform local anaesthetic (LA) thermal balloon endometrial ablation in the \nconscious patient without sedation at any time in the menstrual cycle and without prior \nendometrial preparation.  \n \nThere is considerable heterogeneity in postoperative pain and duration of hospital stay \nreported for LA and GA endometrial ablations. This may be partly explained by differences \nin peri-operative analgesic regimens adopted by such studies. Even if such confounding \ninfluences are minimized, it remains unclear whether women experience higher levels and/or \nprolonged duration of pain during and after LA TBEA compared to GA TBEA. This \ninformation would be particularly important when counseling women on their choices \nbetween LA and GA TBEA. Given the paucity of robust data to answer this concern64;69;70, \nwe conducted a prospective study to compare rescue analgesia requirement and duration of \nhospital stay in LA and GA TBEA.   \n \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n113 \n \nMATERIALS AND METHODS: \nPatient Population Pre-menopausal women with subjectively defined heavy menstrual \nbleeding were referred by primary care (GP) and / or by secondary care physicians for \nassessment in our menstrual disorders clinic. Our routine practice was to offer a first line trial \nof medical treatments for at least 6 months if there was no clinical suspicion of underlying \npathology. The medical treatments included Levonorgestrel-releasing intrauterine hormone \nsystem (LNG-IUS; Mirena®, Schering Health Care), combined oral contraceptive, \nprogestogens (oral and long-acting), tranexamic acid and /or mefenamic acid.  \n \nAll women underwent transvaginal pelvic sonography, hysteroscopy and outpatient \nendometrial Pipelle biopsy (Laboratoire C.C.D, Paris, France) investigations. Any \nsignificantly sized intrauterine polyps (greater than 2cm in size) were excised by either blind \npolyp forcep avulsion or Versapoint [Gynecare, Ethicon Inc. USA) resection. Women were \nexcluded from the study if there were significantly sized uterine fibroids (fibroids greater \nthan 3cm size in any uterine location), enlarged uterine size (uterine cavity length greater \nthan 10cm), abnormally shaped uterine cavity (e.g. bicornuate uterus), endometrial \nhyperplasia or cancer, or active pelvic infection. \n \nWomen with normal sized uteri (less than 10cm cavity size), no underlying structural uterine \npathology and unresponsive to medical therapy commenced by their GP or secondary care, \nwere offered thermal balloon endometrial ablation TBEA (either under LA or GA) and \nhysterectomy as second-line treatments.  Those women who opted for TBEA were given the \nchoice of undertaking the procedure under LA or GA. \n \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n114 \n \nStudy design Recruitment for the study occurred in a prospective continuous manner \nbetween June 2003 and June 2005. During this time period, two prospective consecutively \nrecruited cohorts were established: LA TBEA and GA TBEA i.e. both cohorts were \nconstructed and evaluated over the same time period in parallel. \nIntervention Endometrial ablation was performed using a Thermachoice III (Gynecare®, \nMenlo Park, California, USA) device according to the manufacturer‘s guidance.  \nLocal  Anaesthetic TBEA\n This was performed in our ambulatory gynaecological clinic \naccording to our previously described protocol 64, which included:- \nPre-procedure analgesic regimen (one to two hours prior to TBEA): \n All women received diclofenac 100mg rectally, oral co-dydramol 10/500 (two tablets) and \noral cyclizine 50 mg. Tramadol hydrochloride 100mg was used if non steroidal analgesia was \ncontra-indicated.  \nLocal anaesthetic: The cervix was directly injected in a circumferential manner with three \n2mL cartridges containing 3% prilocaine hydrochloride (30 mg/mL) and felypressin \n0.03 unit/mL (citanest with octapressin®, Dentslply, UK) using a 27G dental syringe.  \nDedicated patient nurse: A particular nurse was allocated to provide continuous supportive \ncare to the patient during the procedure. The nurse engaged the patient in conversation \n(‗distraction‘ analgesia effect termed ―vocal local‖) and often held the patient‘s hand \nthroughout the procedure.  \nPost ablation day case bed stay: All women recovered in a day case bed and were allowed \nhome after a minimum stay of 2 hours. A strict protocol of post-procedure pain relief was \nadhered to. A patient information leaflet was provided detailing expected symptoms and \nanalgesic advice post LA TBEA.  \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n115 \n \nGeneral Anaesthetic TBEA    Women, fasted for at least 6 hours, were admitted to hospital \non the day of the procedure. In a minority of cases, women with high risk medical disorders \n(e.g. diabetes) were admitted the day before the planned procedure.  TBEA was carried out in \ngynaecology theatres after induction of general anaesthesia.  All women received diclofenac \n100mg and 1g paracetamol rectally (or paracetamol alone if diclofenac was contraindicated) \njust prior to performing TBEA. Infiltration of the cervix with a local anaesthetic was not done \nin these women.  The TBEA surgical procedure, post-procedure analgesia regimen and day \ncase bed stay for GA TBEA were identical to the LA TBEA procedure described above. \n \nOutcome measures Initial baseline data recorded were: age, body mass index, menorrhagia \nalone or combined with dysmenorrhoea, menstrual cycle phase, ultrasound and hysteroscopy \nfindings and uterine axis. In relationship to TBEA procedure, the following data were \nrecorded: mean intrauterine ablation pressure, successful completion of TBEA procedure, \nprocedure related complications (e.g. vasovagal episodes for LA TBEA) and duration of \nhospital stay following the TBEA procedure. All women were asked to record the pain they \nexperienced immediately following LA TBEA on a graduated Visual Analogue Scale (VAS), \nranging from 0 (no pain) to 10 (worst imaginable pain), which had been validated in our \nprevious study 64. \n \nRescue analgesia\n This refers to analgesia that was administered post TBEA that was \nadditional to the routinely supplied peri-operative analgesia regimen.  Rescue analgesia was \nadministered at the request of the woman following nurse-led enquiry. The amount of rescue \nanalgesia was determined according to the woman‘s VAS score at the time of enquiry and \nhierarchy of analgesia that was available on a standardised ‗as required‘ drug prescription \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n116 \n \nchart. To quantify the amount of rescue analgesia we utilised a numerical (morphine \nequivalent dose) and an ordinal (mild, moderate, severe) scale was created according to the \nfollowing: \na) Morphine Equivalent Dose. This is the estimated bioequivalent dose of morphine sulphate \n(in milligrams) that corresponds to the oral analgesic preparations (such as codeine \nphosphate, dihydrocodeine) given according to an accepted validated conversion scale 71. \nb) An ordinal ranking scale of none, mild, moderate, strong, very strong rescue analgesia. \nThis scale was created by the study authors, and recorded as mild (paracetamol <2g or \ndiclofenac  <100mg only), moderate (paracetamol <2g and diclofenac <100mg or low \nmorphine equivalent dose <15mg) or strong (paracetamol>2g or diclofenac>100mg or high \nmorphine [>15mg] dose) or very strong (paracetamol> 2g and diclofenac>100mg and \nmorphine>15mg or high morphine [>30mg] dose) grading for strength of analgesia usage. \n \nPost TBEA procedure (common to both LA and GA TBEA cohorts)  \nAll women recovered in a daycase bed. Women were discharged home according to a Nurse-\nled care plan that required patients to have tolerated oral diet, voided urine, and have \nadequate pain control.  All women were discharged with a patient information leaflet that \ndescribed expected postoperative symptoms and were given instructions to take regular \nanalgesics for the first 24 hours (diclofenac 50 mg three times daily and/or co-dydramol \n10/500 two tablets four times daily).  In addition, all women were contacted by telephone at \nhome the following day to check on their progress. \n \n \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n117 \n \nStatistical analysis:  \nDichotomous data were presented as simple proportions.  SPSS version 13 was used to \nundertake univariate linear regression and multivariate regression analysis and to conduct \nChi-square and Mann-Whitney U test for comparing the difference between the two groups.  \nP < 0.05 was considered statistically significant.   \nSample size & Power calculation  \nThere was no pilot data of the expected mean and standard deviation values for the amount of  \nanalgesia used or hospital stay.  Hence, a sample size calculation was not performed a priori \nto study commencement. However, if we assume that a clinically significant difference of the \nmean between two groups is 0.5 Standard Deviations, then the sample size required for an \nalpha of 0.05 and a power of 80% is 64 in each group. Hence, as our study recruited 101 \nsubjects, it approaches the power required to detect this accepted clinically significant \ndifference.  \n \nRESULTS: There were 51 and 50 women in LA and GA TBEA cohorts respectively.  \nBaseline characteristics are depicted in Table 2.9.  The procedure was completed \nsuccessfully in all women in both cohorts.  There was no serious morbidity in either cohort.  \nIndividual requirements for different analgesics are shown in Table 2.10.  The strength of \nrescue analgesia was found to be statistically significantly lower in the LA compared to GA \ncohort: 8/51 compared to 47/50 women required moderate to strong analgesia, respectively \n(Table 2.10).   \n \n \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n118 \n \nOn univariate analysis, duration of hospital stay correlated to strength of rescue analgesia and \ntype of TBEA; significantly lower in LA (11 hours; 95% CI 9 - 13 hours) compared to GA \n(17 hours; 95% CI 14 - 20 hours) cohorts (Tables 2.10, 2.11 and Figure 2.3). However, \nmultivariate regression, correcting for identifiable confounding influences (listed in footnotes \nof Table 2.11), showed that duration of hospital stay was independent of strength of rescue \nanalgesia and type of TBEA (Table 2.11). In the LA cohort, there were no postoperative \ncomplications in 44 (86%) women but 7 (14%) stayed overnight; 2 (4%) due to excessive \nvomiting and 5 (10%) due to pain.  In the GA cohort, there were no postoperative \ncomplications in 36 (72%) patients but 19 (38%) stayed overnight; 2 (4%) due to excessive \nvomiting, 4 (8%) due to pain, 3 (6%) due to urinary retention, 4 (8%) due to dizziness and 6 \n(12%) due to medical reasons unrelated to the ablation procedure (such as hypotension, \nhypertension, transient oxygen requirement). \n \n \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n119 \n \nTable 2.9.  Baseline and procedural characteristics of LA vs GA TBEA \n LA TBEA \nN=51 \nGA TBEA \nN=50 \nOverall \nN=101 \n \nMean age years (Range) \n \n44.1 (30-54) \n \n42.6  (29-55) \n \n43.4 (29-55) \nMean BMI (Range) 30.3 (19-55) 27.7 (14-45) 28.9 (14-55) \nPresenting complaint \nMenorrhagia \nMenorrhagia & dysmenorrhoea \n \n46 \n5 \n \n40 \n10 \n \n86  \n15  \nPhase of cycle \nMenstrual \nProliferative \nMid-cycle \nSecretory \n \n9 \n12 \n5 \n25 \n \n1 \n19 \n14 \n16 \n \n10 \n31 \n19 \n41 \nUterine Scan findings \nNormal  \nPolyp or fibroid \n \n43 \n8 \n \n41 \n9 \n \n84 \n17 \nUterine axis \nAnteverted \nRetroverted \nAxial \n \n36 \n7 \n8 \n \n37 \n13 \n0 \n \n73 \n20 \n8 \nHysteroscopic Uterine findings \nNormal  \nPolyp or fibroid \n \n44 \n7 \n \n47 \n3 \n \n91 \n10 \nIntrauterine Ablation pressures (mmHg)  \n(95% CI intervals) \n170 \n(164-175) \n171 \n(168-174) \n170 \n(168-173) \n \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n120 \n \nTable 2.10.  Outcomes of LA vs. GA TBEA \n LA-TBEA \nN=51 \nGA-TBEA \nN=50 \nOverall \nN=101 \nDifference \nbetween LA \nand GA \nP value \n \nMean duration of stay (hours) \n(95% CI intervals) \n \n11 \n(9-13) \n \n17 \n(14-20) \n \n14 \n(12-16) \n \n0.001 \nStrength of analgesia \nNone \nMild  \nModerate \nStrong \nVery strong \n \n1 \n42 \n7 \n1 \n0 \n \n0 \n3 \n7 \n33 \n7 \n \n1 \n45 \n14 \n34 \n7 \n \n0.001 \nParacetamol   Used (mean dose, mg) \n                       Not used \n25 (617) \n26 \n42 (1760) \n8 \n67 (1206) \n34 \n0.001 \nDiclofenac     Used (mean dose, mg) \n                       Not used \n0 \n51 \n 44 (101) \n6 \n44 (52) \n57 \n0.001 \nMorphine Mean Equiv.Dose (mg) \n(95% CI intervals) \n13.8 \n(11.5-16.1) \n14.2 \n(11.0-17.3) \n14.0 \n(12.0-15.9) \n0.940 \n \nFootnotes \nStatistical tests include Chi-square and Mann-Whitney U test. \n \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n121 \n \nnone m ild m oderate strong very strong\nStrength of rescue analgesia\n-500\n0\n500\n1000\n1500\n2000\n2500\nDuration of hospital stay (minutes)\n \n \n\nTable 2.11. Regression analysis \n Duration of Hospital Stay $ \nLA vs. GA Thermachoice  \nUnivariate ^ \nMultivariate* \n \n0.001  \n0.786 \nStrength of Rescue analgesia \nUnivariate ^ \nMultivariate* \n \n0.001 \n0.303 \nFootnotes \n$ Univariate Linear Regression modelling \n* Multivariate Regression corrected for the presence of fixed categorical factors [LA or GA; presenting \ncomplaint; uterine axis; scan findings; hysteroscopic findings; menstrual phase] and covariates [strength of \nrescue analgesia; intrauterine ablation pressure; uterine length; age; BMI]. \n^ All statistical models were statistically significant (P<0.001) apart from final multivariate regression model.  \n \nFigure 2.3.  Correlation of duration of stay with strength of analgesia for combined LA \nand GA TBEA cohort \n \nStatistically significant correlation (Pearson P=0.001; Kendall P=0.001)  \n \n \n \n \n \n \n \n \n \n \n \nFootnotes        Central box dot shows Mean.  Error Bars show 95% Confidence Interval of Mean. \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n122 \n \nDISCUSSION: \nThis preliminary study suggests that duration of hospital stay is independent of setting \n(outpatient or daycase) of endometrial ablation or amount of rescue analgesia administered. \nEven though on direct observation it appears that there may be shorter post-recovery times \nand lower rescue analgesia with outpatient compared to daycase ablation. This information \nmay be useful for preoperative counselling, but its unexpected result deserves to be validated \nin future confirmatory trials. \n \nTo date, there is a dearth of evidence comparing outpatient LA and GA daycase \nhysteroscopic based treatments, including endometrial ablation 55;57-61. We believe our study \nis the largest sized comparison of LA and GA endometrial ablation, and exceeds the size of \nthe recently published RCT comparison of outpatient and daycase Thermachoice 65. \nIntroduction of study bias was minimized by the prospective continuously recruited cohort \nstudy design and adopting standardized regimens for perioperative analgesia and post-\noperative care. The study was conducted in a pragmatic manner and therefore our findings \nare applicable to current practice.  \n \nHowever, we accept there may be limitations that may make our conclusions less reliable. \nWe did not utilise any specific method of reliably identifying women‘s individual pain \nthresholds (e.g. able to either tolerate outpatient endometrial Pipelle or outpatient \nhysteroscopy procedure) prior to ablation and so are uncertain to the prevalence of women \nwith low-to-high pain thresholds in our two cohorts. Women who opted for LA TBEA may \nhave an inherently higher pain threshold, received more detailed pre-procedure counselling, \nand be more motivated to successfully complete and recover from this procedure, than \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n123 \n \nwomen who opted for GA TBEA. Likewise, higher analgesia in the GA cohort may relate to \nthe higher prevalence of reported dysmenorrhoea and retroverted uterus (possibly indicating \nmore significant pathology such as endometriosis) compared to the LA cohort. Alternatively, \nit is conceivable that the local anaesthetic itself induces highly effective peri-operative \nanalgesia and its effects are sustained over several hours.  We attempted to correct for this \nconfounding using multivariate regression. However, overall, our study is non-randomised \nand likely to be underpowered; the use of regression methodology in such circumstances may \nhave led to spurious interpretation. It would have been useful to record patient satisfaction \nwith pre-procedure counselling, as well as their original preferences for TBEA setting (even \nif they ultimately had a different TBEA setting) prior to the procedure and explore how these \nfactors could impact on both short (post procedure analgesia and recovery time) and long-\nterm outcomes (e.g. surgical re-intervention rates).   \n \nWe achieved successful completion of outpatient TBEA in all our cases [100%, 51/51], \nwhich exceeds that reported by the recently published RCT [87%, 34/39] 65. Our mean \noutpatient recovery time of 11 hours (which includes 7/51 overnight admissions) is \nconsiderably greater than the trial‘s 1 hour 40 minutes 65.  We believe these differences arise \ndue to fundamentally differing patient selection criteria and protocols for perioperative \nanalgesia and nurse-led discharge.   \n \nAt first glance, and in agreement with a recent RCT 65, we showed that LA may result in a \nlower analgesia requirement and shorter recovery time period, indicating from both a cost-\neffective and patient‘s perspective that TBEA should be preferentially performed in the \noutpatient LA rather than currently favoured daycase GA setting. However, our \n―multivariate‖ regression, which corrected for all potential confounders and was not \n\nChapter 2.3 Outpatient vs. Daycase Thermachoice \n124 \n \nundertaken by the previous trial 65, showed that there was no statistically significant \nassociation between setting (outpatient LA or daycase GA) or amount of rescue analgesia \nupon duration of hospital stay. This contradicts the earlier stated hypotheses that there may be \ninherent differences between LA and GA groups in relation to women‘s pain thresholds or of \na ―superior‖ analgesic effect induced through use of LA compared to GA technique.  In order \nto define the optimum role for outpatient ablation, we recommend further RCTs directly \ncomparing outpatient against daycase treatments. It is important that these trials are \nsufficiently powered, and are able to correct for the confounding influences we have \ndiscussed earlier. \n \n \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n125 \n \n2.4. Outpatient Thermachoice endometrial balloon ablation: long-term, \nprognostic and quality of life measures \n \nSTUDY OBJECTIVE: To compare short and long term treatment outcomes of outpatient local \nanaesthetic thermal balloon endometrial ablation (LA-TBEA) and identify any prognostic factors. \nDESIGN: Prospective observational study  \nDESIGN CLASSIFICATION: II-2  \nSETTING: U.K. teaching hospital. \nPATIENTS: 102 menorrhagic women undergoing LA-TBEA between 2001-2005. \nINTERVENTIONS: Thermachoice I (n=51) and Thermachoice III (n=51) TBEA performed under \nlocal anaesthesia without conscious sedation. \nMEASUREMENTS: Treatment completion, pain and analgesia, duration of stay (from admission to \ndischarge), duration of follow up, need for secondary treatment (repeat ablation, hysterectomy or \nLNG-IUS), menstrual symptoms and amenorrhoea, patient satisfaction, and quality of life.  \nRESULTS: TBEA was completed in 97.1% of women. Mean duration of stay was 8.0 hours (95% CI \n6.6-9.3). Mean follow up was 29 months (95% CI 26-32). Secondary treatment occurred in 19/102 \n(19%) and was more likely in Thermachoice I (15/51, 29%) than Thermachoice III (4/51, 8%). \nOverall, 50% of surgical re-interventions occurred by 19 months. There were high rates of \namenorrhoea (29%) and treatment satisfaction (76%). Higher mean intrauterine ablation pressure was \nassociated with increased treatment satisfaction.  \nCONCLUSION:  Endometrial ablation can be successfully performed in the outpatient setting with \nbetter success rates achieved with Thermachoice III.  Higher ablation pressures improve long term \noutcomes. \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n126 \n \nINTRODUCTION \nThere has been considerable expansion in the establishment of Outpatient ‗One Stop‘ ‗See \nand Treat‘ ambulatory clinics in the management of women with abnormal uterine bleeding \n72.  Endometrial ablation is being increasingly used as a treatment option 56 and is endorsed \nby National Institute for Health and Clinical Excellence, NICE 55. Outpatient therapy has \nobvious advantages to the patient in terms of safety, convenience and short discharge time \nafter treatment.  The health provider gains by avoidance of costs associated with in patient \nadmission and general anaesthesia.  There is wide variation in the preferred endometrial \nablation device 73;74 and whether treatment should be performed in the outpatient (using local \nanaesthetic and/or sedation) or daycase general anaesthesia setting 69;75. \n \nWe 64, along with other groups 65-68, have had considerable experience and success in \nperforming outpatient thermal balloon endometrial ablation (TBEA).  We perform local \nanaesthetic thermal balloon endometrial ablation (LA-TBEA) in the conscious patient \nwithout sedation at any time in the menstrual cycle and without prior endometrial \npreparation.  \n \nIn relation to TBEA, there are particular prognostic factors associated with favourable \noutcome following ablation; these include: anteverted compared to retroverted uterus, older \nage, shorter uterine length, lower (<10ml) intrauterine balloon volumes and higher \nintrauterine pressures 68;76-79. Our aim was to compare the short and long term (minimum 12 \nmonths follow-up) treatment outcomes for outpatient LA-TBEA using Thermachoice I and \nThermachoice III devices and identify any prognostic factors that may influence treatment \noutcome. \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n127 \n \nMATERIALS AND METHODS \nStudy population Recruitment for the study occurred prospectively, in a continuous \nmanner, between February 2001 and August 2005.  During this time period, we upgraded our \nThermachoice device: at study commencement we used Thermachoice I (Gynecare®, Menlo \nPark, California, USA) and this was replaced with Thermachoice III (Gynecare®, Menlo \nPark, California, USA) from August 2003 onwards. Thermachoice III contained an impeller \nfan that provided a more even temperature gradient within the balloon and on its surface. \nAccepting that there may be differences in the treatment outcomes between the different \ndevices, we have compared outcomes between Thermachoice I and III, as well as reported \noverall combined outcomes. \n \nPre-menopausal women with subjectively defined heavy menstrual bleeding were referred by \nprimary care (GP) and / or by secondary care physicians for assessment in our menstrual \ndisorders clinic. In the clinic, all patients are assessed on the need for treatment based on the \nimpact of heavy menstrual bleeding (HMB) on the patient‘s quality of life, reported \nmenstrual symptoms, presence of gynaecological pathology (all women routinely had pelvic \nultrasound), fertility requirements, and proven anaemia. Our routine practice was to offer a \nfirst line trial of medical treatments for at least 6 months if there was no clinical suspicion of \nunderlying pathology. The medical treatments included Levonorgestrel-releasing intrauterine \nhormone system (LNG-IUS; Mirena®, Schering Health Care), combined oral contraceptive, \nprogestogens (oral and long-acting), tranexamic acid and / or mefenamic acid. This method \nof practice has been endorsed by the National Institute of Clinical Excellence (NICE) \nguideline on HMB 55.All women were investigated by transvaginal pelvic sonography, \noutpatient endometrial Pipelle (Laboratoire C.C.D, Paris, France)  and outpatient \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n128 \n \nhysteroscopy. Women with normal sized uteri (less than 10cm cavity size), no underlying \nstructural uterine pathology and unresponsive to medical therapy commenced by their GP or \nsecondary care, were offered endometrial ablation (either under general anaesthetic or local \nanaesthetic), or hysterectomy as second-line treatments.  Those women who opted for LA-\nTBEA were invited to participate and included in this study. This population included women \nwith regular and irregular menstrual cycles who expressed a desire for further treatment. No \nspecific screening test (e.g. able or unable to tolerate endometrial Pipelle® biopsy without \nlocal anaesthesia) was undertaken prior to LA-TBEA in order to minimise potential bias in \npatient selection and maintain the pragmatic nature of the study.  \n \nIntervention LA-TBEA was undertaken in our ambulatory gynaecological clinic according \nto our previously described treatment protocol 64. Essential elements of the protocol include:- \nTiming of TBEA: Ablation was performed at any time during the menstrual cycle and without \nany prior endometrial preparation.  \nPre-medication: All women received diclofenac 100mg rectally, oral co-dydramol 10/500 \n(two tablets) and oral cyclizine 50 mg. Tramadol hydrochloride 100mg was used if non \nsteroidal analgesia was contra-indicated.  \n Conscious patient: no intravenous cannulation was present.  There was no use of \nsedation. \n Local anaesthetic: The cervix was directly injected in a circumferential manner with \nthree 2mL cartridges containing 3% prilocaine hydrochloride (30 mg/mL) and felypressin \n0.03 unit/mL (\ncitanest with octapressin®, Dentslply, UK) using a 27G dental syringe.  \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n129 \n \n Dedicated patient nurse: A particular nurse was allocated to provide continuous \nsupportive care to the patient during the procedure. The nurse engaged the patient in \nconversation (‗distraction‘ analgesia effect termed ―vocal local‖) and often held the patient‘s \nhand throughout the procedure.  \n Pre-ablation hysteroscopy: All women underwent an outpatient hysteroscopy check \nprior to LA-TBEA. An endometrial biopsy had usually been carried out prior to the \nscheduled TBEA. A zero degree microhysteroscope with a 2.5-mm rigid outer sheath (Karl \nStorz, Tuttlingen, Germany) was used. Between 10-100mL of Normal Saline via a nurse \ncontrolled syringe was used as intrauterine distension medium. Any significantly sized \nintrauterine polyps (greater than 2cm in size) were excised by either blind polyp forcep \navulsion or Versapoint [Gynecare, Ethicon Inc., Somerville, NJ, USA]) resection prior to LA-\nTBEA (Table 2.12). Women were excluded from the study if there were significantly sized \nuterine fibroids (fibroids greater than 3cm size), enlarged uterine size (uterine cavity length \ngreater than 10cm), abnormally shaped uterine cavity (e.g. bicornuate uterus), endometrial \nhyperplasia or cancer, or active pelvic infection.  \n Type of Thermachoice device: Thermachoice I (February 2001-July 2003) and \nThermachoice III (August 2003-August 2005) devices were used. \n Intrauterine ablation pressure: The manufacturer recommends this is maintained \nbetween 160mmHg and 180mmHg.  However at the discretion of the operator, the upper \nlimit of pressure was controlled in manner so that it did not exceed 195mmHg. This was \nconsistently applied in both Thermachoice I and III groups (Table 2.13). \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n130 \n \nTable 2.12. Baseline demographic data for outpatient TBEA \n  Thermachoice \nI    (N=51) \nThermachoice \nIII (N=51) \nThermachoice \nI and III \nN=102 \nP-value \n(Thermachoice \nI vs. III)* \n     \nMean Age   (95% CI; St Dev)  43.4 (41.9-44.8;4.7) 44.1 (42.4-45.7;5.6) 43.7 (42.6-44.8;5.2) 0.118 \nMean BMI  (95% CI; St Dev) 28.8 (26.5-31.1;7.4) 30.3 (28.0-32.6;7.8) 29.6 (28.0-31.2;7.6) 0.318 \nIndication for Ablation     \nMenorrhagia alone  43 46 89 0.373 \nMenorrhagia & severe dysmenorrhoea 8 5 13  \nCycle phase^    0.167 \nProliferative 18  11 29  \nMid-cycle 11  15 26  \nSecretory  17  25 42  \nUterine axis^    0.539 \nAnteverted 20 36 56  \nAxial 2 8 10  \nRetroverted 5 7 12  \nUltrasound scan findings    0.029 \nNormal 34 45 79  \nPolyp 5 1 6  \nFibroid $ 12 5 17  \nHysteroscopy findings     \nNormal 44 44 88 0.020 \nPolyp 0 5 5  \nFibroid $ 7 2 9  \nMedian uterine size cm \nAnd (Range) \n8.0 \n(7-10) \n8.0 \n(7-13) \n8.0 \n(7-13) \n0.964 \n \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n131 \n \nFootnotes \n* Statistical tests include Chi-square, Mann-Whitney U \n^ Data not reported in all cases, calculation based on cases that were reported \n$ Fibroid corresponds to identification of any submucous, intramural or subserosal fibroids \nby either ultrasound or hysteroscopy that are less than 3cm in size \nThe comparisons in BOLD are those that are statistically significant with a P value <0.05 .  \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n132 \n \nTable 2.13. Peri-procedure outcomes of outpatient TBEA \nFootnotes \n* Statistical tests include Chi-square, Chi-Square trend, Mann-Whitney U \n^ Data not reported in all cases, calculation based on cases that were reported \n$ Excluding 13/102 cases that stayed overnight, the mean duration of hospital stay (minutes) with 95% \nConfidence limits are: 294 [257-330], 364 [326-402] and 329 [303-357; standard deviation 120] for \nThermachoice I, III and overall combined I and III respectively.  \nThe comparisons in BOLD are those that are statistically significant with a P value <0.05 .  \n Thermachoice \nI \nThermachoice \nIII \nThermachoice \nI+III \nP-value \n(Thermachoice \nI vs. III)* \n     \nMean Volume of fluid in mL    \n(95% CI; SD) \n24.1  \n(16.5-31.8;13.3) \n \n19.2 \n(15.8-22.7 ; 11.6) \n20.4 \n(17.3-23.5 ; 12.1) \n0.07 \nAverage Intrauterine pressure  \n(95% CI ; SD ) \n157 \n(147-166 ; 16) \n169 \n(164-176 ; 19) \n167 \n(162-172; 19) \n0.004 \nMean hospital stay minutes (hours)$ \n(95% CI; SD in minutes) \n(95% CI; SD in hours) \n433 (7.2h) \n(318-547; 382) \n(5.3-9.1; 6.4) \n522 (8.7h) \n(405-639; 398) \n(6.8-10.7; 6.6) \n478 (8.0h) \n(397-559; 391) \n(6.6-9.3; 6.5) \n0.277 \nMean Visual Analogue Pain  \n(95% CI; SD) \n5.6 \n(4.7-6.6; 1.6) \n5.9 \n(5.1-6.8; 2.9) \n5.8 \n(5.2-6.5; 2.7) \n0.541 \nRescue Analgesia ^     \nParacetamol  (frequency) 32 25 57 0.412 \nDiclofenac (frequency) 3 0 3 0.074 \nMean Morphine Equivalent Dose (mg)$$ \n (95% CI; SD) \n4.8 \n(3.0-6.5; 6.2) \n13.0 \n(10.7-15.3; 8.0) \n8.9  \n(7.3-10.5; 8.3) \n0.001 \nOverall strength of analgesia $ \nNo rescue analgesia \nMild \nModerate \nStrong \n \n12 \n30 \n4 \n5 \n \n3 \n25 \n6 \n17 \n \n15 \n55 \n10 \n22 \n \n0.005 \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n133 \n \nPost ablation day case bed stay: All women recovered in a day case bed and were allowed \nhome after a minimum stay of 2 hours. A strict protocol of post-procedure pain relief was \nadhered to: supplementary analgesia (termed Rescue Analgesia) was provided according to \nthe patient‘s pain relief scores and patient request from a standardised ‗as required‘ written \ndrug prescription. A patient information leaflet was provided detailing expected symptoms \nand analgesic advice post LA-TBEA. All women were contacted by telephone at home the \nfollowing day to check on their progress. \nStrength of rescue analgesia. To quantify the amount of rescue analgesia utilised a numerical \n(morphine equivalent dose) and ordinal (mild, moderate, severe) scale was created according \nto the following methods:   \na) Morphine Equivalent Dose. This is the estimated bioequivalent dose of morphine \nsulphate (in milligrams) that corresponds to the oral analgesic preparations (such as codeine \nphosphate, dihydrocodeine) given according to an accepted validated conversion scale 71. \nb) An ordinal ranking of mild, moderate, severe rescue analgesia. This scale was created \nby the study authors, and recorded as mild (paracetamol or diclofenac only), moderate \n(paracetamol and diclofenac or low morphine equivalent dose) or strong (paracetamol \n/diclofenac / morphine, or high morphine [>15mg] dose) grading for strength of analgesia \nusage.  \n \nOutcome measures Initial baseline data recorded were: age, body mass index, menorrhagia \nalone or combined with dysmenorrhoea, menstrual cycle phase, ultrasound and hysteroscopy \nfindings, uterine axis and uterine size. Procedure-related data recorded were: types of \nThermachoice device, total fluid volume used, mean intrauterine ablation pressure, \ncompletion of TBEA procedure and any complications (e.g. vasovagal episodes).  \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n134 \n \nAll women were asked to record the pain they experienced immediately following LA-TBEA \non a graduated Visual Analogue Scale, ranging from 0 (no pain) to 10 (worst imaginable \npain), which had been validated in our previous study 64. Amount and type of patient-initiated \npost procedure analgesia and duration of hospital stay (from initial admission to hospital and \nactual discharge) were also recorded.  \nA postal questionnaire was sent to all women post LA-TBEA to determine the effectiveness \nof therapy between July-September 2006. Questionnaire response was maximised by re-\ncontacting women (by phone and letter) with non-returned forms in accordance with accepted \npractice 80. Patient completed data recorded were: menstrual improvement (amenorrhoea, \nlighter periods), menstrual worsening or no change; satisfaction with treatment result or \ndissatisfaction; need for secondary treatment and type (e.g. LNG-IUS, repeat TBEA or \nhysterectomy); usage of HRT; Menorrhagia-specific and generic quality of life measures. \nBoth menorrhagia disease-specific (Shaw) 81 and generic (EuroQol-5D) 82 Quality of Life \ntools were utilised to improve the sensitivity and accuracy in determining this outcome; both \nthese tools had been validated in previous related studies 83;84. The clinical case records of \nwomen undergoing hysterectomy secondary treatment were accessed to determine uterine \nhistology. Similarly, the case records for women with missing questionnaires were accessed \nto determine if any secondary treatment had been necessary.  \n  \nStatistical analysis All statistical analysis was performed using SPSS 13.0 statistical \nsoftware (release 1 Sept 2004, ©SPSS Inc., USA). Categorical data was analysed by Chi-\nsquare and Chi-square trend testing. Continuous data was analysed by Mann-Whitney U test. \nA P-value less than 0.05 was considered statistically significant.  Multivariate regression \n(binary logistic, ordinal and linear) was used to explore the significance of various baseline \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n135 \n \nand procedure related factors (i.e. prognostic factors or independent variables) on key \noutcome measures (duration of hospital stay, satisfaction, amenorrhoea and quality of life i.e. \ndependent variables). We accept the risk of increased overall Type I error (the probability of \nincorrectly rejecting a null hypothesis) when performing multiple hypothesis tests in \nmultivariate regression. The Bonferroni method lowers the observed significance level \nbecause of multiple testing and provides a method to achieve an overall study error rate of \n0.05 using a corrected p-value derived by 1- (1-α)1/n, where α=0.05 and n=number of \nhypothesis tests. However, given this was an exploratory statistical analysis, rather than a \nformal confirmatory study, then correcting for multiple testing procedures is not always \nconsidered necessary 85;86. We have therefore reported both uncorrected and Bonferroni \ncorrected P-values to enable readers to interpret the true significance of any p<0.05 result in \nline with other factors (e.g. consistency of finding, biological plausibility and clinical \nrelevance) 85;86. \n \nEthics A formal application to a Research Ethics Committee was made and they \nrecommended that ethics approval was not required as the study was classified as service \nevaluation according to established Central Office for Research Ethics Committees (COREC) \nguidelines. The study was conducted in accordance with basic ethical principles and \ncomplying with the Data Protection Act 2000 (e.g. informed consent, maintaining patient \nconfidentiality, anonymizing patient held data, secure electronic storage of data).  \n \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n136 \n \nRESULTS \nBaseline and peri-procedure outcomes There were no significant differences in the \nbaseline characteristics between Thermachoice I (n=51) and Thermachoice III (n=51) cohorts \napart from differences in ultrasound and hysteroscopic findings (Table 2.12).  Of 105 \nconsecutively recruited women that underwent planned LA-TBEA, the procedure was \nsuccessfully completed in 102 (97%). Of the 3 failures, two were technical failures (one \nequipment breakdown, one severe cervical stenosis), and due to abandonment of the LA-\nTBEA at 3 minutes due to severe patient discomfort. These 3 failures all occurred in the first \n10 cases of Thermachoice I cohort and may therefore be related to a learning curve effect of \nthe operator and nursing team (i.e. develop better patient reassurance and analgesic regimes). \nOther complications included: 1 case of severe vasovagal syncope (not requiring atropine), 1 \ncase of endometritis, 3 cases of severe vomiting, and 9 cases of severe pain requiring \novernight admission. Fully completed questionnaires were returned by 88/102 participants \n(86%), and partially completed in a further 7 participants (95/102; 93%).  \n \nAll peri-procedure outcomes are depicted in Table 2.13. The overall (n=102) mean duration \nof hospital stay following Thermachoice I and III was 8.0 hours (95% CI  6.6 to 9.3 hours; \nStandard Deviation 6.5 hours) [Table 2.13].  However, this mean has been skewed due to the \ninclusion of a small proportion of women (n=13/102; 12.8%) who required overnight \nadmission. Exclusion of this subgroup (6/51 Thermachoice I, 7/51 Thermachoice III) leads to \nan overall mean duration of stay of 5.5 hours (95% CI 5.1 to 6.0 hours; Standard Deviation \n2.0 hours) [Table 2.13]. Thus, outpatient LA-TBEA was successfully completed as an \nambulatory day case (under 6 hours hospital stay) procedure in the vast majority.  \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n137 \n \nThe amount of morphine rescue analgesia used directly correlated to the post ablation VAS \nscore and duration of hospital stay (Figure 2.4). Univariate analysis showed that \nThermachoice III was associated with greater use of rescue analgesia, but was also performed \nat higher mean intrauterine pressure than Thermachoice I (Table 2.13). \n \nLong term outcomes: Table 2.14 shows the long term outcomes in Thermachoice I \nand III procedures (performed between February 2001-July 2003, follow up range 26-54 \nmonths and August 2003-August 2005, follow up range 12-29 months respectively).  Overall, \ndespite the majority of women reporting improvement in their menstrual symptoms \n(amenorrhoea 29%, lighter periods 55%; total 84%), not all of these were satisfied (overall \nsatisfaction rate 78%).  Further treatment (repeat TBEA, hysterectomy or LNG-IUS) was \nrequired in 19/102 cases (19%). Of the 14 hysterectomies performed as secondary treatment \naround two-thirds had adenomyosis or fibroids on uterine histopathology.  Satisfied \ncompared to dissatisfied women reported higher levels of quality of life and menstrual \nimprovement (Table 2.15). Overall, 50% of surgical re-interventions (n=16, 14 \nhysterectomies, 2 repeat ablations) occurred by 19 months (Range 10-46 )(Figure 2.5).  \nKaplan-Meier survival analysis suggested a statistically significant trend to earlier surgical \nre-intervention  with Thermachoice III than Thermachoice I (Log Rank Mantel-Cox  \np=0.024). However, Cox regression showed that this was a non-significant (p=0.056) trend \nwhen corrected for identified confounders (duration of follow up; intrauterine pressures; \nmorphine equivalent dose) (Figure 2.5).  \n \n \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n138 \n \nFigure 2.4. Correlation of morphine usage to post ablation VAS Score and duration of \nhospital stay \n1086420\nVAS pain immediate post TBEA\n50\n40\n30\n20\n10\n0\nmorphine usage\nR Sq Cubic =0.258\n \nFootnote: Morphine rescue analgesia directly correlates to VAS score immediately post TBEA \n(Pearson p=0.001, Kendall‘s tau p=0.006). \n150010005000\nDuration of post procedure hospital stay (mins)\n50\n40\n30\n20\n10\n0\nmorphine usage\nR Sq Cubic =0.245\n \nFootnote: Morphine rescue analgesia directly correlates to duration of hospital stay post TBEA \n(Pearson p=0.001, Kendall‘s tau p=0.001), and this relationship remains statistically significant after \nmultivariate analysis (see Figure 2.5). \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n139 \n \nTable 2.14. Long-term outcomes of outpatient TBEA \n Thermachoice \nI \nThermachoice \nIII \nThermachoice \nI+III \nP-value \n(Thermachoice \nI vs. III)* \nMean follow up time (months) \n(95% CI ; SD) \n41 \n(38-43; 8) \n18 \n(16-19; 5) \n29 \n(26-32; 13) \n0.001 \nFurther treatment (Repeat ablation, \nHysterectomy or LNG-IUS)  \nNo \nYes \n \n \n36 (70.6%) \n15 (29.4%) \n \n \n47 (92.2%) \n4   (7.8%) \n \n \n83 (81%) \n19 (19%) \n \n \n0.005 \nAll Types of further treatment \nNo further treatment \nLNG-IUS \nDrugs (including HRT) \nRepeat Endometrial Ablation \nHysterectomy ($$ histology) \n \n30 \n3 \n6 \n1 \n11 \n \n44 \n0 \n3 \n1 \n3 \n \n74 (73%) \n3 \n9 \n2 \n14 (14%) \n \n0.024 \nPeriods at review ^ \nAmenorrhoea \nLighter \nNo change or worse \n \n11 (23%) \n23 (49%) \n13 (28%) \n \n16 (35%) \n28 (61%) \n2 (4%) \n \n27 (29%) \n51 (55%) \n15 (16%) \n \n \n0.009 \nDysmenorrhoea at review ^ \nPain free or Less \nNo change \nWorsening \n \n27 \n6 \n14 \n \n37 \n5 \n4 \n \n64 \n11 \n18 \n \n0.027 \nSatisfaction \nSatisfied \nDissatisfied \n \n35(69%) \n16 (31%) \n \n43 (84%) \n8   (16%) \n \n78 (76%) \n24 (24%) \n \n0.062 \nMean EuroQoL VAS score (95% CI; SD) 76 (69-84; 19) 80 (73-87; 19) 78 (73-83; 19) 0.420 \nMean EuroQoL Index (95% CI; SD) 0.81(0.74-0.88; \n0.19) \n0.87 \n(0.80-0.96; 0.21) \n0.84 \n(0.79-0.90; 0.20) \n0.022 \nMean Shaw QOL (95% CI; SD) 83 (73-92; 25) 87 (78-96; 24) 84 (78-91; 25) 0.504 \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n140 \n \nFootnotes \n* Statistical tests include Chi-square, Chi-Square trend, Mann-Whitney U \n^ Data not reported in all cases, calculation based on cases that were reported \n$$ Histology of the 14 hysterectomies reported adenomyosis, fibroids and normal uterus in 4, \n5 and 5 cases respectively \nThe comparisons in BOLD are those that are statistically significant with a P value <0.05 .  \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n141 \n \nTable 2.15. Patient satisfaction and its relationship to quality of life and other treatment \noutcomes following endometrial ablation \nOutcomes Satisfied \nN=78 \nDissatisfied \nN=24 \nP Value \nStatistical significance $ \n \nNo further treatment \nFurther treatment \nLNG-IUS \nDrugs (including HRT) \nRepeat Endometrial Ablation \nHysterectomy  \n \n69 \n9 \n0 \n9 \n0 \n0 \n \n5 \n19 \n3 \n0 \n2 \n14 \n \n0.001 \nPeriods now \nAmenorrhoea \nLighter  \nNo change or worsening \n \n25 \n43 \n1 \n \n2 \n8 \n14 \n \n0.001 \nQuality of life    \nMean EuroQOL VAS score  \n(95% CI ; SD)    \n80.5 \n(75.8-85.2, 16.8) \n63.1 \n(35.5-84.5, 26.5) \n0.077 \nMean EuroQoL Index            \n(95% CI ; SD)    \n0.89 \n(0.86-0.92, 0.13) \n0.50 \n(0.22-0.79, 0.31) \n0.001 \nMean Shaw QOL          \n(95% CI ; SD)     \n87.0 \n(80.6-93.3, 22.6) \n68.3 \n(37.6-99.0, 33.2) \n0.009 \n \nFootnotes \n* Statistical tests include Chi-square, Mann-Whitney U \nThe comparisons in BOLD are those that are statistically significant with a P value <0.05.  \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n142 \n \nFigure 2.5. Survival analysis for likelihood of surgical re-intervention post TBEA \n Mean (95% CI; SD) Median (Range) P value \ncomparison \nAll surgical re-\ninterventions (n=16) \n21.6 (15.8-27.5; 11.0) 19.0 (10-46) \n \nNot applicable \nHysterectomy(n=14)  \nvs.  \nRepeat ablation (n=2) \n20.8 (14.5-27.1) \nvs. \n27.5 \n18.5 (10-46) \nvs. \n27.5 (18-37) \n*0.721 \nThermachoice I (n= 12) \nvs.  \nThermachoice III (n=4) \n24.3 (17.9-30.8)  \nvs.  \n13.5 (10.5-16.5) \n \n23.0 (18.0-28.0)  \nvs. \n12.0 (10.0 vs. 14.0) \n*0.024 \n \n** 0.056 \n*    Log Rank (Mantel-Cox) \n** Cox Regression analysis (corrected for duration of follow up; intrauterine pressures; \nmorphine equivalent dose) \nTime (months) from original endometrial \nablation\n40.0030.0020.0010.000.00\nCumulative probabilty of \nsurgical re-intervention\n1.0\n0.8\n0.6\n0.4\n0.2\n0.0\nThermachoice III\nThermachoice I\n \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n143 \n \nRegression analysis for prognostic factors  \n \nUnivariate analysis showed that Thermachoice III compared to Thermachoice I was more \nlikely to be associated with primary treatment success, menstrual improvement, \ndysmenorrhoea improvement and improved generic Quality of Life (EuroQol-5D index) \n(Table 2.14). However, multivariate regression analysis (Table 2.16; corrected for all \nbaseline and peri-procedure characteristics that utilized a P value less than 0.05 as indicative \nof statistical significance) showed:- \n Morphine dosage in rescue analgesia, but not the overall strength of analgesia \n(combining non-steroidal, Paracetamol and opiates) was independently associated \nwith a longer duration of hospital stay. \n Thermachoice III compared to Thermachoice I increased the likelihood for \namenorrhoea, but was not associated with increased hospital stay, satisfaction or \nquality of life. \n Regardless of the type of Thermachoice device, higher mean intrauterine ablation \npressures and/or higher morphine rescue analgesia correlated to better long term \npatient satisfaction.  \n Neither uterine axis, age nor uterine length was associated with any of the outcomes. \nAs previously stated in our methods section, there is an increased risk of identifying a falsely \npositive statistical finding due to multiple testing. For the analysis shown in Table 2.16 the \nBonferroni corrected p value for a significant factor is 0.001. This meant that only type of \nThermachoice would be considered statistically significant (p=0.001) amongst all factors \ntested if using Bonferroni correction. \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n144 \n \nTable 2.16. Prognostic outcomes for endometrial ablation (using multivariate regression \nanalysis) \n \nIndicator variable Multivariate statistical P values  \nof indicator variable in predicting outcome \nDuration of \nHospital \nStay$ \nAmenorrhoea~ Satisfaction* EuroQoL \nIndex $ \nAge 0.604 0.571 0.078 0.675 \nBMI 0.825 0.322 0.778 0.489 \nIndication (prior dysmenorrhoea) 0.925 0.526 0.315 0.487 \nCycle phase 0.958 0.982 0.522 0.274 \nUterine axis 0.614 0.234 0.146 0.942 \nUltrasound findings 0.754 0.267 0.169 0.331 \nHysteroscopy findings 0.479 0.742 0.593 0.864 \nUterine size 0.155 0.342 0.918 0.415 \nIntrauterine pressure 0.956 0.350 0.014 0.070 \nVolume of fluid 0.402 0.425 0.682 0.621 \nPost procedure pain VAS 0.792 0.335 0.369 0.064 \nStrength of analgesia 0.820 0.791 0.486 0.205 \nMorphine dose 0.042 0.595 0.030 0.394 \nType of Thermachoice 0.591 0.001 0.697 0.387 \nIs model statistically significant No Yes Yes No \nFootnotes \n* Binary Logistic Regression; $ Univariate Linear Regression ; ~ Ordinal Logistic Regression \nThe Bonferroni method lowers the observed significance level because of multiple testing and \nprovides a method to achieve an overall study error rate of 0.05 using a corrected p-value derived by \n1- (1-α)1/n, where α=0.05 and n=number of hypothesis tests.  For the entire table (14 by 4 tests, giving \nn=56 and α=0.05) the Bonferroni corrected p value that is 0.001.This means that only type of \nThermachoice is statistically significant (p=0.001) if using Bonferroni corrected interpretation.  \n \nThe comparisons in BOLD are those that are statistically significant with a P value <0.05.  \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n145 \n \nDISCUSSION \nLocal anaesthetic TBEA can be carried out as an outpatient daycase procedure and is an \neffective treatment option. The vast majority (76%) were satisfied with their treatment at a \nmean 2½ years follow up. Upgrading to Thermachoice III, compared to Thermachoice I was \nassociated with improved rates of amenorrhoea, although overall, both devices achieved \nsimilar rates of patient satisfaction and quality of life. We found higher intrauterine ablation \npressures to be associated with improved long term treatment satisfaction. Overall, 50% of \nsurgical re-interventions occurred by around 1½ years.  \nThis study is an important advancement to the published knowledge in outpatient TBEA 64-\n68;87.  This study‘s principal attribute is that it is of pragmatic design and reflects actual \nclinical management of menorrhagia. We believe this study to be the largest published cohort \nof outpatient TBEA under local anaesthetic without sedation. Apart from two other studies of \n4-6 year follow up 67;88, this study represents the longest follow up of outpatient TBEAs \n(mean of 30 months, range of 12 to 54 months). Importantly, this study is the first to utilize \nboth menorrhagia-specific 81 and generic 82 quality of life tools which has been advocated as \nthe preferred way to measure these outcomes 83;84. Furthermore, our study‘s long term \noutcomes are derived from a high response rate (86%) which improves the accuracy of our \ndata collection.  \n \nWe agree there may be caveats when interpreting the results from our prospective study, \nwhich may lessen the reliability of our conclusions. Our study population may be \nheterogeneous, as we did not use objective criteria to define heavy menstrual bleeding or \nstratify according to differing bleeding patterns. Because of temporal differences between \nThermachoice I and III cohorts, this has inevitably led to differences in follow up between \nThermachoice I (mean 41 months) and III cohorts (mean 18 months) at our time of \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n146 \n \nquestionnaire enquiry. This precluded the reliability of any survival regression analytical \ntechniques, although by opting to use regression analysis we have attempted to correct for \nconfounding influences. Intrauterine polyp or fibroid removal may have exerted an \nindependent curative effect, although their combined proportions were similar in both \nThermachoice I and III cohorts and thus any distinguishing influence minimized. Our study \ndid not collect baseline quality of life data, and therefore we were unable to quantify a change \nin quality of life following LA-TBEA at specified time intervals. Finally, our study is likely \nto be under-powered (Type II error). Based on unpaired student‘s T-test and a minimal \ndifference of interest between means of 0.3 Standard Deviations, we estimate a sample of size \nof 178 for each group would be required to show any significant difference in quality of life \noutcomes.  \nThe mean duration of stay in our study was around 5.5 hours, which is significantly longer \nthan an analogous outpatient Thermachoice LA-TBEA study that reported a mean total time \nspent in hospital of 1 hour 40 minutes 65. This discrepancy may be partly explained by \ndifferences in pre- and post-operative analgesia and nursing-led or physician-led discharge \npractices. However, duration of stay is likely to be even more multifactorial than this (see \nTable 2.16), and any attempt to explain such differences would be frank conjecture.  We feel \nthat a mean duration of stay shown by our study represents a  realistic recovery period before \ndischarge. \nFor some outcome measures, the extent of incomplete questionnaire responses or prolonged \ntime interval from original treatment may have had a greater effect in over-estimating or \nunder-estimating their frequency. This may be particularly pertinent to our reported rates of \namenorrhoea; if we assume those with missing responses were not truly amenorrhoeic then \nthe rates of amenorrhoea for Thermachoice I and III would be 23% and 31% instead of the \nreported 23% and 35%. However, our rates of amenorrhoea are consistent to those reported \n\nChapter 2.4 Outpatient Thermachoice long term outcomes \n147 \n \nby other studies 66-68;88. Nonetheless, we accept there may be a tendency to under-report \nsatisfaction and amenorrhoea rates in Thermachoice I because assessment at longer follow up \nmay have enabled women to have regeneration of the endometrial lining and symptomatic \nrecurrence. Several studies have explored prognostic factors on TBEA success 68;76-79. Unlike \nprevious studies, we have shown no adverse prognostic effects due to a retroverted uterus, \nlarge uterine size or young age 68;76-79. However, it is notable that our association of increased \nintrauterine pressure and improved outcome has been consistently identified in other studies \n76-79.  Furthermore, this study showed associations for amenorrhoea (influenced by type of \nThermachoice), satisfaction (influenced by intrauterine pressure, rescue morphine usage \nand/or post procedure VAS) and Quality of Life (influenced by intrauterine pressure) \noutcomes. This would reinforce the logical notion that the higher the intrauterine ablation \npressure, and/or the more painful the TBEA procedure is, perhaps by inducing a greater depth \nuterine ablation, the more likely it is to achieve a successful long term outcome. \n \nThis study reports on the safety and effectiveness of outpatient LA-TBEA which is clinically \nrelevant to improve patient selection and preoperative counselling. Furthermore, there is \ncontinued expansion in this area, as evidenced by a growing body of literature which includes \na randomised trial comparing outpatient and general anaesthetic TBEA 65. In order to \ndetermine the optimum role for outpatient endometrial ablation in treating women with heavy \nperiods, further trials are needed to determine the clinical and cost-effectiveness of second \ngeneration ablation techniques (e.g. microwave, TBEA and radiofrequency ablation devices) \nagainst each other, against general anaesthetic and local anaesthetic settings, and against \nappropriate treatment alternatives (e.g. LNG-IUS). \n \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n148 \n \n2.5. Long term outcomes following hysteroscopic myomectomy for \nabnormal uterine bleeding  \n \n \nOBJECTIVE: To evaluate the long term effectiveness of hysteroscopic submucous myomectomy for \nwomen with abnormal uterine bleeding and explore any prognostic factors associated with treatment \nsuccess. \nDESIGN: Prospective observational study. \nSETTING: University teaching hospital in U.K. \nPATIENT(S): 92 women symptomatic of abnormal uterine bleeding with submucous myomas. \nINTERVENTION(S): Hysteroscopic myomectomy performed as outpatient local anaesthetic (38%) \nor daycase general anaesthesia (62%) using VersapointTM  . \nMAIN OUTCOME MEASURES: Need for secondary surgical or medical re-intervention, \nmenstrual improvement and patient satisfaction over a minimum 12 month period. Other outcome \nmeasures include: successful completion of primary resection, type of secondary treatment and any \nprognostic factors. \nRESULT(S):  Mean follow up was 2.6 years (95% CI 2.3-2.9). Complete fibroid excision and \nremoval was achieved in 66%. Secondary surgical re-intervention was required in 27 (29%) of which \n11 (12%) were repeat hysteroscopic myomectomy and 10 (11%) were hysterectomy procedures. \nMultiple uterine fibroids and adenomyosis were identified in 80% of hysterectomies. Overall, \nimproved menstrual symptoms and patient satisfaction were reported by 91% and 86% at follow up. \nSize of the submucous fibroid or presence of any intramural or subserosal fibroids were not related to \ntreatment success.  \nCONCLUSION(S): Women with abnormal uterine bleeding diagnosed with submucous myomas \nmay be successfully treated by removing the submucous myoma component, irrespective of co-\nexistent intramural or subserosal fibroids. This effect is sustained over the long term. \n \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n149 \n \nINTRODUCTION \nUterine fibroids are present in 25-40% of women presenting with abnormal uterine bleeding \n89. Although a direct cause-effect relationship has not been completely established, there is \nsufficient observational data to suggest that shrinkage or removal of any identified uterine \nfibroids is beneficial in alleviating menstrual bleeding abnormalities in most symptomatic \nwomen. \nHysteroscopic myomectomy is considered the first-line conservative surgical therapy for the \nmanagement of symptomatic submucous fibroids89-93. Data, from mainly observational \nstudies, has suggested beneficial effects in treating both menstrual abnormalities and \ninfertility with this procedure. The few studies that have reported on long term outcomes for \nfibroid-related menstrual abnormalities, indicate that hysteroscopic myomectomy is \nassociated with a 10-35% risk of surgical re-intervention, including repeat myomectomy, \nopen myomectomy or hysterectomy90;92;93. However, such a high re-intervention rate may \nalter the cost effectiveness of hysteroscopic myomectomy compared to other uterus-\nconserving treatment options and hysterectomy. \nPresently, there is insufficient evidence on reliable selection criteria and long term outcomes \nfor women with symptomatic fibroids who opt for hysteroscopic myomectomy. This \nknowledge would be particularly important for preoperative counselling and appropriate \npatient selection. We therefore wished to evaluate long term efficacy of this treatment, and \nidentify whether there were any adverse (e.g. co-existence of intramural or subserosal \nfibroids) or favourable (e.g. submucous myoma less than 5cm size, completeness of lesion \nexcision) peri-operative prognostic factors. \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n150 \n \nMATERIALS AND METHODS: \nPatient Population Women symptomatic of abnormal uterine bleeding (i.e. mainly with \nheavy menstrual bleeding [HMB]) were referred by primary care (GPs) or secondary care to \nour ―One Stop‖ ―See and Treat‖ menstrual disorders clinic. All women underwent \ntransvaginal pelvic sonography, outpatient hysteroscopy and endometrial Pipelle biopsy \n(Laboratoire C.C.D, Paris, France) investigations. Women who were considered suitable for \nhysteroscopic myomectomy were included in this study. Women were excluded from the \nstudy if an abnormally shaped uterine cavity (e.g. bicornuate uterus), endometrial \nhyperplasia, cancer or active pelvic infection were present. \nStudy design Prospective patients presenting between June 2003 and November 2006 were \nincluded in this study.  \nIntervention All hysteroscopic myomectomies were performed using Versapoint [Gynecare, \nEthicon Inc. USA] according to the manufacturer‘s recommended guidance and as previously \nreported by our group 94;95. We defined a submucous intracavity fibroid at hysteroscopy as \nhaving characteristic appearances (sessile or pedunculated, superficial large blood vessels) \nand non-mobility with intrauterine fluid or gentle hysteroscopic tapping of the lesion. In all \ncases, our preoperative suspicion of intracavity fibroid was confirmed on histological analysis \nof the excised lesion.  All women were offered to have the intervention under local \nanaesthetic (LA) outpatient setting or general anaesthetic (GA) daycase setting. Factors that \ninfluenced the final decision included: patient preference, how she tolerated outpatient \nhysteroscopy, intrauterine location and size of intracavity fibroids. Preoperative preparation \nwith a 3 month course of GnRHa prior to myomectomy was deemed necessary in women \nwith intracavity fibroids greater than 5cm in size. A patient information leaflet was provided \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n151 \n \ndetailing the procedure, expected symptoms and analgesic advice post hysteroscopic \nmyomectomy. \nLA hysteroscopic myomectomy This was performed on a ―See and Treat‖ basis with no \nfasting prior to the procedure. Other elements of the treatment included:- \n Local anaesthetic: The cervix was directly injected in a circumferential manner with \nthree 2mL cartridges containing 3% prilocaine hydrochloride (30 mg/mL) and felypressin \n0.03 unit/mL (Dentsply, UK) using a 27G dental syringe.  \n Dedicated patient nurse: A particular nurse was allocated to provide continuous \nsupportive care to the patient during the procedure. The nurse engaged the patient in \nconversation (‗distraction‘ analgesia effect termed ―vocal local‖) and often held the patient‘s \nhand throughout the procedure.  \n Post procedure analgesic regimen: All women received diclofenac 100mg rectally \nand oral co-dydramol 10/500 (two tablets). All women were recovered in a dedicated patient \nwaiting area and allowed home after a minimum 30 minute stay. A strict protocol of post-\nprocedure pain relief was adhered to.  \nGA hysteroscopic myomectomy:   Women, fasted for at least 6 hours, were admitted to \nhospital on the day of the procedure. In a minority of cases, women with high risk medical \ndisorders (e.g. diabetes) were admitted the day before the planned procedure.  Hysteroscopic \nmyomectomy was carried out in gynaecology theatres after induction of general anaesthesia.  \nAll women received diclofenac 100mg and 1g paracetamol rectally (or paracetamol alone if \ndiclofenac was contraindicated) just prior to the procedure. Infiltration of the cervix with a \nlocal anaesthetic was not done in these women.  The hysteroscopic myomectomy surgical \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n152 \n \nprocedure, post-procedure analgesia regimen and day case bed stay for GA was identical to \nthe LA hysteroscopic myomectomy procedure described above. \nComplete excision, partial excision and devascularisation at hysteroscopic myomectomy \nIn all cases, a standardised technique was adopted in order to completely excise and remove \nthe fibroid. The submucous fibroid was resected at the junction between the fibroid and \nuterine wall using a shearing technique. To facilitate this it was occasionally necessary to \nbisect, trisect or quadrisect the fibroid lesion to access this fibroid-uterine wall interface.  \nComplete excision was achievable in most pedunculated (Type 0) and in those superficially \nmyometrially invading (type 1) intracavity fibroids. Occasionally, where the hysteroscopic \nview became obscured following commencement of the procedure, one of two modified \nprocedures was performed: \n Partial excision and removal of the fibroid was performed. The percentage of the fibroid \nremoved relative to entire intracavity lesion was clinically estimated and recorded. \n Devascularisation of the intracavity without its excision. This entailed multiple scoring of \nthe fibroid lesion (e.g. trisecting the lesion in a ―hot cross bun‖ technique) at or near its \nvascular attachment base. The percentage of the fibroid devascularised relative to the \nentire intracavity lesion was clinically estimated and recorded. \n \nOutcome measures  Initial baseline data recorded were: age, body mass index, parity, \nmenstrual bleeding abnormality, ultrasound and hysteroscopy findings, and use of pre-\nprocedure GnRHa. The size of the intracavity uterine fibroid selected for myomectomy was \ndetermined using ultrasound (objective) data in most cases. Where ultrasound had failed to \nidentify the intracavity fibroid prior to myomectomy the practitioner recorded their clinical \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n153 \n \nestimate of intracavity fibroid size. In relationship to the hysteroscopic myomectomy \nprocedure, the following data were recorded: LA or GA setting, completeness of excision, \noperation length, procedure related complications (e.g. vasovagal episodes) and duration of \nhospital stay. \nA postal questionnaire was sent to all women post procedure between June-November 2007, \nensuring there was a minimum 12 month follow up period. Questionnaire response was \nmaximised by re-contacting women (by phone and letter) with non-returned forms. Patient \ncompleted data recorded were: need for and nature of any secondary treatment, improvement \nin their menstrual bleeding pattern and dysmenorrhoea (ordinal Likert scales), and patient \nsatisfaction at that time (ordinal Likert scale). Secondary treatments were categorised \naccording to medical (LNG-IUS, oral progestins, combined oral contraceptive, tranexamic \nacid) and surgical (repeat hysteroscopic myomectomy, open myomectomy, endometrial \nablation, hysterectomy) interventions. Primary treatment success was defined as the absence \nof any type of medical or surgical secondary treatment following the primary treatment of \nhysteroscopic myomectomy. The case records and histology results of all study participants \nwere reviewed and recorded. \n \nStatistical analysis: Dichotomous data were presented as simple proportions.  SPSS version \n13 was used to undertake multivariate regression analysis and to conduct Chi-square tests.  P \n< 0.05 was considered statistically significant.   \n \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n154 \n \nRESULTS \nNinety-two women participated in the study and their baseline characteristics are depicted in \nTable 2.17.  The characteristics associated with hysteroscopic myomectomy procedure are \ndepicted in Table 2.18. Of the 35 (38%) women undergoing LA procedure, none were \nadmitted for overnight stay. Of the 57 (62%) women undergoing GA procedure, 20 (35%) \nwere admitted for overnight stay. \nIn relation to menstrual symptom improvement and patient satisfaction outcomes, only 2 \nwomen (2%) failed to return their questionnaire, representing a 98% follow-up. Examination \nof the clinical case notes and contacting their GPs confirmed that neither of these two women \nhad undergone secondary treatment following hysteroscopic myomectomy. The mean follow \nup was 2.6 years (95% CI 2.3-2.9; Range 1-7.3 years; St Dev 1.5).  Overall, greater than, or \nequal to, 12 months, 24 months and 36 months outcome data were available for 90 (98%), 52 \n(57% ) and 31 (34%) women.  \nThe menstrual and secondary treatment outcomes are depicted in Table 2.19. Surgical re-\nintervention was necessary in 27 (29%) women, and this involved hysterectomy in 10 cases \nand their characteristics are depicted in Table 2.20. Seven hysterectomies (70%) were \nperformed by 12 months of the primary hysteroscopic myomectomy, and of these, 2 \nhysterectomies were performed for unexpectedly identified gynaecological pathology (one \ncase complex hyperplasia, one case leiomyosarcoma). Adenomyosis and multiple fibroids \nwere the commonest histological findings at hysterectomy (8/10 cases).  Multivariate analysis \nof the need for secondary treatment identified no statistically significant prognostic factor \n(Table 2.21).   \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n155 \n \nTable 2.17. Baseline characteristics for 92 women undergoing hysteroscopic \nmyomectomy \n \nPatient characteristics N=92 \nFrequency (Percentage) \nAge \n20-30  years \n30-40 years \n40-50 years \n>50  years \n \n4 (4) \n33 (36) \n42 (46) \n13 (14) \nBMI Mean 28.0 (95% CI 26.4-29.7) \nRange 20-52; St Dev 6.9 \nMenopausal status at presentation \nPre-menopausal \nPost-menopausal \n \n84 (91) \n8 (9) \nMenstrual Bleeding abnormality \nHeavy Menstrual Bleeding (HMB) \nUnscheduled bleeding on HRT \n \n84 (91) \n8   (9) \nScan findings \nSubmucous \nSubmucous & Intramural \nSubmucous & Intramural & Subserosal \n \n \n41 (45) \n47 (51) \n4   (5) \n \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n156 \n \nTable 2.18. Characteristics associated with hysteroscopic myomectomy procedure \n \nProcedure setting \nLA Local anaesthetic outpatient \nGA General anaesthetic daycase  \n \n35  (38) \n57  (62) \nPreop GnRHa \nYes \nNo \n \n20 (22) \n72 (78) \nLength of operation* \n<30 minutes \n30-60 minutes \n> 60 minutes \n \n77 (84) \n14 (16) \n1 (1) \nSize of uterine fibroid (u/s and by clinical estimation) \n<3cm \n3-5 cm \n>5cm \n \n22 (24) \n53 (58) \n17 (19) \nPrimary treatment performed \nComplete excision and removal \n>50% excision and removal \n>50% devascularised and left in situ \nComplete excision and removal and endometrial ablation \nComplete excision and removal and insertion of Mirena \n \n48 (52) \n13 (14) \n18 (20) \n11 (12) \n2 (2) \nComplications \nNone \nBleeding requiring balloon tamponade \nCervical trauma \n \n83 (90) \n8 (9) \n1 (1) \nLength of hospital stay \nDaycase \nOvernight \n \n72 (78) \n20 (22) \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n157 \n \nTable 2.19 Outcomes after hysteroscopic myomectomy \nFootnotes for Table 2.19          * Missing questionnaire responses for menstrual (2) and \ndysmenorrhoea (2) characteristics and patient satisfaction (2). Case notes and GPs were \ncontacted and no secondary treatments were undertaken in the 2 non-returned questionnaire \nresponses. \nOutcome measure  Entire cohort, \nincluding 10 women \nwho had \nhysterectomy (n=92) \nExcluding 10 \nwomen who had \nhysterectomy \n(n=82) \nMenstrual bleeding characteristics at enquiry \nAmenorrhoea \nBrown discharge \nMuch lighter \nMarginally lighter \nNo change \nHeavier  \nUnknown \nOverall menstrual symptoms improved \n \n28 (30) \n3 (3) \n40 (43) \n13 (14) \n4 (4) \n2 (2) \n2 (2) \n84 (91) \n \n18 (22) \n3 (4) \n40 (49) \n13 (16) \n4 (5) \n2 (3) \n2 (2) \n74 (90) \nDysmenorrhoea characteristics at enquiry  \nNone \nLess \nNo change \nWorse \nUnknown \nOverall dysmenorrhoea symptoms improved \n \n50 (54) \n26 (28) \n11 (12) \n3 (3) \n2 (2) \n76 (83) \n \n40 (49) \n26 (31) \n11 (13) \n3 (4) \n2 (2) \n66 (81) \nDegree of satisfaction at enquiry \nVery satisfied  \nSatisfied \nDissatisfied  \nVery Dissatisfied  \nUnknown \nOverall satisfied \n \n55 (60) \n24 (26) \n6 (7)  \n5 (5)  \n2 (2) \n79 (86) \n \n54 (66) \n20  (24) \n4 (5) \n2 (2) \n2 (2) \n74 (90) \nSecondary treatment \nMyomectomy (open) \nThermal Balloon Endometrial Ablation \nLNG-IUS (Mirena) \nHysterectomy \nRepeat hysteroscopic myomectomy \nRepeat hysteroscopic myomectomy and ablation \nRepeat hysteroscopic myomectomy and  Mirena \nOral progestins \nNo  secondary treatment \nSecondary treatment (all types) required \nNo surgical re-intervention \nOverall repeat surgical treatment required \n \n4 (4) \n2 (2) \n7 (8) \n10 (11) \n8 (9) \n1(1) \n2 (2) \n2 (2) \n56/92 (61) \n36/92 (39) \n65/92 (71) \n27/92 (29) \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n158 \n \nTable 2.20 Women (n=10) undergoing hysterectomy following hysteroscopic \nmyomectomy \n \nCharacteristic Value \nNumber of  hysterectomies 10 \nAverage time to Hysterectomy Mean 14.4  months (95% CI 4.5-24.3) \nMedian 9.5 months; Range 1-41 months; St Dev 13.9 \n \nTime from procedure and cumulative \nrate of hysterectomy \n \nBy 6 months:  4/10 cases   [one for leiomyosarcoma] \nBy 12 months: 7/10 cases  [one for complex hyperplasia] \nBy 24 months: 8/10 cases \nBy 48 months: 10/10 cases  \nVery satisfied \nSatisfied \nDissatisfied \nVery dissatisfied \nOverall satisfied \nOverall dissatisfied \n1 \n4  \n2 \n3  \n5 (50%) \n5 (50%)  \nHistology  \nAdenomyosis and fibroids (multiple) \nFibroids (multiple) \nLeiomyosarcoma \nComplex Hyperplasia \n \n6 \n2  \n1 (identified on resection histology and reason for TAH) \n1 (identified at resection histology and reason for TAH) \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n159 \n \nTable 2.21. Multivariate analysis of prognostic factors \n \n Need for secondary treatment \n (all types) \nNeed for secondary \nsurgical re-intervention \nPrognostic factor \nand its  \nsignificance \n(p-value) \nMenstrual pattern  \n \nPrimary treatment  \n \nScan findings        \n \nSize of fibroid  \n[p=0.90]  \n \n[p=0.10]  \n \n[p=0.61]  \n \n[p=0.35] \nMenstrual pattern  \n \nPrimary treatment  \n \nScan findings \n \nSize of fibroid \n \n[p=0.09]  \n \n[p=0.12]  \n \n[p=0.66]  \n \n[p=0.84] \n \n \n \nFootnotes \n \n1. Multivariate regression corrected for the following confounding factors, including: age, \nBMI, parity, menopausal status, type of menstrual bleeding abnormality, scan findings, \npreoperative GnRHa, size of uterine fibroid, type of primary myomectomy treatment. \n \n2. Binary logisitic regression models for secondary treatment and secondary surgical re-\nintervention were statistically significant (p<0.001) \n \n \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n160 \n \nDISCUSSION \nWomen with abnormal uterine bleeding diagnosed with submucous myomas may be \nsuccessfully treated by removing the submucous myoma component, irrespective of co-\nexistent intramural or subserosal fibroids or size of fibroid that has been resected. The \nbeneficial effects of hysteroscopic myomectomy persist long term (mean follow up around \n2½ years), and with the secondary surgical re-intervention rate of 29% this suggests that the \nremoval of the submucous component can avoid hysterectomy in 70% of cases.  The majority \nof women who underwent hysterectomy as secondary treatment were identified to have \nadenomyosis and multiple uterine fibroids.  \n \nTo date,  studies published on hysteroscopic myomectomy have utilised various technical \napproaches, been mainly performed under GA in daycase settings, have mixed retrospective \nand prospective observational designs, and have minimal data on long term follow up, \nparticularly patient satisfaction and surgical re-intervention rates94;96-104. Our study adds to \nthis published literature by exclusively utilising a modern Versapoint bipolar system; has \nbeen successfully undertaken in both outpatient and daycase setting; has a prospective design; \nhas long term follow up incorporating patient satisfaction; has evaluated peri-operative \nfeatures that may have prognostic value; and expands on our previously published work94. \nOur surgical re-intervention rate of 29% (over mean 2½ years) was lower than that reported \nby a previous study of 35% (over mean 5 years)101. This study has been pragmatic in design, \nensuring our results are applicable to actual clinical practice.  \n\nChapter 2.5. Hysteroscopic myomectomy: long term outcomes \n161 \n \nHowever, we accept our study may have limitations that may make our conclusions less \nreliable. Our study sample size, although at 92 with a low dropout rate (2%), may be \nunderpowered to identify all peri-operative prognostic factors. Because our follow up \nintervals varied between patients, there may be a tendency to overestimate or underestimate \nthe beneficial effects of hysteroscopic myomectomy at the extremes of follow up. The \nvariation in follow up outcome data also precluded our use of survival analysis techniques to \nasses both efficacy and durability of the hysteroscopic procedure.  \n \nGiven the 29% risk of surgical re-intervention following submucous myomectomy, there is a \nneed to identify significant peri-operative prognostic factors that could be usefully employed \nduring preoperative counselling. Even though our study did not identify any specific \nprognostic factor previous studies have identified enlarged uterine size, three or more \nintracavity myomas, fibroid size>3cm and increased depth of myometrial penetration to be \nadverse prognostic factors 100;101 .  In fact, our study reinforces the widely held opinion that it \nis only the presence of the submucous fibroid itself that appears to be responsible for the \nheavy menstrual bleeding 92;93. Furthermore, our study showed that adenomyosis was \nfrequently identified in those women who required hysterectomy as secondary treatment. \nThere is insufficient evidence on the ultrasonographic criteria predictive of adenomyosis and \nwhether adenomyosis should be routinely screened for in women with menstrual \ndisorders105;106. Future studies are needed to identify the clinical efficacy and optimal patient \nselection criteria for submucous myomectomy, and whether preoperative imaging suspicion \nof adenomyosis may be usefully employed in the treatment decision making process.  \n \n\nChapter 3  Systematic reviews \n162 \n \nChapter 3. SYSTEMATIC REVIEWS AND SYSTEMATIC LITERATURE \nAPPRAISAL \n \nSystematic reviews performed for two clinical queries: \n Do screening-preventative interventions in asymptomatic pregnancies reduce the risk \nof preterm delivery. \n Is Levonorgestrel-releasing hormone system (LNG-IUS) effective therapy for a \nvariety of non-contraceptive indications . \n \nPublications arising from the chapter \nChapter Manuscript title Reference \n3.1 Do screening-preventative interventions in asymptomatic \npregnancies reduce the risk of preterm delivery--a critical appraisal \nof the literature. \n \nAntibiotic treatment of bacterial vaginosis in pregnancy: Multiple \nmeta-analyses and dilemmas in interpretation \n1 \n \n \n \n2 \n3.2 Non-contraceptive uses of levonorgestrel releasing hormone system \n(LNG-IUS)- a systematic enquiry and overview \n3 \n \n\nChapter 3  Systematic reviews \n163 \n \nIntroduction \n \nBackground \nSystematic reviews are considered to provide the definitive evidence-based answer as to \nwhether a particular intervention or test is clinically effective and to quantify the strength of \nthis effect. The term ―systematic review‖ is widely considered to be synonymous to the \nhighest level of evidence-based medicine available for that cited topic. The key strength of \nthe research methodology is its ability to produce a more reliable measure of effectiveness \nthrough mathematically pooling outcomes of clinical trials rather than using an outcome \nascertained from an individual trial. The methodology is further underpinned through a \nrigorous systematic search, with strict quality control of studies that are eligible to be \nincluded or excluded in the final meta-analysis stage. \n \nThe methodology conforms to established standards which are, by convention, explicity \nstated prior to the systematic review being accepted by peer-reviewed publications or the \nCochrane collaboration4;5.  Hence, systematic reviews, analyzing the same clinical question, \nought to be consistent and reproducible. Importantly, inherent to the transparency of the \nmethods and trial selection process, systematic reviews are relatively easily updated as newer \ntrials are published; the process of periodic update is mandatory for all Cochrane reviews and \nensures the review evidence is continually up-to-date and reliable.  Despite a multitude of \npublished systematic reviews, mostly presented through the Cochrane collaboration, there \nremains several unanswered clinical questions within our specialty of obstetrics and \ngynaecology.  \n\nChapter 3  Systematic reviews \n164 \n \nAims  \nUndertake a systematic review, incorporating the elements of : systematic literature search ; \nappraisal of studies for rejection or inclusion and meta-analyses. This will be performed in \naccordance to standardized methodology as set out by the Cochrane collaboration and \nothers4;5.  \nFor each therapeutic intervention, appraise the quality of supporting evidence and assign a \ngrade to the strength of recommendation that can be derived according to the evidence by \nusing established evidence appraisal tools (Royal College of Obstetricians and \nGynaecologists guideline development criteria (Table 3i  and GRADE evaluation of \nevidence Table 3ii) 6;7.  These appraisal tools will be discussed at greater length in Chapter \n4 (Clinical Guideline Development). \n \nThe specific clinical queries that will be used as examples of the systematic review research \nmethodology process are: \n1. Do screening-preventative interventions in asymptomatic pregnancies reduce the risk \nof preterm delivery. \n2. Is Levonorgestrel-releasing hormone system (LNG-IUS) effective therapy for a \nvariety of non-contraceptive indications . \n\nChapter 3  Systematic reviews \n165 \n \n Table 3i.  Classification of evidence used by RCOG Guideline development (originate \nfrom US Agency for Health Care Research and Quality)  \nClassification of Evidence Levels  \nIa Evidence obtained from meta-analysis of randomised controlled trials.  \nIb Evidence obtained from at least one randomised controlled trial.  \nIIa Evidence obtained from at least one well-designed controlled study without \nrandomisation.  \nIIb Evidence obtained from at least one other type of well-designed quasi-experimental \nstudy.  \nIII Evidence obtained from well-designed non-experimental descriptive studies, such as \ncomparative studies, correlation studies and case studies.  \nIV Evidence obtained from expert committee reports or opinions and/or clinical \nexperience of respected authorities.  \nGrades of Recommendations  \n \nRequires at least one randomised controlled trial as part of a body of literature of \noverall good quality and consistency addressing the specific recommendation. \n(Evidence levels Ia, Ib) \n \nRequires the availability of well controlled clinical studies but no randomised \nclinical trials on the topic of recommendations. (Evidence levels IIa, IIb, III) \n \nRequires evidence obtained from expert committee reports or opinions and/or \nclinical experiences of respected authorities. Indicates an absence of directly \napplicable clinical studies of good quality. (Evidence level IV) \nGood Practice Point    \n \nRecommended best practice based on the clinical experience of the guideline  \ndevelopment group \n \n \n\nChapter 3  Systematic reviews \n166 \n \nTable 3ii. GRADE approach (http://www.gradeworkinggroup.org/index.htm)  \nThe Grading of Recommendations Assessment, Development and Evaluation (GRADE) \nGRADE: Quality of evidence \nThe GRADE system classifies the quality of evidence in one of four levels: \nHigh quality— Further research is very unlikely to change our confidence in the estimate of \neffect \nModerate quality— Further research is likely to have an important impact on our confidence in \nthe estimate of effect and may change the estimate \nLow quality— Further research is very likely to have an important impact on our confidence in \nthe estimate of effect and is likely to change the estimate \nVery low quality— Any estimate of effect is very uncertain \nEvidence based on randomised controlled trials begins as high quality evidence, but our \nconfidence in the evidence may be decreased for several reasons, including:  \nStudy limitations  \nInconsistency of results  \nIndirectness of evidence  \nImprecision  \nReporting bias.  \nAlthough observational studies (for example, cohort and case-control studies) start with a \"low \nquality\" rating, grading upwards may be warranted if the magnitude of the treatment effect is \nvery large, if there is evidence of a dose-response relation or if all plausible biases would \ndecrease the magnitude of an apparent treatment effect.  \nGRADE: Strength of recommendation \nThe GRADE system offers two grades of recommendations: \"strong\" and \"weak\" depending \non whether effects of intervention clearly outweigh the undesirable effects, or clearly do not. If \ntrade-offs are less certain—either because of low quality evidence or because evidence \nsuggests that desirable and undesirable effects are closely balanced—weak recommendations \nbecome mandatory. \nFactors that affect the strength of a recommendation  \nFactor Examples of strong \nrecommendations \nExamples of weak \nrecommendations \nQuality of \nevidence \nMany high quality randomised \ntrials have shown the benefit of \ninhaled steroids in asthma \nOnly case series have examined the \nutility of pleurodesis in \npneumothorax \nUncertainty about \nthe balance \nAspirin in myocardial infarction \nreduces mortality with minimal \nWarfarin in low risk patients with \natrial fibrillation results in small \n\nChapter 3  Systematic reviews \n167 \n \nbetween desirable \nand undesirable \neffects \ntoxicity, inconvenience, and cost stroke reduction but increased \nbleeding risk and substantial \ninconvenience \nUncertainty or \nvariability in \nvalues and \npreferences \nYoung patients with lymphoma \nwill invariably place a higher \nvalue on the life prolonging effects \nof chemotherapy than on treatment \ntoxicity \nOlder patients with lymphoma may \nnot place a higher value on the life \nprolonging effects of chemotherapy \nthan on treatment toxicity \nUncertainty about \nwhether the \nintervention \nrepresents a wise \nuse of resources \nThe low cost of aspirin as \nprophylaxis against stroke in \npatients with transient ischemic \nattacks \nThe high cost of clopidogrel and of \ncombination dipyridamole and \naspirin as prophylaxis against stroke \nin patients with transient ischaemic \nattacks \n\nChapter  3.1: Prevention of preterm delivery \n 168 \n3.1. Do screening-preventative interventions in asymptomatic pregnancies \nreduce the risk of preterm delivery: a systematic review and meta-analysis \n \nBackground: Recent research has suggested that women who experience preterm delivery (PTD) \nmay be identified earlier in pregnancy and before onset of symptoms.   Interventions commenced at \nthis earlier asymptomatic stage may offer an opportunity to prevent PTD or lengthen gestation \nsufficiently to reduce adverse perinatal outcome.     \nObjectives: To examine the evidence that supports or refutes interventions that prevent PTD. To \nexamine whether interventions are effective in all women or only women at high risk of PTD. To \ngenerate clinical recommendations for each intervention according to evidence appraisal tools 6;7.   \nMethods: A systematic search, meta-analysis and evidence-based appraisal of the identified literature.  \nResults: There is evidence that introducing screening-preventative strategies for asymptomatic \npregnancies may reduce the rate of PTD.   Evidence for screening and selective treatment exists for : \nasymptomatic bacteriuria in all women; bacterial vaginosis in low-risk population groups; elective \ncervical cerclage in high-risk pregnancies; indicated cervical cerclage in women with short cervical \nlength on ultrasound; prophylactic progesterone supplementation in high-risk pregnancies; Smoking \ncessation in all women. However, for most other strategies, such as increased antenatal attendance, or \nroutine administration of prophylactic micronutrients, the evidence is inconsistent and conflicting.  \nConclusion: The review suggests an antenatal management plan designed to prevent PTD based on \ncurrent practice and the evidence identified. Data on neonatal outcomes apart from PTD (such as \nserious neonatal morbidity and mortality) were lacking in most studies.  It was therefore not possible \nto establish whether preventing PTD or prolonging gestation would correlate to improved perinatal \noutcome. This lessened the potential clinical usefulness of any proposed preventative strategy. \nFurthermore, no studies were found that evaluated the effectiveness of combining screening-\npreventative strategies.  \n\nChapter  3.1: Prevention of preterm delivery \n 169 \nINTRODUCTION  \nPreterm delivery (PTD) is defined in the UK as delivery after 24 completed weeks‘ gestation \nand before the onset of 37 weeks‘ gestation.   In the United States, the lower limit of PTD is \n20 weeks, which is used for all US Perinatal statistics. PTD affects 6%-15% of deliveries and \nrepresents a major worldwide health concern 8.  PTD has multifactorial aetiology. The causes \nand subgroups associated with PTD include : spontaneous preterm labour (PTL), 31-50%; \nmultiple pregnancy, 12-28%; preterm premature rupture of membranes (PPROM), 6%-40%; \nmedically indicated (e.g. hypertensive disorders of pregnancy, intrauterine growth restriction, \nantepartum haemorrhage and chorioamnionitis), 20%-25%; miscellaneous (cervical \nincompetence, uterine malformation), 10% 9. Some clinicians believe that cervical \ninsufficiency and PPROM/chorioamnionitis have moreover similar origins and have \ncombined such subgroups when reporting studies of PTD. PTD accounts for 50%-70% of all \nneonatal morbidity and mortality.   Importantly, the earlier the gestation at delivery, the \ngreater the risk of adverse perinatal outcome (Table 3.1) 10.  Consequently, there is a need to \nprevent PTD and any proposed strategy should ideally aim to target PTDs that occur before \n34 weeks gestation, as this group contributes most to perinatal morbidity and mortality.    \nTable 3.1.   Gestation-specific perinatal mortality  \nGestational age (weeks’) Survival (%) \n22-24 5-40 \n25-27 55-75 \n28-30 80-85 \n31-33 95-100 \n34-36 100 \n \n\nChapter  3.1: Prevention of preterm delivery \n 170 \nIn trying to reduce rates of PTD, the emphasis has been on applying screening-preventative \ninterventions to women symptomatic of PTL or PPROM.   However, these have had limited \nefficacy 11.   Recent research has suggested that women who experience PTD, PTL and \nPPROM may be identified earlier in pregnancy and before onset of symptoms12;13. It has thus \nbeen suggested that prophylactic and therapeutic interventions commenced at an earlier \nasymptomatic stage of pregnancy, either as specific measures or general measures, may offer \nan opportunity to prevent PTD or lengthen gestation sufficiently to reduce adverse perinatal \noutcome.  We have therefore conducted a systematic search and critical appraisal of the \nliterature to identify the evidence that supports or refutes this approach to reducing the rate of \nPTD and related perinatal morbidity and mortality.  In particular, this review considers health \napproaches that address all risk factors that affect the entire population of pregnant women, \nas well as those screening-preventative strategies directed only at high-risk asymptomatic \nwomen.   The review concludes with a suggested an antenatal management plan designed to \nprevent PTD based on current practice and the evidence presented in this article. \nMETHODS An electronic search of MEDLINE (1966- October 2005), EMBASE (1980-\nOctober 2005), and the Cochrane library (2005) was conducted using combinations of \nprinciple MeSH terms and text words: ―preterm labour‖, ―preterm birth‖, ―preterm labor‖, \n―labor, premature‖, ―infant mortality‖, ―infant premature‖, ―infant, premature, diseases‖, \n―cerclage, cervix‖, ―cervical incompetence‖, ―vaginosis, bacterial‖, ―fibronectins‖, \n―glucocorticoids‖ and ―tocolysis‖.   The reference lists of all known primary, review and \nclinical evidence-based guidelines were also examined to identify cited articles not captured \nby electronic searches.   Articles cited frequently were used in the Science Citation Index to \nidentify additional citations. A meta-analysis was conducted to examine whether differences \nin outcome occur when the intervention is applied to high or low risk of PTD study \npopulations.\n\nChapter  3.1: Prevention of preterm delivery \n 171 \nDEFINITIONS   \nSeveral studies, albeit based on varying population groups and competing risks, have \nconsistently shown that women with a previous history of PTD, PPROM, medically indicated \nPTD, delivery of a small-for-gestational-age infant, second trimester pregnancy loss, \ncongenital uterine anomalies, or suspected cervical incompetence are at increased risk of \nsubsequent PTD 14-17.   A selection of these and other risk factors for PTD, with supporting \nodds ratios is depicted in Table 3.2.  Associations for a particular risk factor are supported \nwith a single reference citation of a high quality study.   \nUnless indicated by the presence of another risk factor,  ‗high-risk‘ groups are defined as \nthose asymptomatic pregnancies who are deemed to be at increased risk of PTD because of \nexperiencing previous PTD.  The pregnancy is asymptomatic if symptoms of PPROM, PTL \nare absent and there are no overt manifestations of obstetric complications (e.g. multiple \npregnancy, hypertension, antepartum haemorrhage).  The review is structured by considering \nscreening-preventative interventions that may be commenced following routine antenatal care \nor antenatal care combined with specialist investigations, and then elaborate on the evidence \nfor the value of strategies in specific high-risk groups as well as population wide health \nstrategies.  \n\nChapter  3.1: Prevention of preterm delivery \n 172 \nTable 3.2.   Risk factors associated with increased risk of preterm delivery.    \nRISK FACTOR Preterm delivery \nunder study \nOdds Ratio Reference \nRoutine Screening    \nWomen aged <18 years at delivery \ncompared to 18-34 years \n<32 weeks‘ 1.41 (1.02-1.90) 18 \nSecond birth in women aged 15-19 years \ncompared to 20-29 years \n24-32 weeks‘ 2.5 (1.5-4.3) 19 \nPrevious singleton PTD < 35 weeks‘ \ncompared to >35 weeks‘ \n<35 weeks‘ 5.6 (4.5-7.0) 20 \nBody mass index<20 <37 weeks‘ 3.96 (2.61-7.09) 21 \nElevated (>90th centile) maternal serum \nalpha- fetoprotein (AFP) \n<35 weeks‘ 3.9 (1.7-8.7) 22 \nSingletons following in vitro fertilisation <37 weeks‘ 2.0 (1.7-2.2) 23 \nGenital bleeding below 24 weeks‘ <37 weeks‘ 2.5 (1.6-3.8) 15 \nPlacenta praevia 24-27 weeks‘ 2.90 (2.46-3.42) 24 \nLoop electrosurgical excision of cervix \n(matched for smoking status) \n<37 weeks‘ 2.53 (1.42-4.49) 25 \nUrinary tract infection <37 weeks‘ 4.4 (1.47-13.34) 21 \nShort inter-pregnancy interval     \n(<6 months) \n24-32 weeks‘ 4.1 (3.2 -5.3) 26 \nEthnicity- Black Afro-Caribbean \n                 Asian \n         vs.   White Europeans (UK Study) \n<37 weeks‘ 1.33 (1.15-1.56) \n1.45 (1.33-1.56) \n27 \nAlcohol (>7 drinks/week)  <32 weeks‘ 3.26 (0.8-13.24) 28 \nSmoking  27-32 weeks‘ 1.7 (1.3-2.2) 29 \nSpecialist Screening    \nBacterial vaginosis< 16 weeks‘ <37 weeks‘ 7.6 (1.8-31.7) 30 \nPositive cervico-vaginal fetal fibronectin  <35 weeks‘ 6.6 (1.7-25.5)  22 \nCervical length ≤25mm <35 weeks‘ 3.9 (1.7-9.2) 22 \nBilateral uterine artery notching <37 weeks‘ 2.38 (1.19-4.75) 31 \nSerum granulocyte colony-stimulating \nfactor (>75th centile) \n<35 weeks‘ 3.1 (1.4-6.9) 22 \n\nChapter  3.1: Prevention of preterm delivery \n 173 \nEvidence for the value of screening-preventative interventions on routine \nantenatal population screening \n \nThe components of routine antenatal care will vary according to country and local resources.   \nWe have used UK‘s NICE guideline as a model for recommended routine antenatal care \npractice 32. \n1. Early pregnancy booking and ultrasound dating (10-13 weeks) This provides an \nopportunity to accurately date the pregnancy, identify multiple pregnancies, and categorise \nthe pregnancy risk based on obstetric history and routine investigations.   There is no direct \nevidence that this care would decrease PTD.   \n2. Psychosocial, work and lifestyle factors There is epidemiological evidence that \nshows that PTD is associated with low maternal weight, poor weight gain during pregnancy, \nand low birth weight 33 (Table 3.2).   Two meta-analyses 34;35 have shown there is insufficient \nevidence of a beneficial reduction in PTD following increased psychosocial support and \nhome visits, preterm delivery education, bed rest, hydration, reducing excess manual labour \nand psychological stress, and ensuring that BMI is greater than 20 before conception.   \nSimilar results were found for interventions undertaken in both high-risk and low-risk \npregnancies 34;35.  \n3. Smoking and drugs avoidanceThe association between smoking or illicit drugs \n(such as heroin or cocaine) and adverse perinatal outcomes is well established.   A Cochrane \nmeta-analysis of 16 trials showed a reduction in low birthweight (RR 0.81, 95% CI 0.70 to \n0.94), a reduction in PTD (RR 0.84, 95% CI 0.72 to 0.98), and an increase in mean \nbirthweight of 33 g (95% CI 11 g to 55 g) with smoking cessation programs36.  Three non-\nrandomised comparative studies have shown improved neonatal outcomes (but not neonatal \n\nChapter  3.1: Prevention of preterm delivery \n 174 \nmortality) with antenatal drug avoidance programmes 37-39, with two of these studies \nsuggesting a modest reduction in PTD38;39. \n4. Screening and treatment of anaemiaThere is epidemiological evidence to support \nan association between low maternal hemoglobin concentration and low birth weight, as well \nas between low maternal haemoglobin concentration and PTD40;41.   However, a meta-\nanalysis 42, and two recent RCTs 43;44 have shown that supplementation of anaemic or non-\nanaemic pregnant women with iron, folic acid, or both, does not appear to increase either \nbirth weight or the duration of gestation.  \n5. Screening and treatment of asymptomatic bacteriuria A meta-analysis has shown \nthat antibiotic treatment in pregnant women with asymptomatic bacteriuria found on \nantenatal screening is effective in reducing the risk of pyelonephritis (OR 0.24; 95% CI 0.19 - \n0.32), and PTD or low birthweight (OR 0.60; 95% CI 0.45-0.80)45, and is thus advocated as \npart of routine antenatal care 32 . \n6. Elective prophylactic cervical cerclage for cervical incompetence  A history that \ncomprises any combination of: second trimester miscarriage, painless and progressive \ndilatation of the cervix, bulging membranes through the cervix prior to onset of labour, or \nprevious cervical surgery (e.g. cone biopsy), may suggest cervical incompetence and an \nincreased risk of PTD in the current pregnancy.   This information would normally be \nidentified through routine antenatal screening.   Presently, overall evidence suggests that \nelective cervical cerclage (defined in Table 3.3) compared to no cerclage or bed rest is likely \nto reduce the risk of PTD in women considered to be 'at very high-risk' of a second trimester \nmiscarriage due to a cervical factor 46-48‖.  There is no consensus on defining this ―very high-\nrisk‖ group, but subgroup analyses suggest this mainly comprises of women with three or \nmore prior preterm births or second trimester losses.  \n\nChapter  3.1: Prevention of preterm delivery \n 175 \nTable 3.3.   Defining elective and indicated types of cervical cerclage \nElective \ncerclage \nCerclage is performed before clinical or ultrasonographic evidence of \ncervical dilatation, using McDonald or Shirodkar techniques.   Usually \nperformed at 12-16w and based on reproductive history, or other criteria \nsuggestive of cervical incompetence.  Also termed as primary cerclage. \n \nIndicated   \n(emergency) \ncerclage \nCerclage is performed following clinical or ultrasonographic evidence of \ncervical dilatation, funneling or shortening.   Also uses McDonald or \nShirodkar techniques.   Theoretically may be performed at any preterm \ngestation below 32 weeks’, but most often undertaken at midtrimester (18-22 \nweeks’) period.  Also termed as secondary cerclage (if scan evidence of \ncervical dilatation) or tertiary cerclage (if clinical evidence of cervical \ndilatation) \n \nOf the three meta-analyses that have clearly distinguished between emergency and elective \ncervical cerclage, one 47 has shown no statistically significant reduction in rates of PTD, \nwhereas the other two meta-analyses 46;48have showed elective cervical cerclage to be \neffective at preventing PTD . Heterogeneity in defining the risk of PTD due to a ―cervical \nfactor‖ has contributed to differences in the findings of the meta-analyses. \n \nThe largest cervical cerclage trial 49 showed that elective cervical cerclage performed \nbetween 12-16 weeks gestation, in women at risk of cervical incompetence based on clinical \nhistory, reduced the risk of PTD (<34 weeks‘) but did not reduce perinatal mortality.   It \nfound that 24 women (95% CI 10-61) would need to undergo elective cervical cerclage to \nprevent one additional PTD before 34 weeks‘.   Importantly, a quasi-randomisation method \nwas adopted, which allowed clinicians to allocate cerclage or no cerclage according to the \nperceived risk of cervical incompetence and only when the clinician felt unsure if such \ncerclage would be beneficial or non-beneficial.   \n \n\nChapter  3.1: Prevention of preterm delivery \n 176 \nEvidence for screening-preventative interventions based on routine antenatal \ncare plus specialist investigations \n \n1. Microbiological screening and treatment of the genital tract  \n Bacterial vaginosis (BV) and trichomonas vaginalis (TV): Bacterial vaginosis (BV) \nis found in 9%-23% of pregnant women.   The presence of BV or trichomonas vaginalis (TV) \nin asymptomatic women in the second trimester is associated with PTD independent of other \nknown risk factors50-52.  Importantly, the earlier in gestation BV is detected, the greater is the \nrisk of an adverse outcome.   For example, BV at 26-32 weeks‘ is associated with PTD odds \nratio (OR) of 1.4 to 2  whereas BV at 7-16 weeks‘ carries an OR of 5 to 7.5 52.     \n \nThere is evidence that screening and treating BV in unselected low-risk population groups, \nrather than a heterogeneous combination of high-risk population groups, is effective at \nreducing the rate of PTD. When considering the combined screening of both low and high-\nrisk populations, six meta-analyses 53-58have shown that screening and treating asymptomatic \nBV, using either oral metronidazole (majority of trials) or vaginal/oral clindamycin, does not \nreduce the risk of PTD. However, our recently published meta-analysis (Figure 3.1) 58 has \nshown that screening and treating BV from a low-risk population does result in a statistically \nsignificant reduction in PTD (nine trials, RR 0.73; 95% CI 0.55-0.98).  \n\nChapter  3.1: Prevention of preterm delivery \n 177 \nFigure 3.1 Meta-analyses for treatment of bacterial vaginosis and reduction in preterm \ndelivery \nScreening and treatment of bacterial vaginosis in all population groups (both high and low \nrisk) and reduction in preterm delivery \n \nScreening and treatment of bacterial vaginosis in high-risk population and reduction in \npreterm delivery \n  \nScreening and treatment of bacterial vaginosis in low risk population and reduction in \npreterm delivery \n \n\nChapter  3.1: Prevention of preterm delivery \n 178 \nIn addition to pregnancy risk stratification, other factors contribute to heterogeneity of the \ntrials and methods used by the meta-analyses of screening and treating BV in pregnancy.  \nThese factors have been highlighted by two recent commentaries58;59 and include: antibiotic \ntypes, dosaging and gestation-specific efficacies; criteria for diagnosing BV; and unexplained \nhigh therapeutic responses observed from both placebo 60 and the screening process itself \nwithout any antibiotics being administered 61. A randomized trial showed that screening and \ntreating with metronidazole asymptomatic pregnant women for TV at 16 to 23 weeks‘ did not \nreduce PTD, and rather worryingly increased the risk of PTD (RR 1.8; 95% CI 1.2 to 2.7; \nP=0.004) 62.  This is the only trial included in the corresponding Cochrane systematic \nreview63. \n \n Chlamydia trachomatis: Chlamydia trachomatis is estimated to infect 2%-37% of \npregnant women.   Data from the NIH Preterm Prediction Study showed that women with \nChlamydia trachomatis infection at 24 weeks‘' gestation were twice as likely as uninfected \nwomen for PTD <37 weeks‘ (OR 2.2; 95% CI 1.03-4.78) and 3 times as likely to have PTD \n<35 weeks‘' gestation (OR, 3.2; 95% CI 1.08-9.57)64.   Only one trial 65 has examined \nscreening and treatment for Chlamydia to prevent PTD, and this showed no statistically \nsignificant reduction.   Information from on-going national opportunistic chlaymdia screening \nprogrammes may provide further evidence in this area.   \n \n Ureaplasma: Ureaplasma genital tract infection is associated with PTD and PPROM.   \nA Cochrane ‗meta-analysis‘ including only one trial, concluded there was insufficient \nevidence to show whether screening and treating women with ureaplasma in the vagina \nwould prevent PTD66. \n\nChapter  3.1: Prevention of preterm delivery \n 179 \n Group B streptococcus (GBS): Around 20% of pregnant women have Group B \nstreptococcus (GBS) urogenital colonization.   This is associated with an increased risk of \nurinary tract infection, PTL, PPROM, and infectious perinatal transmission67.   Only one trial \nwas identified, and this showed no reduction in PTD when pregnant women were routinely \nscreened and treated (using erythromycin) for GBS in the third trimester 68.   Based on this \ntrial, and other observational studies, both the RCOG and CDC have stated that routine \nscreening and antenatal treatment of women with GBS does not reduce the risk of PTD 67;69.  \nHowever, screening of high-risk pregnancy groups is recommended by the UK 67, and is \nuniversally undertaken during the third trimester in United States69 and Canada70.   This \nfacilitates the policy of prophylactic antibiotic treatment to carriers of GBS in labour, which \nhas been shown to reduce the incidence of GBS-related neonatal morbidity and mortality \n67;69. \n \n2. Screening for cervical length by ultrasound or clinically and subsequent \nindicated cervical cerclage There is considerable evidence to show that in the absence of \nuterine contractions transvaginal sonographic measurement of cervical length is an effective \nway of identifying pregnancies at high-risk of PTD, and has greater predictive value than \nother ultrasonographic measurements of the cervix such as dilatation of the internal os or \nfunneling of the internal os 71-78. A systematic review showed for asymptomatic women at or \nbelow 20 weeks gestation, a cervical length of 25 mm or less had a test positive likelihood \nratio of 6.29 (95% CI, 3.29-12.02) and negative test likelihood ratio of 0.79 (95% CI, 0.65-\n0.95) for predicting spontaneous PTD before 34 weeks gestation 76.    \nCervical cerclage may be performed electively (prophylactic procedure discussed earlier) or \nas an indicated (emergency) procedure (defined in Table 3.3) following clinical or \nultrasonographic evidence of cervical dilatation, funneling or shortening. \n\nChapter  3.1: Prevention of preterm delivery \n 180 \n Cerclage vs. no cerclage for short cervical length by ultrasound A meta-analysis has \nshowed no statistically significant effect of midtrimester indicated cerclage compared to no \ncerclage on the rates of PTD (four studies) or neonatal mortality (three studies) in women \nwith shortened cervical length on transvaginal ultrasound scanning 79. However, a meta-\nanalysis of four trials using individual patient-level data has shown that indicated \nmidtrimester cervical cerclage prevents PTD before 35 weeks in women with singleton \npregnancies and a short cervical length (RR 0.74, 95% CI 0.57-0.96), and this risk reduction \nis greater in singleton gestations with prior PTD or prior second-trimester loss 80. This meta-\nanalysis included two recently published trials 81;82 that had individually shown no beneficial \neffect of cerclage, as well as the CIPRACT trial 83which was the only trial to show any \nbeneficial effect of cerclage on preventing PTD.    \n Emergency cerclage in women with cervical incompetence on physical \nexamination\n In women with cervical incompetence on physical examination, with \nmembranes at or beyond a dilated external cervical os, a small RCT (n=23) showed that a \ncombination of emergency cerclage, bed rest, antibiotics and indomethacin was more \neffective at reducing PTD <34 weeks than bed rest with antibiotics alone 84.  \n\nChapter  3.1: Prevention of preterm delivery \n 181 \n3. Elective first trimester cerclage vs. cervical ultrasound surveillance and \nindicated emergency cerclage  Four retrospective comparative studies 85-88 have shown no \ndifference in obstetric outcomes with either strategy, whereas a recent prospective study \nshowed better outcome with cervical surveillance and indicated cerclage 89.   Reliable \ninterpretation of these studies, as well as comparison with the trials of cerclage (elective and \nindicated) discussed earlier, is markedly hampered due to variation in the definition and \nmagnitude of the risk in the population under study. Ultrasonographic cervical length, \ncombined with previous obstetric and reproductive history, has been successfully \nincorporated into a risk scoring system for predicting PTD 90. However, there is no evidence \nfrom any robust studies that indicates whether such a pregnancy risk stratification strategy \nfollowed by indicated cervical cerclage in those at most risk would reduce the rate of PTD.  \nThe accumulated evidence therefore suggests that a combination of assessment of risk \nfactors, obstetric history and serial follow-up of cervical length is more likely to identify the \ngroup of women who would benefit most from cervical cerclage. \n \n4. Positive fetal fibronectin testing followed by antibiotic treatment Fetal fibronectin \n(fFN) is a basement membrane protein produced by fetal membranes which functions as an \nadhesive factor of the placenta and membranes to the decidua.   It is normally present in \ncervical secretions until 16-20 weeks gestation.    Before testing for the presence of \ncervicovaginal fFN the following criteria must be met: intact amniotic membranes; minimal \ncervical dilatation (<3cm); sampling between 22 and 34 weeks gestation. A meta-analysis of \ncervicovaginal fFN testing in asymptomatic pregnancies to predict PTD before 34 weeks \nshowed a test positive likelihood ratio of 4.01, and a test negative likelihood ratio of 0.78 91.    \nSimilar results were found by another meta-analysis92, where for the prediction of outcomes \nof delivery <37 and <34 weeks‘' gestation, a positive fFN had overall sensitivity rates of 52% \n\nChapter  3.1: Prevention of preterm delivery \n 182 \nand 53%, and overall specificity rates of 85% and 89%, respectively.   For the outcomes of \ndelivery within 7, 14, and 21 days, the sensitivities were 71%, 67%, 59% and specificities \nwere 89%, 89% and 92% respectively. \n \nA positive midtrimester fFN test has been associated with an increased risk of subsequently \ndiagnosed maternal and fetal infection.    A primary analysis of the trial conducted by MFMU \nshowed that metronidazole plus erythromycin treatment of asymptomatic women with a \npositive midtrimester fFN (screened between 21 and 26 weeks‘) did not reduce the risk of \nPTD as hypothesized, but caused a non-statistically significant increase in PTD <37 weeks \nand <32 weeks‘93.   Furthermore, a subgroup analysis in women with previous PTD showed a \nstatistically significant increased risk of PTD when the treated group was compared to \nplacebo (46.7% versus 23.9%, P =0.039).  Whereas,  a secondary analysis of the MFMU \nstudy showed that women with both BV or TV and a positive fFN, who were treated with \nmetronidazole, had a non-significant reduction in spontaneous PTD from 14.6% to 8.3% 94. \n \nIf the detection of fFN does not alter the natural history of PTD through earlier antibiotic \ntreatment, could there still be a beneficial role for fFN testing in the asymptomatic ‗low‘ or \n‗high‘ risk woman? A positive fFN may have clinical benefit by lowering the threshold for \nadmission, or in utero transfer, or administering antenatal corticosteroids.  Conversely, a \nnegative fFN may have clinical value in avoiding unnecessary, costly and potentially harmful \ninterventions.  However, clinical trials examining improvements in perinatal outcomes \nfollowing such risk assessment with fFN were not identified in the literature.   \n\nChapter  3.1: Prevention of preterm delivery \n 183 \nEvidence for the value of screening-preventative strategies in specific high risk \ngroups \n \nMultiple Pregnancies Overall, there is a paucity of RCTs that have evaluated screening-\npreventative interventions in women with multiple pregnancies.   A retrospective study \nshowed no difference in perinatal outcomes between multiple weekly prophylactic \nadministration and single course antenatal corticosteroids in women with twin pregnancies 95.    \nProphylactic corticosteroids have no proven benefit in twin or higher order multiple \npregnancies, and may in fact be associated with increased harm such as decreased birthweight \nand increased risk of infection 96.   A short cervical length (less than or equal to 25mm), with \nor without funneling, at midtrimester screening is predictive for PTD in twin pregnancy, \nalbeit at lower sensitivity than when the same test is applied to singleton pregnancies97-100.  A \nmeta-analysis of trials using individual patient data 80showed a significant increase in PTD \n(RR 2.15, 95% CI 1.15-4.01) at less than 35 weeks when indicated cervical cerclage was \nperformed  in twin gestations with short cervical length.  However, a non-randomized \nprospective trial showed that indicated midtrimester cerclage in multiple pregnancies does \nnot alter the risks of PTD, PPROM or low birth weight101.   One retrospective study showed \nthat prophylactic elective cerclage in triplet and higher order multiple pregnancies had no \nbeneficial effect on obstetric or perinatal outcome102, although this was contradicted by \nanother retrospective study103. \n \n \n\nChapter  3.1: Prevention of preterm delivery \n 184 \nAntiphospholipid syndrome (APLS Antiphospholipid syndrome (APLS) in pregnancy is \ncharacterized by the presence of autoantibodies (anticardiolipin and/or lupus anticoagulant) in \nassociation with recurrent fetal loss, maternal thrombocytopenia and other pregnancy \ncomplications. Systemic lupus erythematosus (SLE), APLS, and thrombophilias have been \nassociated with similar pregnancy complications of early and late fetal loss, abruption, pre-\neclampsia and intrauterine growth restriction in three meta-analyses 104-106.  Evidence from \nobservational studies of rates of PTD in women with SLE, APLS, or thrombophilias is \nconflicting, and the analysis is complicated by complex co-morbidities of maternal disease \n(hypertension, renal impairment), fetal compromise (growth restriction), spontaneous PTL, \nPPROM and medically-indicated PTD. A retrospective study suggested that actively treated \nSLE (requiring prednisone or other immunosuppresants), or the presence of anticardiolipin \nantibodies, are predictive of a higher risk of PTD than inactive disease107; implying a \npotential beneficial role in suppressing active SLE disease in pregnancy to reduce the risk of \nPTD. \n \nSystematic reviews of therapeutic trials for treating APLS in pregnancy conclude that there is \ncurrently only weak evidence for a role of low dose aspirin and low-molecular-weight \nheparin in preventing adverse outcomes 108;109, despite this being the recommended treatment.  \nHowever, recent preliminary pilot studies suggests the beneficial effect of such prophylaxis \nhas been underestimated, and further research in this area is currently being actively pursued \n105;110;111.  Current consensus is that thrombophilia screening is recommended for women \nwith the following previous complications: fetal loss including three or more first trimester \nloss, two or more second trimester loss, or any stillbirth; early, severe or recurrent \npreeclampsia and severe intrauterine growth restriction 112. \n\nChapter  3.1: Prevention of preterm delivery \n 185 \nPre-eclampsia and uterine artery Doppler There is evidence that women with a previous \nhistory of pre-eclampsia-related PTD have a greater risk of pre-eclampsia-related PTD in a \nsubsequent pregnancy as compared with women with a previous PTD113;114.   A systematic \nreview showed that low dose aspirin (150mg) reduces the risk of perinatal death, pre-\neclampsia and PTD in women with a history of previous pre-eclampsia, and should therefore \nbe strongly advocated 115.   The reduction of recurrent pre-eclampsia and perinatal death was \ngreater in women with previous severe early-onset (second trimester) pre-eclampsia. There is \nevidence showing an association between impaired midtrimester uterine artery Doppler \nvelocimetry and/or uterine artery notching and subsequent pre-eclampsia116.  However, there \nare no data from any individual trial or meta-analysis demonstrating any direct reduction in \nPTD following low dose aspirin administration in women with impaired with impaired \nuterine artery Doppler characteristics that have been identified by either selective or \nunselective population Doppler screening117-121.  \nGestational Diabetes and Impaired Glucose Tolerance Overall, both gestational diabetes \n(GDM) and impaired glucose tolerance (IGT) affect 3%- 6% of pregnancies, and are \nassociated with PTD, PTL, PPROM, and numerous other pregnancy complications.   Once \nidentified, women are usually intensively managed with increased obstetric surveillance, \ndietary regulation, insulin therapy and instructed to maintain tight glycaemic control.   \nHowever, evidence to support this intensive treatment is lacking.   Cochrane meta-analyses \nhave concluded there is insufficient evidence to determine any beneficial or non-beneficial \neffect of dietary therapy, tight glycaemic control, or other treatments for GDM and IGT, upon \npregnancy outcomes122-124.   A non-randomised comparative study has suggested that \nuniversal glucose tolerance screening performed at the first antenatal visit compared to later \nscreening (24-28 weeks‘) resulted in a reduced risk of PTD and polyhydramnios 125.    \n\nChapter  3.1: Prevention of preterm delivery \n 186 \nEvidence for the value of population-wide preventative strategies in high and low \nrisk groups \n \nIncreased antenatal care and attendance There is conflicting opinion whether increased \nantenatal attendance reduces rates of PTD, and robust RCTs in this area are lacking.   \nNonetheless, lack of antenatal care has been associated with increased rates of PTD in the \npresence, as well as absence, of high-risk conditions 126.   In an attempt to reduce PTD, many \nhealth organizations such as Canada 127 and France128 have adopted a population wide health \nstrategy that integrates disease prevention, health promotion, improvements in socioeconomic \nstandards and increased attendance to antenatal care.   Observational studies examining \nvariations of this approach have shown modest reduction in rates of PTD when applied to the \ngeneral pregnant population 128;129.   However, two meta-analyses 130;131 have shown that \nincreased antenatal attendance without specific specialist investigations (such as fetal \nbiophysical or microbiological surveillance)  does not reduce the risk of PTD, low birth \nweight or perinatal mortality in low-risk women. \n \nProphylactic micronutrients e.g. fish oil, magnesium, vitamins  \nAn overview of trials and systematic reviews concluded that there was insufficient evidence \nto show that antenatal prophylactic micronutrient supplementation reduced the risk of PTD in \neither low or high-risk pregnancy groups42.   Small studies have shown limited reductions in \nPTD when using fish oil, omega-3 fatty acids, calcium, zinc, magnesium, or multivitamin \ncombinations132-139, although vitamin C supplementation might even increase the risk of PTD \n140. These interventions need to be further explored in larger RCTs along with other important \nperinatal outcomes such as growth restriction and pre-eclampsia. \n \n\nChapter  3.1: Prevention of preterm delivery \n 187 \nProphylactic tocolytics Three meta-analyses evaluating prophylactic or maintenance oral \ntocolytics (mainly beta-mimetics) in high-risk pregnancies (women with threatened PTL or \nprevious PTD) have not shown any reduction in PTD, PTL, perinatal morbidity or perinatal \nmortality141-143.   Importantly, newer tocolytics such as nifedipine and atosiban have not \nundergone evaluation in this manner144. A recent trial showed no beneficial effect on PTD, \nand a potential harmful effect on fetal renal function and the ductus arteriosus, when \nrofecoxib (a COX-2-specific prostaglandin inhibitor) was administered prophylactically to \nwomen at high-risk of PTD between 16-32 weeks gestation 145. \n \nProphylactic corticosteroids Meta-analysis has shown maternal antenatal administration of \na single course of corticosteroids is associated with a significant reduction in perinatal \nmortality (OR 0.60, 95% CI 0.48 to 0.75), respiratory distress syndrome (OR 0.53, 95% CI \n0.44 to 0.63) and intraventricular haemorrhage in preterm infants 146.   Consequently, a single \ncourse of antenatal corticosteroids is recommended in women symptomatic of PTL or \nPPROM or threatening to deliver preterm because of an obstetric disorder 147.   No beneficial \neffect has been reported following corticosteroids given before 28 weeks‘ or if infants are \ndelivered more than seven days after initiation of treatment.    However, there are no \nprospective trials on the prophylactic use of corticosteroids (single or multiple courses) in \nhigh-risk asymptomatic pregnancies (e.g. growth restricted fetuses, pre-eclampsia, multiple \npregnancies, previous recurrent PTD) not at imminent risk of PTD.  Their use in these \ncircumstances remains controversial and unproven 147-150. In particular, many of these women \nmay remain at risk of PTD seven days after the first course, which creates the clinical \ndilemma of whether to administer a repeat course of antenatal corticosteroids.   Repeated \ncourses of antenatal corticosteroids may have a lower rate of neonatal lung disease according \nto one meta-analysis 148.  However, an extensive review performed by the NIH 151 reported \n\nChapter  3.1: Prevention of preterm delivery \n 188 \nthat there was insufficient evidence to conclusively show any marked adverse or beneficial \nchange with repeated courses of corticosteroids for important neonatal outcomes like small-\nfor-gestational-age at birth, perinatal death, periventricular haemorrhage, periventricular \nleucomalacia, infectious morbidity, and neonatal lung disease.   Absence of beneficial effect \nof repeated weekly vs. single course antenatal corticosteroids in women at risk of PTD was \nshown in a recently published trial 152. Notably, a subgroup analysis of the trial in women \nwith PPROM153 showed that there was no difference in neonatal morbidity but an increased \nrisk of chorioamnionitis in women who received weekly courses of corticosteroids. \n \nProphylactic antibiotics  A meta-analysis has shown that prophylactic antibiotics given \nduring the second and third trimester of pregnancy in unselected pregnancies reduces the risk \nof PPROM (OR 0.32, 95% CI 0.14-0.73) 154.  There was a risk reduction in PTD in pregnant \nwomen with previous PTD associated with bacterial vaginosis (BV) but there was no risk \nreduction of PTD in pregnant women with previous PTD unrelated to BV.   This observation \ncomplements our meta-analysis 58discussed earlier, that showed screening and treating BV in \nunselected low-risk populations was beneficial in reducing PTD. \n \nProphylactic progesterone Two recently published meta-analyses have shown a beneficial \nrole for prophylactic progesterone supplementation in the prevention of PTD 155;156. Despite \ndifferences in the pregnancy risk status of the population included, and the number of \nincluded trials  [seven trials 155 and ten trials 156] both meta-analyses have reported similar \nrates of risk reduction of  PTD: OR 0.58, 95% CI 0.48-0.70 155 and OR 0.45, 95% CI 0.25-\n0.80 156 . Based on increasing research in this area, a supportive but cautionary statement was \nreleased by the ACOG 157 in 2004, which recommended that prophylactic progesterone to be \n\nChapter  3.1: Prevention of preterm delivery \n 189 \nused only in women with a history of previous PTD. Two recently published RCTs are \nincluded in the meta-analysis. One RCT 158.   (n=142) showed that daily administration of \nprophylactic vaginal progesterone (100mg) compared to placebo between 24 and 34 weeks‘ \nin high-risk pregnancies (women with previous PTD) reduced the frequency of uterine \ncontractions and the rate of PTD (OR 0.40, 95% CI 0.17-0.94).  The other RCT (n=463) 159 \nshowed that women with a history of previous PTD, who received weekly injections of 17 \nalpha-hydroxyprogesterone caproate (17P) from recruitment (16-20weeks‘) to 36 weeks‘ \ngestation, had a reduced risk of PTD before 37 weeks‘ (OR 0.66, 95% CI 0.54 to 0.81), \nnecrotizing enterocolitis, intraventricular hemorrhage, and need for supplemental oxygen.  A \nsecondary analysis of this study showed the risk reduction in PTD is greatest in the subgroup \nof women whose previous PTD was before 34 weeks 160. Further research on the correct \nprogesterone formulation, mechanism of action, efficacy, and risk-benefit profile is needed \nbefore prophylactic progesterone may become an accepted clinical intervention in high-risk \nasymptomatic pregnancies. \n \n \n\nChapter  3.1: Prevention of preterm delivery \n 190 \nAntenatal management plan and role of specialist antenatal prematurity clinics  \nSpecialist antenatal clinics for women with multiple pregnancy, diabetes, epilepsy, and \nhaematological disorders are widespread and well established.   Likewise, women at high-risk \nof prematurity may also benefit from such specialised antenatal care with individualised risk \nassessment and application of general and specific screening-preventative measures to \nprevent PTD or reduce adverse perinatal outcome.   These clinics are common in many \nuniversity teaching hospitals161, although rigorous evaluation of their exact beneficial role in \nreducing PTD is pending.   The exact antenatal design, resources needed, and timing of \nscreening interventions remain a controversial issue and have little supporting evidence. \nNevertheless, we suggest an antenatal management plan (Table 3.4) that may prevent PTD \nbased on established practice and evidence presented in this review that may be considered a \nbasis for further modification and research. \n\nChapter  3.1: Prevention of preterm delivery \n 191 \nTable 3.4.    Suggested antenatal strategy to prevent preterm delivery \nANTENATAL \nVISIT AND \nPURPOSE \nInfection \n(Screen and \ntreat  \nBV, UTI) \nCervico-\nvaginal \n fFN \n \nUltrasound \nAbdominal \nand \nTransvaginal \n \nOther interventions to be \nconsidered \nPre-pregnancy \nCounselling on recurrence \nrisk and any modifiable \npredisposing factors \nYes \n \nNo No Cessation smoking and illicit drugs \nImprove BMI>25 \nThrombophilia screen if history \nsuggests \nOptimise control of diabetes, high BP \nChange anticoagulation or \nantihypertensive drugs \n8 weeks’ \nRoutine booking bloods \n  \n \nYes  No Dating \npregnancy \n \n  \n \nThrombophilia screen and commence \naspirin & LMWH if positive. \nLow dose aspirin if previous pre-\neclampsia (consider use if previous \nstillbirth, abruption, severe IUGR) \nProphylactic progesterone \nGeneral preterm birth education, \nsupport, and risk factor avoidance. \nScreen and treat BV, UTIs \nLow threshold for GTT testing \n12, 16, 20, 24, 28 weeks’ \nNuchal Translucency(12w) \nand/or  Triple Test or msAFP \n(15-18w) \nNo No Serial Cervical \nassessments in \nwomen at high \nrisk of PTD \nEmergency or elective (12-16w) \ncervical cerclage based on ultrasound \nfindings and/or reproductive history \nEmergency cervical cerclage is not \nindicated if above 32 weeks‘ \nLow threshold for GTT testing \n22 weeks’ Yes  No Detailed fetal \nsurvey \nUterine artery \nDoppler \nLow dose aspirin if suspect pre-\neclampsia or IUGR due to uterine \nartery notching and/or previous \nhistory \nScreen and treat BV and UTIs \n\nChapter  3.1: Prevention of preterm delivery \n 192 \n \nFootNote: Bacterial vaginosis, BV; BMI, body mass index; BP, blood pressure; fFN, fetal \nfibronectin; GBS, Group B streptococcus; GTT, glucose tolerance test; LMWH, IUGR, \nintrauterine growth restriction; low-molecular weight heparin; msAFP, maternal serum alpha-\nfetoprotein; NICU, neonatal intensive care unit; PPROM, preterm premature rupture of \nmembranes; PTL, preterm labour; UA, uterine artery; UTI, urinary tract infection. \n24, 28, 32, 36 weeks’ \nGTT at 28 weeks‘ \nNo Only if \nsymptomati\nc \nFetal growth \nand umbilical \nartery Doppler \n \nProphylactic corticosteroids, \nantibiotics if symptomatic of PTL or \nPPROM. \nIn utero transfer to unit with NICU if \nsymptomatic with positive fFN \n \nLabour \nSpontaneous or induced \nYes  Helps \nconfirm \nLikelihood \nof \nPTL, \nPPROM \nAsses fetal well-\nbeing, and \npresentation \nProphylactic corticosteroids, \nantibiotics (especially GBS \nprophylaxis).    \nTocolytics if in utero transfer to unit \nwith NICU is needed. \nPost-partum \n6 week antenatal check \nNo No No Review antenatal events and delivery \nIdentify modifiable factors for future \nprevention of PTD \n\nChapter  3.1: Prevention of preterm delivery \n 193 \nDiscussion \n \nThere is evidence that introducing screening-preventative strategies for asymptomatic \npregnancies may reduce the rate of PTD.   Evidence for screening and selective treatment \nexists for: asymptomatic bacteriuria (meta-analysis: OR 0.60; 95% CI 0.45-0.80); bacterial \nvaginosis in low-risk population groups (meta-analysis: RR 0.73; 95% CI 0.55-0.98, figure \n3.1); elective cervical cerclage in high-risk pregnancies; indicated cervical cerclage in women \nwith short cervical length on ultrasound (meta-analysis: RR 0.74, 95% CI 0.57-0.96); \nprophylactic progesterone supplementation in high-risk pregnancies (meta-analysis: OR 0.45, \n95% CI 0.25-0.80). A summary of the quality of evidence and grading of recommendation \nfor these interventions are depicted in Table 3.5.  \n \nHowever, for most other strategies, such as increased antenatal attendance, or routine \nadministration of prophylactic micronutrients, the evidence is inconsistent and conflicting. \nInformation on neonatal outcomes apart from PTD (such as serious neonatal morbidity and \nmortality) was found to be lacking in most studies.  It was therefore not possible to establish \nwhether preventing PTD or prolonging gestation would correlate to improved perinatal \noutcome, and so lessened the potential clinical usefulness of any proposed preventative \nstrategy. No studies were found that evaluated the effectiveness of combining screening-\npreventative strategies. \n  \n\nChapter  3.1: Prevention of preterm delivery \n 194 \nTable 3.5  Summary of screening and preventative strategies that may reduce the risk of \npreterm delivery \nStrategy for preventing preterm \ndelivery \nRCOG \nLevel of \nEvidence \nGRADE \nQuality \nof \nEvidence \nGRADE \nStrength of \nRecommendation \nAsymptomatic bacteriuria in all women \n \nIa High Strong \nBacterial vaginosis in low-risk population \ngroups \nIa, Ib Moderate Weak \nElective cervical cerclage in high-risk \npregnancies \nIb, IIa, \nIIb \nModerate Strong \nIndicated cervical cerclage in women with \nshort cervical length on ultrasound \nIb, IIa, \nIIb \nModerate Strong \nProphylactic progesterone \nsupplementation in high-risk pregnancies \nIa, Ib High Strong \nSmoking cessation in all women IIb, III Very \nLow \nWeak \n \nReviews discussing screening-preventative interventions for preventing PTD often consider \nboth symptomatic (symptoms of PTL or PPROM) and asymptomatic pregnancies.   We have \nfocused solely on asymptomatic pregnancies and adopted a rigorous systematic review \nmethodology to provide the best possible analysis of the data available.   The review is \nweakened by over-reliance on conclusions drawn from meta-analyses and underpowered \nRCTs. We have identified considerable heterogeneity in the studies and methodologies \nadopted by the meta-analyses, in particular, the groups of women considered to be ‗high‘ and \nlow‘ risk of PTD, the magnitude of their risk of PTD, and gestation-specific timing of the \nintervention differs considerably for each trial and meta-analysis (e.g differences in types of \nantibiotic, dosage, method of administration, and gestation when given).  This heterogeneity \nwould propagate any potential omission, de-emphasis or misinterpretation of the results of \nRCTs.  \n\nChapter  3.1: Prevention of preterm delivery \n 195 \nThe poor clinical efficacy of the proposed screening-preventative strategies is not \nunexpected.   Firstly, current routine antenatal screening is relatively ineffective at identifying \nthe majority of pregnancies at risk for PTD, even if combined with specialist investigations.   \nSecondly, most of the preventative interventions discussed appear to have, at best, only mild \nefficacy at preventing PTD.  Importantly, adverse effects of increasing the risk of PTD were \nnoted for some of the interventions.  Examples include antibiotic treatment for women \nscreened to be positive for fFN or trichomonas vaginalis, and inherent surgical risks \nassociated with cervical cerclage.   \n \nFurther trials are needed to identify the optimum gestation and subgroups that may benefit \nmost from such screening and therapeutic interventions.   Currently on-going meta-analyses \nof individual patient data 162;163 may provided further evidence for the roles of elective and \nindicated cerclage on preventing PTD, and aspirin on the prevention of pre-eclampsia related \nconsequences. \n \nIt was surprising to show a reduction in PTD following screening and treating BV in the low-\nrisk (RR 0.73; 95% CI 0.55-0.98) (Figure 3.1) but not the high-risk group, as one would \nnormally expect an opposite relationship and treatment to exert greater risk reduction in the \nhigher risk group.  The differences in antibiotic sensitivity between high and low risk groups \nmay suggest differing causal contributions of the infectious process to PTD. The evidence, \nalong with prior knowledge of differing predisposing factors and prognosis between these \nrisk groups 52;164, supports the hypothesis that PTD in high and low risk pregnant women \nare different entities and not linear extremes of the same syndrome; a view shared by \nothers 8, and deserving of further confirmatory research. \n\nChapter  3.1: Prevention of preterm delivery \n 196 \nThis review has provided a structured approach to addressing the complex issue of preventing \nPTD.  By elaborating on the use of both specific and general measures this review should \nappeal to all health care professionals (General Practitioners, Health Visitors, Midwives, \nObstetricians) involved in the care of pregnant women, as well as colleagues involved in \ndelivering public health care strategies.   We have proposed an antenatal care strategy that \nadopts a gestation-specific approach to assessing risk and intervening as needed (Table 3.4) \nthat may be commenced at initial antenatal booking.  However, the efficacy and cost-\neffectiveness of these approaches (Tables 3.4 & Table 3.5) needs to be rigorously evaluated \nbefore routine clinical implementation.  Differences in the prevalence of infection and other \nobstetric and reproductive factors means that any proposed preterm prevention strategy \nshould be individualised to the population and health care setting.  Specialist antenatal clinics \nfor women deemed at high-risk of PTD may provide an opportunity to carry out this research \nand perform this clinical role.  \n \nThe recent NICE UK antenatal care guideline 32 has stated pregnant women should not be \noffered routine screening for BV, Chlamydia, group B streptococcus, cervical \nultrasonography, or cervical fFN 32.   Our review has presented preliminary evidence that \nsome of these strategies may actually be beneficial, and as such, adds to the current debate in \nthis important clinical area. \n \n \n\nChapter  3.1: Prevention of preterm delivery \n 197 \nDiscussion on bacterial vaginosis meta-analyses  \n \nMeta-analyses are liable to numerous biases despite quality control measures, and their \nresults may not necessarily be trusted 165;166.  Concerning screening and treating BV in \npregnancy, five meta-analyses [Cochrane 53, 13 trials; Riggs 54, 11 trials; Leitich 55, 10 trials; \nGuise 56, 7 trials, and Okun 57, 11 trials] have been published in the last four years.  All have \nshowed no reduction in PTD.  The authors of all five meta-analyses have reported significant \nclinical, methodological and statistical heterogeneity of the included studies, and have \nutilised different fixed or random effects pooling.  Only two meta-analyses 53;57 undertook a \ncomprehensive investigation of the reasons for heterogeneity or attempted strategies to \ncounter this effect.  Nevertheless, without undertaking this process, the summary estimate \nproduced by each meta-analysis may not be valid.  Our ―repeat‖ meta-analysis , which \nconsidered the classification of population risk and therefore addressed the issue of \nheterogeneity to some extent, showed an unexpected beneficial effect of screening in women \nthat were at low-risk rather than high risk of PTD .   \n \nIn summary, we wish to emphasise why it is important to consider the individual primary \nstudy as well as the methodology criteria adopted by meta-analyses, particularly when \nincluded trials are underpowered, few in number, and exhibit marked heterogeneity.  These \nfactors may contribute to why meta-analyses to date have reported evidence of lack of \neffectiveness, but in fact may be subject to the bias of varying study methodologies, mixing \nhigh and low risk pregnancies groups, and a confounding effect introduced by the screening \nprocess itself that is difficult to distinguish from antibiotic treatment of bacterial vaginosis.\n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   198 \n \n3.2. Non-contraceptive uses of levonorgestrel releasing hormone system \n(LNG-IUS)- a systematic enquiry and overview \n \nAbstract \nLevonorgestrel releasing intrauterine systems (LNG-IUS) were originally developed as a \nmethod of contraception in the mid 1970‘s.   The only LNG-IUS approved for general public \nuse is the Mirena® LNG-IUS, which releases 20mcg of levonorgestrel per day directly in to \nthe uterine cavity.  However, new lower dose (10mcg and 14mcg per day) and smaller sized \nLNG-IUS (MLS, FibroPlant-LNG) are currently under clinical development and \ninvestigation.  Research into the non-contraceptive uses of LNG-IUS is rapidly expanding.  \nIn the UK, LNG-IUS is licensed for use in menorrhagia and to provide endometrial \nprotection to perimenopausal and postmenopausal women on estrogen replacement therapy.  \nThere is limited evidence to suggest that LNG-IUS may also be beneficial in women with \nendometriosis, adenomyosis, fibroids, endometrial hyperplasia and early stage endometrial \ncancer (where the patient is deemed unfit for primary surgical therapy).  This systematic \nenquiry and overview evaluates the quality of evidence relating to the non-contraceptive \ntherapeutic uses of LNG-IUS in gynaecology. \n \nAdditional point relating to date listed in tables: For all studies listed in tables, we have \nreported the sample sizes originally recruited by the studies. Where the study drop out rate \nexceeds 10% we have stated this rate to provide the reader with an impression of the number \nof subjects actually evaluated by the study where this drop out rate is exceeded. \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   199 \n \nIntroduction \nThe only levonorgestrel-releasing intrauterine system (LNG-IUS) approved for general \npublic use is the Mirena® (Schering AG), which is a T-shaped plastic intrauterine device \n(IUD) that releases levonorgestrel (20mcg per day) directly into the uterine cavity.  The mean \nsystemic levels of levonorgestrel with this LNG-IUS (425pg/mL at 1 month, 330 pg/mL at 6 \nmonths, mean age of subjects was 31 years (range 18-42) 1 are less than those achieved with \ntherapeutic oral or parenteral doses of progestogens (hence minimizing systemic side effects) \nand exceeds the critical value of 200 pg/mL below which ovulation occurs 2.  Mirena was \nfirst launched in Finland in 1990 and has been marketed in the UK since 1995 as a \ncontraceptive device.  Two new lower levonorgestrel dose and smaller sized LNG-IUS \ndevices are currently under clinical development and investigation:  FibroPlant™-LNG \n(frameless device, Contrel Research, Belgium) and MLS system, releasing 14mcg and 10mcg \nlevonorgestrel per day respectively3;4.  \nMirena® LNG-IUS is currently licensed in the UK as a 5-year contraceptive agent (license \nawarded 1995), treatment for idiopathic menorrhagia (license awarded 2001), and to provide \nuterine protection during estrogen replacement therapy in peri- and postmenopausal women \n(license awarded 2005). The latter two applications for Mirena® LNG-IUS are not licensed \nin USA or Canada. The fertility control provided by LNG-IUS is comparable with that of \nfemale sterilisation, and is completely reversible 5.  There are many other non-contraceptive \nbeneficial effects of LNG-IUS that have important public health implications.   These have \nbeen summarized by several reviews 6-9 and policy statements 10, and incorporated within one \nsystematic review examining all types of intrauterine device 11.  However, there has since \nbeen a considerable expansion of publications in this area, many of which have contrasting \nmethodological quality and results.  This article expands on past reviews by incorporating \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   200 \n \nthese recent advances and performs an up-to-date systematic review focused entirely on \nLNG-IUS.  Furthermore, this review evaluates the quality of supporting evidence, and where \navailable, presents information relating to adverse effects, cost-effectiveness, and health \nrelated quality of life (HRQL) issues. \n \nMaterials and Methods  All observational and experimental studies examining the use of \nLNG-IUS in Gynaecology were retrieved from MEDLINE (1996-2005), EMBASE (1996-\n2005 week 08), Cochrane Central Register of Controlled Trials, Cochrane Database of \nSystematic Reviews, Database of Abstracts of Reviews of Effects (DARE), The National \nResearch Register NRR (http://www.update-software.com/National/), Medical Research \nCouncil's Clinical Trials Register, and details on reviews in progress collected by the NHS \nCentre for Reviews and Dissemination were searched.  Schering HealthCare (UK) were also \ncontacted for further information on licensing and any unpublished controlled clinical trials. \nThe following search terms and word variants were used: ‗exp Intrauterine Devices, \nMedicated/‘, ‗levonorgestrel releasing‘, ‗levonorgestrel-releasing‘,‗LNG-IUS‘, ‗LN-IUS‘, \n‗LN-IUD‘, ‗LNG-IUD‘, ‗mirena.tw.‘  ‗Levonorgestrel adj5 (intrauterine or device or coil or \nsystem). tw, ‗progest$ adj5 (intrauterine or device or coil or system).tw‘, ‗intra-uterine \nprogestogen‘  combined with ―AND‖ to ‗gyne$‘, ‗therapy‘ ‗endometriosis‘, \n‗endometrio$.mp‘, ‗genital neoplasms, female‘, ‗dysmenorrhoea‘, ‗pelvic pain‘, ‗estrogen \nreplacement therapy‘, ‗hormone replacement therapy‘, or ‗genital diseases, female‘.  The \nsearch was completed in March 2005.  Obtained data were qualitatively and quantitatively \nanalysed. If trials are deemed suitable (similar population groups, trial methodology and \noutcome measures) meta-analysis will be performed. \n \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   201 \n \nResults  \nA summary of the studies identified describing the non-contraceptive therapeutic use of \nLNG-IUS according to the therapeutic indication is shown in Table 3.6.  The associated level \nof evidence and strength of recommendation for each indication is also indicated according to \naccepted criteria 12. \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   202 \n \nTable 3.6.  Summary of studies that assess LNG-IUS use in various non-contraceptive \ntherapeutic indications as primary study outcome measures \nTherapeutic use of LNG-\nIUS \nRCTs Cohort \nStudies \n \nProspective \nor \nRetrospective \nObservational \nStudies \nCase \nReport \nor small \ncase \nseries \n**Level \nof \nevidence \n*** \nStrength \nof \nrecomme\nndation \nMore than \n50 women \n in \nLNG-IUS \narm  \nof trial \nLess than \n 50 women  \nin  \nLNG-IUS \narm \n of trial \nMenorrhagia 1 9 2 5 0 I, II, III A \nFibroids/Fibroid related \nmenorrhagia \n1# 2# 1 6 1 II, III B \nEndometriosis 0 2 0 3 0 I, III C \nAdenomyosis 1 0 0 1 1 I, III, III C \nUterine protection with \nestrogen replacement \ntherapy in per- and \npostmenopausal women \n3 4 3 7 0 I, II, III A \nUterine protection with \ntamoxifen in \npostmenopausal women \n1 0 0 1 0 I, III A \nEndometrial hyperplasia 0 0 1 3 2 II, III C \n \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   203 \n \nFootnotes to Table 3.6 \n# Trial(s) exist, but therapeutic outcome was not assessed as a priori primary outcome \nmeasure in the RCT comparison \n \n**Classification of Evidence Levels  \nIa Evidence obtained from meta-analysis of randomised controlled trials.  \nIb Evidence obtained from at least one randomised controlled trial.  \nIIa Evidence obtained from at least one well-designed controlled study without \nrandomisation.  \nIIb Evidence obtained from at least one other type of well-designed quasi-experimental study.  \nIII Evidence obtained from well-designed non-experimental descriptive studies, such as \ncomparative studies, correlation studies and case studies.  \nIV Evidence obtained from expert committee reports or opinions and/or clinical experience of \nrespected authorities.  \n \n***Strength of Recommendation  \nA Directly based on category I evidence \nB Directly based on category II evidence or extrapolated recommendation from category \nI evidence \nC Directly based on category III evidence, or extrapolated recommendation from \ncategory I or II evidence \nGPP Directly based on category IV evidence, or extrapolated recommendation from \ncategory I, II or III evidence \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   204 \n \nMenorrhagia \nEarly RCTs and cohort studies evaluating the contraceptive efficacy of LNG-IUS against Cu-\nIUCD showed women who received LNG-IUS reported less dysmenorrhoea and menstrual \nblood loss (MBL)35;36.  This provided a basis to examine whether LNG-IUS would also \ndecrease menstrual blood loss in women with idiopathic menorrhagia (dysfunctional uterine \nbleeding DUB) and compare its efficacy against established medical and surgical treatments \nfor menorrhagia.  In total, approximately 670 women with menorrhagia have used LNG-IUS \nas part of a comparative or non-comparative study (Table 3.7) evaluating the efficacy of \nLNG-IUS in treating menorrhagia. Women using the frameless FibroPlant-LNG™ or \nFemilstrade LNG-IUS (20mcg/24hr) devices for contraception 33;37 or treatment of \nmenorrhagia 30-33 also reported decreased MBL, however study sample sizes were limited \n(n=76 menorrhagia cases) and the devices remain under clinical development.   \n \nTwo incomplete trials were identified in the search, SMART (Satisfaction with Mirena and \nAblation: a Randomised Trial) 38 and TALIS (Thermo-Ablation versus the Levonorgestrel \nIntrauterine System)39. Furthermore, our unit is about to commence the ECLIPSE trial \n(Effectiveness and Cost-effectiveness of Levonorgestrel containing Intrauterine system in \nPrimary care against Standard treatment, ISRCTN 86566246) in the UK. \n \nOverall, for all listed studies, LNG-IUS use in women with menorrhagia reduces menstrual \nblood loss by 79% to 97%.  No RCTs have compared LNG-IUS with placebo or no treatment \nin women with menorrhagia.  Importantly, studies have used various outcome measures, \nwhich precludes pooled meta-analysis.  These include: indirect (pictorial blood loss \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   205 \n \nassessment chart, PBAC) or direct (alkaline haematin method) measures of menstrual blood \nloss (MBL); patient willingness to continue with treatment; or patient preference to abandon \nLNG-IUS treatment in favour of hysterectomy or endometrial resection.  There are \ninsufficient participants to show long term therapeutic effect with LNG-IUS, as most studies \ndid not extend beyond one year follow up. The total number of participants continuing with \nLNG-IUS by 3-year 19;34 and 5-year follow up14 was 96 cases. Of the ten trials depicted in \nTable 3.7, seven 13;16-18;20;23;24 have been incorporated in two Cochrane reviews 40;41 and one \nsystematic review 42.  Three recent RCTs 15;21;22 and two quality cohort studies 25;26 not \nincluded in the prior published meta-analyses have been listed in Table 3.7.  The high patient \nsatisfaction (72-94%) and overall continuation rates (65- 88%) obtained in these RCTs are \nconsistent with those identified in observational studies of LNG-IUS use for treating \nmenorrhagia 29;43;44. Interpreting the evidence from Table 3.7, LNG-IUS system is at least \ncomparable or more effective than oral progestogens.  Similar rates of patient satisfaction and \nquality of life are reported when comparing LNG-IUS against transcervical endometrial \nresection or balloon ablation.  However, surgical methods are significantly more effective in \nreducing menstrual bleeding or inducing amenorrhoea within one year follow up.  However, \none trial with longer follow up of three years 19 showed no significant difference between the \nLNG-IUS and TCRE in the reduction of menstrual blood loss.   \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   206 \n \nTable 3.7. LNG-IUS studies assessing therapeutic effect in women with menorrhagia \nAuthor Year  \nof  \nPublication \nStudy \nType \nSample \nSize of women \nwith \nmenorrhagia \nComparison  Outcomes (within one year \nfollow up unless stated \notherwise) \nHurskainen 13;14 2001, 2004 RCT 236 119 LNG-IUS  vs. \n117 hysterectomy  \n \nFor the LNG-IUS \ngroup: at one year \n81/119 and at five \nyear 57/119 \ncontinued to have \nLNG-IUS in situ \n5 year follow up \n \nOf the LNG-IUS group by one \nyear 68% continued with LNG-\nIUS and 20% had TAH. \nBoth treatments had comparable \nimprovements in HRQL \nSoysal 15 2002 RCT 72 36 LNG-IUS  \nvs.  \n36 thermal balloon \nablation \n \n14% drop out from \nLNG-IUS \nGreater reductions in PBAC with \nablation than LNG-IUS. \nComparable improvements in \nhaemoglobin \nAblation group perceived greater \nimproved HRQL than LNG-IUS \nCrosignani 16 1997 RCT 70 35 LNG-IUS vs.  \n35 TCRE \n \n14% drop out from \nLNG-IUS \nMarginally greater reductions in \nPBAC with TCRE \nComparable satisfaction rates \nKittelsen 17 1998 RCT 60 30 LNG-IUS vs.  \n30 TCRE \n \n12% drop out rate \nComparable reductions in PBAC \nComparable satisfaction rates \nIstre 18 \nRauramo 19 \n2001,2004 RCT 59 30 LNG-IUS vs.  \n29 TCRE \n \n31% drop out rate \n3 year follow up \n \nGreater reductions in PBAC with \nTCRE than LNG-IUS (90% cure \nvs. 67% cure) at one year , but \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   207 \n \n comparable reductions of MBL \nnoted at 3 years. \nIncreased haemoglobin and \nferritin with both treatments \nLahteenmaki 20 1998 RCT 56 28 LNG-IUS vs.  \n28 medical \ntreatment whilst \nawaiting \nhysterectomy \n \n25% drop out from \nLNG-IUS \nAt 6m, 64% LNG-IUS cancelled \nTAH whilst 14% cancelled TAH \nin medical treatment group \nReid 21 2005 RCT 51 25 LNG-IUS vs 26 \nmefenamic acid \n \n16% drop out from \nLNG-IUS \nGreater reductions in MBL, \nPBAC and total menstrual fluid \nloss with LNG-IUS (90% vs \n23%) at 6 months. \nBarrington 22 2003 RCT 50 25 LNG-IUS vs. 23 \nballoon ablation \n \n12% drop out rate \nComparable reductions in PBAC \n \nIrvine \n23 \n1998 RCT 44 22 LNG-IUS vs. 22 \noral norethisterone \n \nNo drop out rate \nComparable reductions in MBL \n(>90%). Greater satisfaction with \nLNG-IUS \nMilson 24 1991 RCT 35 20 LNG-IUS vs.  \n15 transexamic acid \n \n20% drop out from \nLNG-IUS \nGreater reduction in MBL with \nLNG-IUS (>90%) \nRomer 25 2000 Prospective \ncohort \n30 LNG-IUS vs. roller \nball endometrial \nablation \nComparable reductions in MBL \nand rates of amenorrhoea \nHenshaw 26 2002 Retrospecti\nve cohort \n62 LNG-IUS vs. \nmicrowave \nendometrial \nmean 14 month follow up \nComparable reductions in MBL \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   208 \n \nablation \n \nand dysmenorrhoea \nComparable patient satisfaction \nrates  \nMansour 27 1998 Prospective  52  No comparison \nLNG-IUS \n91% of women had improved \ndysmenorrhoea and menorrhagia \n83% continued with treatment \nbeyond one year \nBarrington 28 1997 Prospective  50  LNG-IUS \nNo comparison. \nWomen were \nawaiting TCRE or \nhysterectomy \nReduced PBAC in 82% \n8% amenorrhoea \nNo change in haemoglobin or \nferritin \nDecreased premenstrual \nsymptoms in 56% \nReduced dysmenorrhoea in 80% \nMonteiro 29 2002 Prospective  44 LNG-IUS  \nNo comparison \nDecreased MBL and increased \nhaemoglobin \n80% continuation rate at one year \nWildemeersch 30 \n31;32 \n2004 Prospective  12 in 2004,  \n32 in 2001 \n \nNo comparison. \nFibroPlant-LNG \nDecreased PBAC (median MBL \ndecreased by 90%) \nDecreased dysmenorrhoea \nWildemeersch 33 2005 Prospective 60 women: \n28 normal \nperiods, \n32 menorrhagia \nNo comparison \nFemilstrade LNG-\nIUS 20mcg/24 hr \nSimilar reductions in MBL (96-\n99%) for both groups \n33% developed amenorrhoea (10 \nwomen in each group) \nXiao 34 2003 Prospective  34 LNG-IUS \nNo comparison \n \n3 year follow up \nDecreased MBL at one year \n(84%) and three (85%) years. \n33% amenorrhoea at 6 months. \nIncreased Hemoglobin and serum \nferritin. \nFootnotes FibroPlant-LNG is a frameless low-dose (releasing 14mcg levonorgestrel/day) frameless \nLNG-IUS; HRQL, Health related quality of life assessments; LNG-IUS releasing 20mcg levonorgestrel /day; \nMBL menstrual blood loss; PBAC pictorial blood loss assessment chart; TCRE transcervical endometrial \nresection\n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   209 \n \nMeta-analyses and RCTs have shown that a significant proportion of women with \nmenorrhagia initially treated with either conservative surgery 45 or LNG-IUS 42 are likely to \nrequire hysterectomy as a definitive treatment.  However, an RCT (n=236) with five year \nfollow up has shown hysterectomy does not improve overall health related quality of life \nsignificantly more than LNG-IUS and it can cause serious complications 14.   Furthermore, \nthe same trial showed that LNG-IUS was more cost-effective than hysterectomy at one-year \n13 (US $ 1530 vs. US $ 4222) and five-years 14 follow up (US $ 2817 vs. US $ 4660 per \nparticipant).  This estimate includes the direct (e.g. operative, costs) and indirect costs (e.g. \nsick leave days) associated with the 42% of the women assigned to the LNG-IUS group who \neventually underwent hysterectomy. Menorrhagia may arise from inherited bleeding \ndisorders (e.g. von Willebrand's Disease).  A prospective study (n=16) has shown reduction \nin menstrual blood loss, improvement in quality of life in women with menorrhagia due to an \ninherited bleeding disorder when treated with LNG-IUS 46.   \n \nUterine fibroids and fibroid related menorrhagia \nOne cohort study, five prospective \nobservational studies, and one case report have directly assessed the use of LNG-IUS in \ntreating fibroids and fibroid related menorrhagia or dysmenorrhoea.  Three RCTs, undertaken \nfor other indications, have described decreased incidence of fibroids following LNG-IUS \ninsertion 35;54;55. All these studies are depicted in Table 3.8.  Apart from one study35, study \nduration and follow up did not exceed one year.  Inclusion criteria were clearly stated in two \nstudies: women with fibroid uterus below 12 weeks gestational size on pelvic examination or \n380ml uterine volume on pelvic ultrasound 47;48. \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   210 \n \nTable 3.8. LNG-IUS studies directly or indirectly assessing therapeutic effect on fibroids \nor fibroid related menorrhagia \nAuthor Year  \nof \nPublication \nStudy \nType \nSample \nSize \nComparison  Outcomes (within one year \nfollow up unless stated \notherwise) \nDIRECT STUDIES      \nSoysal 47 2005  \n \nProspective \nand \nretrospective \ncohort \n64 32 LNG-IUS vs. \n32 thermal \nballoon ablation \n(historical \nmatched group) \nComparable effective reductions \nin PBAC (around 90%) \nComparable increases in \nhaemoglobin \nFibroid size change not assessed \nGrigorieva 48 2003 Prospective \nand \nretrospective  \n67  No comparison \n \nEffective reductions in PBAC.  \nImproved ferritin and \nhaemoglobin \n40% amenorrhoea at 12 months \nDecrease in fibroid size (33%) \nMercorio 49 2003 Prospective  19  No comparison Reduced PBAC, but 14/19 still \nhad persistent menorrhagia \nWildemeersch 50 2002 Prospective  14  No comparison \nFibroPlant-LNG \nReduction in MBL in 13/14 \nNo reduction in fibroid size \nStarczewski 51 2000 Prospective  12  No comparison Reduction in MBL 11/12 cases. \nAmenorrhoea 50% cases \nImproved Haemoglobin \nNo change in fibroid size \nSinger 52 1994 Prospective  5  No comparison Reduction in MBL \nReduction in fibroid size \nFollow up to 18 months \nFong 53 1999 Case report 1  No comparison Reduction in MBL and fibroid \nsize \nINDIRECT STUDIES      \nGardner 54 2000 RCT 122 64 LNG-IUS and \ntamoxifen against \n58 tamoxifen \n13% reduction in fibroids from \nbaseline in LNG-IUS group \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   211 \n \n \n27% drop out rate \nfrom LNG-IUS \ngroup \nInki 55 2002 Prospective \nstudy \n(examine \none arm of \nRCT) \n38 117 had LNG-\nIUS for \nmenorrhagia, of \nthis 38/119 (32%) \nhad uterine \nfibroids \nNo ultrasonographic change in \nuterine fibroids, but decreased \nendometrial thickness. \nIncreased risk of ovarian cysts \ncompared to hysterectomy \nSivin 35 1994  \n \nRCT 2226 \nrecruited, \n1125 had \nLNG-IUS,  \n1121 had Cu-\nIUCD. \n \nBaseline \nfibroid \nincidence: \nunclear. \nIdentified 15 \nfibroids at \nend of study \nLNG-IUS vs. Cu-\nIUCD (TCu \n380Ag) \n \nParous women \naged 18-38, all \ndesiring \ncontraception. \n \n7 year study \nfollow up (3416 \nwomen years in \nLNG-IUS and \n3975 women \nyears in Cu-\nIUCD) \n \n11.4% drop out \nrate from LNG-\nIUS \n7 year follow up \nLNG-IUS compared to Cu-\nIUCD has decreased incidence \nof dysmenorrhoea, vaginitis, \nfibroids, but higher rates of \namenorrhoea, follicular ovarian \ncysts, acne, mastalgia, weight \ngain, and headache. \n \nLNG-IUS: 50% amenorrhoea or \noligoamenorrhoea by end of \nstudy, compared to 9% with Cu-\nIUCD \n \n \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   212 \n \nAll studies directly assessing LNG-IUS in women with fibroids reported decreased menstrual \nblood loss (84-90%) and similar increases in haemoglobin of 2-3 g/dl 47;48;51. However, there \nwas inconsistency on whether LNG-IUS is associated with decreased fibroid size 48;52;53 or no \nchange in fibroid size 50;51;55.  Fibroid size following LNG-IUS was not assessed in one \ncohort study 47. Regarding the indirect studies, one large RCT suggested there may be \ndecreased incidence of uterine fibroids with LNG-IUS compared to Cu-IUCD 35.  A similar \nobservation of 13% decreased incidence of fibroids was observed in a RCT comparing LNG-\nIUS and tamoxifen against tamoxifen alone 54. \n \nEndometriosis  \nTwo RCTs and three prospective observational studies were identified.  All studies had \nlimited sample sizes (range 11 to 39 participants in LNG-IUS arm of study), and their \nfeatures are shown in Table 3.9.  Population groups differed considerably between studies \nand included women with early stage and late stage endometriosis, rectovaginal \nendometriosis, immediately surgically treated endometriosis, prior history of endometriosis \ndiagnosis, chronic pelvic pain and/or dysmenorrhoea.  This heterogeneity of population, \ncombined with small sample size, limits the strength and validity of the findings. Two studies \nfrom the same group 57;60 report approximately 40% absolute risk reduction in \ndysmenorrhoea by one year with LNG-IUS use. This is consistent with a three year \nprospective study59 and a one year RCT 56 that reported similar magnitude reductions in \ndysmenorrhoea and chronic pelvic pain . A prospective study reported decreasing severity of \nendometriosis on AFS staging following LNG-IUS insertion 58. \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   213 \n \nTable 3.9. LNG-IUS studies assessing therapeutic effect in women with endometriosis \nAuthor Year  \nof  \nPublication \nStudy \nType \nSample \nSize \nComparison  Outcomes (within one year \nfollow up unless stated \notherwise) \n*Petta 56 \n*electronic \npublication \nahead of written \npublication \n2005 RCT 82 with \nendometriosis, \ndysmenorrhoea and \nchronic pelvic pain \n39 LNG-IUS \nvs 43 GnRH \nanalogue \n6 months follow up \nComparable reductions in \npelvic pain and improved \nquality of life measures. \nGreater amenorrhoea with \nGnRH than LNG-IUS (98% vs \n70%) \nVercellini 57  2003  RCT 40 parous women , \nnot desiring fertility, \nwith endometriosis \nassociated \ndysmenorrhoea and \nreceiving \nconservative surgical \ntreatment of \nendometriosis \n20 Post \noperative \nLNG-IUS \nand \nendometriotic \nsurgery vs. \n20 \nendometriotic \nsurgery alone \n10% drop out \nfrom LNG-\nIUS group \nDecreased recurrence of \ndysmenorrhoea in LNG-IUS \nvs. surgery alone group (10% \nvs. 45%, p=0.03) \n28% or 50% LNG-IUS users \nhad amenorrhoea or \noligoamenorrhoea \nComparable levels of patient \nsatisfaction (75% and 50%) \nLockhat 58;59 2004,2005 Prospective  34 with symptomatic \nmild-moderate \nendometriosis \nNo \ncomparison \n(1 yr and 3yr \nfollow up) \nDecreased dysmenorrhoea \nand/or non-cyclical pelvic pain \nand AFS staging of \nendometriosis. \n68% continuation rate at one \nyear  \n56% continuation rate at 3 \nyears. \nVercellini 60  1999 Prospective \n \n18 \nParous women who \nhad history of \nprevious \nendometriotic \nsurgery and had \nrecurrent \ndysmenorrhoea \n \nNo \ncomparison \n \nAmenorrhoea in 24% \nOligoamenorrhoea in 47%    \nDecreased dysmenorrhoea by  \n45% \nDecreased menstrual blood \nloss by 76% \n75% Satisfaction rates \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   214 \n \nFedele 61 2001 Prospective  11 symptomatic \nwomen with \nrectovaginal \nendometriosis \nNo \ncomparison \nDecreased pelvic pain, \ndyspareunia, dysmenorrhoea \nrelated to endometriosis \nDecreased size of \nendometriosis lesions \n(ultrasound) \n \nAdenomyosis  \nOne non-blinded RCT (n=95), one prospective observational study and one case report were \nidentified.   The features of the studies are listed in Table 3.10.  All studies showed a \nreduction in adenomyosis related dysmenorrhoea and menorrhagia, and this effect was \nstatistically significant in the RCT 62 that compared LNG-IUS against expectant treatment in \nwomen following TCRE for adenomyosis. However, dysmenorrhoea and menorrhagia \nobserved in the trial may not necessarily be due to adenomyosis.  \n \n \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   215 \n \nTable 3.10. LNG-IUS studies assessing therapeutic effect in women with adenomyosis \nAuthor Year  \nof  \nPublication \nStudy \nType \nSample \nSize \nComparison  Outcomes within one year \nfollow up \nMaia 62 2003 \n \n \n \nRCT \nNon-blinded \n95 women post TCRE \nfor adenomyosis \n53 LNG-IUS \nvs. 42 \nexpectant \n \nNo drop out  \nreported. \n19% of expectant group \nneeded second treatment for \nuterine bleeding and pain \ncompared to none in LNG-IUS \n \nSignificantly lower rate of \ndysmenorrhoea in LNG-IUS \n(10%) than expectant (80%) \ngroup \n \nSignificantly higher rate of \namenorrhoea in LNG-IUS \ngroup (100% vs. 9%) at one \nyear \nFedele 63 1997 Prospective  25 with adenomyosis \nrelated menorrhagia \nNo \ncomparison \nFor all cases, reduction in \nPBAC, dysmenorrhoea. \nImproved haemoglobin and \nferritin \nFong 64 1999 Case report 1 enlarged \nadenomyosis uterus \nNo \ncomparison \nReduction in uterine size, \ndysmenorrhoea, MBL \n \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   216 \n \nEndometrial protection during oestrogen replacement therapy or tamoxifen in peri-\nmenopausal women \nSeven RCTs, three cohort studies, and seven observational studies have described the use of \nLNG-IUS to protect the endometrium from endometrial hyperplasia or malignant \ntransformation during exogenous estrogen replacement therapy (ERT) in peri- and \npostmenopausal women.  One RCT 54 and one observational study (n=6)65 have examined the \nendometrial protective effect of LNG-IUS during tamoxifen therapy in postmenopausal \nwomen.  The characteristics of these studies are summarised in Table 3.11.  The tamoxifen \nRCT 54 showed that 91% women had endometrial suppression (histological decidual or \natrophic response) in the LNG-IUS and tamoxifen group (n=47) compared to 75% in the \ntamoxifen only group  (n=52) 54.   \n \nRCTs differed in population subgroups (peri-menopausal and post-menopausal women), \nmethods of ERT administration (such as implant, oral, transdermal gel, vaginal ring) \ncomparisons (cyclic oral estrogen/progestogen HRT, continuous combined \nestrogen/progestogen HRT, vaginal progestogen, subdermal progestogen, low dose LNG-IUS \n[10mcg or 14mcg systems] vs. higher does LNG-IUS [20mcg]) and methods of assessing \nendometrial suppression outcome (clinical, histological, ultrasonographic, MRI).  A meta-\nanalysis of discrete groups of studies may be less informative than individually listing the \nstudy design and outcomes, and was therefore not performed. \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   217 \n \nTable 3.11. LNG-IUS studies assessing use to provide uterine protection during oestrogen \nreplacement or tamoxifen therapy \nAuthor Year  \nof \nPublicati\non \nStudy \nType \nSample \nSize \nComparison  \ngroup \nOutcomes \n(within one \nyear of \nfollow up \nunless stated \notherwise) \nTAMOXIFEN \nSTUDIES \n     \nGardner 54 2000 \n \nRCT Initial \nrecruitment of \n122 \nPostmenopau\nsal breast \ncancer \nwomen  \n \n \n64 LNG-IUS and \ntamoxifen group vs. 58 \ntamoxifen group only \n \n27% drop out rate \nfrom LNG-IUS arm \nAll LNG-IUS \nhad \nendometrial \nsuppression \n(histological \ndecidual \nresponse) \nDecreased \nendometrial \npolyps and \nsubmucous \nfibroids in \nLNG-IUS \ngroup \nTurnbull 65 1998 Prospectiv\ne  \n6 \npostmenopau\nsal breast \ncancer \nwomen with \nirregular \nthickened \nendometrium \non tamoxifen \ntherapy \nNo comparison. \nInserted LNG-IUS \nNo change in \nendometrial \nthickness \nwith TV \nultrasound \nA reduction \nin sub-\nendometrial \ncysts and \nendometrial \nvolume with \nMRI by 6 \nmonths \nESTROGEN \nREPLACEME\nNT STUDIES \n     \nBoon 66 2003 RCT 200 \nperimenopaus\nal women \n \n100 LNG-IUS and 100 \noral estradiol vs. \ncyclic/combined oral \nestrogen and \nprogestogen HRT \n(Trisequens) \n2 year follow \nup \n \nendometrial \nsuppression \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   218 \n \n  \n18% drop out rate from \nLNG-IUS group \n(atrophic or \ninactive) \ngreater with \nLNG-IUS \nthan oral \nHRT: \n100% vs 6% \n \nLNG-IUS:  \ninitial erratic \nbleeding, \n62% \namenorrhoeic \nby 2 years. \nCyclic HRT: \nnormal \nregular \nmonthly \nbleeds in 70-\n80% \nWolter-Sven. 67 1997 RCT 112 \nPerimenopaus\nal women \nsymptomatic \nof menopause \n \n \n51 LNG-IUS 10 \nmcg/24hr plus estrogen \n(oral/transdermal) vs. \n45 LNG-IUS \n5cmg/24hr plus \nestrogen \n(oral/transdermal) \n \n11% drop out rate \n95/96 cases \nhad \nhistological \nendometrial \nsuppression \n \nAmenorrhoea \nin most cases \n(62% for \n5mcg and \n61% for \n10mcg \ngroups) \n \nSatisfactory \nrelief of \nmenopausal \nvasomotor \nsymptoms \nRaudaskoski 68 2002 RCT 163  \npostmenopau\nsal women \n \nOral estrogen \nDifferent progestogen \nformulations of HRT \ncombining oral \nestradiol with \nHigh or low dose \nEndometrial \nsuppression \n(histologicall\ny) and \namenorrhoea \nin >98% of \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   219 \n \nwith: \n \n54  \n10mcg/24hr \nLNG-IUS \n(MLS) or \n56 \n20mcg/24hr \nLNG-IUS  \nor \n53 oral \nprogestogen. \n \nLNG-IUS or cyclical \noral progestogen \n \n7% drop out from \ncombined LNG-IUS \n10mcg and 20mcg \ngroups \nLNG-IUS \ncases.  \n \nProliferative \nendometrium \nand regular \nwithdrawal \nbleeds with \noral \nprogestogen  \nRaudaskoski 69 1995 RCT 40 \npostmenopau\nsal \n20 LNG-IUS plus \ntransdermal estrogen vs \n20. continuous oral \nestrogen and \nprogestogen \n \n12% drop out from \nLNG-IUS group \nComparable \nendometrial \nsuppression \n(histological \nand \nultrasound) \n \nComparable \nimprovement \nof \nmenopausal \nsymptoms \nAndersson 70 1992 RCT 40 \nperimenopaus\nal \n \n20 LNG-IUS and oral \nestrogen vs. \n20 Cyclic HRT (oral \nestrogen 3 weeks, oral \nprogestogen 1 week) \n83%of LNG-\nIUS became \namenorrhoeic\n, but cyclic \nHRT had \nregular \nwithdrawal \nbleeds. \nBoth groups \nhad \nendometrial \nsuppression \nSuhonen 71 1995 RCT  36 \npostmenopau\nsal \n16 LNG-IUS and one \nsubdermal estrogen \nimplant vs \n20 LNG-IUS and three \nsubdermal estrogen \nEndometrial \nsuppression \nin all cases \n \n72% had \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   220 \n \nimplants \n \nNo drop out reported \namenorrhoea \nor spotting by \nthree months \nSuhonen 72 1995 RCT 19 \npostmenopau\nsal \n10 oral estrogen and \nLNG-IUS vs. 9 oral \nestrogen and subdermal \nlevonorgestrel-\nreleasing implant  \n \nNo drop out reported \nComparable \nendometrial \nsuppression \nSuvanto-Luuk. \n73-75 \n1997, \n1998, \n1999 \nProspectiv\ne cohort \n \n \n60 \npostmenopau\nsal women \n \n \n20 received \nLNG-IUS  \n \n21 oral \nprogesterone \n \n19 vaginal \nprogesterone \n \nAll received \ntransdermal estrogen \ngel \n \n25% drop out rate of \nLNG-IUS group at 5 \nyears \n5 year  \nfollow up for \n20 cases in \nLNG-IUS \ngroup \n \nAt one year \nvarying \ndegrees of \namenorrhoea: \n80%, LNG-\nIUS; 67%, \noral \nprogesterone; \n53% in the \nvaginal \nprogesterone. \n \nAt five years \n80% \namenorrhoea \nin LNG-IUS \n \nEndometrial \nsuppression \n(histological, \nultrasound) in \nall LNG-IUS  \ncases \nAntoniou 76 1997 Prospectiv\ne cohort \n56 \npostmenopau\nsal women \n28 women with LNG-\nIUS plus daily \ntransdermal estrogen \nComparable \nendometrial \nsuppression \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   221 \n \nwith \nurogenital \nsymptoms \nvs. 28 women with \nestradiol-releasing \nvaginal ring plus \nvaginal progesterone \n(ultrasound) \nKalogirou 77 1996 Prospectiv\ne cohort \n56 \npostmenopau\nsal  \nLNG-IUS and \ntransdermal estrogen \nvs. \nEstrogen releasing \nvaginal ring and oral \nprogestogen \nComparable \nendometrial \nsuppression \n(ultrasound \nand \nhistological) \nSturdee 3 2004 Prospectiv\ne  \n294 \npostmenopau\nsal  \n \n \nNo comparison \nLNG-IUS 10mcg/24hr \n(MLS device) and \ntransdermal estrogen \n \nInterim 1 \nyear results \nfrom 3 yr \nstudy \n \n67% \namenorrheic  \nat one year. \n9/294 \ndiscontinued \nbecause of \nbleeding. \nWildemeersch \n78 \n2003 Prospectiv\ne  \n83 \nperimenopaus\nal and 58 \npostmenopau\nsal \n \n* Mixed \ngroup of \nwomen-\ncontraception \nneeds, \nmenorrhagia, \nvasomotor \nsymptoms, \nfibroids  \nNo comparison  \nUsed FibroPlant-LNG \nwith transdermal \nestrogen gel \n \nUp to 3 year \nfollow up \n \nAll effective \nendometrial \nsuppression \n(ultrasound) \n \n64% \namenorrhoea \nin \nperimenopau\nsal group and \n100% in \npostmenopau\nsal group \n \n5 cases of \nfibroid \nrelated \nmenorrhagia \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   222 \n \nimproved \n \n \nHampton 79 2005 Prospectiv\ne \n82 \nperimenopaus\nal  \nNo comparison \nUse LNG-IUS with \noral estrogen \n5 year follow \nup \n96-98% non-\nproliferative \nendometrium \n55% \namenorrhoea \nat one year \n93% \namenorrhoea \nby fifth year \n \n80 per 100 \nwomen \ncontinuation \nrate at 5 years \nVarila 80 2001 Prospectiv\ne \n  \n \n40 \npostmenopau\nsal \n \n \nNo comparison \nUsed LNG-IUS with \noral or transdermal \nestrogen \n \n5 year follow \nup \n39 completed \n12 mths \n29 completed \n5 years \n \nAll cases had \nendometrial \nsuppression \n(histological \nand \nultrasound) \n51% \namenorrhoea \nor only \nspotting at 5 \nyears  \nWildemeersch 81 2000 Prospectiv\ne  \n22 \nperimenopaus\nal, 8 \npostmenopau\n19 cases had FibroPlant \nLNG 14mcg/24hr and \n11 cases had \nUp to 2½ \nyears follow \nup \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   223 \n \nsal 10mcg/24hr doses \nAll with transdermal \nestrogen gel \nAll effective \nendometrial \nsuppression \n(ultrasound) \n \n77% \namenorrhoea \nin \nperimenopau\nsal group and \n100% in \npostmenopau\nsal group \nSuhonen 82 1997 Prospecti\nve  \n \n29 peri- and \npostmenopau\nsal women \nNo comparison \nLNG-IUS and \ntransdermal/subdermal/\noral estrogen \n \n3 year follow \nup \nAll cases had \nendometrial \nsuppression \n(ultrasound, \nhistology)  \n79% \namenorrhoea \nat 3 years \nWildemeersch 83\n  \n2004 Prospectiv\ne  \n 24 \npostmenopau\nsal women \nNo comparison  \nUsed FibroPlant-LNG \nwith oral estradiol or \nestrogen patches \n \n3 year follow \nup \nAll effective \nendometrial \nsuppression \n(histologicall\ny and \nultrasound) \nand clinical \namenorrhoea  \n \n \n Footnotes \nFibroPlant-LNG is a frameless low-dose LNG-IUS (releasing 14mcg levonorgestrel/day)  \nMLS is a low dose smaller sized LNG-IUS (releasing 10mcg levonorgestrel/day) \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   224 \n \nEndometrial suppression and symptomatic improvement of menopausal symptoms (e.g. hot \nflushes) was achieved in all studies examining LNG-IUS use in women receiving ERT.  \nFrom the study outcomes, amenorrhoea appeared to be more common in postmenopausal \nwomen receiving LNG-IUS (studies ranging from 61% to 100% of subjects) than peri-\nmenopausal women (studies ranging from 38% to 83% of subjects), although this was not \nformally statistically tested due to study heterogeneity.  Seven studies have reported follow \nup beyond one year 66;75;78-83, three reported up to a maximum of five-years 75;79;80, and one \nstudy published its interim one year results from a proposed three year study duration 3.  \nThere was no statistically significant difference between LNG-IUS 10mcg and LNG-IUS \n5mcg in one RCT (n=108)67. Participants in three separate publications 78,81,83 are likely to be \nfrom the same study cohort. \n  \nEndometrial hyperplasia \nNo RCTs were identified.  Characteristics of the one cohort, three prospective observational \nstudies, and two case reports/case series are shown in Table 3.12.  Most studies examined \nwomen with non-aypical endometrial hyperplasia, but three studies have included women \nwith atypical hyperplasia 86;87;89.  Hyperplasia of all types was regressed in all cases treated \nwith LNG-IUS. \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   225 \n \nTable 3.12.  LNG-IUS studies assessing therapeutic effect in women with endometrial \nhyperplasia  \nAuthor Year of \nPublication \nStudy \nType \nSample Size Comparison  Outcomes within one year \nfollow up \nVereide84 2003 Retrospecti\nve cohort \n57 endometrial \nhyperplasia \nLNG-IUS vs. oral \nprogestogen  \nGreater regression with \nLNG-IUS that with oral \nprogestogens (100% vs. \n55%) at 3 months \nScarselli 85 1988 Prospective 31 (4 atypical \ntypes) \nNo comparison  Endometrial regression in \nall cases \nPerino 86 1987 Prospective 14 (1 case \natypical type) \nNo comparison Endometrial regression in \n29/31 cases at 16 months \nfollow up \nWildemeersch \n87 \n2003 Prospective  12 (non-atypical \nand atypical \ntypes) \nNo comparison Endometrial regression in \nall cases by three years \nRose 88 2001 Case report 1 No comparison Endometrial regression \nBahamondes 89 2003 Case report 1  No comparison Endometrial regression \n \nEndometrial cancer   \nThe preferred primary treatment for early stage endometrial cancer is surgical hysterectomy, \nwith systemic progestins used palliatively or as adjuvant treatments for higher stage cancers.  \nA literature review of limited sized case series and cohort studies (n=81 cases, 27 articles) has \nshown safe and effective treatment (overall 76% cure) with systemic progestin therapy in \nwomen with well differentiated stage 1 endometrial cancer 90.  This evidence, although \nlimited in quality, establishes a plausible role for LNG-IUS in early stage disease, particularly \nin those women medically unfit for surgical therapy.  One case report describes successful \nreversion of the cancer on endometrial biopsy when using a combination of oral progestogens \nand LNG-IUS in such an indication 91.  However, another case series (2 patients) showed no \nregression of the endometrial cancer when treated with LNG-IUS alone in patients awaiting \ndefinitive surgical treatment 89.  A comparative study performed in 14 women with early \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   226 \n \nstage endometrial cancer considered high risk for surgery showed successful reversion of \ncancer on endometrial biopsy in 75% of cases at 12 months 92.  However, a case series has \nidentified two cases of endometrial carcinoma that were diagnosed following insertion of \nLNG-IUS 93.  Clearly, further cases, controlled trials, and longer follow up are required in \norder to obtain more valid conclusions. \n \nDysmenorrhoea and pain Only one observational study has formally examined the \ntherapeutic use of LNG-IUS in women with primary and secondary dysmenorrhoea 32. The \nstudy is of poor quality (limited sample size, n=18, and non-comparative) which makes \ninterpretation of the observed beneficial response difficult.  However, reductions in \ndysmenorrhoea have been reported in numerous LNG-IUS trials 26;35;94-97 and observational \nstudies 28;98;99, albeit not being an a priori primary outcome measure in the vast majority. \nAn RCT (n=236) that compared LNG-IUS with hysterectomy for women with menorrhagia \nevaluated pain as an outcome using a RAND-36 health survey 13;14. The trial showed greater \nimprovement in pain by the hysterectomy group than LNG-IUS at one year. However, by five \nyears, both LNG-IUS and hysterectomy groups had achieved almost identical reductions in \npain. Most studies have failed to distinguish dysmenorrhoea from co-existent pelvic pain \ndisorders (e.g. endometriosis, chronic pelvic pain, chronic pelvic inflammatory disease) in \ntheir subgroup analyses. This may cause confounding. However, the fact the association is \nreproducible in so many studies suggests the effect is real even though the magnitude cannot \nbe accurately ascertained. \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   227 \n \n LNG-IUS and effect on pelvic inflammatory disease \nNo RCTs have examined whether the incidence of pelvic inflammatory disease (PID) is \nmodified following introduction of LNG-IUS as a primary outcome measure.  One RCT 100 \nand reviews of the early LNG-IUS trials 101;102 has suggested a lowering of PID rates when \nusing LNG-IUS compared to Cu-IUCD.  Whereas, two early RCTs 35;103, a recent 5-year \nstudy 104,  and a systematic review 105 of all the contraceptive trials have shown comparable \nrates of PID during the use of the LNG-IUS or a copper IUD. \n \nOther non-contraceptive therapeutic indications Large multicentre studies have not \ndetected differences in cervical cytology or breast cancer incidence between copper IUD and \nLNG-IUS users, and non-users 35;101;102.  Long-term epidemiological studies are needed to \nconfirm this finding, and whether these may represent alternative therapeutic indications.   \n \nAdverse effects Irrespective of study design and indication all studies have reported \nadverse side effects following insertion of LNG-IUS, although a direct causal relationship to \nLNG-IUS cannot always be confirmed.  Around 15-20% of LNG-IUS users experience at \nleast one or more unwanted side effects5;106;107. The most frequent (around 10-15% of users) \nis unscheduled erratic menstrual bleeding, which usually occurs during the first 3-4 months \nfollowing LNG-IUS insertion but tends to subside thereafter. Erratic irregular menstrual \nbleeding is cited by women as the most common reason for discontinuing LNG-IUS \ntreatment.  During LNG-IUS use, 17.5% of women had a cyst at 6 months (diameter over \n3cm) and 21.5% at 12 months 55. The vast majority of these were asymptomatic and \nfunctional, and exhibited a high rate (94%) of spontaneous resolution by six months55. Other \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   228 \n \nless common side effects include mastalgia, migraine, acne, weight gain, oedema, labile \nmood, abdominal pain, pelvic pain, nausea and coil-related (infection, perforation, \nspontaneous expulsion) complications 35;101;102. Nevertheless, the continuation and patient \nsatisfaction rates in women using LNG-IUS for contraception remains over 75% 98;108-111. \n \nStudies conflict on whether the induction of amenorrhoea is considered a desired effect 98 or \nan unwanted side effect 5;35 that may lead to LNG-IUS discontinuation. This determination is \nbased on the individual‘s clinical symptomology pre-LNG-IUS insertion. Amenorrhoea \noccurs following LNG-IUS insertion in 20-60% of normally menstruating women using the \ndevice for contraception, between 50-75% in women with menorrhagia, and 61%-100% in \npostmenopausal women using the device to protect the uterus during estrogen replacement \ntherapy9;35;42;75;82;97;107;108;112.   \n \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   229 \n \nDiscussion  \n \nOur systematic review has shown strong evidence that LNG-IUS is effective in treating \nwomen with idiopathic menorrhagia and in providing uterine protection for women receiving \nestrogen replacement therapy or tamoxifen.  There is preliminary evidence that shows LNG-\nIUS may be therapeutic in women with fibroids, endometriosis, adenomyosis, endometrial \nhyperplasia, early stage endometrial cancer and dysmenorrhoea, and may reduce the risk of \npelvic inflammatory disease.  The grading of evidence is depicted in Table 3.6.  The \nincidence of adverse effects, in particular initial period of erratic menstrual bleeding, is \nunaffected by the indication for the use of LNG-IUS.  The incidence of amenorrhoea \nfollowing LNG-IUS insertion appears to be influenced by age and independent of underlying \ngynaecological pathology: the incidence is greater as the woman approaches her menopause.  \n \nThis review has been original in systematically collecting and presenting the data relating to \nLNG-IUS use in HRT, tamoxifen, endometrial hyperplasia, endometrial cancer, \nendometriosis and adenomyosis. The systematic search strategy employed was \ncomprehensive and methodological analysis followed standardized criteria.  This review has \nupdated and expanded on studies listed in the Cochrane database  40;41;113 and a previous \nrelated systematic review 11.  Our findings complement the recently published Cochrane \nprotocol on post-operative LNG-IUS in endometriotic surgery 113, and supplements the \nevidence reported in a Cochrane review of pre- and post-operative medical therapy for \nendometriotic surgery which had excluded LNG-IUS usage  114.  Our review has included \nrecent developments such as data from lower dose LNG-IUS devices currently under \ndevelopment (e.g. FibroPlant™-LNG) and health related quality of life assessments for \nwomen using LNG-IUS14.  \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   230 \n \nWe observed a general paucity of RCTs, varying study methodologies and outcome \nmeasures, which made the interpretation of study data difficult and prevented us from \nperforming a meta-analysis. We had intended to perform a systematic review of LNG-IUS, \nand instead this review is a narrative assimilation of the available literature. Furthermore, our \nsystematic search strategy may have missed relevant studies.  However, by maintaining a \nsensitive keyword search, contacting the manufacturer Schering for unpublished studies, and \nchecking registered clinical trials databases, we believe this loss has been minimized.  Apart \nfrom menorrhagia and HRT therapeutic indications, the published literature mainly consists \nof limited sample-sized (below 50 participants in LNG-IUS arm of study) non-controlled \nobservational studies with less than one year follow up, which although showing consistent \ntrends, are likely to be subject to information and selection biases. Consequently, no firm \nconclusions can be inferred from these studies (evidence grading C).  However, these studies \nmay provide a basis to estimate minimum numbers needed to be recruited to demonstrate \nclinically significant results in future therapeutic trials using LNG-IUS. \nThere is strong evidence demonstrating the efficacy, cost-effectiveness, and safety of LNG-\nIUS in menorrhagia. This evidence has been translated to clinical practice through recent \nlicensing (2001) of LNG-IUS for women with menorrhagia.   A similar abundance of RCTs, \ncohort and observational evidence, demonstrating efficacy and endometrial safety, exists for \nthe use of LNG-IUS in providing endometrial protection during estrogen replacement \ntherapy.   Research in to this modality of HRT has been abundant since its inception in the \nlate 1980s115;116. However, unlike menorrhagia, the license for HRT use has not been \nforthcoming in many countries, and was only awarded in 2005 by the UK.    \n \nThe Women‘s Health Initiative and Million Women Study, showed HRT use increased the \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   231 \n \nrisk of stroke, pulmonary embolism, and breast cancer, but decreased risk of hip fracture, \nwith no effect on coronary heart disease incidence 117-120.   Incidence of breast cancer was \nsignificantly increased for users of hormone replacement therapy containing estrogen only \n(1.30 [1.21-1.40]), estrogen-progestogen (2.00 [1.88-2.12]), and Tibolone (1.45 [1.25-1.68]), \nbut the magnitude of the associated risk was substantially greater for estrogen-progestogen \nthan for other types of HRT.   The reluctance to use LNG-IUS may be based on concerns that \nstable systemic levels of levonorgestrel (330-350 pg/mL)1 may be sufficient through its \nprogestogenic effect to promote tumourigenesis in the breast (particularly if given with \nexogenous estrogen) or blunt the anti-tumour effect of tamoxifen on the breast. Similarly, it is \nplausible to extrapolate the endometrial suppression data observed in the perimenopausal \nhormone replacement therapy, tamoxifen and endometrial hyperplasia studies, and \nhypothesize that the risk of endometrial cancer may be reduced in long-term users of LNG-\nIUS. However, we found no data relating LNG-IUS use to an increased or decreased risk of \nbreast or endometrial cancer risk. However, absence of publications showing association does \nnot necessarily indicate a lack of association between LNG-IUS and cancer. We believe this \nto be an important safety issue that remains to be addressed, either through long-term follow \nup and re-analysis of published studies or further prospective trials.  \n \nDespite promising findings, further trials are needed to establish efficacy, safety, cost-\neffectiveness, and quality of life measures before recommending LNG-IUS in most of the \nnon-contraceptive indications discussed.   Studies need to identify which population groups \nbenefit most from LNG-IUS use, and this is made difficult due to the varying spectrum of \ndisease, co-existence of multiple gynaecological pathology, and whether LNG-IUS is being \ntested as a first line or second line treatment following failed medical or surgical intervention.  \nFor example, subgroup analysis of trial data has shown that the magnitude of baseline \n\nChapter 3.2. Systematic review of LNG-IUS (Mirena) \n   232 \n \nmenstrual blood loss was negatively predictive of successful treatment with LNG-IUS121. The \nauthors and manufacturers of the newer lower-dose and smaller sized LNG-IUS devices \nassert they are easier to insert, have less adverse side effects and greater patient acceptability \n3;67;67;68;83 than conventional 20mcg/24hr LNG-IUS. However, there is little supporting \nevidence for this assertion, and these devices need to be rigorously evaluated in robust head-\nto-head comparisons with conventional LNG-IUS to validate this viewpoint. \n \nThere is a paucity of data on patient preference and decision analysis strategies in the use of \nLNG-IUS 8.   This research should accompany future trials, particularly given the number of \ncompeting similar efficacy therapeutic medical and surgical interventions.   A recent \nquestionnaire study highlighted how patient‘s choice of treatment is influenced by several \nfactors.   These may include the likelihood of whether the treatment will be completely \nsuccessful, prolonged hospital stay and convalescence, and preservation of future fertility.   \nThe majority of women scheduled for an endometrial ablation or LNG-IUS for menorrhagia \nwere inclined to take a risk of 50% likelihood of treatment failure to avoid a hysterectomy 122. \nLNG-IUS can no longer just be considered suitable for women with menorrhagia who wish \nreversible contraception.   The fact that so many conditions in Gynaecology are likely to be \namenable to LNG-IUS underlies the importance of progestogens in the normal and \npathological female genital tract.   This review‘s findings complement the current resurgence \nof basic science research interest in this area and clinical trials evaluating potential \ntherapeutic use of selective progesterone receptor modulators in the conditions discussed in \nthis review123.   This review has provided a foundation to undertake robust research trials in \nthis area that could potentially show greater therapeutic benefit and lesser patient harm when \nusing LNG-IUS compared to currently available medical and surgical therapies.    \n \n\nChapter 4. Clinical Guideline Development \n233 \n \nChapter 4: CLINICAL GUIDELINE DEVELOPMENT  \nContents \nIntroduction    Aims of clinical guidelines and their value in clinical practice \nMethodology   Development and appraisal of clinical guideline methodology: \n   as illustrated through case studies. \nResults  Clinical guidelines for four topics (Table 4.1)- \nConclusion  Considerations that may improve guideline development process \nTable 4.1 Guideline publications arising from chapter 4: \nChapter Manuscript title Reference \n4.1 Varma, R, Gupta, J.K., Smith, G.C. Birth after previous caesarean section. Royal \nCollege of Obstetricians and Gynaecologists Clinical Green top guideline No.45. \nFebruary 2007. http://www.rcog.org.uk/index.asp?PageID=1913  \nVarma R, Smith GC. Management of women with previous caesarean section. In \nPress. In: Warren R, Arulkumaran S, editors. Best Practice in Labour and \nDelivery. Cambridge University Press, Cambridge, UK.; 2008. \n1 \n \n \n2 \n4.2 Varma R, Gupta JK. Ectopic Pregnancy. \nhttp://clinicalevidence.bmj.com/ceweb/conditions/pac/1406/1406_background.jsp \n. BMJ Clinical Evidence . 2006. \n3 \n4.3 Varma R, Gupta JK. Laparoscopic entry techniques: clinical guideline, national \nsurvey, and medicolegal ramifications. Surg Endosc 2008; 22(12):2686-2697. \n4 \n4.4 Varma R, Gupta JK. Failed sterilisation: evidence-based review and medico-legal \nramifications. BJOG 2004; 111(12):1322-1332. \nVarma R, Gupta JK.  Minimizing the risk of sterilization faliure: An evidence \nbased approach.  In: Complications in Gynecological Surgery.  Editor: \nO'Donovan P.  Spinger-Verlag, London 2008.  Chapter 12; pages 106-126 \n5 \n \n6 \n \n\nChapter 4. Clinical Guideline Development \n234 \n \nIntroduction \n \nClinical guidelines are designed to be educational aids that will promote Good Clinical \nPractice.  Guideline development and practice has become widespread in modern healthcare. \nIn the UK, both national bodies (National Institute of Clinical Excellence, Scottish \nIntercollegiate Guideline Network) and specialty based professional organisations (Royal \nCollege of Obstetricians and Gynaecologists) have active programmes of clinical guideline \ndevelopment and publication. There are many clinical topics that lend themselves to \nguideline development, although topics that have the greatest ‗clinical impact‘ are prioritised \nby guideline development bodies (Table 4i). The impetus for the continued proliferation of \nguidelines is the drive to ensure best clinical practice is achieved for both the patient (such as \ndesired clinical outcomes, reduction of clinical risk) and health care provider (optimum use of \nhealthcare resources and consideration of costs). Furthermore, although not tested, there are \nlikely to be medico-legal ramifications in cases where clinical harm has occurred and the \nclinician has not followed established clinical guidelines (either national or at local Trust \nlevel) or has not clearly justified their rationale for adopting alternative clinical decision \nmaking. There are established methodologies utilised in the production of clinical guidelines; \nfour essential criteria have been defined by the Appraisal of Guidelines for Research and \nEvaluation in Europe (AGREE) guidelines 7 and include: \n1. Systematic review of the literature  \n2. Graded recommendations with explicit links to the evidence  \n3. Input of a multidisciplinary working group  \n4. Quality control; for example, input by an independent advisory board or by \nindependent peer review.  \n\nChapter 4. Clinical Guideline Development \n235 \n \nTable  4i.  Assessment criteria for selecting topics for clinical guideline development: \nhigh clinical impact topics \nAssessment criteria  \n Areas where there are high rates of mortality, morbidity or disability.  \n Areas where improved standards of care would reduce rates of mortality, morbidity or \ndisability.  \n Areas where there is uncertainty, as evidenced by a wide variation in clinical practice and \nservice delivery.  \n Areas where new high-quality clinical evidence has been published.  \n Areas where there are resource implications: either high cost and low turnover or low cost \nand high turnover.  \n Areas where there are implications across the primary–secondary care interface.  \n Areas where there is a frequent chance of litigation \n \nHowever, not all clinical guidelines incorporate all of these criteria. Furthermore, concerns \nhave been raised on the ‗practical‘ value of clinical guidelines to real life clinical practice. \nGuidelines place considerable weight on the evidence originating from randomised controlled \ntrials. Nevertheless, in practice, there is considerable patient heterogeneity, the clinical \nenvironment is less well controlled, patient compliance is less reliable and resources are more \nrestricted. than the trial setting. Consequently, the anticipated benefits of the guideline may \nnot be fully realised in an everyday setting. There has been no robust research that has \ndemonstrated clear superiority of clinical guideline direct practice over conventional practice. \nAims of chapter \n To undertake systematic reviews and develop clinical guidelines in topics in obstetrics \nand gynaecology that are assessed to be of high clinical importance and impact (see earlier \ndefinition). Case examples selected are: Vaginal Birth after caesarean, Ectopic Pregnancy, \nLaparoscopic entry (Table 4.1). \n\nChapter 4. Clinical Guideline Development \n236 \n \n To explore the value of utilising differing methodological approaches to clinical \nguideline development (RCOG, SIGN, GRADE approaches) (Tables 4ii, 4iii, 4iv). 7-9 \n To identify if there are any potential improvements to the guideline development \nprocess based on appraisal of the guideline methodology. Evidence to justify improvements \nto be acquired through 1) the methodological and practical problems encountered during the \ncase examples and 2) any published evidence. \nTable 4ii.  Classification of evidence used by RCOG Guideline development (originate \nfrom US Agency for Health Care Research and Quality) 7 \nClassification of Evidence Levels  \nIa Evidence obtained from meta-analysis of randomised controlled trials.  \nIb Evidence obtained from at least one randomised controlled trial.  \nIIa Evidence obtained from at least one well-designed controlled study without \nrandomisation.  \nIIb Evidence obtained from at least one other type of well-designed quasi-experimental study.  \nIII Evidence obtained from well-designed non-experimental descriptive studies, such as \ncomparative studies, correlation studies and case studies.  \nIV Evidence obtained from expert committee reports or opinions and/or clinical experience of \nrespected authorities \nGrades of Recommendations  \n \nRequires at least one randomised controlled trial as part of a body of literature of \noverall good quality and consistency addressing the specific recommendation. \n(Evidence levels Ia, Ib) \n \nRequires the availability of well controlled clinical studies but no randomised \nclinical trials on the topic of recommendations. (Evidence levels IIa, IIb, III) \n \nRequires evidence obtained from expert committee reports or opinions and/or \nclinical experiences of respected authorities. Indicates an absence of directly \napplicable clinical studies of good quality. (Evidence level IV) \nGood Practice Point    \n \nRecommended best practice based on the clinical experience of the guideline  \ndevelopment group \n \n\nChapter 4. Clinical Guideline Development \n237 \n \nTable 4iii.  Classification of evidence used by Scottish Intercollegiate Guidelines \nNetwork (SIGN) Grading System 9 \nLevels of evidence \n1++ High quality meta analyses, systematic reviews of RCTs, or RCTs with a very low risk of \nbias \n1+ Well conducted meta analyses, systematic reviews of RCTs, or RCTs with a low risk of \nbias \n1 - Meta analyses, systematic reviews of RCTs, or RCTs with a high risk of bias \n2++ High quality systematic reviews of case-control or cohort studies \nHigh quality case-control or cohort studies with a very low risk of confounding, bias, or \nchance and a high probability that the relationship is causal \n2+ Well conducted case control or cohort studies with a low risk of confounding, bias, or \nchance and a moderate probability that the relationship is causal \n2 - Case control or cohort studies with a high risk of confounding, bias, or chance and a \nsignificant risk that the relationship is not causal \n3 Non-analytic studies, e.g. case reports, case series \n4 Expert opinion \nGrades of recommendation \nA At least one meta analysis, systematic review, or RCT rated as 1++, and directly applicable to \nthe target population; or \nA systematic review of RCTs or a body of evidence consisting principally of studies rated as \n1+, directly applicable to the target population, and demonstrating consistency of results \nB A body of evidence including studies rated as 2++, directly applicable to the target population, \nand demonstrating overall consistency of results; or \nExtrapolated evidence from studies rated as 1++ or 1+ \nC A body of evidence including studies rated as 2+, directly applicable to the target population \nand demonstrating overall consistency of results; or extrapolated evidence from studies 2++ \nD Evidence level 3 or 4; or Extrapolated evidence from studies 2+ \nGPP Good practice points: Recommended best practice based on the clinical experience of the \nguideline development group \n \n\nChapter 4. Clinical Guideline Development \n238 \n \nTable 4.iv. GRADE approach 8 (http://www.gradeworkinggroup.org/index.htm)  \nThe Grading of Recommendations Assessment, Development and Evaluation (GRADE) \nGRADE: Quality of evidence \n The GRADE system classifies the quality of evidence in one of four levels: \n High quality— Further research is very unlikely to change our confidence in the estimate of effect \n Moderate quality— Further research is likely to have an important impact on our confidence in the \nestimate of effect and may change the estimate \n Low quality— Further research is very likely to have an important impact on our confidence in the \nestimate of effect and is likely to change the estimate \n Very low quality— Any estimate of effect is very uncertain \nEvidence based on randomised controlled trials begins as high quality evidence, but our confidence in \nthe evidence may be decreased for several reasons, including:  \n Study limitations  \n Inconsistency of results  \n Indirectness of evidence  \n Imprecision  \n Reporting bias.  \nAlthough observational studies (for example, cohort and case-control studies) start with a \"low \nquality\" rating, grading upwards may be warranted if the magnitude of the treatment effect is very \nlarge, if there is evidence of a dose-response relation or if all plausible biases would decrease the \nmagnitude of an apparent treatment effect.  \nGRADE: Strength of recommendation \nThe GRADE system offers two grades of recommendations: \"strong\" and \"weak\" depending on \nwhether effects of intervention clearly outweigh the undesirable effects, or clearly do not. If trade-offs \nare less certain—either because of low quality evidence or because evidence suggests that desirable \nand undesirable effects are closely balanced—weak recommendations become mandatory. \nFactors that affect the strength of a recommendation  \nFactor Examples of strong \nrecommendations \nExamples of weak \nrecommendations \nQuality of evidence Many high quality randomised \ntrials have shown the benefit of \ninhaled steroids in asthma \nOnly case series have examined the \nutility of pleurodesis in pneumothorax \n\nChapter 4. Clinical Guideline Development \n239 \n \nUncertainty about the \nbalance between \ndesirable and \nundesirable effects \nAspirin in myocardial infarction \nreduces mortality with minimal \ntoxicity, inconvenience, and cost \nWarfarin in low risk patients with \natrial fibrillation results in small \nstroke reduction but increased \nbleeding risk and substantial \ninconvenience \nUncertainty or \nvariability in values and \npreferences \nYoung patients with lymphoma \nwill invariably place a higher \nvalue on the life prolonging effects \nof chemotherapy than on treatment \ntoxicity \nOlder patients with lymphoma may \nnot place a higher value on the life \nprolonging effects of chemotherapy \nthan on treatment toxicity \nUncertainty about \nwhether the \nintervention represents \na wise use of resources \nThe low cost of aspirin as \nprophylaxis against stroke in \npatients with transient ischemic \nattacks \nThe high cost of clopidogrel and of \ncombination dipyridamole and aspirin \nas prophylaxis against stroke in \npatients with transient ischaemic \nattacks \n \n\nChapter 4. Clinical Guideline Development \n240 \n \nTables 4v \nSummary of evidence for each clinical guideline according to RCOG  and \nGRADE guideline development tools 7; 8;9 \nAll tables exclude recommendations that have been generated from Level IV Evidence (absence of  \ndirectly applicable clinical studies of good quality; evidence generated from committee reports or \nexpert opinion). \nChapter 4.1 Birth after previous caesarean RCOG  \nLevel of \nEvidence \nGRADE \nQuality of \nEvidence \nGRADE Strength of \nRecommendation \nWomen with a single previous caesarean section and \nuncomplicated pregnancy may be offered VBAC \nIIb, III Moderate Weak \nWomen with previous uterine rupture, classical \ncaesarean, two previous caesarean sections, should not \nbe offered VBAC \nIII Very Low Weak \nThe probability of successful planned VBAC is around \n75% \nIIb Moderate Strong \nThe probability of uterine scar rupture during planned \nVBAC labour is around 0.5% \nIIb Moderate  Strong \nPlanned VBAC may increase the risk of uterine \nendometritis and requirement for blood transfusion \nIIb Low Weak \nPlanned VBAC is associated with a 10 per 10,000 risk of \nantepartum stillbirth beyond 39 weeks and a 4 per \n10,000 risk of delivery related perinatal death \nIIb, III Moderate Weak \nPlanned VBAC carries an 8 per 10,000 risk of the infant \ndeveloping hypoxic ischaemic encephalopathy (HIE) \nIIb Low Weak \nPlanned VBAC reduces the risk of neonatal respiratory \nafter birth: rates are 2 to 3% with planned VBAC and 3 \nto 4% with ERCS. \nIIb, III Low Weak \nThe risk of subsequent placenta praevia and accreta is \nlinearly associated with the number of previous \ncaesarean deliveries \nIIb, III Moderate Strong \nIn women with previous caesarean delivery, there is a 2 \nto 3-fold increased risk of uterine rupture and around \n1.5-fold increased risk of caesarean section in induced \nand/or augmented labours compared to spontaneous \nlabours \nIIb, III Very Low Weak \n\nChapter 4. Clinical Guideline Development \n241 \n \n \nChapter 4.2 Ectopic pregnancy RCOG \nLevel of \nEvidence \nGRADE \nQuality of \nEvidence \nGRADE  \nStrength of \nRecommendation \nSalpingectomy in women not desiring future \nfertility is beneficial compared to \nsalpingotomy or methotrexate in achieving \nprimary treatment success \nIIa, IIb Moderate Strong \nProphylactic methotrexate (systemic) \nfollowing salpingotomy compared to \nsalpingotomy alone is beneficial in reducing \nthe risk of persistent trophoblast \nIb, IIa Moderate Strong \nIn women desiring future fertility, systemic \nmethotrexate (single or multiple dose) and \nsalpingotomy achieve similar primary \ntreatment success and subsequent fertility \noutcomes \nIa, Ib, IIa Moderate Strong \nIn women desiring future fertility, there is \nmarginally improved subsequent fertility rate \nby performing salpingotomy compared to \nsalpingectomy \nIII Very Low Weak \nSingle dose methotrexate may result in higher \nrates of treatment failure in women with \nectopic pregnancies compared with multiple \ndose regimens.  \nIa, Ib, IIb Low Weak \nIn selected cases, expectant management has \nsimilar primary treatment success and future \nfertility outcomes to salpingectomy or \nsalpingotomy \nIII Very Low Weak \nMethotrexate plus mifepristone is no more \neffective at increasing treatment success rates \ncompared with methotrexate alone but it \nseems this combination may be more effective \nin increasing treatment success rates in women \nwith high levels of progesterone. \n  \nIb Moderate Weak \n \n\nChapter 4. Clinical Guideline Development \n242 \n \n \nChapter 4.3 Safe Laparoscopic Entry RCOG \nLevel of \nEvidence \nGRADE \nQuality of \nEvidence \nGRADE Strength \nof \nRecommendation \nIn high risk women (previous abdominal surgery; obesity, \nextremely thin or known abdominal adhesions), an \nalternative to close umbilical entry (e.g. Palmer‘s point or \nopen (Hasson) technique) may reduce the risk of \nlaparoscopic entry related injury  \nIIb,III Very Low Weak \nThe Veress needle should be inserted at the deep umbilical \npit, at 90º to the skin, with or without stabilising or \nelevating the umbilical sheath/fascia or anterior abdominal \nwall.  \nIIb, III Low Weak \nA safety check of correct Veress placement is most reliably \nachieved by using a Veress Intra-Abdominal Pressure (IAP) \nof less than 10mmHg. \nIIa Moderate Weak \nA safety check of intra-abdominal pressure of at least \n25mmHg should preceded vertical insertion of the primary \ntrocar \nIIa, IIb Moderate Weak \nSecondary trocars should be inserted under direct \nvisualisation \nIII Moderate Strong \n \nChapter 4.4 Preventing sterilisation failure RCOG \nLevel of \nEvidence \nGRADE \nQuality of \nEvidence \nGRADE Strength \nof \nRecommendation \nPre-sterilisation pregnancy testing and ensuring the woman \nhas taken adequate contraceptive precautions prior to the \nprocedure \nIII Very low Strong \nSterilisation performed at the time of abortion or immediate \npost-partum period is associated with increased risk of \nfailure and regret compared to interval sterilisation  \nIII Very low Weak \nSterilisation performed by laparoscopy is equivalent to \nmini-laparotomy in terms of primary treatment success but \nis superior in terms of patient recovery and shorter \noperative time \nIIb, III Low Weak \nLaparoscopic tubal occlusion using mechanical devices \nhave the lowest risk of sterilisation failure. \nIb, IIa Low Weak \nA second operating surgeon that counter checks the \nsterilisation procedure has been correctly performed may \nreduce the risk of sterilisation failure \nIIb Very Low Weak \nSterilisation failure occurred significantly earlier in \nnegligent than non-negligent failure mechanisms \nIIb Low Weak \n\nChapter 4.1 Birth after previous caesarean section \n243 \n \n4.1. Birth after previous caesarean section  \n \nAim  To provide evidence- based information to inform the care of women \nundergoing either planned vaginal birth after previous caesarean section (VBAC) or elective \nrepeat caesarean section (ERCS).  \nIntroduction and background There is widespread public and professional concern \nabout the increasing proportion of births by caesarean section 10. Increasing rates of primary \ncaesarean section have led to an increased proportion of the obstetric population who have a \nhistory of prior caesarean delivery. Pregnant women with a previous section may be offered \neither planned VBAC or ERCS. The proportion of women who decline VBAC is, in turn, a \nsignificant determinant of overall rates of caesarean delivery 11-14 . New evidence is emerging \nto indicate that VBAC is not as safe as originally thought 15;16. These factors, along with \nmedico-legal fears, have led to a recent decline in clinicians offering, and women accepting, \nplanned VBAC in the UK and North America 11-14 . This guideline presents the best available \nevidence to facilitate antenatal counselling in women with prior caesarean delivery and \nintrapartum management of women undergoing planned VBAC. Prior to this guideline, the \nNICE/RCOG Caesarean Section guideline (April 2004) provided the only UK generated \nguidance on the management of childbirth after caesarean 17. Our guideline supports the \nrecommendations made in the NICE/RCOG Caesarean Section guideline but addresses \nVBAC in more detail. \nIdentification and assessment of evidence Electronic searches were performed in \nMEDLINE (Ovid version 1996-October 2006), EMBASE (Ovid version 1996-October 2006) \nusing relevant medical subject headings and text words. Evidence based reviews and \nguidance from ACOG 18;19, SOGC 20 , ARHQ USA 21, New Zealand Guidelines Group 22 and \n\nChapter 4.1 Birth after previous caesarean section \n244 \n \nThe Cochrane Library (2006) 23 were identified and used in the development of this \nguideline. The definitions of the types of evidence used in this guideline originate from the \nUS Agency for Health Care Research and Quality (Table 4.ii)7. Where possible, \nrecommendations are based on and explicitly linked to the evidence that supports them. Areas \nlacking evidence are highlighted and annotated as ‗Good Practice Points‘. The definition of \nthe terms used in this guideline is shown in Tables 4.2 and 4.3. \nLimitations of data used in guideline Presently, there are no published randomised \ncontrolled trials (RCTs) comparing planned VBAC against planned ERCS.  Evidence for \nthese interventions is obtained mainly from retrospective non-randomised studies 1 making \ntheir conclusions less reliable. However, a study by the National Institute of Child Health and \nHuman Development (NICHD) Maternal–Fetal Medicine Units Network15 has overcome \nsome of the shortcomings of previous studies by combining a large sample size, a prospective \ncohort design and utilisation of standardised definitions for assessing outcomes. Where \npossible, data on various risks and benefits of VBAC and ERCS reported in this chapter \noriginate from this study. Further robust data on maternal and infant health outcomes will \nbecome available following completion of the BAC trial (Birth After Caesarean) 24. \nOptions for Delivery: VBAC or ERCS Pregnant women with a history of previous \ncaesarean section may be offered either planned VBAC (vaginal birth after caesarean) or \nERCS (elective repeat caesarean section) for their delivery. Such women would have \nconsultant-led antenatal care and typically would follow an antenatal strategy that is depicted \nin Figure 4.1. \n \n\nChapter 4.1 Birth after previous caesarean section \n245 \n \nTable 4.2. Definition of obstetric terms \nPlanned VBAC \n \nPlanned VBAC (vaginal birth after caesarean) refers to \nany woman who has experienced a prior caesarean birth \nwho plans to deliver vaginally rather than by elective \nrepeat caesarean section (ERCS). \nSuccessful and unsuccessful \nplanned VBAC \nA vaginal delivery (spontaneous or assisted) in a woman \nundergoing planned VBAC indicates a successful VBAC. \nDelivery by emergency caesarean section during the \nlabour indicates an unsuccessful VBAC. \nUterine rupture  \n \nDisruption of the uterine muscle extending to and \ninvolving the uterine serosa or disruption of the uterine \nmuscle with extension to the bladder or broad ligament. \nUterine dehiscence  Disruption of the uterine muscle with intact uterine serosa \n \nTable 4.3. Definition of perinatal terms \nTerm perinatal mortality  \n \nCombined number of stillbirths (antepartum and \nintrapartum) and neonatal deaths (death of a live born \ninfant from birth to age 28 days) per 10,000 live births and \nstillbirths at or beyond 37 weeks gestation. Term perinatal \nmortality rates exclude deaths due to fetal malformation \nunless otherwise stated. \nTerm delivery-related perinatal \ndeath  \nCombined number of intrapartum stillbirths and neonatal \ndeaths per 10,000 live births and stillbirths at or beyond \n37 weeks gestation. Delivery-related perinatal mortality \nrates exclude antepartum stillbirths and deaths due to fetal \nmalformation unless otherwise stated. \nNeonatal respiratory morbidity  \n \nCombined rate of transient tachypnoea of the newborn \n(TTN) and respiratory distress syndrome (RDS). \n \n \n\nChapter 4.1 Birth after previous caesarean section \n246 \n \nFigure 4.1 Plan of care for singleton uncomplicated pregnancy with previous LSCS \n \nBooking Fetal\nanomaly scan\n39w: ERCS \n36w: assess and \ndecide mode of \ndelivery\n41w and no onset of \nlabour: assess and \ndecide mode of delivery \n(consider chance of \nVBAC success, priority \nattached to vaginal \nbirth and antepartum \nstillbirth risk) \nPlacental \nlocalisation\n40w20w 12w -16w\nProvide patient \ninformation \nleaflet on VBAC \nand ERCS \noptions\nRe-assessment \nof low lying \nplacenta\n32w 36w \n 41w\n36w to 41w: await\nonset of VBAC labour \n41w to 42w:\nERCS or Induction\n42w\nDetermining the mode of delivery For some women, the decision to attempt VBAC may \nbe very clear on the basis of their first antenatal visit. In such cases, it may be acceptable, \nfollowing thorough counselling, to have their next review in the consultant clinic post-dates, \nto discuss elective delivery in the event that they do not go into labour spontaneously. For all \nother women, it has been suggested that the final decision on mode of delivery should be \nestablished at a 36 week gestation antenatal visit. However, it would be prudent to at least \ndocument an initial preference by the woman at her hospital booking visit (12-16 weeks) \ntogether with provision of a patient information leaflet detailing VBAC and ERCS options. \nThis approach would provide her with the opportunity to consider her options and help guide \ndecision making should she go into labour prior to her 36 week review (Figure 4.1\n).  \n\nChapter 4.1 Birth after previous caesarean section \n247 \n \n Suitability for planned VBAC:  \nWomen with a prior history of one uncomplicated lower segment transverse caesarean \nsection, in an otherwise uncomplicated pregnancy at term, with no contraindication for \nvaginal delivery should be able to discuss her options for planned VBAC, and should \nalso be offered information about the alternative of a repeat caesarean section (ERCS). \n \nThere is limited evidence on whether maternal or neonatal outcomes are significantly \ninfluenced by the number of prior caesarean deliveries or type of prior uterine scar 15;25-29. \nNonetheless, due to higher absolute risks of uterine rupture or unknown risks, planned VBAC \nis contraindicated in women with:- \n Previous uterine rupture- risk of recurrent rupture is unknown 27;30. \n Previous high vertical classical caesarean section (200-900 per 10,000 risk of uterine \nrupture)- where the uterine incision has involved the whole length of the uterine corpus 27;30. \n More than two previous caesarean deliveries (reliable estimate of risks of rupture unknown)\n                                              Evidence Levels IIIb, III and IV \nHowever, it is recognised that in certain extreme circumstances (e.g. miscarriage, intrauterine \nfetal death), for some women in the above groups, the vaginal route (although risky) may not \nnecessarily be contraindicated. A number of other variants are associated with an increased \nrisk of uterine rupture. These include: women with a prior inverted T or J incision (190 per \n10,000 rupture risk) 15 and women with prior low vertical incision (200 per 10,000 rupture \nrisk) 15.                   Evidence Level IIa  \nThere is insufficient and conflicting information on whether the risk of uterine rupture is \n\nChapter 4.1 Birth after previous caesarean section \n248 \n \nincreased in women with previous myomectomy or prior complex uterine surgery31-33. \n                   Evidence Level III \nTherefore, women with a previous uterine incision other than an uncomplicated low \ntransverse caesarean section incision who wish to consider vaginal birth should be assessed \nby a consultant with full access to the details of the previous surgery.        Evidence Level IV \n \nWomen with a prior history of two uncomplicated low transverse caesarean sections, in \nan otherwise uncomplicated pregnancy at term, with no contraindication for vaginal \ndelivery who have been carefully counselled and selected, may be considered suitable \nfor planned VBAC. This should be a Consultant-led decision.                                \n                \n  \nA multivariable analysis of the NICHD study, showed that there was no significant difference \nin the rates of uterine rupture in VBAC with two or more previous caesarean sections (9/975, \n92 per 10,000) compared to women with a single previous caesarean section (115/16,915, 68 \nper 10,000) 34. However, the rates of hysterectomy (60 per 10,000 vs. 20 per 10,000) and \ntransfusion (3.2% vs.1.6%) were increased in the former group 34. These findings concur with \nother observational studies, which overall, have shown similar rates of VBAC success with \ntwo previous caesarean deliveries (VBAC success rates of 62%-75%) and single prior \ncaesarean delivery 26;35-37 . Therefore, provided the woman has been adequately counselled \nregarding these increased risks and a comprehensive individualised risk analysis of the \nindication for - and the nature of - the previous caesarean sections has been undertaken, then \nplanned VBAC may be allowed in women with two previous low transverse caesarean \ndeliveries. This counselling process should be Consultant-led.  \nEvidence Levels IIa,IIb and III \n\nChapter 4.1 Birth after previous caesarean section \n249 \n \nAntenatal counselling: \nThe antenatal counselling of women with a prior caesarean delivery should be \ndocumented in the notes. There should be provision of a patient information leaflet with \nthe consultation. \n           \n \nAll women who have experienced a prior caesarean birth should be counselled about the \nmaternal and perinatal risks and benefits of planned VBAC and ERCS when deciding the \nmode of delivery. The trade off between risks and benefits for VBAC and ERCS is highly \nindividualised. Women differ in the magnitude of risks they are willing to expose either \nthemselves or their unborn child to during delivery 38. For example, women who wish to \nminimize the risk of rare, but severe adverse outcome for their child may prefer ERCS in \npreference to VBAC. Conversely, there are many reasons why a woman might prefer to \nattempt vaginal birth and these may lead them to accept a small degree of risk to both \nthemselves and their infant during labour and to choose VBAC in preference to ERCS.  \n            Evidence Level IV \n \nThe risks and benefits should be discussed in the context of the woman's individual \ncircumstances, including her personal motivation and preferences to achieve vaginal birth or \nERCS, her attitudes towards the risk of rare but serious adverse outcomes, her plans for \nfuture pregnancies and her chance of a successful VBAC (principally whether she has \npreviously had a vaginal birth - see below). In addition, where possible, there should be \nreview of the operative notes of the previous caesarean to identify the indication, type of \nuterine incision and any peri-operative complications. Decision making should be a shared \n\nChapter 4.1 Birth after previous caesarean section \n250 \n \nprocess between the woman and her obstetrician. Items that should be discussed and \ndocumented during the consultation are listed in Tables 4.4 and 4.5, and are expanded on \nbelow. Decision aids and specific patient information literature may facilitate this process 39.  \n                 Evidence Levels II and IV  \nA final decision for mode of delivery should be agreed between the woman and her \nobstetrician before the expected/planned delivery date (ideally by 36 weeks gestation). \nHowever, as up to 10% of women scheduled for ERCS go into labour before the 39th week, it \nis good practice to have a plan for the event of labour starting prior to the scheduled date15.  \n                Evidence Level IIa \nTable 4.4. Items to be discussed when determining mode of delivery \nItems  \n \nSpecial considerations \n1 Her understanding of the maternal and \nperinatal risks and benefits of VBAC \ncompared to ERCS \nParticularly her attitude towards the risk of rare \nbut serious adverse outcomes. \n2 Any contraindications to VBAC \n \nAny complicating obstetric factors e.g. placenta \npraevia, fetal malpresentation, obstructing \ncervical fibroid, maternal medical disorders. \n \nAssessment of previous caesarean delivery and \nany peri-operative complications. A classical scar \nor more than two previous lower segment \nincisions or previous uterine rupture would be \nabsolute contraindications to VBAC.  \n3 The likelihood of a successful VBAC  \n \nParticularly if she has had a previous vaginal \nbirth or successful VBAC \n4 Her plans for future pregnancies \n \n \n5 Her personal preference and motivation  \nto achieve vaginal birth or ERCS \n \n\nChapter 4.1 Birth after previous caesarean section \n251 \n \nTable 4.5 Risks and Benefits of opting for VBAC or ERCS  \n ^Planned VBAC ERCS at 39 weeks \n \nMother \nBenefits \n \n72%-76% chance of successful VBAC \nIf successful, shorter hospital stay and \nconvalescence  \nIncreases likelihood that future \npregnancies may be delivered vaginally \nAble to plan to known delivery date \n**Lower risk of blood transfusion (1%) and \nendometritis (1.8%) \n*Essentially zero risk of uterine scar rupture \nNo risk of vaginal tears and no worsening of \npelvic floor support and continence mechanisms \nAble to be surgically sterilised at the same time \n \n \nMother \nRisks \n \n \n*Around 50 per 10,000 (0.5%) risk of \nuterine scar rupture-if occurs associated \nwith maternal morbidity and fetal \nmorbidity/mortality \n24-28% chance of emergency caesarean \n10-15% chance of instrumental delivery \nand/or perineal tear requiring suturing \n**Higher risk of blood transfusion (1.7%) \nand endometritis (2.9%) \n0.1%-2% risk of serious surgical complications \nsuch as injury to bladder \nLonger stay and convalescence   \nFuture pregnancies would require caesarean \ndelivery \nIncreased risk of surgical complications with \neach subsequent caesarean delivery due to \nadhesions, placental praevia/accreta \nInfant \nBenefits \n \n1% risk of transient respiratory morbidity \n \nAvoids the 10 per 10,000 prospective risk of \nantepartum stillbirth as delivery is undertaken at \ncommencement of 39th week \n1 per 10,000 (0.01%) risk of delivery-related \nperinatal death or hypoxic ischaemic \nencephalopathy (HIE) at delivery \n \nInfant \nRisks \n10 per 10,000 (0.1%) prospective risk of \nantepartum stillbirth beyond 39 weeks \nwhilst awaiting spontaneous labour \n4 per 10,000 (0.04%) risk of delivery-\nrelated perinatal death \n$8 per 10,000 (0.08%) risk of hypoxic \nischaemic encephalopathy (HIE) during \nlabour \n1-3% risk of transient respiratory morbidity \n[6% risk if delivery performed at 38 instead of 39 \nweeks] \n \n\nChapter 4.1 Birth after previous caesarean section \n252 \n \nFootnotes Table 4.5 \n^ The estimates of risk for adverse maternal or fetal events in VBAC are based on women \nreceiving continuous electronic monitoring during their labour. The relative and absolute \nrisks of such events in the absence of continuous electronic fetal monitoring are unknown. \n*Uterine rupture in an unscarred uterus is extremely rare at 0.5 to 2 per 10,000 deliveries, and \nthis risk is mainly confined to multiparous women in labour 60. \n**In the NICHD study there was no statistically significant difference between planned \nVBAC and ERCS groups in relation to hysterectomy (23 per 10,000 vs. 30 per 10,000), \nthromboembolic disease (4 per 10,000 vs. 6 per 10,000) or maternal death (17/100,000 vs. \n44/100,000)15 \n$Approximately half of the increased risk of HIE in planned VBAC arises due to the \nadditional risk of HIE caused by uterine rupture (4.6 per 10,000)15 \n \nWomen considering their options for birth after a single previous caesarean should be \ncounselled that overall, the chances of successful planned VBAC are 72%-76% \n \nIndividual studies report success rates of 72%-76% 15;16;40 for planned VBAC after a single \nprevious caesarean, which concurs with pooled rates derived by systematic and summative \nreviews [Table 4.5]  41-43.        Evidence Levels IIa  and IIb \n\nChapter 4.1 Birth after previous caesarean section \n253 \n \nA number of factors are associated with successful VBAC. Previous vaginal delivery, \nparticularly previous VBAC, is the single best predictor for successful VBAC and is \nassociated with an approximately 87%-90% planned VBAC success rate 29;44;45. Risk factors \nfor unsuccessful VBAC are induced labour, no previous vaginal delivery, body mass index \ngreater than 30 46-48 and previous caesarean for dystocia 29. When all these factors are present, \nsuccessful VBAC is achieved in only 40% of cases 29. There are numerous other factors \nassociated with a decreased likelihood of planned VBAC success 29;44;49-52: VBAC at or after \n41 weeks gestation; birth weight >4000g; no epidural anaesthesia; previous preterm \ncaesarean delivery; cervical dilatation at admission less than 4cm; less than 2 years from \nprevious caesarean delivery; advanced maternal age, non-Caucasian ethnicity, short stature \nand a male infant. Where relevant to the woman‘s circumstances, this information should be \nshared during the antenatal counselling process to enable the woman to make the best \ninformed choice.      Evidence Levels IIa, IIb and III \nThere is limited and conflicting evidence on whether the cervical dilatation achieved at the \nprimary caesarean for dystocia impacts on the subsequent VBAC success rate53;54. \nUnfortunately, the NICHD study was unable to address this concern as data relating to the \nlabour of the primary caesarean were not collected during the study29 \n                       Evidence Levels IIb and III \nSeveral pre-admission and admission based multivariate models have been developed to \npredict the likelihood of VBAC success 44;53;55-58 or uterine rupture 59 in planned VBAC. \nHowever, their usefulness in assisting women to make the decision about whether VBAC or \nERCS is the best choice in their personal situation remains to be determined.  \n                  Evidence Level IIb \n\nChapter 4.1 Birth after previous caesarean section \n254 \n \nWomen considering their options for birth after a previous caesarean should be \ncounselled that planned VBAC carries a risk of uterine rupture of 22 to 74 per 10,000. \nThere is virtually no risk of uterine rupture in women undergoing ERCS. \n \nUterine rupture in an unscarred uterus is extremely rare at 0.5 to 2 per 10,000 deliveries, and \nthis risk is mainly confined to multiparous women in labour 60. The NICHD study reported \nthe overall risk for symptomatic uterine rupture at term was 74 per 10,000 planned VBACs \n15. There was zero risk in women undergoing ERCS 15. Studies with differing methodological \ndesigns and definitions of scar rupture report similar estimates for risk of uterine rupture per \n10,000 planned VBACs: systematic and non-systematic reviews of 39 43, 43 61 and 62 41 ; \nretrospective studies of 22 62 , 33 63 , 35 64 and 65 40 per 10,000. For counselling purposes a \nmean uterine rupture risk of 50 per 10,000 may be utilised (as depicted in Table 4.5). \nAlthough a rare outcome, uterine rupture is associated with significant maternal and perinatal \nmorbidity and perinatal mortality (see below).   \nEvidence Levels IIa and IIb \nThere is limited evidence from a case control study that women who experienced both \nintrapartum and postpartum fever in their prior caesarean delivery were at increased risk of \nuterine rupture in their subsequent planned VBAC labour (OR 4.02; 95% CI 1.04-15.5)65 . \nThere is conflicting evidence on whether single-layer compared with double-layer uterine \nclosure may increase the risk of uterine rupture in subsequent planned VBAC 17;66.    \nEvidence levels IIb and \nIII \n\nChapter 4.1 Birth after previous caesarean section \n255 \n \nWomen considering their options for birth after a previous caesarean should be \ncounselled that planned VBAC compared to ERCS carries around 1% additional risk \nof either blood transfusion or endometritis. \n           \n  \nWomen undergoing planned VBAC compared to ERCS are at greater risk of blood \ntransfusion requirement (170 per 10,000 vs. 100 per 10,000) and endometritis (289 per \n10,000 vs. 180 per 10,000) )[Table 4.5] 15. There was no statistically significant difference \nbetween planned VBAC and ERCS groups in relation to hysterectomy (23 per 10,000 vs. 30 \nper 10,000), thromboembolic disease (4 per 10,000 vs. 6 per 10,000) or maternal death \n(17/100,000 vs. 44/100,000)15. The vast majority of cases of maternal death in women with \nprior caesarean section arise due to medical disorders (such as thromboembolism, amniotic \nfluid embolism, pre-eclampsia and surgical complications).                       Evidence Level IIa \nMaternal death due to uterine rupture in planned VBAC occurs in less than 1 in 100,000 \ncases in the developed world, and this estimate is based on information from case reports \n40;67.                     Evidence Level III \nThe increased risk of morbidity overall among women attempting VBAC is due to higher \nrates among women who attempt VBAC and are unsuccessful. The NICHD study 15 showed \nthat unsuccessful planned VBAC compared to successful VBAC is associated with an \nincreased risk of uterine rupture (231 per 10,000 vs. 11 per 10,000), uterine dehiscence (210 \nper 10,000 vs. 14.5 per 10,000), hysterectomy (46 per 10,000 vs. 14.5 per 10,000), \ntransfusion (319 per 10,000 vs. 116 per 10,000) and endometritis (767 per 10,000 vs. 116 per \n10,000). Similar trends were identified in a retrospective study from a Canadian dataset 40. \n                    Evidence Level IIa\n\nChapter 4.1 Birth after previous caesarean section \n256 \n \nWomen considering planned VBAC should be counselled that this decision carries a 2 to \n3 per 10,000 additional risk of delivery-related perinatal death compared to ERCS, but \nthat the absolute risk of such delivery-related perinatal loss is comparable to the risk for \nwomen having their first birth. \n \nIn the NICHD study 15, perinatal mortality at term was significantly greater among women \nhaving a planned VBAC than ERCS. Overall perinatal mortalities for planned VBAC vs. \nERCS respectively were 32 per 10,000 vs. 13 per 10,000 (RR 2.40, 95% CI 1.43 to 4.01) and \nperinatal mortalities after excluding fetal malformation were 24 per 10,000 vs. 9.3 per 10,000 \n(RR 2.52, 95% CI 1.37-4.62). The increased risk of perinatal mortality is largely attributable \nto the statistically significantly increased risk of antepartum stillbirth beyond 37 weeks in \nplanned VBAC compared to ERCS (19.6 per 10,000 vs. 8.0 per 10,000; RR 2.45, 95% CI \n1.27-4.72) in infants without fetal malformation. Approximately 43% of such stillbirths in \nplanned VBAC were at or after 39 weeks gestation (approximately 9 per 10,000 women \ndelivering at or after 39 weeks), and may have been prevented by ERCS at 39 weeks. A \nsimilar estimate was identified from analysis of a Scottish data set which showed that the \nabsolute risk of antepartum stillbirth at or after 39 weeks among women with one prior \ncaesarean section was 10.6 per 10,000 68.         \nEvidence Level IIa \nIn the NICHD study, rates of delivery-related perinatal death were 4 per 10,000 for planned \nVBAC and 1.4 per 10,000 for ERCS 15. A report of data for the whole of Scotland \ndemonstrated higher overall rates of delivery-related perinatal death associated with \nattempted VBAC of 12.9 per 10,000 whereas the risk of death associated with ERCS was \n\nChapter 4.1 Birth after previous caesarean section \n257 \n \ncomparable to the US study at 1.1 per 10,000 16. The reason for the higher rate of delivery-\nrelated deaths among women attempting VBAC in Scotland may reflect the fact that these \nwere population-based data whereas the US data were exclusively from tertiary centres. \nConsistent with this interpretation, a further study of data from Scotland demonstrated a \nlower risk of perinatal death due to uterine rupture in larger centres 64.      Evidence Level IIa \nAccepting the limitations of using these observational data, a reasonable summary is that \nplanned VBAC is associated with a 10 per 10,000 risk of antepartum stillbirth beyond \n39 weeks and a 4 per 10,000 risk of delivery related perinatal death (if conducted in a \nlarge centre) [Table 4.5]. It is likely that these risks can be reduced by ERCS at the start of \nthe 39th week, but direct evidence to support this is lacking. It may be helpful to emphasise to \nwomen that the absolute risks of delivery-related perinatal death associated with VBAC are \ncomparable to the risks for nulliparous women 16;69.            Evidence Level IIa \nWomen considering their options for birth after a previous caesarean should be \ncounselled that planned VBAC carries an 8 per 10,000 risk of the infant developing \nhypoxic ischaemic encephalopathy (HIE). The effect on the long term outcome of the \ninfant upon experiencing HIE is unknown. \n            \n \nThe incidence of intrapartum hypoxic ischaemic encephalopathy (HIE) at term is \nsignificantly greater in planned VBAC (7.8 per 10,000) compared to ERCS (zero rate)[Table \n4.5] 15. Approximately half of the increased risk in planned VBAC arises due to the \nadditional risk of HIE caused by uterine rupture (4.6 per 10,000)15. The definition used and \ndistribution of severity of HIE is not stated in the NICHD study 15. Severe neonatal metabolic \nacidosis (pH<7.00) occurred in 33% of term uterine ruptures 15. There is no information \n\nChapter 4.1 Birth after previous caesarean section \n258 \n \ncomparing long term outcome, such as cerebral palsy, associated with VBAC and ERCS. \nGiven that cerebral palsy following term birth is very rare (approximately 10 per 10,000) and \nonly 10% of cases are thought to be related to intrapartum events 70, appropriate analysis of \nthis question would require a scale involving hundreds of thousands of women. No adequate \nstudy has currently been reported.               Evidence Level IIa \nWomen considering their options for birth after a previous caesarean should be \ncounselled that attempting VBAC reduces the risk that their baby will have respiratory \nproblems after birth: rates are 2 to 3% with planned VBAC and 3 to 4% with ERCS. \n        \n  \nThree observational studies, pooling data from around 90,000 deliveries, have shown an \nincreased risk of neonatal respiratory morbidity (defined earlier) among term infants \ndelivered by elective caesarean (3.5%-3.7%) compared to vaginal delivery (0.5%-1.4%) 71-73. \nThe NICHD study 15 (n=30,352 deliveries) reported a similar trend in women with prior \ncaesarean section, where the incidence of TTN in ERCS vs. planned VBAC was 3.6% vs. \n2.6% (RR 1.40, 95% CI 1.23-1.59; NNT -98)[Table 4.5]. These rates concur with rates of \nTTN derived from a smaller data set that examined women with prior caesarean section (2 \nstudies, n=4,478 deliveries) of 2.4%-6% vs. 1.3%-3% 73;74 for ERCS vs. planned VBAC \nrespectively. The NICHD study did not report rates of RDS, however the smaller data set \nreported RDS as 0.4%-0.6% vs. 0%-0.05% for ERCS vs. planned VBAC respectively 73;74.    \nEvidence Level IIa \n\nChapter 4.1 Birth after previous caesarean section \n259 \n \nWomen considering ERCS should be counselled that delaying delivery by one week \nfrom 38 to 39 weeks reduces the risk of respiratory morbidity, but this delay may be \nassociated with a 5 per 10,000 risk of antepartum stillbirth. \n \nEvidence from observational studies 71-73 and a recently published trial 75 has shown a \nbeneficial effect on reducing respiratory morbidity by delaying elective caesarean section to \nat least 39 weeks. The trial reported respiratory morbidity was 11.4%, 6.2% and 1.5% at 37, \n38 and 39 weeks gestation respectively 75. Thus, delaying delivery by one week from 38 to 39 \nweeks enables around a 5 per 100 reduction in the incidence of respiratory morbidity, but this \ndelay may be associated with a 5 per 10,000 increase in the risk of antepartum stillbirth 68;69 . \n                 Evidence Levels Ib and IIa \nFurthermore, the trial 75 demonstrated an approximate 50% reduction in respiratory morbidity \n(for both TTN and RDS components) by administering prophylactic Betamethasone to \nwomen having elective caesarean deliveries beyond 37 weeks (steroid vs. control; 2.4% vs. \n5.1%; RR 0.46, 95% CI 0.23-0.93), and this treatment effect was still apparent at 39 weeks \n(steroid vs. control; 0.6% vs. 1.5%). However, it has been suggested that even a single course \nof antenatal steroids may have long term consequences for the baby 76 and therefore it may be \nsafer to delay ERCS until 39 weeks rather than give steroids and deliver at 38 weeks. The \nroutine use of prophylactic Betamethasone in ERCS is beyond the scope of this guideline.  \nEvidence Level Ib \n\nChapter 4.1 Birth after previous caesarean section \n260 \n \nWomen considering their options for birth after a previous caesarean should be \ncounselled that the risk of anaesthetic complications is extremely low, irrespective of \nwhether they opt for planned VBAC or ERCS. \n            \n  \nAnaesthetic procedure-related complications are extremely rare 77. Of the women undergoing \ncaesarean section (emergency and elective) in the NICHD study (n=37,142), 93% received a \nregional anaesthetic and only 3% of regional procedures failed. There was one maternal death \n(2.7 per 100,000) attributed to an anaesthetic problem (failed intubation) 78.  \n                    Evidence Level IIa \nWomen considering their options for birth after a previous caesarean should be \ncounselled that ERCS may increase the risk of serious complications in future \npregnancies.  \n       \n   Evidence Levels IIa, IIb & III \nWhen considering mode of delivery, women should be advised about the effect of their \ndecision on future pregnancies. The following risks significantly increase with increasing \nnumber of previous caesarean deliveries: \n Placenta praevia. Overall placenta praevia occurs in 0.5% of deliveries. However, \npraevia is present in 0.38%, 0.63% and 0.72% after single vaginal delivery, single \ncaesarean, and two consecutive caesareans, respectively 79. \n Placenta accreta. Overall placenta accreta between 0.25-2 per 1000 deliveries80. \nHowever, accreta is present in 0.24%, 0.31%, 0.57%, 2.13%, 2.33% and 6.74% of women \n\nChapter 4.1 Birth after previous caesarean section \n261 \n \nundergoing their first, second, third, fourth, fifth, and sixth or more caesarean deliveries, \nrespectively 81. The risk that placenta accreta coexists with placenta praevia is 3%, 11%, \n40%, 61%, and 67% for first, second, third, fourth, and fifth or more repeat caesarean \ndeliveries, diagnosed to have placenta praevia 81. \n Placental abruption. Overall placenta abruption occurs in 1% of deliveries. However, \nabruption is present in 0.74%, 0.95% and 1.06% after single vaginal delivery, single \ncaesarean, and two consecutive caesareans, respectively 79. \n Injury to bladder, bowel or ureter. A retrospective study of approximately 3000 \nwomen from Saudi Arabia showed a linear increase in the risk of bladder injury (0.3%, \n0.8%, 2.4%), with a history of two, three and five caesarean sections, respectively 82. \n Ileus \n Need for postoperative ventilation \n Intensive care unit admission \n Hysterectomy-required in 0.65%, 0.42%, 0.90%, 2.41%, 3.49% and 8.99% of women \nundergoing their first, second, third, fourth, fifth, and sixth or more caesarean deliveries. \n Blood transfusion (requiring 4 or more units) \n Duration of operative time and hospital stay. \nGiven the high absolute risks of serious complications, caesarean delivery of women with \nhigh numbers of previous caesarean sections requires the immediate availability of senior \nsurgical staff. \n         Evidence levels IIb and III \n\nChapter 4.1 Birth after previous caesarean section \n262 \n \nPlacenta praevia and accreta: preoperative investigations \nIt is widespread practice in the UK, and endorsed by a RCOG guideline83, that women \nidentified to have low lying placentas at the routine mid-pregnancy fetal anomaly scan should \nbe re-scanned in the third trimester. Provided the woman is asymptomatic (not bled), it is \nsuggested that re-scan be conducted at 32 or 36 weeks gestation depending on whether the \nmid-pregnancy scan suggested major or minor praevia, respectively (30). However, given the \nstrong association between placenta praevia, placenta accreta and prior caesarean birth, and \nthe importance of their pre-operative identification, then re-scan and placental localisation \nassessment should commence at 32 weeks (and repeated at 36 weeks) for women with prior \ncaesarean delivery. Furthermore, those women identified to have praevia (especially anterior \nplacenta praevia) should undergo further antenatal imaging (such as power amplitude \nultrasonic angiography, MRI or colour flow Doppler) to help clarify the risk of accreta 83;84. \n \nIdentification of placenta accreta prior to delivery enables instigation of specific management \nstrategies to minimise adverse outcome at delivery. These include: consultant anaesthetist and \nobstetrician conducting the delivery; access to crossed matched blood; colleagues from other \nspecialties/subspecialties to be on standby to attend as needed; discussing the risk of \nhaemorrhage, transfusion and hysterectomy with the women as part of the consent procedure. \nIn addition, advance planning and consideration could be given to: prophylactic or \ntherapeutic uterine artery embolisation; internal iliac artery ligation at the same time as initial \nsurgery; methotrexate treatment following delivery, and expectant management (placenta left \nin place at the end of the caesarean section)83;84. \nEvidence Level III and IV \n\nChapter 4.1 Birth after previous caesarean section \n263 \n \nPlanned VBAC in special circumstances \nWomen who are preterm and considering their options for birth after a previous \ncaesarean should be counselled that planned preterm VBAC has similar success rates to \nplanned term VBAC but with a lower risk of uterine rupture. \n            \n \nA retrospective cohort study showed women who were preterm (24-36 weeks gestation) and \nundergoing planned VBAC had higher success rates when compared with term patients \nundergoing planned VBAC (82% vs. 74%) and non-significantly lower risks of uterine \nrupture 85. The prospective NICHD study showed planned VBAC success rates for preterm \nand term pregnancies were similar (72.8% vs. 73.3%), however, the rates of uterine rupture \n(34 per 10,000 vs. 74 per 10,000, respectively) and dehiscence (26 per 10,000 vs. 67 per \n10,000, respectively) were significantly lower in preterm compared with term VBAC 86. \nThromboembolic disease, coagulopathy and transfusion were more common in women \nundergoing preterm than term VBAC, although overall combined absolute risks were less \nthan 3% in the preterm VBAC group. Perinatal outcomes were similar with preterm VBAC \nand preterm ERCS 86. Therefore, following appropriate counselling and in a carefully \nselected population, planned VBAC may be offered as an option to women undergoing \npreterm delivery with a history of prior caesarean delivery.        Evidence Levels IIa and IIb \nA cautious approach should be adopted when considering planned VBAC in women \nwith twin gestation, fetal macrosomia and short inter-delivery interval as there is \nuncertainty in the safety and efficacy of planned VBAC in such situations. \n \n\nChapter 4.1 Birth after previous caesarean section \n264 \n \nStudy sample sizes are underpowered to provide reliable evidence suitable for any clinical \npractice recommendation in relation to twin gestation, fetal macrosomia and short inter-\ndelivery interval.  \n Twin Gestation: The NICHD study 87 (n=186 twins), US retrospective study 88 \n(n=535 twins) and a review 27 (7 studies, n=233 twins) have reported similar successful rates \nof VBAC in twin pregnancies to that in singleton pregnancies (65%-84%). However, a \npopulation based study reported a lower VBAC success rate (45%) but a comparable risk of \nuterine rupture (90 per 10,000) 89                 Evidence Levels IIa, IIb and III  \n Fetal Macrosomia: A review 27 of four retrospective studies, and the NICHD study \n29, has reported a significantly decreased likelihood of successful trial of VBAC for \npregnancies with infants weighing 4000g or more (55-67%) compared to smaller infants (75-\n83%). The risk of uterine rupture was reported in one of the retrospective studies to be only \nincreased in those who did not have previous vaginal delivery (relative risk, 2.3; P <.001)90 . \nA subgroup analysis of the NICHD study showed that women with previous caesarean \ndelivery for dystocia, greater birth weight in the subsequent planned VBAC labour relative to \nthe first birth weight decreased the likelihood of VBAC success 91. However, in reality, birth \nweight cannot be accurately predicted by antenatal ultrasound which limits the clinical \nusefulness of discussing these observations when counselling women for planned VBAC and \nERCS.        Evidence Levels IIa, IIb and III \n Short inter-delivery interval: Three observational studies of limited size 92-94 \nhave shown a two-to-three fold increased risk of uterine scar rupture for women with a short \ninter-delivery interval (below 12-24 months) from their previous caesarean section. In the \nNICHD study, women undergoing planned VBAC whose previous caesarean delivery was \nwithin 2 years of their labour had an increased risk of caesarean delivery compared to women \n\nChapter 4.1 Birth after previous caesarean section \n265 \n \nwhose labour was more than 2 years from their previous caesarean (32% vs. 25% \nrespectively)29 . Although this information is useful antenatally, it should also be shared with \nwomen postnatally to enable them to plan their preferred spacing intervals for subsequent \npregnancies.                      Evidence Levels IIa and III  \nIntrapartum support and intervention during planned VBAC \nPlanned VBAC should be conducted in a suitably staffed and equipped delivery suite, \nwith continuous intrapartum care and monitoring, and available resources for \nimmediate caesarean section and neonatal resuscitation. \n                  \n  \nObstetric, midwifery, anaesthetic, operating theatre, neonatal and haematological support \nshould be continuously available throughout planned VBAC and ERCS.   Evidence Level IV \nA retrospective study of Canadian data showed that the relative risk of uterine rupture when \ncomparing planned VBAC with ERCS increased two fold in low-volume obstetric units \n(<500 births per year) than high-volume (>500 births per year) units, even though lower \nvolume units had lower-risk obstetric population 40. A retrospective study of Scottish data \nshowed that planned VBAC in low-volume hospitals (<3000 births/year) was not associated \nwith an increased risk of uterine rupture overall but was associated with an increased risk of \nuterine rupture that led to perinatal death 64. It is likely that the availability of resources for \nimmediate delivery and neonatal resuscitation may reduce the risk of infant morbidity and \nmortality due to uterine rupture.               Evidence Level IIa \nEpidural anaesthesia is not contraindicated in planned VBAC. \n                        \n \n\nChapter 4.1 Birth after previous caesarean section \n266 \n \nIn the NICHD study, planned VBAC success rates were higher among women receiving \nepidural analgesia than those not receiving epidural analgesia (73.4% vs. 50.4%) 29. The \nauthors suggested that this difference may relate to the disproportionate use of spinal \nanaesthesia in short planned VBAC labours or opting for non-epidural analgesia in cases with \nnon-reassuring fetal well being.                Evidence Level IIa  \nA smaller observational study showed comparable rates of unsuccessful VBAC and operative \ndelivery in those women receiving epidural analgesia compared to those not receiving \nepidural, even when correcting for oxytocin usage 95.             Evidence Level III  \nFurthermore, concerns that epidural analgesia might mask the signs and symptoms associated \nwith uterine rupture were based on a single case report 96, and VBAC is not a \ncontraindication for epidural analgesia 77. A retrospective comparative study showed that \nwithin the planned VBAC group, infants of mothers who received epidural analgesia were \nmore likely to be subjected to diagnostic tests and therapeutic interventions (including sepsis \nevaluation and antibiotic treatment) compared to infants from a matched no-epidural \nanalgesia group 97.               Evidence Levels III and IV  \nMonitoring in Labour\n \nWomen should be advised to have continuous electronic fetal monitoring following \nonset of uterine contractions for the duration of planned VBAC.  \n            \n  \nAn abnormal CTG is the most consistent finding in uterine rupture and is present in 55%-\n87% of these events [Table 4.6] 61.                Evidence Level IIb \n\nChapter 4.1 Birth after previous caesarean section \n267 \n \nTable 4.6 Clinical features associated with uterine scar rupture \nAbnormal CTG \nSevere abdominal pain, especially if persisting between contractions \nAcute onset scar tenderness \nAbnormal vaginal bleeding or haematuria \nCessation of previously efficient uterine activity \nMaternal tachycardia, hypotension or shock \nLoss of station of the presenting part \n \nFootnotes Table 4.6. An abnormal CTG is the most consistent finding in uterine scar rupture \nand is present in 55%-87% of these events61. \nMoreover, continuous CTG is generally used among women during planned VBAC and thus \nthe estimates of risk of both lethal and non-lethal perinatal asphyxia associated with VBAC \nare in this context. The relative and absolute risks of severe adverse events in the absence of \ncontinuous electronic fetal monitoring are unknown.              Evidence Level IV \nContinuous intrapartum care is necessary to enable prompt identification and \nmanagement of uterine scar rupture. \n                    \n \nEarly diagnosis of uterine scar rupture followed by expeditious laparotomy and resuscitation \nis essential to reduce associated morbidity and mortality in mother and infant. There is no \nsingle pathognomic clinical feature that is indicative of uterine rupture but the presence of \nany of the factors listed in Table 4.6 occurring in the peripartum period should raise the \nconcern of the possibility of this event 30. The diagnosis is ultimately confirmed at emergency \ncaesarean section or postpartum laparotomy.           Evidence Levels III and IV \n\nChapter 4.1 Birth after previous caesarean section \n268 \n \nThere is insufficient evidence to support the use of intrauterine pressure catheters in the \nearly detection of uterine scar rupture \n \nObservational studies, with varying methodology and case mix, have shown intrauterine \npressure catheters may not always be reliable and are unlikely to add significant additional \nability to predict uterine rupture over clinical and CTG surveillance 98-100. Furthermore, \nintrauterine catheter insertion may be associated with risk 101. However, some clinicians may \nprefer to use intrauterine pressure catheters in special circumstances (e.g. in obese women to \nlimit the risk of uterine hyper-stimulation) - this should be a Consultant-led decision. \nEvidence Level III \nInduction and Augmentation\n \nParticular caution should be applied to women requiring induction or augmentation \nwith prior caesarean delivery. \n \nWomen should be informed of the 2 to 3-fold increased risk of uterine rupture and \naround 1.5-fold increased risk of caesarean section in induced and/or augmented \nlabours compared to spontaneous labours.  \n \n\nChapter 4.1 Birth after previous caesarean section \n269 \n \nThe risks of induction and/or augmentation should be weighed against the advantages \nof a successful VBAC, avoiding the risks that may occur whilst awaiting spontaneous \nlabour and avoiding the short and long-term risks of repeat caesarean delivery.  \n \nThere should be careful serial cervical assessments, preferably by the same person, for \nboth augmented and non-augmented labours, to ensure there is adequate cervicometric \nprogress thereby allowing the planned VBAC to continue.  \n \nThe decision to induce, the method chosen, the decision to augment with oxytocin, the \ntime intervals for serial vaginal examination, and the selected parameters of progress \nthat would necessitate discontinuing VBAC labour, should be Consultant-led decisions. \n \nThe risk of adverse maternal and perinatal outcomes are lower among women in spontaneous \nVBAC labour not requiring induction or augmentation (Table 4.7). Although augmentation \nand induction are not contraindicated in women with prior caesarean delivery, there remains \nconsiderable disagreement amongst clinicians on their use. Systematic reviews 102-105 \nexamining induction and augmentation of labour for women with previous caesarean birth \nhave found no RCTs comparing induction/augmentation in planned VBAC against ERCS. In \nthe NICHD study, the risks of uterine rupture per 10,000 planned VBACs were 102, 87 and \n36 per 10,000 for induced, augmented and spontaneous labour groups, respectively (Table \n4.7) 15. This compares to an overall risk of uterine rupture of 2 per 10,000 in women with \n\nChapter 4.1 Birth after previous caesarean section \n270 \n \nunscarred uteri, and this risk includes the combined risks of women undergoing induction, \naugmentation and spontaneous labour 60. In the NICHD study, the increased risk of uterine \nrupture after labour induction was found only in women with no prior vaginal delivery106. In \nthe NICHD study the rates of caesarean section in women undergoing planned VBAC were \n33%, 26% and 19% for induced, augmented and spontaneous labour groups respectively \n(Table 4.7) 29.                     Evidence Level IIa  \nProstaglandin vs. Non-Prostaglandin induction methods \nTwo studies have expanded on the differences in adverse outcomes between PG and non-PG \nbased induction regimens 15;64. In the NICHD study, PG induction compared to non-PG \ninduction incurred a non-significantly higher rupture risk of uterine (140 per 10,000 vs. 89 \nper 10,000; p=0.22) 15. In an analysis of nationally collected data from Scotland, PG \ninduction compared to non-PG induction was associated with a statistically significantly \nhigher uterine rupture risk (87 per 10,000 vs. 29 per 10,000) and a higher risk of perinatal \ndeath due to uterine rupture (11.2 per 10,000 vs. 4.5 per 10,000) 64. This compares to 6 per \n10,000 risk of perinatal death in women with an unscarred uterus induced by prostaglandin \nidentified by a Cochrane review 107.                 Evidence Level IIa  \nGiven these risks, and the absence of direct robust evidence, it is important not to exceed the \nsafe recommended limit for prostaglandin priming in women with prior caesarean delivery \n102. Moreover, due consideration could be given to restricting the dosaging and adopting a \nlower threshold of total prostaglandin dose exposure. Importantly, the decision to induce and \nthe method chosen (e.g. prostaglandin or non-prostaglandin methods such as intracervical \nFoley catheter) should be Consultant-led.               Evidence Level IV \n\nChapter 4.1 Birth after previous caesarean section \n271 \n \nTable 4.7. Risks of planned VBAC labours from NICHD study (N=17,898 planned \nVBACs) 15;29 \n Induced Augmented Spontaneous Overall All Planned \nVBAC s \nUterine \nrupture \nOverall \n102 per 10,000 \n(1.0%) \n87 per 10,000 \n0.9% \n36 per 10,000 \n0.4% \n69 per 10,000 \n0.7% \nPG method  \n140 per 10,000 \n1.4% \n   \n Non-PG method  \n89 per 10,000 \n0.9% \n   \nCaesarean \nsection 33% 26% 19% 27% \n \nTable 4.8. Management of augmentation in established VBAC labour \n Clinical management issues \n1 The decision for augmentation should follow careful obstetric assessment, maternal \ncounselling and be Consultant-led.  \n2 Oxytocin augmentation should be titrated such that it should not exceed the \nmaximum rate of contractions of 4 in 10 minutes. Particular caution is necessary \nwhen using high oxytocin augmentation doses as there is a ―\"dose response\" for \nmaximum oxytocin amount and uterine rupture. \n3 Careful serial cervical assessments, preferably by the same person, are necessary to \nshow adequate cervicometric progress, thereby allowing augmentation to continue. \nThese intervals should not exceed 4 hours. \n4 If there was less than 2 cm progress after 4 hours of oxytocin then caesarean \nsection should be considered. A more conservative threshold of inadequate \nprogress after 2 hours of augmentation may also justify consideration for caesarean \nsection depending on the woman‘s individual circumstances. \n5 If there was 2 cm or more progress, augmentation could be continued and vaginal \nexaminations performed 4-hourly. \n \n\nChapter 4.1 Birth after previous caesarean section \n272 \n \nPost dates induction \nThe RCOG Induction of labour guideline suggests induction for post dates be offered from \n41weeks as this reduces perinatal mortality without an increase in caesarean section rates. \nThere are no adequate data which directly address this issue among women with a previous \ncaesarean section. However, there are some specific issues about women with a previous \ncaesarean delivery which may influence the decision making process. First, these women are \nat increased risk of antepartum stillbirth 68;108. Hence, the reduction in risk of perinatal death \nassociated with post-dates elective delivery may be even greater among women with a \nprevious caesarean. However, it is also possible that the effect of routine post-dates induction \non the risk of emergency caesarean section may be different among women with a previous \ncaesarean delivery. These women have a higher background risk of emergency intrapartum \ncaesarean section and the risk of a failed VBAC is increased both post-dates and with \ninduction of labour. These issues lead some women to decide to attempt VBAC if they labour \nspontaneously prior to 41 weeks but to have a planned caesarean section if their pregnancy \nproceeds post-dates. The choice about the method of elective delivery post-dates will also be \ninformed by other factors determining the likelihood of a successful VBAC (favourable \ncervix and previous vaginal birth) and by the priority attached to achieving vaginal birth \n(such as plans for many future pregnancies).  \n \nThere is no direct evidence to recommend what is acceptable or unacceptable cervicometric \nprogress in women being augmented with a previous caesarean section 109-113. Amongst \nwomen with unscarred uteri, the NICE Intrapartum guideline defines delay in the established \nfirst stage of labour as cervical dilatation of less than 2 cm in 4 hours 114. For women with \nintact membranes, an amniotomy would then be recommended and repeat vaginal \n\nChapter 4.1 Birth after previous caesarean section \n273 \n \nexamination performed 2 hours later: if progress was still less than 1cm then diagnosis of \ndelay would be confirmed. If there was less than 2 cm progress after 4 hours of oxytocin, \nfurther obstetric review would be required to consider caesarean section. If there was 2 cm or \nmore progress, augmentation could be continued and vaginal examinations performed 4-\nhourly.  \nIf, in the presence of adequate (strength and frequency) uterine contractions, there is a \nslowing down of a previously normally progressing labour, augmentation may increase the \nrisk of uterine rupture. A small sized retrospective study suggested that early recognition and \nintervention for labour dystocia (specifically, not exceeding two hours of static cervicometric \nprogress) may have prevented a proportion of uterine ruptures among women attempting \nVBAC113. Awareness of the increased risk of uterine rupture in scarred uteri, particularly if \nthere is labour dystocia, implies that a more conservative threshold to the upper time limit \n(such as 2 hours instead of 4 hours) of oxytocin augmentation without progress may be \njustified. Furthermore, a retrospective multicentre study showed a \"dose response\" for \nmaximum oxytocin amount and uterine rupture, with a uterine rupture rate of 2.07% at the \nhighest dosages 115. Therefore, particular caution is necessary when using high oxytocin \naugmentation doses. A summary of the key management issues relating to augmented VBAC \nis shown in Table 4.8.  \nEvidence Level III \nThe key management issues relating to augmented VBAC labour are listed in Table 4.8, and \nalthough not based on robust evidence, are considered to be helpful in minimising additional \nharmful risks that are consequent to augmentation. When counselling women for induction \n(prostaglandin or non-prostaglandin methods) and/or augmentation clear information should \nbe provided on all potential risks and benefits of such a decision and how this may impact on \n\nChapter 4.1 Birth after previous caesarean section \n274 \n \nher long term health. For example, women who are contemplating many future pregnancies \nmay be prepared to accept the short-term additional risks associated with induction and/or \naugmentation in view of the reduced risk of serious complications in future pregnancies if \nthey have a successful VBAC.        \nEvidence Level IV \nAuditable standards \nStandards for audit of practice should include the following: \n Use of continuous electronic fetal monitoring during VBAC labour. \nStandards for audit of documentation could include the following:  \n Documented discussion of risks and benefits of VBAC and ERCS \n Documentation of Consultant involvement in:  \nDeciding to induce or augment labour  \nEstablishing a plan for induction or augmentation (e.g. preferred vaginal examination \ninterval, expected minimal cervicometric progress, and the criteria needed to discontinue \nlabour and proceed to emergency caesarean section). \nFuture research \n1. Development, validation and pragmatic clinical evaluation of a scoring system to identify \nwomen at high or low risk of unsuccessful VBAC that is antenatally and/or intrapartum \nbased. \n2. The clinical effectiveness of differing induction and augmentation regimens, perhaps \nindividualised according to clinical features rather than standardised strategies. \n3. Identify if there are differences in long-term maternal and infant outcomes between \nplanned VBAC and ERCS e.g. subfertility, depression, pelvic floor dysfunction, \nincontinence, psychosexual problems, respiratory illness, and neurodevelopmental \ndisorders (…this list is not exhaustive). \n\nChapter 4.1 Birth after previous caesarean section \n275 \n \n4. Investigate the aetiology and prevention (e.g. specific antenatal monitoring strategies) of \nthe increased risk of stillbirth in women with previous caesarean delivery, in the presence \nor absence of other previous complications (e.g. pre-eclampsia, preterm delivery, small \nfor gestational age) 68;116. \n5. Research in to factors that may explain the regional and unit-based variation in uptake of \nVBAC, and which factors impact most on women accepting or declining VBAC (e.g. \npatient information leaflet, previous childbirth experiences, desired family size, \nunderstanding the risk analysis during counselling, how to reduce any decisional conflict, \nvariation in case mix)117-127. \n6. Assess maternal satisfaction 128-130, quality of life measures and health-state utilities in \nwomen following VBAC and ERCS to undertake robust economic modelling \nassessments. \nPending relevant trials \n BAC Birth After Caesarean - Planned vaginal birth or planned caesarean section for \nwomen at term with a single previous caesarean birth. ISRCTN 53974531, Prof C \nCrowther, University of Adelaide, Australia24. \n The Twin Birth Study- a multicentre RCT comparing planned caesarean section with \nplanned vaginal birth for twins at 32-38 weeks gestation, ISRCTN 74420086, Dr J \nBarrett, Toronto, Canada \n DiAMOND-Decision Aids for Mode Of Next Delivery, ISRCTN 84367722, Dr A \nMontgomery, Bristol, UK \n CAESAR-Caesarean Section Surgical Techniques, ISRCTN 11849611, Dr P \nBrocklehurst, National Perinatal Epidemiology Unit, Oxford, UK \n \n\nChapter 4.2.Ectopic Pregnancy \n276 \n \n4. 2. What treatments improve outcomes in women with unruptured tubal \nectopic pregnancy? \n \nSUMMARY OF TREATMENT OPTIONS (see also Tables 4v)  \nBeneficial Salpingectomy in women not desiring future fertility \nLikely to be beneficial Prophylactic methotrexate (systemic) following salpingotomy \n \nSystemic methotrexate (single or multiple dose) \nUnknown effectiveness \n \nExpectant management of a subgroup of unruptured ectopic \npregnancies \n \nSalpingotomy compared to salpingectomy in the presence of a \nhealthy contralateral tube for those women desiring future fertility \n \nSalpingotomy in women with contralateral tubal disease who \ndesire future fertility \nUnlikely to be \nbeneficial \nSystemic methotrexate combined with mifepristone versus \nsystemic methotrexate alone \n \nKEY POINTS Approximately one in a hundred pregnancies are ectopic, with the \nconceptus usually implanting in the fallopian tube. Some ectopic pregnancies can resolve \nspontaneously, but others continue to grow and lead to rupture of the tube. \nRisks are higher in women with damage to the fallopian tubes due to pelvic infections, \nsurgery, or previous ectopic pregnancy or abortion, and in smokers. The intrauterine \ncontraceptive device does not increase the absolute risk, but a pregnancy that does occur with \nIUD use is more likely to be ectopic than intrauterine. \nExpectant management of unruptured ectopic pregnancies may lead to similar subsequent \nintrauterine pregnancy rates compared with surgery, but few studies have been done. \nOngoing surveillance is required as part of expectant management, but tubal rupture can \noccur despite falling beta hCG levels. \n\nChapter 4.2.Ectopic Pregnancy \n277 \n \nMethotrexate , as single or multiple dose regimens, seems to be as likely as salpingotomy to \nremove trophoblast material and leave a patent fallopian tube in women with non-invasive, \nsmall ectopic pregnancies with no tubal rupture or bleeding, no sign of fetal cardiac activity \nand low beta human chorionic gonadotrophin (hCG) levels. About 15–40% of ectopic \npregnancies may be suitable for such non-surgical management.  \nSystemic or intratubal methotrexate may also reduce persistent trophoblast after \nsalpingotomy.Adding mifepristone to systemic methotrexate seems unlikely to increase \ntreatment success compared with methotrexate alone, other than in women with higher \nprogesterone levels. \n \nDEFINITION Ectopic pregnancy is defined as a conceptus implanting outside the uterine \nendometrium. The most common implantation site is within the fallopian tube (95.5%), \nfollowed by ovarian (3.2%) and abdominal (1.3%) sites. The sites of tubal implantation in \ndescending order of frequency are ampulla (73.3%), isthmus (12.5%), fimbrial (11.6%), and \ninterstitial (2.6%).131 Population: In this systematic review, we will consider \nhaemodynamically stable women with unruptured tubal ectopic pregnancy, diagnosed by \nnon-invasive or invasive techniques. All terms used in this chapter are defined and  listed in \nTable 4.9 \n \n\nChapter 4.2.Ectopic Pregnancy \n278 \n \nTable 4.9 Glossary of terms used in ectopic pregnancy guideline \nβhCG is the pregnancy hormone beta-human chorionic gonadotrophin. \nContralateral tube denotes the opposite tube to that affected by the ectopic pregnancy. \nCompare with homolateral or ipsilateral tube. \nDiscriminatory zone denotes a serum hCG level at which it is assumed that all intrauterine \npregnancies will be visualised by transvaginal ultrasound. This may vary according to \nsonographic expertise but is often between 1000 and 1500 IU/L. \nExpectant management is where ectopic pregnancy treatment involves a watch and wait \npolicy in conjunction with close clinical, ultrasonographic, and serum hCG surveillance. \nFecundity rate ratio (FRR) The fecundity rate represents the probability of spontaneous \nintrauterine pregnancy (IUP) per time unit elapsed derived from analysing the cumulative \nprobability of pregnancy over the study duration. Only women trying to conceive are \nincluded in the calculation, and women who have conceived using additional treatments (e.g. \nIVF) are excluded up and till the start of their additional treatment. The fecundity rate ratio \n(FRR) is the ratio of fecundity between the test treatment (e.g. salpingotomy) against the \nreference treatment (e.g. salpingectomy). A significant treatment difference between \nsalpingotomy compared to salpingectomy is indicated if 1 is not included in the 95% CI for \nthe FRR of salpingotomy compared to salpingectomy. Thus a FRR of 1.9 for intrauterine \npregnancy indicates that the probability of intrauterine pregnancy is 90% higher with \nsalpingotomy than salpingectomy. \nFertility outcome reports the rates of subsequent intrauterine pregnancy, repeat ectopic \npregnancy, and live birth rate. Such pregnancies may either be spontaneous or achieved \nthrough assisted reproductive technology, and this should be stated clearly in the fertility \noutcome. Furthermore, fertility outcome rates differ according to the ectopic pregnancy \nassociated reproductive and pathological characteristics, and treatment method chosen. The \ndenominator will differ in those women who desire future fertility and who are trying to \nconceive compared to those women taking contraceptive measures. \nHomolateral or ipsilateral tube denotes the tube that is affected by the ectopic pregnancy. \nCompare with contralateral tube. \nPersistent trophoblast is defined as suboptimal falling, increasing, or plateauing serum \nhCG concentrations following initial ectopic pregnancy treatment for which additional \ntreatment (surgical or medical) is needed. This rarely occurs following salpingectomy, but \nmay arise following salpingotomy, methotrexate, or expectant management. \nPregnancy of unknown location is defined as absence of pregnancy localisation (either \nintrauterine or extrauterine) by transvaginal sonography when serum hCG levels are below \nthe discriminatory zone (1000–1500 IU/L). If there is an absence of pregnancy localisation \nwith the serum βhCG above the discriminatory zone then this, along with other clinical, \nultrasonographic, and serum βhCG features increases the likelihood of ectopic pregnancy. \nPrimary treatment success is defined as progressive decline of serum hCG to undetectable \nlevels following initial treatment without reintervention (surgical or medical) for persistent \ntrophoblast or supervening clinical sequelae (e.g. tubal rupture or worsening clinical \n\nChapter 4.2.Ectopic Pregnancy \n279 \n \nsymptoms). \nSalpingotomy is where the ectopic conceptus is removed from the affected tube through a \nlinear incision of the tube overlying the ectopic. This incision is not surgically closed and is \nallowed to heal through secondary intention. This surgical treatment conserves the affected \ntube. \nTreatment failure denotes the sum of the reintervention rates for persistent trophoblast and \nsupervening clinical sequelae (e.g. tubal rupture or worsening clinical symptoms). \nTubal excision or salpingectomy is defined as the surgical removal of the tube affected by \nthe ectopic pregnancy. \nTubal preservation is a treatment approach designed to preserve the tube affected by the \nectopic. This involves expectant, medical (e.g. systemic methotrexate) or salpingotomy \ntreatment approaches. \nTubal patency examines the homolateral tube for the passage of dye at hysterosalpingogram, \nor at second look laparoscopy, or the passage of contrast media at transvaginal ultrasound. \nOnly those cases that have been managed by tubal preservation, rather than salpingectomy, \nare eligible for tubal patency testing. \n \n\nChapter 4.2.Ectopic Pregnancy \n280 \n \nINCIDENCE/PREVALENCE  Around 10,000 ectopic pregnancies are \ndiagnosed annually in the UK. The incidence of ectopic pregnancy in the UK (11.0 per 1000 \npregnancies) is similar to other countries like Norway (14.9 per 1000) and Australia (16.2 per \n1000).132-134 Since 1994, the overall rate of ectopic pregnancy and mortality rate (0.4 per \n1000 ectopic pregnancies) has been static in the UK. 134 Until recently, most epidemiological \nstudies have failed to distinguish between ectopic pregnancies occurring in women who did \nnot use contraception (reproductive failure) and women who used contraception \n(contraceptive failure). 135;136 A French population study undertaken from 1992 to 2002 \nfound that, over the duration of the study, the rate of reproductive failure ectopic pregnancies \nincreased by 17%, whilst the rate of contraceptive failure ectopic pregnancies decreased by \n29%.136 Increasing rates of Chlamydia infection, smoking, and assisted reproductive \ntechnology usage may have contributed to the disproportionate increase in reproductive \nfailure ectopic pregnancy rate over contraceptive failure ectopic pregnancy rate. Widespread \nuse of dedicated early pregnancy assessment units and non-invasive diagnostic algorithms are \nlikely to have contributed to increasing rates of ectopic pregnancy diagnosis.137;138 \n \nAETIOLOGY/RISK FACTORS  The aetiology of ectopic pregnancy is unclear. \nEctopic pregnancy arising from reproductive failure or contraceptive failure should be \nconsidered separate entities with differing aetiology, risk factors and reproductive outcomes. \n135;136;139The main risk factors for reproductive failure are a history of pelvic inflammatory \ndisease, previous ectopic pregnancy, pelvic and tubal surgery, infertility, smoking, and \nassisted conception. 135;140The main risk factor for contraceptive failure ectopic is intrauterine \ncontraceptive device (IUD) failure. IUDs do not increase the absolute risk of ectopic \npregnancy, but a pregnancy occurring with IUD is more likely to be ectopic than intrauterine. \n\nChapter 4.2.Ectopic Pregnancy \n281 \n \nOther risk factors for ectopic include prior spontaneous abortion, prior induced abortion, \nendometriosis, uterotubal anomalies, and prior in utero exposure to diethylstilbestrol. \nHowever, less than half of the ectopic pregnancies diagnosed are associated with risk factors. \n141 \nPROGNOSIS OF ECTOPIC PREGNANCIES  As the pregnancy advances, tubal \npregnancies may either diminish in size and spontaneously resolve, or increase in size and \neventually lead to tubal rupture with consequent maternal morbidity and mortality. There are \nno reliable clinical, sonographic or biological markers (e.g. serum hCG or serum \nprogesterone) that can predict rupture of tubal ectopic pregnancy.142;143 Maternal mortality \nfollowing ectopic pregnancy is an uncommon short-term outcome in developed countries. \nThe recent UK Confidential Enquiry into Maternal Deaths cited ectopic pregnancy as a cause \nof 11 maternal deaths (0.4 per 1000 ectopic pregnancies). 134 Short-term maternal morbidity \nrelates to pain, transfusion requirement and operative complications.  \nPrimary treatment success and long-term fertility outcomes depend on the clinical \ncharacteristics of the ectopic pregnancy (e.g. whether the ectopic occurred in a woman using \ncontraception or not, tubal rupture or not, contralateral tubal disease) and the type of surgical \nor medical treatment chosen. A ten-year follow up of ectopic pregnancies showed the rate of \nrepeat ectopic pregnancy was much higher in women who had a IUD at the time of the index \nectopic pregnancy compared to women whose ectopic was not associated with IUD use. In \ncontrast, the rate of intrauterine pregnancy was 1.7-fold higher (Fecundity Rate Ratio 1.7, \n95% CI 1.3-2.3) in women who had a IUD at the time of the index EP compared to women \nwhose index ectopic was not associated with IUD use.139Short and long-term consequences \non health-related quality of life and psychological issues (e.g. bereavement) are also \nimportant, but are rarely quantified.  \n\nChapter 4.2.Ectopic Pregnancy \n282 \n \nPREGNANCIES OF UNKNOWN LOCATION  Pregnancy of unknown location is \ndefined as the absence of pregnancy localisation (either intrauterine or extrauterine) by \ntransvaginal sonography when serum hCG levels are below the discriminatory zone (1000–\n1500IU/L).  An observational study of pregnancies of unknown location has shown 55% \nspontaneously resolve, 34% are subsequently diagnosed as viable, and 11% are subsequently \ndiagnosed as ectopic pregnancies. 144  \n \nAIMS OF INTERVENTION \nShort-term: primary treatment success; to reduce maternal morbidity and mortality related to \nectopic pregnancy (tubal rupture and haemorrhage) and/or treatment method used (e.g. \nsurgical complications, medical drug toxicity).  \nLong-term (all women): to reduce risk of recurrent ectopic pregnancy.  \nLong-term (for subgroup of women desiring future fertility): to maximise chance of \nfuture intrauterine pregnancy and live birth rate from unassisted spontaneous conception, or \nfollowing use of assisted reproductive technology techniques (e.g. in vitro fertilisation). \n \nOUTCOMES \nPrimary outcomes: primary treatment success (eradication of the ectopic pregnancy without \nthe need for secondary treatment arising from persisting trophoblast and/or tubal rupture \nand/or worsening clinical symptoms and signs); persistent trophoblast.  \nSecondary outcomes: future fertility-spontaneous intrauterine pregnancy, live birth rate, and \nrepeat ectopic pregnancy in women desiring future fertility (this should ideally be expressed \n\nChapter 4.2.Ectopic Pregnancy \n283 \n \nas fecundity rate ratios over specific time intervals corrected for known confounders [e.g. \nhistory of infertility and contraception usage at time of index ectopic pregnancy]).  \nOther outcome measures: tubal rupture; ipsilateral tubal patency following tubal preserving \ntreatment (salpingotomy, methotrexate, or expectant management); maternal morbidity and \nmortality (prior to ectopic treatment [natural history of ectopic pregnancy] and following \ntreatment alternatives); harms of treatment alternatives; complications of surgery [injury, \ninfection, thromboembolism]; drug toxicity; health-related quality of life assessments. \n \nMETHODS \nClinical Evidence search and appraisal June 2006. Given that there are limitations in \nperforming RCTs comparing medical and surgical ectopic pregnancy treatments, and limited \ntrial numbers, the search was extended to incorporate large sample sized quality cohort \nstudies (either prospective or retrospective, with control or comparison treatment groups). \nWhere appropriate, evidence from quality observational studies is utilised, when RCT \nevidence is lacking. The following databases were used to identify studies for this chapter: \nMedline 1966 to June 2006; Embase 1980 to June 2006; and The Cochrane Library 2006, \nissue 2. Additional searches were carried out using the \nfollowing websites: NHS Centre for Reviews and Dissemination (CRD), Database of \nAbstracts of Reviews of Effects (DARE), Health Technology Assessment (HTA), Turning \nResearch into Practice (TRIP), and National Institute of Health and Clinical Excellence \n(NICE) guidance. Abstracts of the studies retrieved were assessed independently by two \ninformation specialists using predetermined criteria to identify relevant studies. Design \ncriteria included: study types — published systematic reviews, meta-analysis, RCTs, \n\nChapter 4.2.Ectopic Pregnancy \n284 \n \ncontrolled clinical trials, cohort studies with a control or comparison group, or case-control \nstudies in any language; open or blinded studies acceptable; studies had to contain 20 or more \nindividuals. There was no maximum loss to follow up or minimum length of follow up. \nFecundity rate ratios have been calculated by the Clinical Evidence author, except where \nindicated. A GRADE (Grading of Recommendations Assessment, Development and \nEvaluation) 8 approach and evaluation of the quality of evidence for interventions is included \nin this review (see Introduction to Chapter 4 ; Table 4.iv; Table 4v\n).  \nTREATMENT OPTION: SALPINGECTOMY \nTreatment failure (persistent trophoblast) \nCompared with salpingotomy: salpingectomy may be more effective at reducing initial \ntreatment failure rates compared with salpingotomy.  (moderate quality evidence) \nCompared with methotrexate: Salpingectomy may be more effective at reducing initial \ntreatment failure rates compared with methotrexate .  (moderate quality evidence) \nSubsequent pregnancy rates \nCompared with salpingotomy: we don't know whether salpingectomy may result in lower \nrates of subsequent intrauterine pregnancies or recurrent ectopic pregnancy rates compared \nwith salpingotomy.       (very low quality evidence) \nCompared with expectant management: Salpingectomy may be no more effective at \nincreasing subsequent pregnancy rates in women with ectopic pregnancies compared with \nexpectant management.      (very low quality evidence)  \n \nSALPINGECTOMY BENEFITS \nSalpingectomy versus salpingotomy:  We found no systematic review or RCTs. We found \none non-systematic review and four cohort studies (and related single follow up publication) \nthat compared salpingectomy versus salpingotomy (see Table 4.10). 139;145-149 \n\nChapter 4.2.Ectopic Pregnancy \n285 \n \nTable. 4.10. Comparison of fertility outcomes of salpingotomy versus salpingectomy  \nStudy Sample size  \n \n \n(sum of salpingectomy and \nsalpingotomy cases unless \notherwise stated) \nSalpingotomy compared with salpingectomy as \nthe reference \ntreatment \nCrude spontaneous \nintrauterine \npregnancy (IUP) rates \nand/or \nFecundity Rate \nRatios* (FRR) \n (95% CI) \nCrude repeat ectopic \npregnancy (REP) rates \nand/or  \nFecundity Rate Ratios* \n(FRR) \n (95% CI) \nNon-systematic \nreview 149  \n \n1774 women (in 9 cohort studies) \nundergoing salpingotomy or \nsalpingectomy for ectopic \npregnancy and desiring future \nfertility \n \n176 women (in 18 cohort studies) \nwith cTD after salpingotomy \n(corresponding results for \nsalpingectomy not reported) \n280/528 (53%) with \nsalpingotomy v \n614/1246 (49%) with \nsalpingectomy \n \n Crude FRRs $ \n1.08 (0.97 to 1.19) \n \nSalpingotomy in \nwomen with cTD then \nIUP is 96/176 (55%) \n78/528 (15%) v 123/1246 \n(10%) \n \nCrude FRRs $ \n1.50 (1.15 to 1.95) \n \n \nSalpingotomy in women \nwith cTD then REP is \n36/176 (21%)  \nProspective \ncohort 148 \n86 women undergoing \nlaparoscopic surgery for ectopic \npregnancy and were attempting \nconception. \ncTD present in 33/60 \nsalpingotomy and 15/26 \nsalpingectomy cases \n \n36/60 (60%) with \nsalpingotomy v 14/26 \nwith salpingectomy \n(53.9%) \nFRR 1.11 (0.77 to \n1.76)* \nIrrespective of the type \nof surgery performed if \ncTD then crude FRR \n0.53 (0.36 to 0.75) \n (based on 20/50, 40% \npregnant with cTD vs. \n27/34, 79.4% not \npregnant with cTD)  \n11/60 (18.3%) v 2/26 \n(7.7%) \nFRR 2.38 (0.67 to 9.30) \nRetrospective \ncohort 147 \n135 women undergoing \nlaparoscopy or laparotomy for \n62% v 38% at 18 \nmonths (numbers not \n28% v 23% at 3 years \n(numbers not reported) \n\nChapter 4.2.Ectopic Pregnancy \n286 \n \nectopic pregnancy. \ncTD present in 15/56 \nsalpingotomy and 38/79 \nsalpingectomy cases \nreported) \nFRR 1.9 (0.91 to 3.8)  \n \nIf cTD, FRR 0.80 (0.13 \nto 4.9) \nIf bilateral tubal \npathology, FRR 1.4 \n(0.13 to 16) \n \nIrrespective of the type \nof surgery performed:  \nIf cTD then FRR 0.48 \n(0.18 to 1.2) \nFRR 2.4 (0.57 to 11) \n \n \n \n \n \n \nIrrespective of the type of \nsurgery performed:  \nIf cTD, FRR 0.79 (0.18 to \n3.4) \nRetrospective \ncohort 145 \n276 women undergoing \nsalpingotomy or salpingectomy \n(by laparoscopy or laparotomy) \nfor their first ectopic pregnancy. \ncTD present in 30/208 \nsalpingotomy and 17/68 \nsalpingectomy cases \n89% v 66% at 7 years \n(numbers not reported); \nP<0.05 \nFRR 1.58 (1.06 to \n2.38)* \nIrrespective of the type \nof surgery performed: \nIf cTD, FRR 0.46 (0.26 \nto 0.82) \nIf previous fertility \nsurgery,  \nFRR 0.74 (0.34 to 1.60) \n17% v 16% at 2 years \n(numbers not reported; \nreported as not significant) \nFRR 1.28 (0.57 to 2.87)* \nIrrespective of the type of \nsurgery performed: \nIf cTD, FRR 2.25 (1.11 to \n4.531) \nIf previous fertility \nsurgery,  \nFRR 2.51 (1.002 to 6.31) \n\nChapter 4.2.Ectopic Pregnancy \n287 \n \nCohort 139;146 \n  \n476 women with tubal ectopic \npregnancy who were not using \ncontraception at conception. \nSalpingotomy (262, cTD in \n236/262 cases); Salpingectomy \n(178, cTD in 159/178 cases); \nMethotrexate (36, cTD in 8/36 \ncases) \n \nSalpingotomy vs. \nSalpingectomy vs. \nMethotrexate \n73% vs. 57% vs. 80%. \n \nIrrespective of the type \nof surgery performed: \nIf cTD, FRR 0.53 (0.33 \nto 0.83) \nWomen with infertility \nfactors: \nSalpingotomy vs. \nSalpingectomy FRR \n1.67 (1 to 2.78) \nMethotrexate vs. \nSalpingectomy FRR 2.5 \n(1.95 to 8.33) \nWomen with no \ninfertility factors: \nSalpingotomy vs. \nsalpingectomy FRR \n1.18 (0.63 to 2.22) \nMethotrexate vs. \nSalpingectomy FRR \n2.12 (0.49 to 9.78) \nSalpingotomy vs. \nSalpingectomy vs. \nMethotrexate \n25% vs. 27% vs. 41% (no \nsignificant difference \nbetween groups, p=0.55) \n \nSalpingotomy vs. \nSalpingectomy FRR 0.93 \n(0.76 to 3.5) \nMethotrexate vs. \nSalpingectomy FRR 1.51 \n(0.25 to 7.08) \n139 \n1595 women with ectopic \npregnancy. \nSalpingotomy (798); \nSalpingectomy (654); \nMethotrexate (143) \nNumbers of cases with cTD for \neach treatment is unstated \n \nSalpingotomy vs. \nSalpingectomy \nFRR 1.25 (1 to 1.67) \nMethotrexate vs. \nSalpingectomy \nFRR 1.25 (0.7 to 2.33) \nIrrespective of the type \nof surgery performed: \nIf cTD, FRR 0.83 (0.67 \nto 1.0) \nSalpingotomy vs. \nSalpingectomy \nFRR 1.25 (0.67 to 2) \nMethotrexate vs. \nSalpingectomy \nFRR 2.25 (0.6 to 7.4) \nIrrespective of the type of \nsurgery performed: \nIf cTD, FRR 1 (0.5 to 2.0) \n \n\nChapter 4.2.Ectopic Pregnancy \n288 \n \nFootnotes  \ncTD contralateral tubal disease. This may be absent or occluded or distorted by pathology \n(hydrosalpinges, adhesions) \nAsterisked FRRs – calculated by author \n*Fecundity Rate Ratio-See glossary. \nFRR are stated for salpingotomy compared to salpingectomy as the reference unless \notherwise stated. FRRs are also stated for the presence relative to absence of confounding \nfactors (e.g. cTD or infertility) disregarding the type of surgery (either salpingotomy or \nsalpingectomy) that was performed. Where studies have calculated FRR using salpingotomy \nas the reference standard the reciprocal of this FRR has been quoted as this provides the FRR \nof salpingotomy compared to salpingectomy as the reference standard.  \nCrude FRRs$ We report an FRR based on the results reported in the meta-analysis. However, \ndue to study heterogeneity and non-adoption of survival analysis techniques by included \nstudies within the meta-analysis, a pooled FRR as we have reported is likely to be crude and \nsubject to bias. \n \nPrimary treatment success: Salpingectomy compared to salpingotomy has higher primary \ntreatment success and a lower risk of persistent trophoblast. Primary treatment success rate of \nsalpingectomy is almost 100%, with the risk of persistent trophoblast less than 1%. 146;149 \nPrimary treatment success rate of salpingotomy ranges from 72%-98% 150;151, with a 3-20% \nrisk of persistent trophoblast.145;146;149;152;153  \nSubsequent pregnancy rates in salpingectomy versus salpingotomy (see Table 4.10): \nSubsequent spontaneous intrauterine pregnancy rates following salpingotomy (range 53%-\n89%) did not significantly differ from those following salpingectomy (range 38-66%) apart \nfrom one cohort study 145  that showed an improved fecundity with salpingotomy [89% vs. \n66%, FRR 1.58 (1.06-2.38)]. A similar trend of no difference between salpingotomy (range \n10%-28%) and salpingectomy (range 8%-23%) was observed for rates of repeat ectopic \npregnancy.  \n\nChapter 4.2.Ectopic Pregnancy \n289 \n \nContralateral tubal disease and infertility factors (see Table 4.10): In the presence of \ncontralateral tubal disease (hydrosalpinx, peritubal adhesions or absent tube) or infertility \nfactors (e.g. previous ectopic, previous tubal surgery, previous pelvic inflammatory disease, \ninfertility factors) there is a trend to decreased subsequent intrauterine pregnancy and \nincreased risk of ectopic pregnancy irrespective of whether salpingotomy or salpingectomy is \nperformed. 145-148;154-156. In the presence of such factors, salpingotomy provides a greater \nprobability of subsequent intrauterine pregnancy than salpingectomy; however this does not \nachieve statistical significance.  \nSalpingectomy versus methotrexate: We found no RCTs or observational studies of \nsufficient quality. \nSalpingectomy versus methotrexate (systemic): One cohort study, and its follow up \npublication, compared three interventions: salpingotomy, salpingectomy, and methotrexate. \n139;146  It found that the rate of treatment failure with salpingectomy was similar to \nsalpingotomy, but less than methotrexate. The study showed no significant difference \nbetween salpingectomy and salpingotomy in rates of subsequent intrauterine pregnancy or \nsubsequent ectopic pregnancy. \nSalpingectomy versus expectant management: see benefits of expectant management. \nSALPINGECTOMY HARMS \nSalpingectomy versus salpingotomy: The non-systematic review and cohort studies listed in \n(see Table 4.10) did not report on harms.  \nSalpingectomy versus methotrexate: We found no RCTs or observational studies of \nsufficient quality.  \n\nChapter 4.2.Ectopic Pregnancy \n290 \n \nSalpingectomy versus salpingotomy or methotrexate management (systemic): We found \nno RCTs or observational studies of sufficient quality. The cohort study gave no information \non adverse effects. 139;146  One cost-effectiveness meta-analysis found rates of 0–22% (mean \n10%) for minor complications (e.g. drug side effects), and 0–11% (mean 7%) for serious \ncomplications (e.g. ruptured ectopic, or other symptoms of persistent trophoblast) in women \nwho had methotrexate. 151 It also found intraoperative complications of 0–8% (mean 2%) and \npostoperative complications of 0–15% (mean 9%) for laparoscopy (either salpingectomy or \nsalpingotomy). \nCOMMENTS ON SALPINGECTOMY EVIDENCE:  All comparisons included here \nhave been based on retrospective or prospective observational cohort designs in women with \nunruptured tubal ectopic pregnancies (see Table 4.10).  Few studies have considered the \nimpact of infertility factors (known infertility, contralateral tubal disease) on treatment choice \n(conservative salpingotomy or radical salpingectomy) and future fertility outcome. \nDifferences in such prognostic factors may not be adequately clarified when comparing \nsalpingotomy with salpingectomy, even when adopting multivariate analysis techniques. \nHowever, further information may be provided by an RCT comparing salpingotomy with \nsalpingectomy that is currently recruiting. This is the ESEP (European Surgery in Ectopic \nPregnancy) study, which represents an international multi-centre Dutch–Swedish–British \ncollaboration. Importantly, any potential benefits of improved intrauterine pregnancy rate \nwith salpingotomy compared with salpingectomy appear to be small, and possibly restricted \nto subgroups with contralateral tubal disease. This effect and its magnitude should be verified \nby RCTs comparing salpingotomy with salpingectomy. \n\nChapter 4.2.Ectopic Pregnancy \n291 \n \n \nTREATMENT OPTION: PROPHYLACTIC METHOTREXATE AFTER \nSALPINGOTOMY \nTreatment failure (persistent trophoblast) \nCompared with no salpingotomy alone: A single prophylactic dose of methotrexate after \nsalpingotomy is more effective at reducing persistent trophoblast compared with \nsalpingotomy alone.       (moderate quality evidence) \n \nBENEFITS PROPHYLACTIC METHOTREXATE AFTER SALPINGOTOMY \nSalpingotomy plus single systemic dose methotrexate versus salpingotomy alone: One \nRCT found that adding a single prophylactic dose of systemic methotrexate (1 mg/kg im) \nafter salpingotomy (by laparoscopy or laparotomy) significantly reduced the incidence of \npersistent trophoblast compared with salpingotomy alone (1/54 [2%] v 9/62 [15%]; RR 0.13, \n95% CI 0.02 to 0.74; NNT 8, 95% CI 4 to 33). 153 \nHARMS AND COMMENTS ON PROPHYLACTIC METHOTREXATE AFTER \nSALPINGOTOMY.   See under methotrexate section. \nTREATMENT OPTION: SYSTEMIC METHOTREXATE \nTreatment failure \nSingle dose methotrexate compared with multiple dose regimens: Single dose methotrexate \nmay result in higher rates of treatment failure in women with ectopic pregnancies compared \nwith multiple dose regimens.       (low quality evidence) \nCompared with salpingectomy: Methotrexate may be less effective at reducing initial \ntreatment failure rates compared with salpingectomy.       (moderate quality evidence) \nSingle dose methotrexate compared with salpingotomy: Single dose methotrexate is no \ndifferent at increasing primary treatment success rates in women with small unruptured tubal \npregnancies compared with salpingotomy by laparoscopy.  (moderate quality evidence) \nMultiple dose methotrexate compared with salpingotomy: Multiple dose methotrexate is no \ndifferent at increasing primary treatment success rates in women with confirmed unruptured \ntubal pregnancy compared with salpingotomy by laparoscopy. (moderate quality evidence) \nSubsequent pregnancy rates \nSingle or multiple dose methotrexate compared with salpingotomy: Single dose methotrexate \nis no   different at increasing tubal patency, subsequent intrauterine or ectopic pregancy rates \nin women with small unruptured tubal pregnancies compared with salpingotomy. \n                     (moderate quality evidence) \n\nChapter 4.2.Ectopic Pregnancy \n292 \n \nSYSTEMIC METHOTREXATE BENEFITS \nSystemic single dose versus multiple dose methotrexate regimens: We found one \nsystematic review (3 RCTs, 23 observational studies, 1327 women with ectopic \npregnancy)157, one RCT (108 women)158 and one cohort study (643 women). 159 The \nsystematic review found that single dose methotrexate had significantly higher primary \ntreatment failure than multiple dose methotrexate (absolute numbers not reported; OR 1.71, \n95% CI 1.04 to 2.82). The review also found a significant difference for the studies \nconsidered high quality (high quality according to the authors‘ own rating system; absolute \nnumbers not reported; OR 1.96, 95% CI 1.07 to 3.60) and for studies that controlled \nconfounding factors (βhCG and fetal cardiac activity: OR 4.74, 95% CI 1.77 to 12.62) \nSystemic methotrexate (single or multiple dose) versus salpingotomy: We found one \nsystematic review (search date 2004, 4 RCTs, 307 women; see Table 4.11).150 The review (3 \nRCTs, 207 haemodynamically stable women with a small unruptured tubal pregnancy) found \nthat single dose methotrexate was significantly less effective than salpingotomy (by \nlaparoscopy) in primary treatment success (elimination of tubal pregnancy), but found no \nsignificant difference in tubal patency, subsequent intrauterine pregnancy, and repeat ectopic \npregnancy rates. The systematic review also found no significant difference between multiple \ndose methotrexate (1 RCT, 100 haemodynamically stable women with a laparoscopically \nconfirmed unruptured tubal pregnancy) compared with salpingotomy (by laparoscopy) in \nprimary treatment success or tubal patency (see Table 4.11). One RCT identified by the \nreview found that physical functioning (measured as part of SF-36, 0 = worst, 1 = best) was \nsignificantly better with single dose methotrexate compared with salpingotomy at 4 and 10 \ndays (4 days: 73 with methotrexate v 43 with salpingotomy, P = 0.001; 10 days: 93 v 70; P = \n0.006). 160 Another RCT identified by the review found that a variety of scores of quality of \n\nChapter 4.2.Ectopic Pregnancy \n293 \n \nlife were significantly lower with multiple dose methotrexate compared with salpingotomy at \n2 weeks (Medical Outcomes Study, 0 = worst, 100 = best; role function: 29 v 51; social \nfunction: 45 v 68; health perceptions: 52 v 63; P < 0.05 for all comparisons). 161 \nMethotrexate (systemic) versus salpingectomy or salpingotomy: See Table 4.11  \nSystemic methotrexate versus expectant management: We found no RCTs. \n \nTable 4.11. RCTs and meta-analyses of surgical and surgical versus medical treatments \nin the management of ectopic pregnancy. 150 \nType of comparison No \nOf  \nRCTs \nSample \nsize \nTrial \n/ including longer \nfollow up of trial in \nseparate publication \nMeta-analysis of trials \nas reported by Cochrane review 150 \nPrimary \ntreatment \nsuccess \nRelative risk \n(95% CI) \nTubal \npatency \nin those \ndesiring \nfuture \nfertility \nRelative risk \n(95% CI) \nSubsequent \nintrauterine \npregnancy \nrate \nRelative risk \n(95% CI) \nRepeat \nectopic \npregnancy \nrate \nRelative risk \n(95% CI) \nLaparoscopic \nsalpingotomy vs. \nlaparotomy \nsalpingotomy194;200-\n202.  \n \n3 105 \n63 \n60 \nLundorff 1991/1992 \nMurphy 1992 \nVermesh 1989  \n91/104 (88%) \nv 121.124 \n(98%) \n(RR 0.90, \n95% CI 0.83 \nto 0.97; NNT \n10, 95% CI 6 \nto 27) \n78% v 87% \n(0.89, 0.74 to \n1.1) \n61% v 53% \n(1.20, 0.88 to \n1.15) \n6% v 15% \n(0.43, 0.15 to \n1.2) \nSystemic MTX \nmultiple dose i.m. \nvs. laparoscopic \nsalpingotomy203;204 \n1  \n \n100 Hajenius 1997/ \nDias Pereira 1999 \n \n82% v 72% \n(1.15, 0.93 to \n1.4) \n55% v 59% \n(0.93, 0.64 to \n1.4) \n36% v 43% \n(0.89, 0.42 to \n1.9) \n9% v 10% \n(0.77, 0.17 to \n3.4) \nSystemic MTX \nsingle-dose i.m vs. \nlaparoscopic \nsalpingotomy.160;205-\n207 \n3 71 \n74 \n62 \nFernandez 1998  \nSaraj 1998  \nSowter 2001/ \nSowter 2001  \n71% v 88% \n(0.83, 0.71 to \n0.97) \n60% v 57% \n(1.1, 0.74 to \n1.5) \n37% v 47% \n(0.99, 0.55 to \n1.8) \n0% v 17% \n(0.27, 0.02 to \n4.5) \n \n\nChapter 4.2.Ectopic Pregnancy \n294 \n \nSYSTEMIC METHOTREXATE HARMS \nSystemic single dose versus multiple dose methotrexate regimens: One systematic review \nfound significantly lower rates of adverse effects (including nausea, vomiting, alopecia) in \nwomen who had single dose compared with multiple dose methotrexate (31% v 41% \n[absolute numbers not reported]; OR 0.44, 95% CI 0.31 to 0.63). However, it found no \nsignificant difference between regimens when it adjusted for serum βhCG (OR 0.79, 95% CI \n0.21 to 3.01). 157  It also found no significant difference between regimens for abdominal pain \nor hospital admission (abdominal pain: 22% v 26%; OR 0.80, 95% CI 0.53 to 1.19; hospital \nadmission: 12% v 11%; OR 1.11, 95% CI 0.83 to 1.47).  \nSystemic methotrexate (single or multiple dose) versus salpingotomy: \nThe systematic review gave no information on adverse effects. 150 The first RCT found that \nwomen who received single dose methotrexate had significantly longer vaginal bleeding than \ndid those who underwent salpingotomy (7.5 days v 3 days; P < 0.001). 160  The second RCT \nfound that pain was greater with multiple dose methotrexate over 16 weeks compared with \nsalpingotomy (results presented graphically; significance assessment not reported). 161 \nSalpingotomy plus single systemic dose methotrexate versus salpingotomy alone: The \nRCT reported that there were no ―clinically significant‖ adverse effects in people who had \nmethotrexate. It also reported that there was no significant difference in laboratory values \n(white blood cell count, haemoglobin, haematocrit, serum creatinine, and transaminase) \nbetween groups at 7 days after surgery (reported at non-significant; P value not reported). 153 \nSystemic methotrexate versus expectant management: We found no RCTs or \nobservational studies of sufficient quality. \n\nChapter 4.2.Ectopic Pregnancy \n295 \n \nCOMMENTS ON SYSTEMIC METHOTREXATE EVIDENCE: The primary treatment \nsuccess rate of systemic methotrexate (single or multiple dose regimens) in treating ectopic \npregnancies has been reported by meta-analyses as 87% (range 75%-90%) 151, 84%162and \n89%. 157 The risk of persistent trophoblast is 18% (range 6%-31%). 150 Despite the term \nsingle dose methotrexate regimen, repeat doses are permitted every 7 days if there is \ninadequate hCG fall and a meta-analysis has shown two or more doses are required in 13.5% \nwomen undergoing single dose regimens. 157A retrospective study (n=93) has shown two-\nyear subsequent cumulative intrauterine pregnancy rates of 67% and repeat ectopic \npregnancy rates of 24%163, which correlates to fertility outcomes obtained by RCTs included \nin the meta-analysis \n(see Table 4.11). 150 \nClinical guide: Multiple dose systemic methotrexate involves a regime of once daily \nmethotrexate 1mg/kg i.m on alternate days (days 1, 3, 5, 7), and leucovorin 0.1mg/kg i.m.on \nalternate days (days 2, 4, 6, 8). The regimen is continued unless βhCG falls by more than \n15% in 48 hours or until four doses of methotrexate are given. A repeat course can be given if \nβhCG is not less than 40% of its initial value by day 14.Single dose systemic methotrexate \nregime involves a single dose of methotrexate i.m. (50mg/m2). The dose is repeated if βhCG \nhas not fallen by at least 15% between day 4 and day 7 of treatment. Up to four doses can be \ngiven if βhCG does not decline by 15% every week. Prospective studies suggest around 25-\n40% of non-invasively diagnosed ectopic pregnancies are suitable for non-surgical (expectant \nor methotrexate) management 160;164-166.The criteria necessary for methotrexate treatment \nhave been agreed by the RCOG and includes: non-invasive diagnosis of ectopic pregnancy; \nhaemodynamically stability with no signs of tubal rupture; ectopic mass is less than 3.5cm in \ndiameter and there is no fetal cardiac activity;  hCG does not exceed 3000 IU/L; no medical \ncontraindications to methotrexate usage; woman consents to frequent outpatient follow up. 167 \n\nChapter 4.2.Ectopic Pregnancy \n296 \n \nObservational (prospective and retrospective) studies have shown higher primary treatment \nsuccess of methotrexate with ectopics that have: low pre-treatment βhCG  (preferably < 1000 \nIU/l); 160;168-174absent fetal cardiac activity 170;175 or absent yolk sac identified by sonography; \n176;177;no prior history of treated ectopic; 175no pelvic pain; 172;no previous history of \ninfertility.163 Therefore, treatment outcomes of methotrexate should be compared against the \nother tubal conserving methods (salpingotomy and expectant management).  \nAdverse effects: The frequency of methotrexate complications is similar to that associated \nwith laparoscopy.151 However, the nature of the complications differs, with serious \ncomplications of laparoscopy having greater morbidity and mortality than those related to \nmethotrexate. Women who experienced side effects were more likely to have successful \ntreatment regardless of single or multiple dose methotrexate regimen.157 Drug adverse effects, \nalthough prevalent, are usually self-limiting and relatively minor and include: nausea, \nvomiting, gastritis, diarrhoea, abdominal pain, oral mucositis, pneumonitis, bone marrow \nsuppression and abnormal liver function. Case reports have described other rare but serious \ncomplications: life-threatening neutropenia and fever,178anaphylaxis, 179 haematosalpinx and \npelvic haematocoele, 180and death due to multi-organ failure.181 A meta-analysis of single \ndose methotrexate showed side-effects in 24% (95% CI 9% to 47%) and 10% (95% CI 7% to \n14%) had a ruptured ectopic pregnancy. 162 \n\nChapter 4.2.Ectopic Pregnancy \n297 \n \n \nTREATMENT OPTION: EXPECTANT MANAGEMENT \nSubsequent pregnancy rates \nCompared with surgery: Expectant management may lead to similar subsequent pregnancy \nrates in women with nonviable embryos (non-invasive with declining hCG levels) compared \nwith salpingectomy or salpingotomy.      (very low quality evidence) \n \nNote: We found no clinically important results about expectant management compared with \nmethotrexate in women with ectopic pregnancies. \nEXPECTANT MANAGEMENT BENEFITS    \nWe found no systematic review or RCTs.  \nExpectant management versus salpingectomy or salpingotomy:  \nOne retrospective cohort study (180 women with ectopic pregnancy) found similar rates of \nexpectant management and salpingectomy or salpingotomy in subsequent intrauterine \nconception rate (19/37 [51%] with expectant management v 31/49 [63%] with surgery). 182 \nThe study did not report on success of treatment or report separate data by type of surgery. \nWe found one small prospective observational study that compared expectant (16 women) \nversus systemic methotrexate (26 women) versus salpingotomy/salpingectomy (46 women); \ninterpretation of outcomes was biased by case selection. 183 The study also only reported \ntreatment success for women who had methotrexate.  \nExpectant management versus methotrexate: We found no RCTs or observational studies \nof sufficient quality.  \nExpectant management in studies with no control group: We found one non-systematic \nreview (15 prospective cohort studies, 482 women with ectopic pregnancy who were \ndescribed as ―stable‖ or ―well‖) which found a mean rate of 67% (range 47–82%) for \n\nChapter 4.2.Ectopic Pregnancy \n298 \n \nsuccessful management of ectopic pregnancy by expectant management. 184 The review also \nreported that rates of tubal patency were 57/74 (77%), subsequent intrauterine pregnancy \n42/62 (68%), and repeat ectopic pregnancy 6/47 (13%). One prospective cohort study (107 \nwomen who were clinically stable with non-viable pregnancies and no signs of \nhaematoperitoneum) found that 75/107 (70%) of ectopic pregnancies resolved spontaneously. \n166 Another prospective cohort study (30 women who wanted to become pregnant again) \nfound tubal patency in 28/30 (93%) of women, subsequent intrauterine pregnancy in 21/24 \n(88%), and repeat ectopic pregnancy in 1/24 (4%).185 \n \nEXPECTANT MANAGEMENT HARMS \nExpectant management versus salpingectomy or salpingotomy: The retrospective cohort \nstudy did not report on harms. 182  \nExpectant management versus methotrexate: We found no RCTs or observational studies \nof sufficient quality.  \nExpectant management in studies with no control group: The meta-analysis reported that \n2.5% of women had a tubal rupture in one of the cohort studies. 184 The two cohort studies \ndid not report on harms. 166;185 \n\nChapter 4.2.Ectopic Pregnancy \n299 \n \nCOMMENTS ON EXPECTANT MANAGEMENT EVIDENCE: Expectant management \nwas confined to a selected subgroup of unruptured ectopic pregnancies. We found no RCTs \nthat compared expectant management with laparoscopic surgery or systemic methotrexate. \nData for expectant management are derived from retrospective studies with different \ninclusion criteria (e.g. ectopic size, serum βhCG, presence of fetal cardiac activity) that \ncontribute to methodological bias and preclude effective statistical comparison. There is \nlimited evidence that expectant management has similar primary treatment success and future \nfertility outcomes to surgically treated ectopic pregnancy.  \nClinical guide on expectant management Cases considered suitable for expectant \nmanagement should conform to strict criteria: non-invasive diagnosis of ectopic pregnancy, \nunruptured ectopic, woman is haemodynamically stable, less than 100 mL of fluid in Pouch \nof Douglas, initial βhCG is below 1000 IU/L (when the success rate increases to 80% 184 ), \nconsecutive serial serum hCG levels show spontaneous decline, no worsening of symptoms \n(especially abdominal pain, vaginal bleeding) during this interval, and woman understands \nthe need for ongoing surveillance. 167These factors have been verified as favourable \nprognostic signs in observational studies. 184Prospective and retrospective observational \nstudies have shown that low serum progesterone (< 20 nmol/L) and increased rate of βhCG \ndecline to be important predictor of successful expectant management in pregnancies of \nunknown location. 144;186-189 . There is no quantifiable harm in expectant management as \nintervention is absent. However, harm would arise should primary treatment fail or tubal \nrupture ensues. Expectant management necessitates regular surveillance until normalisation \nof clinical, ultrasound, and hCG variables. The risks of tubal rupture and persistent \ntrophoblast remain despite adequately declining serum hCG concentrations. Tubal rupture \nhas been reported with serum hCG levels below 50 IU/L.190;191 \n\nChapter 4.2.Ectopic Pregnancy \n300 \n \nTREATMENT OPTION: SALPINGOTOMY \nTreatment failure (persistent trophoblast) \nSalpingotomy by laparoscopy compared with salpingotomy by laparotomy: Salpingotomy by \nlaparoscopy is less effective at increasing primary treatment success rates compared with \nsalpingotomy by laparotomy.                    (high quality evidence) \nCompared with salpingectomy: Salpingotomy may be less effective at reducing initial \ntreatment failure rates compared with salpingectomy        (moderate quality evidence) \nCompared with single or multiple dose methotrexate Salpingotomy is no different at \nincreasing primary treatment success rates compared with single or multiple dose \nmethotrexate                                                                                  (moderate quality evidence) \nSubsequent pregnancy rates \nSalpingotomy by laparoscopy compared with salpingotomy by laparotomy: Salpingotomy by \nlaparoscopy is as effective at increasing tubal patencies, subsequent intrauterine pregnancy \nrates, and decreasing subsequent ectopic pregnancies compared with salpingotomy by \nlaparotomy.                      (high quality evidence) \nCompared with salpingectomy: We don't know whether salpingotomy may result in lower \nrates of subsequent intrauterine pregnancies or recurrent ectopic pregnancies compared with \nsalpingectomy.              (very low quality evidence) \nCompared with expectant management: Salpingotomy may be no more effective at increasing \nsubsequent pregnancy rates in women with ectopic pregnancies compared with expectant \nmanagement.               (very low quality evidence) \nCompared with single or multiple dose methotrexate: Salpingotomy by laparoscopy is no \ndifferent at increasing tubal patency, subsequent intrauterine or ectopic pregancy rates in \nwomen with small unruptured tubal pregnancies compared with single dose methotrexate. \n                  (high quality evidence) \n \nSALPINGOTOMY BENEFITS \nSalpingotomy (via laparoscopy) versus salpingotomy (via laparotomy):  \nWe found one systematic review (search date 2004, 3 RCTs, 228 women haemodynamically \nstable women with a small unruptured tubal pregnancy) that compared laparoscopic \nsalpingotomy with laparotomy salpingotomy (see Table 4.11). 150 It found that significantly \nfewer women have primary treatment success with salpingotomy by laparoscopy compared \n\nChapter 4.2.Ectopic Pregnancy \n301 \n \nwith salpingotomy by laparotomy (see Table 4.11) due to a higher rate of persistent \ntrophoblast (RR 3.6, 95% CI 0.63 to 21), but found no difference in tubal patency (see Table \n4.11). In those women desiring future fertility (145/228 [64%]), there was no significant \ndifference between salpingotomy by laparoscopy and salpingotomy by laparotomy in \nintrauterine pregnancy rate and rate of repeat ectopic pregnancies (see Table 4.11).  \nSalpingotomy versus salpingectomy: See benefits of salpingectomy. \nSalpingotomy versus expectant management: See benefits of expectant management.  \n \nSALPINGOTOMY HARMS \nSalpingotomy (by laparoscopy) versus salpingotomy (by laparotomy): The systematic \nreview gave no information on adverse effects.150 \nSalpingotomy versus salpingectomy: See harms of salpingectomy.  \nSalpingotomy versus expectant management: We found no RCTs or observational studies \nof sufficient quality.  \nCOMMENTS ON SALPINGOTOMY EVIDENCE: \nThe surgeon‘s preference and operative experience, as well as patient related factors (e.g. \nobesity, previous abdominal surgery, known pelvic adhesions, haemodynamic instability) \ndictates whether laparoscopy or laparotomy is preferred. These confounding factors may lead \nto an overestimation of laparotomy related complications in high operative risk groups.192 See \nalso comment on salpingectomy. \nLaparoscopy or laparotomy surgical treatment of ectopic pregnancy  Evaluation of these \ntrials showed that laparoscopy compared with laparotomy treatment of ectopic pregnancy \n\nChapter 4.2.Ectopic Pregnancy \n302 \n \nincurs less blood loss and analgesic requirement, and has a shorter duration of operation time, \nhospital stay, and convalescence time.150 A reduced prevalence of pelvic adhesions has been \nsuggested as a mechanism for the potential higher future fertility rate observed with \nlaparoscopy compared to laparotomy. 193;194 A multicentre observational study reported major \nsurgical complication rates of 2.7/1000 for diagnostic laparoscopic procedures, and 17.9/1000 \nfor operative laparoscopy.195 The major complications arise following laparoscopic bowel \n(0.4-0.7/1000 cases) and major vessel (0.2/1000 cases) injury.196 Apart from possible \ndifferences in primary treatment success and future fertility, there are no additional maternal \nharms between laparoscopic salpingotomy and laparoscopic salpingectomy. \n \nTREATMENT OPTION: METHOTREXATE (SYSTEMIC) PLUS MIFEPRISTONE \nTreatment failure \nCompared with methotrexate alone: Methotrexate plus mifepristone is no more effective at \nincreasing treatment success rates compared with methotrexate alone but it seems this \ncombination may be more effective in increasing treatment success rates in women with high \nlevels of progesterone.           (moderate quality evidence) \n \nMETHOTREXATE (SYSTEMIC) PLUS MIFEPRISTONE BENEFITS \nSystemic methotrexate plus mifepristone versus systemic methotrexate alone: \nOne RCT found no significant difference between methotrexate plus mifepristone and \nmethotrexate in the number of women who had initial treatment success (22/25 [88.0%] v \n18/25 [72.0%]; OR 2.85, 95% CI 0.54 to 19.17). 197  However, the median time to resolution \nof the ectopic pregnancy was quicker with the combined treatment (14 days v 21 days; \nsignificance assessment not reported). A second RCT also found no significant difference \nbetween methotrexate plus mifepristone and methotrexate in the number of women who had \ninitial treatment success (90/113 [79.7%] v 72/97 [74.0%]; RR 1.07, 95% CI 0.92 to 1.25). 198 \n\nChapter 4.2.Ectopic Pregnancy \n303 \n \nFor women with higher levels of progesterone (greater than or equal to 10nmol/L), it found \nthat treatment success was significantly more successful with the combined treatment than \nwith methotrexate alone (15/18 [83.3%] v 5/13 [38.5%]; RR 2.16, 95% CI 1.06 to 4.44). One \nprospective cohort study found that there were less treatment failures with methotrexate plus \nmifepristone compared with methotrexate alone (1/30 [3.3%] with methotrexate plus \nmifepristone v 11/42 [26.2%] with methotrexate). 199 \nMETHOTREXATE (SYSTEMIC) PLUS MIFEPRISTONE HARMS \nSystemic methotrexate plus mifepristone versus systemic methotrexate alone: \nThe first RCT found that two women in each group reported mild nausea. 197 The second \nRCT found the same rate of gastritis in both groups (34/113 [30.1%] v 30/99 [30.3%]; P = \n1.00). 198  The cohort study gave no information on adverse effects. 199 \nCOMMENTS ON EVIDENCE: See comment under methotrexate. \n\nChapter 4.3. Safe Laparoscopic entry \n304 \n \n4. 3. Laparoscopic entry techniques: clinical guideline, national survey and \nmedicolegal ramifications \n \nINTRODUCTION Although complications associated with laparoscopic surgery are \nfortunately rare, a significant proportion of these occur at the time of laparoscopic abdominal \nwall entry 208-233.Meta-analyses and large multicentre centre studies have provided pooled \nrisks of vascular and bowel injury at the time of laparoscopic entry as 0.2 per 1000 and 0.4 \nper 1000 respectively 208;226;234-245. Such complications may incur serious morbidity and \nmortality, and this is compounded if such injuries are not detected at the time of original \nsurgery, particularly in the case of bowel injury 208;212;246-254. \nTwo laparoscopic entry methods are principally used in gynaecology and general surgery \nrespectively: \n Closed entry and creation of a pneumoperitoneum at the umbilicus (or Palmer‘s \npoint). \n Open laparoscopy (Hasson) 255-257. \nOther techniques, used less frequently and with limited supporting evidence213, are direct  \nentry 258-266, optical access trocars 267-277 and radially expanding trocars 278-283.  \nOn current evidence, mainly based on observational studies, no one laparoscopic entry \nmethod has demonstrated clear superiority over another. This has led to wide variation \namongst clinicians as to which entry method should be recommended 213;284;285. It has been \nsuggested that open (Hasson)  entry is superior as vascular injury is less likely to occur \ncompared to closed entry techniques 235;239;256;257;264;284;286-296, although this viewpoint has \nbeen challenged 235.  \n\nChapter 4.3. Safe Laparoscopic entry \n305 \n \nThere is significant variation in laparoscopic entry practice in the UK 297-300 and International  \nlocations 234;301;302. In an attempt to minimise the risks of laparoscopy and unify clinical \npractice, several international bodies [International Middlesbrough Consensus 303, RCOG \n(draft version only)304 , SOGC 285, RANZCOG 305, EAES 306, SAGES 307, French society of \nEndoscopic Gynecology 308, Netherlands 309] and experts 310-312 have recommended specific \n―safe laparoscopic entry‖ principles. In fact, several small-sized studies 313-315 have shown \nthat adopting a recommended technique 285;303 can reduce the incidence of laparoscopic entry \nrelated complications.  \nWe wanted to evaluate the status of gynaecological laparoscopic entry in the UK, bearing in \nmind that litigated cases normally consider both what should be (published \nrecommendations) and what is (questionnaire enquiries) occurring in clinical practice. To \nachieve this we planned to:  \n1. Establish evidence based criteria for safe laparoscopic entry through a systematic \nliterature search and critical appraisal of the literature. \n2. Currently identify what laparoscopic entry techniques are used in the UK, and explore \nany factors that may influence the preference for a particular technique. This was \ndetermined through a UK wide questionnaire survey. \n3. Identify the current judicial viewpoint on laparoscopic entry injuries from the published \nliterature. \n\nChapter 4.3. Safe Laparoscopic entry \n306 \n \nMETHODS \nEstablish criteria for safe laparoscopic entry: Electronic searches were performed in \nMEDLINE (Ovid version 1996-December 2007), EMBASE (Ovid version 1996-December \n2007) using relevant combinations of medical subject headings (laparoscopy; gynecological \nsurgical procedures; intraoperative complications; postoperative complications; \npneumoperitoneum, artificial; malpractice; risk assessment; legal liability; judicial role; \njurisprudence) and text words. International guidelines were identified by interrogating \nspecialised electronic repositories (e.g. national guideline clearinghouse, national electronic \nlibrary for health, OMNI, TRIP database, E guidelines and GFMER databases) and searching \nnational Collegiate (e.g. RCOG, ACOG, RANZCOG, SOGC) and specialist international \nlaparoscopy organisation websites (e.g. AAGL, SLS, ISGE, BSGE, SAGES, EAES, \nASERNIP-S). Literature was critically appraised according to established evidence-based \ncriteria [see Scottish Intercollegiate Guidelines Network (SIGN) recommendations Table 4.ii-\n4iv 7;316] to generate a list of key steps necessary for safe laparoscopic entry (see Table 4.v \nand 4.12).  For each step, we denoted a level of evidence and grade of recommendation and \ndiscussed their derivation from the supporting literature. \n \n\nChapter 4.3. Safe Laparoscopic entry \n307 \n \nTable 4.12:  Evidence-based criteria for safe laparoscopic entry: 10 steps \nStep Intervention \nLevel of evidence and \nGrade of \nrecommendation  \nSupporting \nreferences \n1 \nSuitability criteria: Consider alternative to close umbilical entry \n(e.g. Palmer‘s point or open (Hasson) technique) in patients with \nrisk factors such as: previous abdominal surgery; obesity, \nextremely thin or known abdominal adhesions. \n2++, B Adhesion risks: \n 324-344 \n2 \nSafety Criteria: Patient should be lying flat with an empty \nbladder. Palpation for the abdominal aorta, any masses and check \nVeress needle for spring action and gas patency. \n4, GPP  \n3 Incision: 10mm vertical intra-umbilical incision starting deep \ninside the umbilicus pit extending caudally.  \n4, GPP  \n4 \nInsertion of Veress: At the deep umbilical pit, at 90º to the skin, \nwith or without stabilising or elevating the umbilical sheath/fascia \nor anterior abdominal wall,  and in a controlled manner with \ninsertion of less than 2cm of the Veress needle tip \n2+, 2-, C \n(Indirect evidence from \nknowledge of abdominal \nanatomy) \n258;317-319;351-355 \n5 No movement of the Veress needle following insertion - avoid \nconverting a possible needlepoint injury into a large complex tear \n4, GPP  \n6 \nSafety Abdominal pressure check of Veress placement: Most \nreliably achieved by using a Veress Intra-Abdominal Pressure \n(IAP) of less than 10mmHg.  \n2+, C 315;356-358 \n7 \nSafety Abdominal pressure check for Primary trocar: The \nintra-abdominal pressure should be 25mmHg to achieve the \nmaximum safe distance between anterior abdominal wall and \nunderlying abdominal contents. \n2+, C 315;359-363 \n8 \nVertical Primary Trocar insertion: \nInserted in a controlled two-handed screwing manner, vertically at \n90° to the skin, with only the tip of the trocar inserted through the \nabdominal wall.   \n2+, C 317-319;351-353 \n9 Injury check: An initial 360º laparoscopic check for intra-\nperitoneal organ injury is performed \n4, GPP  \n10 \nNo Epigastric for Secondary trocar(s) insertion \nInserted under direct vision in a controlled two-handed manner at \n90° to the skin, avoiding inferior Epigastric vessels. \n2+, C \n(Indirect evidence from \nknowledge of abdominal \nanatomy) \n364-368. \nFootnotes:We suggest an acronym, SCIIN  SAVE SAVING, for the 10 steps: Suitability, Criteria, Incision, \nInsertion, No movement, Safety Abdominal VEress, Safety Abdominal pressure (Trocar), Vertical trocar, Injury \ncheck, No epiGastrics. \n\nChapter 4.3. Safe Laparoscopic entry \n308 \n \nQuestionnaire survey: The questionnaire was developed in collaboration with the \nBritish Society of Gynaecological Endoscopy (BSGE) which recorded:  \n1. Clinician grade \n2. Method of entry in the uncomplicated woman and the high-risk woman (defined as any \nwoman with previous suprapubic or midline laparotomy, very thin or obese)   \n3. Angle of entry for Veress needle and primary trocar \n4. Criteria used to test for correct placement of Veress and adequacy of CO2 \npneumoperitoneum prior to primary trocar insertion.   \n5. Whether the clinician routinely inspected the abdomen for laparoscopic injury at the \nbeginning or end of the laparoscopy procedure  \n6. Whether the clinician had experienced (personally or through witnessing) any \nlaparoscopic entry-related bowel or vascular injury. \n7. Awareness of Middlesbrough Consensus and RCOG sourced information on \nrecommended laparoscopic entry practice. \nIn contrast to previous questionnaire studies, we wished to compare practice amongst trainee \ngrades as well as consultant specialists. The study population comprised of three groups. \n1. Registered BSGE members at May 2006. The questionnaire (and pre-paid postage reply \nenvelope) was included in the BSGE May 2006 quarterly newsletter, which was sent to all \n180 registered BSGE members. \n2. Specialist Registrar trainees. The questionnaire was distributed to all trainees who \nattended regional study days at Birmingham Women‘s Hospital, UK.  \n\nChapter 4.3. Safe Laparoscopic entry \n309 \n \n3. Attendees at the joint RCOG/BSGE conference held on Friday 8th December 2006 at \nRoyal College of Obstetricians and Gynaecologists, London, UK where a questionnaire \nthrough an electronic-audience participation format was used. Audience members responded \nthrough handheld devices and instantaneous feedback on the entire audience was \nelectronically displayed after each question.  \nRESULTS \nA. Evidence based criteria for safe laparoscopic entry  \n \nThe original systematic literature review identified 276 primary studies relating to \nlaparoscopic techniques and complications, 21 secondary studies (13 meta-analyses and 8 \nclinical guidelines) and 12 citations relating to medicolegal aspects of laparoscopy entry \nrelated complications. A further 17 relevant citations were identified through the bibliography \nof primary and secondary (clinical guidelines, reviews) studies. Through a process of critical \nappraisal of the literature a 10 step evidence-based criteria for safe closed umbilical \nlaparoscopic entry was constructed and shown in Table 4.12. The level of evidence \njustifying each step is outlined below. \nSuitability criteria (Step 1):  Women who are extremely thin 317-319 or obese 320-323 or known \nto have abdominal adhesions are at increased risk of laparoscopic entry related injury at the \numbilical entry point. The estimated risks of umbilical and/or anterior abdominal wall \nadhesions in women with no prior laparoscopic surgery, previous suprapubic laparotomy and \nprevious midline laparotomy are 0-5%, 20%-30% and 50-65%, respectively 324-344.  \nProspective observational studies suggest the risk of laparoscopic entry related injury may be \nconsiderably reduced by adopting alternative entry (e.g. left upper quadrant Palmer‘s point or \nopen Hasson technique) in women with such risk factors.  However, the actual relative risk \n\nChapter 4.3. Safe Laparoscopic entry \n310 \n \nreduction is not quantified as the studies have no comparator. Left upper quadrant Palmer‘s \nlaparoscopic entry could also be considered if there has been failure to achieve \npneumoperitoneum at the umbilicus. Of significance, there is limited evidence that testing for \nreduced (<1cm) visceral slide (ultrasound\n-visualised movement of the underlying bowel or \nomentum) may be helpful in detecting sub-umbilical adhesions, thereby allowing \nconsideration of an alternative laparoscopic entry strategy 345-350.    \nSupine patient positioning, safety checks and umbilical incision (Steps 2 and 3): Reliable \ndata on appropriate patient positioning and location/type of umbilical incision were not \nidentified. Consequently, we suggest the patient should be laid flat at commencement of \nlaparoscopy to avoid the theoretical risk that ―pelvic‖ bowel being displaced towards the \numbilicus, thereby exposing the bowel to entry related injury. On a similar stance, adopting \nan alternative entry technique is advisable if a prominent abdominal aorta pulsation is \nidentified in close proximity to the undersurface of the umbilicus.  Current consensus among \nclinicians is for a 10mm vertical intra-umbilical incision extending caudally. \n \nControlled vertical (90 degree) Veress needle entry (Steps 4 and 5):  There are no \ncomparative studies assessing the optimum angle of Veress needle entry. The fusion of the \nparietal peritoneum and linea alba at the pit of the umbilicus logically dictates that a vertical \n(90 degree to the horizontal abdomen) Veress insertion represents the shortest skin-to-\nperitoneum anatomical distance to enable direct peritoneal entry. According to CT abdominal \nmapping 317;318;351 and actual laparoscopy 319;352;353, this skin-to-peritoneum distance at the \numbilical pit is consistently no greater than 2cm, irrespective of abdominal obesity. \nNevertheless,  it has been suggested that the Veress angle of entry should vary (45 degrees in \nnon-obese women and 90 degrees in obese women) as CT abdominal imaging 318, and \n\nChapter 4.3. Safe Laparoscopic entry \n311 \n \nvisualisation at laparoscopy 319, has shown that the location of the underlying aortic \nbifurcation (which may be prone to Veress injury) tends to be directly under or 2-3cm caudal \nto the umbilicus in non-obese and obese women, respectively. The umbilicus pit (and \nunderlying parietal peritoneum) may also be stabilised and/or successfully elevated away \n(either by grasping the lower abdominal wall or by applying tissue forceps/towel clips within \n2cm from the umbilicus) from underlying abdominal viscera during Veress insertion 258;353-\n355. However, a reasonable summary of the indirect evidence stated is that, traversing the \nthinnest portion of the abdomen by controlled 90 degree vertical entry, with insertion of no \ngreater than 2cm of the Veress needle tip, with selective umbilical stabilisation/elevation, is \nlikely to be safest route of Veress insertion for the vast majority of women, regardless of any \ncaudal displacement of their umbilicus. \n \nLess than 10mmHg IAP safety test for correct Veress placement (Step 6):\n A variety of \nsafety tests for correct intra-peritoneal placement of the Veress needle are employed in \nclinical practice, and include: double-click, aspiration, and hanging drop tests. Prospective \nstudies in women undergoing laparoscopy have shown that a Veress intra-abdominal pressure \n(IAP) less than, or equal to, 10 mm Hg, reliably indicates correct Veress placement at \numbilical 315;356;357 and Palmer‘s point entry 358 locations. The Veress IAP pressure correlates \npositively with the weight and BMI and negatively with the parity of women 357. \n \nControlled vertical (90 degree) primary trocar insertion at 25mmHg IAP (Steps 7, 8, 9): \nProspective observational studies have shown higher intra-abdominal CO2 insufflated \npressures achieve greater anterior abdominal wall splinting and intra-abdominal CO2 gas \n\nChapter 4.3. Safe Laparoscopic entry \n312 \n \nbubble space 315;359-361. An IAP of 25mmHg has been shown to achieve a maximum safe \ndistance between anterior abdominal wall and underlying abdominal contents, without \ncompromising cardio-respiratory function 362;363. A two-handed screwing manner controlled \nvertical (90 degree) entry of only the primary trocar tip utilises the safe CO2 bubble depth \nafforded through an IAP of 25mmHg and is highly unlikely to injure underlying vessels \nbased on actual laparoscopy 319;352;353 and abdominal vasculature CT mapping studies \n317;318;351.  Although there is no direct supporting evidence, an initial check for bowel and \nvascular injury, immediately after primary trocar insertion, is recommended to avoid missing \nthis complication and exposing the women to serious morbidity. \nControlled insertion of secondary trocars under direct vision (Step 10):\n   Epigastric \nvessels can be reliably identified through a combination of direct visualisation [vessels lie 1-2 \ncm lateral to the medial umbilical ligaments (obliterated umbilical arteries)], \ntransillumination and external anatomical landmarks \n364-368. In most women, a useful and safe \npoint of insertion is 2 cm from the anterior superior iliac crest along an imaginary line \nconnecting the iliac crest to the umbilicus.  The controlled insertion, at a 90 degree angle to \nthe skin, using a two-handed screwing manner of the secondary trocar (analogous to that used \nto insert the primary trocar), should be observed under direct vision to ensure no inadvertent \ninjury of abdominal organs. \n \n\nChapter 4.3. Safe Laparoscopic entry \n313 \n \nB. Questionnaire survey \n \nThere was a 62% (n=112) response rate from the postal questionnaire, and 100% response \nrates from SpR registrars (n=82) and attendees at the RCOG/BSGE meeting (n=32). Analysis \nwas performed on all 226 total respondents. \nEntry technique in uncomplicated vs. high risk women: The vast majority would perform \na closed umbilical laparoscopic entry in uncomplicated women and a Hasson or Palmer‘s \npoint entry in women with previous midline laparotomy (Table 4.13).  However, there was \ninconsistency when selecting entry technique in women with previous suprapubic \nlaparotomy, obesity, or who were extremely thin (Table 4.13). \nVeress and primary trocar entry: Only 18% would use the recommended 90º/90º Veress \nand primary trocar entry method (Table 4.14). Safety checks performed to ensure correct \nVeress placement and prior to primary trocar insertion are depicted in Tables 4.15 and 4.16, \nrespectively. The proportion of respondents aware of evidence-based guidance, or who have \nprevious experience of laparoscopic injury, is depicted in Table 4.17. \n \n \n\nChapter 4.3. Safe Laparoscopic entry \n314 \n \nTable 4.13 Laparoscopic entry technique in uncomplicated vs.  high-risk women \n  \nVeress entry technique Uncomplicat\ned women \nHigh risk women \nWomen \nwith \nPrevious \nSuprapubi\nc \nlaparotomy \nWomen \nwith \nprevious \nmidline \nlaparotom\ny \nWomen \nwith obesity \nWomen \nwho are \nextremely \nthin \nClosed umbilical (umb.) 213 (94%) 193 (85%) 37 (16%) 179 (79%) 189 (84%) \nOpen (Hasson)  5 (2%) 14 (6%) 49 (22%) 13 (6%) 15 (7%) \nPalmer's point 1 (<1%) 8 (4%) 102 (45%) 4 (2%) 2 (<1%) \nSuprapubic point 3 (1%) 1 (<1%) 4 (2%) 9 (4%) 6 (3%) \nDirect entry 3 (1%) 2 (<1%) 0 3 (1%) 2 (<1%) \ntransvaginal culdoscopy  1 (<1%) 1 (<1%) 1 (<1%) 1 (<1%) 1 (<1%) \nClosed umb.  or suprapubic 0 0 1 (<1%) 4 (2%) 1 (<1%) \nHasson or Palmer's 0 3 (1%) 23 (10%) 0 1 (<1%) \nClosed umb. or Palmer's 0 2 (<1%) 5 (2%) 0 4 (2%) \nClosed umb. or Hasson or \nPalmer‘s \n0 0 3 (1%) 1 (<1%) 0 \nClosed umb. or Hasson  0 2 (<1%) 1 (<1%) 9 (4%) 6 (3%) \n \nFootnotes \nUmb. Refers to umbilical \nDirect entry would be gasless direct primary trocar abdominal entry and would not utilise \nVeress. \n \n\nChapter 4.3. Safe Laparoscopic entry \n315 \n \nTable 4.14 Frequency of angle of entry for Veress and Primary Trocar \n \n  Angle of Primary Trocar entry [Count] Total \n  90  60  45  30  Z angle b   \nAngle of \nVeress \nentry \n[Count] \n90  40 \n(18%) \n28 \n(12%) \n24 \n(11%) \n1 1 94 \n60  6 34 \n(15%) \n9 0 2 51 \n45  1 11 57 \n(25%) \n1 3 73 \n30  0 1 1 0 0 2 \nNot used a 3 1 1 1 0 6 \nTotal 50 75 92 3 6 226 \n \nFootnotes \na Veress angle not determined as practitioner prefers to use either Hasson or direct entry \nmethod for insertion of primary trocar. \nb Z angle system corresponds to initial shallow angle <30 then a steeper angle >60. \nThe five most frequent Veress/Primary trocar combinations are shaded in grey and bolded \nfont. \n\nChapter 4.3. Safe Laparoscopic entry \n316 \n \nTable 4.15. Safety checks performed to ensure correct Veress placement \n \nTESTS  SpR 1-3 \nn=63 \nCount \nSpR 4-5 \nn=41 \nCount \nConsultant \nn=122 \nCount \nTotal \nCount (%) \nPressure & saline aspiration & two Veress clicks  14 11 28 53 (23%) \nPressure & saline aspiration 13 11 24 48 (21%) \nSaline aspiration 20 6 10 36 (16%) \nPressure & two Veress clicks 6 5 21 32 (14%) \nPressure  3 4 14 21 (9%) \nSaline aspiration & two Veress clicks 6 1 4 11 (5%) \nPressure & freely moving Veress & two Veress \nclicks  \n0 0 7 7 (3%) \nTwo Veress clicks 1 1 3 5 (2%) \nFreely moving Veress & two Veress clicks 0 1 2 3 (1%) \nPressure and freely moving Veress 0 0 3 3 (1%) \nFreely moving Veress 0 0 2 2 (<1%) \nNot use Veress  0 1 4 5 (2%) \n \nFootnotes \nPressure refers to pre-insufflation intra-abdominal pressure recorded as below 8mmHg \nTwo Veress clicks refers to the audible or tactile impression of two Veress clicks on \nabdominal insertion \nSaline aspiration refers to the four-component saline aspiration, injection, aspiration, drop test \ncommonly known as Palmer‘s test. \n \n\nChapter 4.3. Safe Laparoscopic entry \n317 \n \nTable 4.16. Safety checks performed prior to primary trocar insertion \n \nTESTS SpR 1-3 SpR 4-5 Consultan\nt \nTotal \nCount (%) \nIAP 25mmHg 29 22 48 99 (44%) \nDistension and  IAP 25mmHg 20 6 14 40 (18%) \nDistension and IAP 12-15mmHg 4 0 21 25 (11%) \nDistension 2 4 13 19 (8%) \nIAP 12-15mmHg 4 4 8 16 (7%) \nDistension,   >3L CO2 , IAP 12-15mmHg 1 3 6 10 (4%) \nDistension,   >3L CO2, IAP 25mmHg 2 2 6 10 (4%) \nDistension, >3L CO2 0 0 5 5 (2%) \nCO2 >3 litres  1 0 1 2 (<1%) \n \nFootnotes \nIAP refers to intra-abdominal pressure \nDistension refers to clinical abdominal wall distension \n\nChapter 4.3. Safe Laparoscopic entry \n318 \n \nTable 4.17. Awareness of evidence-based guidance and previous experience of \nlaparoscopic injury  \n \n SpR 1-3 \nn=63 \nCount \nSpR 4-5 \nn=41 \nCount \na Consultant \nn=122 \nCount \nTotal \nn=226 \nCount (%) \nAwareness of Middlesbrough Consensus  \nYes \n \n27 \n \n22 \n \n100 \n \n149 (66%) \nNo 36 19 22 77 (34%) \nAwareness of RCOG Guidance  \nYes \n \n55 \n \n33 \n \n90 \n \n178 (79%) \nNo 8 8 32 48 (21%) \nPrevious experience laparoscopic injury  \nYes, bowel injury \n \n14 \n \n13 \n \n57 \n \n84 (37%) \nYes, vascular injury 7 4 7 18 (8%) \nYes, both vascular and bowel injury 4 5 32 41 (18%) \nNo 38 19 26 83 (37%) \nRoutine Inspection of abdomen  \nYes \n \n48 \n \n38 \n \n110 \n \n196 (87%) \nNo 15 3 12 30 (13%) \n \n \nFootnotes \na Consultant category includes 4 Staff Grades, 5 Associate Specialists and 113 Consultants.  \n \n \n\nChapter 4.3. Safe Laparoscopic entry \n319 \n \nC. Medico-legal ramifications  \n \nThe civil standard of law is used in UK medico-legal litigation. This means it is the \nresponsibility of the claimant (woman patient) to prove that, it is more likely than not (greater \nthan 51% probability), that the injury she incurred arose through a negligently performed \nrather than non-negligently performed surgical technique by the defendant (Surgeon). \nLaparoscopic entry related complications have contributed significantly to medical litigation \nin gynaecological surgery 233;247;248;369-376. Until recently, there had been inconsistency in the \njudicial viewpoint in awarding negligent or non-negligent verdicts. However, the case of \nPalmer v Cardiff & Vale NHS Trust 377 has now set judicial guidance in this area. The court \nruled that the likelihood of laparoscopic related bowel injury occurring in an uncomplicated \ncase if there had been good surgical technique, was highly unlikely. If there was no \nalternative plausible non-negligent explanation for the complication then the defendant was \nliable - complying with the legal maxim res ipsa loquitir (―the thing speaks for itself‖). This \noverruled the defendant‘s viewpoint that injury was a recognized complication of \nlaparoscopy and therefore its occurrence was not proof of negligence per se. The judicial \nguidance accepted that given a woman without risk factors, and a surgeon following a safe \ntechnique (i.e. correctly inserting Veress needle, it‘s position checked, insufflation of the \nperitoneal cavity to 25mmHg, controlled insertion of the primary trocar with penetration of \nthe cavity by just the trocar tip), then the risk of injury was highly improbable. Thus the \noccurrence of any injury under these circumstances would imply a negligent technique. \n\nChapter 4.3. Safe Laparoscopic entry \n320 \n \nDiscussion \n \nBased on our systematic literature search, critical appraisal of the published literature and \navailable guidelines, we have constructed a ten step evidence based guideline necessary for \nsafe closed laparoscopic entry. Our findings are analogous to Semm‘s original 11 safety steps \n312.  However, we have updated these steps in line with current evidence-based literature and \nhave ascribed the level of evidence to each step supported by the literature citation(s) for that \nstep. We feel that these 10 steps represent the current most up to date evidence to enable \nclinicians to practice safe closed laparoscopic entry (Table 4.12). \nOur national questionnaire study revealed considerable heterogeneity in laparoscopic entry \npractice despite widespread awareness of the Middlesbrough Consensus or RCOG sourced \nguidance. The inconsistency was inherent throughout every step of the laparoscopic entry \nprocedure, and has been identified by previous UK based surveys 297-300. Fundamentally, \nthere was a failure to appreciate risk factors that would justify a change in entry technique, as \nwell as not adopting the correct safety checks following Veress insertion and prior to primary \ntrocar insertion. Even if there was authoritative guidance on safe laparoscopic entry technique \nit is unclear how many practitioners would actually change their clinical practice accordingly.  \nHowever an Australian based questionnaire study suggests that this would be supported by \nthe majority of minimally invasive surgeons 301.  \nWe acknowledge that we have a limited sample size and have surveyed a highly selected \ngroup. On one hand it is reassuring that we have shown no real differences between trainees \nand specialists. However, it is of great concern that even in the ―expert‖ specialist group there \nis such a wide variation in entry technique.  It is possible that a survey of general \ngynaecologists may identify an even wider and more alarming variation in practice. \n\nChapter 4.3. Safe Laparoscopic entry \n321 \n \nWe strongly feel that safe laparoscopic entry guidance should be disseminated widely such as \nthe 10 steps shown in Table 4.12.  However, we accept that following such guidance would \nnot necessarily negate the risk of laparoscopic entry related injury nor would it protect the \nclinician against any negligent ruling should a complication occur. We believe that written \nguidance should be reinforced through simulated training 378;379, structured formal assessment \nand consistent clinical direction by specialists. Unless practice concurs with recommended \nguidance, women undergoing laparoscopy will be exposed to increased unnecessary \noperative risk.   \n \n\n4.4 Preventing Sterilisation Failure \n322 \n \n4.4.  Minimising the Risk of Sterilisation Failure-an evidence-based \napproach \n \nIntroduction Female sterilization is one of the commonest procedures performed \nworldwide. In 1999 around 50,000 female sterilisations were performed in England in the \nNHS and charitable sectors 380 .  The procedure is performed on mainly healthy women at \ntheir request, and the intention is to occlude each fallopian tube. This may be achieved \nthrough tubal surgical excision, application of a mechanical device or electrocautery \ncoagulation (Table 4.18). Where resources permit, the preference, and most widely \nestablished technique, is laparoscopic tubal occlusion, which has moreover replaced the \nearlier technique of performing female sterilisation via mini-laparotomy. In the UK, the \nRCOG 380 recommends that laparoscopic sterilisation should be performed using either \nFilshie clip or ring. Tubal excision and separation and related techniques (e.g. Pomeroy \nprocedure) are preferred if sterilisation is performed at caesarean delivery.  \n \nHysteroscopic sterilisation may be considered a non-incisional, non-surgical form of \npermanent contraception, and is a promising alternative to laparoscopic tubal occlusion. The \nprocedure involves the insertion of a small flexible titanium microinsert into each of the \nfallopian tubes through the cervix using a guidewire and a hysteroscope (ESSURE®, \nConceptus Inc.).  The procedure is usually performed under local anaesthesia and/or \nintravenous sedation. Despite being licensed in the UK, NICE considers hysteroscopic \nsterilisation to still be under evaluation and should only be performed in accordance with \nspecific NICE guidance (particularly on patient consent and coordinated follow up.381 This is \nmainly because there is insufficient evidence on long term efficacy (single case report of \nfailure382 and tubal perforation383) and safety of hysteroscopic sterilisation, with the \nmanufacturer reporting 99.8% effectiveness at preventing pregnancy at 2 year follow up \n\n4.4 Preventing Sterilisation Failure \n323 \n \n(http://www.essure.co.uk). 384-386Furthermore, there are no published randomised controlled \ntrials comparing ESSURE directly with commonly used female tubal occlusion methods.387 \nTable 4.18: Female surgical sterilisation techniques \nMethod Techniques Comments \nLigating tube with \npartial or complete \ntubal excision \nPomeroy \nFimbriectomy \nSalpingectomy \nPreferred option at  mini-\nlaparotomy, but laparoscopic \nsalpingectomy is an alternative \n \nMechanical occlusion \nof the tubal lumen \nFilshie clip \nHulka-Clemens clip \nFalope ring  \nSilastic ring \nLess of the tube is damaged \nincreasing the chance of \nreversibility \n \n \nCoagulation induced \ntubal closure \nUnipolar diathermy \nBipolar diathermy \nNot recommended as the first line \nmethod in the UK by the RCOG \n \nHysteroscopic tubal \nocclusion \nExpanding metal tubal \nmicro-insert implant \n(ESSURE) \nLicensed in UK and under \nevaluation. Guidance for usage in \naccordance to NICE.  \nVirtually no possibility of reversal. \nContraceptive precautions to \ncontinue for at least 3 months post \nprocedure and X-Ray HSG \nconfirmation of tubal occlusion \n \n\n4.4 Preventing Sterilisation Failure \n324 \n \nRates of sterilisation failure  Conception occurring after sterilisation is termed \nsterilisation failure, and can occur several years after the procedure. Publications have \nreported differences in rates in sterilisation failure rates, even amongst the same sterilisation \nmethod. Such variation is due to differences in: the characteristics of the women undergoing \nsterilisation; operator experience; operating centre workload; sterilisation method chosen, and \nthe time interval to resuming sexual activity post sterilisation and its frequency.5 \n \nThe two largest studies that have examined failed sterilisation have reported the ten-year \ncumulative probability of pregnancy of 18.5 per 1000 procedures  (US CREST study) 388 and \n8 per 1000 procedures (Canada) 389 . The reason for the lower sterilisation failure rate in the \nCanadian study compared to the US CREST study may be due to predominant use of the \nFilshie clip and incorporation of non-teaching hospitals in the Canadian dataset.  However, \nboth studies were also significant in: \n Utilizing the superior and preferred life table analysis method (cumulative probability of \npregnancy at serial time intervals since sterilisation) for reporting sterilisation failure, \nrather than the less accurate crude failure or Pearl index outcomes that were reported by \nprevious studies.  \n Obtaining follow up data for at least 5 to 15 years following the sterilisation \nThis concept of cumulative risk of pregnancy is particularly important for those women \nsterilized at a young age (who will be exposed to a risk of pregnancy for a greater time \nperiod) and who have been sterilisation by methods of low short and long term efficacy \n(because such methods, over certain time frames, will acquire a greater percentage of total \nfailures than other more effective methods). \n \n\n4.4 Preventing Sterilisation Failure \n325 \n \nBoth US and Canadian dataset studies 388;389 validated this concept of cumulative risk of \npregnancy. In the Canadian dataset 389 the cumulative probability of pregnancy increased \nfrom 0.3% at 1 year, to 0.7% by 5 years and 0.9% by 15 years.389  This is depicted in Figure \n1.  It is therefore important to quote women a 10 year risk of sterilisation failure, \nindividualised to each method and patient age, when counselling them for the sterilisation \nprocedure.  Bearing in mind that as long as a woman is fertile, and sexually active, she may \ncontinue to be at risk for sterilization failure.The RCOG has recommended a 10-year \nsterilisation failure rate of 2-3 per 1000 procedures be used for the Filshie clip method. \nHowever, this rate is predominantly drawn from a retrospective questionnaire study, of 5 year \nfollow up, with an exaggerated denominator.390 Given this information, and considering the \nother reported Filshie clip studies (listed in Table 4.19), 2-3 per 1000 risk is more likely to \ncorrespond to the first year or even annual non-cumulated absolute risk of sterilisation failure.  \n\n4.4 Preventing Sterilisation Failure \n326 \n \nTable 4.19.  Filshie Clip: reported sterilisation failure rates  \nStudy Period \ndata are \ncollected \nfrom \nSterilisations  \nPerformed \nSterilisation \nmethod \nOutcome Type of  \nstudy \nPeterson388 \nUS Collaborative \nreview of \nSterilisation \n(CREST) \n1978-1986 \n \n10,685 \nFilshie clip was \nnot used- as it \nwas not \nlicensed in \nUSA until 1996 \n392 \nVarious methods. \nHulka spring clip \n(1595) \nSilicone Rubber \nband (3329) \nOverall 18.5 per 1000 \nover 10 years \nHulka 36.5 per 1000 \nSilicone rubber band \n17.7 per 1000 \nProspective cohort \nmulticentre \nTrussell 389 1980-1999 311,960 Mainly \nLaparoscopic \nFilshie clip  \n \n8 per 1000 \n[2496 failures] \nRetrospective \nmulticentre  \nKovacs 390 1994-1998 30,000 \n(estimate) \nAll Filshie 2.4 per 1000 \n[73 failures]a \nRetrospective \nmulticentre \nFilshie 424 1982-1992 First 202 \nresponders \nfrom a series of \n434 \nAll Filshie 2.3 per 1000  \n[1 failure at 6 months] \nCase series \nBirdsall 415 1988-1989 1094  Mainly \nLaparoscopic \nFilshie clip  \n12 per 1000 at 12 \nmonths b  \nCase series \nSokal 423 1984-1990 2746 Filshie clips vs. \nRings \n[2 in each group \nbecame pregnant] \n1.7 per 1000 for both \nRing and Filshie clip \ngroups at 12 months \nRCT \nDominik 422 1984-1990 2126 Filshie clips vs. \nHulka clips \n[11 pregnancies \noccurred:  \n9 Hulka, 2 Filshie] \nAt 12 months \n1.1 per 1000 for Filshie \nClip  \n6.9 per 1000 for Hulka \nClip group.  \nAt 24 months, 9.7 per \n1000 for Filshie and \n28.1 per 1000 for Hulka \nRCT \nFootnotes  \na   Kovacs: Of the 73 failures, 14 cases were due to operator error, 29 were properly applied clips and 30 cases \nhad unknown reason for failure   b    Birdsall: Registrars had a 1.3% failure rate, consultants 1.9% and when both a consultant and registrar \nperformed the procedure a failure rate was 0.7%. Eighty-six percent (6/7) of failed sterilisations were due to \noperator error (wrong structure, initial non-occlusion). \n\n4.4 Preventing Sterilisation Failure \n327 \n \nKey factors [excluding operator error] identified to alter cumulative probability of \npregnancy \nThe failure rate for each sterilisation method tends to stabilize over the long term and may \nthus be represented as a constant lifetime risk of sterilisation failure (1 in 200 is quoted for \nthe Filshie clip380). However, a more precise estimate would also be based upon her age at \nsterilisation and the subsequent number of fertile years during which she is at risk of \npregnancy. The Canadian dataset 389 showed that  sterilisation of young women (< 30years of \nage) compared to older women (>35 years age) was associated with an overall increased \nabsolute risk of pregnancy after sterilisation (1.5% vs. 0.4% ), and that this cumulative risk \nstabilized later in the younger age group. This is depicted in Figure 4.2. Multivariate \nregression analysis of the CREST study 388 showed the following factors were associated \nwith an increased risk of sterilisation failure: \n Sterilisation method used. Most effective were postpartum partial salpingectomy and \nlaparoscopic unipolar coagulation at 7.5 pregnancies per 1000 procedures, but \nlaparoscopic spring clip application had the highest risk of failure at 36.5 pregnancies per \n1000 procedures (see Figure 4.3) \n Age at sterilisation. The probability of failure for women sterilized at ages <28 years is \ngreater than that for women sterilized at ages >34 years for all methods of sterilization \nexcept interval partial salpingectomy. \n Race-ethnicity. Black, non-Hispanic women were at significantly greater risk for \nsterilization failure than were white, non-Hispanic women. \n Operating centre. Substantial differences in procedure specific failure rates between \nsites, likely representing variation in operator experience, requirements to teach juniors \nand volume of sterilisation operations. \n\n4.4 Preventing Sterilisation Failure \n328 \n \nFigure 4.2: Clinico-pathological mechanisms proposed in sterilisation failure based on \nCanadian dataset 389 \n \n \n\n4.4 Preventing Sterilisation Failure \n329 \n \nFigure 4.3.  Cumulative risk of pregnancy by method from US CREST study 388 and \nFilshie clip references \n \nYears since \nsterilisation  \n  \nCumulative risk of pregnancy per 1000 sterilisation procedures \n \nBipolar Unipolar Silicone \nBand \nHulka \nclip \npost partum \nsalpingectomy \nFilshie \nclip \n(estimate \nonly) \n1 2.3 0.7 5.9 18.2 0.6 2.5 \n2 4.6 2.3 7.6 23.8 3.9 2.5 \n3 6.7 2.3 8.3 29.1 4.6 2.5 \n4 13.1 2.3 9 30.7 5.4 2.5 \n5 16.5 2.3 10 31.7 6.3 2.5 \n6 18.3 2.3 10 31.7 6.3 2.5 \n7 20.7 2.3 13 31.7 6.3 2.5 \n8 22 2.3 16.1 31.7 6.3 2.5 \n9 23.3 4 16.1 34 7.5 2.5 \n10 24.8 7.5 17.7 36.5 7.5 2.5 \n0\n5\n10\n15\n20\n25\n30\n35\n40\n1 2 3 4 5 6 7 8 9 10\nYears since sterilisation\nCumulative risk of pregnancy\nBipolar\nUnipolar\nSilicone Band\nHulka\nSalpingectomy\nFilshie\n \n\n4.4 Preventing Sterilisation Failure \n330 \n \nNotably, the US Crest study showed no statistically significant associations between risk of \nsterilisation failure and a history of pelvic inflammatory disease, history of previous \nabdominal or pelvic surgery, or presence of any adhesions recorded at sterilization. Although \nthese factors have been assumed empirically by practitioners to affect the risk of sterilisation \nfailure.  \nSterilisation failure and subsequent intrauterine or ectopic pregnancy\n \nOverall, for all sterilisation methods, studies have shown ectopic pregnancy may occur in  \n4.3–76.0% of failed sterilisations.380 The relative risk of intrauterine to ectopic pregnancy \noccurrence in failed sterilisation varies according to the sterilisation method and time interval \nfrom the sterilisation procedure. Women who have been sterilized have a considerably lower \nabsolute risk of an ectopic pregnancy compared to non-sterilised fertile women (as \nsterilisation protects against both intrauterine and ectopic pregnancies). However, should \npregnancy occur, the relative risk of it being ectopic rather than intrauterine is higher in \npregnant women who have been sterilized. Women should be counselled about such risks \nwhen deciding the method of sterilisation. \n \nThere were 47 ectopic pregnancies in the 10,685 sterilised women in the US CREST study, \nwhich equates to a 10-year cumulative probability of ectopic pregnancy for all sterilisation \nmethods combined of 7.3 per 1000 procedures. 391 Women sterilized by bipolar tubal \ncoagulation before the age of 30 years had a probability of ectopic pregnancy that was 27 \ntimes as high as that among women of similar age who underwent postpartum partial \nsalpingectomy (31.9 vs. 1.2 ectopic pregnancies per 1000 procedures).391 \n\n4.4 Preventing Sterilisation Failure \n331 \n \nClassification of causes of sterilisation failures: the role of operator error [negligent \nmechanism] \nThe mechanism of failure should be identified through a systematic assessment of fallopian \ntube histology, X-ray hysterosalpingography and direct pelvic visual inspection. Neither of \nthe major observational studies on sterilisation failure reported on  the underlying mechanism \nof sterilisation failure.388;389 Our systematic review identified only 81 cases in the world \nliterature where the mechanism of sterilisation failure had been confirmed by such systematic \nmethodology. 5 Sterilisation failure may be classified as arising from negligent or non-\nnegligent mechanisms, which may be dependent or independent of the sterilisation method \nutilised (Table 4.20).  If the mechanism of failure is due to ‗tubal non-occlusion‘ or ‗wrong \nstructure sterilisation ‘, these are considered negligent mechanisms, whereas ‗spontaneous \ntubal recanalisation‘ or ‗fistula formation‘ mechanisms of failure are considered non-\nnegligent. \nSeveral studies have shown operator error to represent a significant (if not the major) cause of \nsterilisation failure. One summative review showed that the overall ten year failure rate for \nworldwide Filshie clip sterilisations was 0.56% in 10,000 women, but fell significantly to \n0.2% when cases caused by operator error were excluded.392 A questionnaire based study \nexamining Filshie clip use in Australia showed of the 73 sterilisation failures from 30,000 \nprocedures, 14 were due to operator error, 30 unknown reason and 29 occurred in the \npresence of a ‗properly applied clip‘.390 Another study, which incorporated participants of the \nUS CREST study, reported that all 20 sterilisation failures using spring clip and silicone \nrubber band arose to improper application of the occlusive devices.393;394 Of  the 81 \nsterilisation failures reported in our systematic review of published literature 5, 57 cases were \ndue to operator error ( wrong structured ‗sterilised‘ and initial tubal non-occlusion, ) and 24 \nnot due to operator error (fistula formation or recanalisation). We have recently published an \n\n4.4 Preventing Sterilisation Failure \n332 \n \nanalysis of 131 cases of sterilisation failure, incorporating our systematic review, where 88 \nwere negligent and 43 non-negligent sterilisation failures (see Chapter 2 and reference 395). \nTable 4.20. Classification system for mechanism of sterilisation failure \nDEPENDENT ON THE STERILISATION METHOD \nNegligent \nInitial tubal non-occlusion (poor operator technique) e.g. slippage or overclosure of \nFilshie clip (see Figure 3). \nWrong structured ‗sterilised‘  \nImproperly maintained equipment (e.g. non-calibrated/serviced Filshie clip applicator) \nwhich contributed to initial tubal non-occlusion. \n \nNon negligent \nInitial tubal non-occlusion  (true method failure)- this is reported extremely rarely \nand occurs despite correctly applied technique \nthe ends of the fallopian tube can reconnect spontaneously (recanalisation) \na fistula can develop at the occluded portion of the tube \nINDEPENDENT OF THE STERILISATION METHOD* \nAlready conceived in the cycle prior to sterilisation \nOr in the case of Filshie clip, conceives following sterilisation in the remainder of the \nmenstrual cycle because the ovulatory ovum is proximal to tubal sterilisation point \n(luteal-phase pregnancy) \nOr in the case of Hysteroscopic sterilisation, conceives within the 3 month interval \npost sterilisation and/or prior to confirmation of effective sterilisation by HSG or \nultrasound. \n*Most studies on sterilisation failure have excluded such pregnancies from their \nreported final analysis \nMechanical tubal occlusive methods have lower rates of tuboperitoneal fistula formation than \ncoagulation based techniques.396-398  This may be because mechanical occlusion methods \ndestroy much less tube (approximately 4 mm for clips and 2 cm for rings) than \nelectrocoagulation methods (3-4 cm).  However the exact aetiology of tubal lumen \nregeneration remains unclear. Other factors such as individual‘s tubal ‗healing‘ response, pre-\n\n4.4 Preventing Sterilisation Failure \n333 \n \nexisting proliferative tubal disease (e.g. endosalpingiosis), degree of tubal avascularity and \ninterval from operation are likely to modify tubal lumen regeneration ability.396;399-403  \nPresently there is no evidence to suggest that operator fault in sterilisation technique \npredisposes to tubal lumen regeneration, and therefore this mechanism of sterilisation failure \nwould be considered to be non-negligent and independent of operator error. \nMedico legal consequences  The psychological and physical morbidity following failed \nsterilisation often leads to litigation.404 A gynaecologist has a duty to inform women of the \nrisk of failure, to carry out the operation in accordance with accepted good medical practice \nand to avoid foreseeable complications. Women who have undergone sterilisation performed \nnegligently are entitled to recover damages according to:  \n Wrongful conception: In addition, an action in contract may also arise if the sterilisation \nprocedure was performed outside the NHS in the private sector. \n Negligence: A breach of duty arises when an operation is not carried out in accordance \nwith practice accepted as proper by a reasonable body of gynaecologists (Bolam test).  \nNegligence also occurs when there is omission in appropriate pre-operative counselling.  \n Wrongful birth: The negligent act deprived the mother of the possibility to prevent the \nconception of a disabled child or to have a lawful abortion.  \nWomen are entitled to recover general damages for pain and suffering during pregnancy and \ndelivery, and loss of earnings during pregnancy. A recent judgment in the Australian High \nCourt 405  led the Australian government to amend the Civil Liberty Act to restrict the amount \nof damages that could be awarded in such situations. In the majority of failed sterilisation \ncases, even those in the advanced stages of litigation, the mechanism of failure remains \nunknown as there is no uniform requirement for such cases to undergo systematic enquiry or \n\n4.4 Preventing Sterilisation Failure \n334 \n \nto be reported to any supervisory national registry. The RCOG should consider this \nrequirement at the time of the sterilisation guideline review in 2006. 380 Thus, a common \nscenario in the legal setting is to cast judgment on the likelihood of negligence or non-\nnegligence in cases with unknown mechanism of sterilisation failure. Based on pooling the 81 \ncases of sterilisation failure with documented interval to pregnancy and mechanism of failure \nwe proposed: \nThat a greater proportion of early (within 12 months from operation) than late (after 12 \nmonths from operation) sterilisation failures occurred by a negligent mechanism. Thus, the \ntime interval to sterilisation failure may be predictive of negligence. In our recent publication \nof 131 cases of sterilisation failure395, we showed sterilisation failure occurred significantly \nearlier in negligent than non-negligent failure mechanisms (mean failure intervals 7.5 vs. 14.2 \nmonths; Hazard Ratio 2.35 [95% CI 1.31-4.21]). \nInitial tubal non-occlusion is more likely to lead to early sterilisation failure (within one \nyear), and as it is less likely to damage the tube, the resulting pregnancy is more likely to be \nintrauterine than ectopic. Conversely, late sterilisation failure arising from tubal re-\ncanalisation or fistula formation is more likely to result in an abnormal lumen predisposing to \na decreased risk of pregnancy, but should it occur there would be an increased risk of ectopic \npregnancy.  This is graphically illustrated in figure 1.   \nIdentification and assessment of evidence \n \nMEDLINE 1966-2006, the Cochrane library, 2006, Royal College of Obstetricians and \nGynaecologists (RCOG, UK) were searched for relevant randomised controlled trials, \nsystematic reviews, meta-analyses, and evidence-based guidelines relating to sterilisation. \nThe searches were performed using the relevant MeSH terms including: sterilization, tubal; \n\n4.4 Preventing Sterilisation Failure \n335 \n \nsterilization; sterilization sexual; surgical instruments; electrocautery; cautery; liability, legal; \njurisprudence; malpractice; medical errors; treatment failure; risk factors. The majority of \npublications were retrospective observational studies, case reports and reviews, with a \npaucity of prospective controlled trials or meta-analyses.5;380;406 The definitions of the types \nof evidence used in this chapter are as denoted in the RCOG Clinical Governance advice.407 \nWhere possible, recommendations on strategies to minimise sterilisation failure are annotated \nwith the level of evidence that supports them (A, B, C or GPP) as indicated ( Table 4.2ii). \nData generated was incorporated in our recently published systematic review in failed \nsterilisation 5 and utilised for this chapter to generate a best evidence based guideline 6. \nClinical Guideline: Minimising the risks of sterilisation failure \n \n1. Patient Selection                    Level GPP  \nThere is limited evidence that pre-existing gynaecological pathology, in addition to \nincreasing the technical difficulty of performing the sterilisation procedure, independently \npredisposes to sterilisation failure. Factors such as pre-existing tubal disease, history of \nabdominal or pelvic surgery, history of pelvic inflammatory disease previous ectopic \npregnancy, pregnancy or post-partum state,, obesity, prior use of an intrauterine contraceptive \ndevice, previous induced abortion, congenital uterine anomalies, fibroids, endometriosis, \nendosalpingoblastosis and adenomyosis.388;389;397;408-412  The myth that sterilisation protects \nagainst pelvic inflammatory disease has recently been challenged.413 \n2. Pre-sterilisation pregnancy testing and timing of sterilisation        Level C\n \nBoth hysteroscopic and laparoscopic tubal occlusion may be performed at any time during \nthe menstrual cycle provided that the clinician is certain that the woman has used effective \ncontraception up until the day of the operation. It is recommended practice that all women \n\n4.4 Preventing Sterilisation Failure \n336 \n \nshould have a urine pregnancy test prior to sterilisation. Routine pre-operative same day \npregnancy testing has been shown to reduce the incidence of pregnancies discovered after the \nsterilisation procedure that have been falsely attributed to presumed negligent sterilisation \nprocedure.414  However, such a test may still be falsely negative in a very early pregnancy. A \nserum hCG pre-operatively may be considered, however, if there is any doubt, then the \nsterilisation should be deferred until the follicular phase of a subsequent cycle.  \n3. Pre-procedure contraception and the need to continue until onset of next menstrual \ncycle (reduce risk of luteal pregnancy)                 Level GPP  \nContraception is immediately effective if using the combined pill (if commenced between \nday 1 and day 5 of period) and Mirena IUS. However, with laparoscopic tubal occlusion, \ncontraception is only likely to be completely effective by the onset of the next menses. \nTherefore, for this method, pre-procedure contraception measures should be continued until \nthe onset of next menses to prevent ―luteal phase‖ pregnancy failure (Table 4.20). This is \nwhere sterilisation has occurred just after ovulation, and the ovum is already ‗proximal‘ to the \ntubal occlusion, enabling pregnancy to occur in this luteal phase through post sterilisation \n‗unprotected‘ intercourse. Studies have identified luteal pregnancy occurring in 0.32% to \n0.6% of sterilisation cases.388;415;416 Women selecting hysteroscopic sterilisation (ESSURE) \nneed to continue with contraceptive precautions for at least three months post procedure and \nmay resume ‗unprotected‘ sexual intercourse only after there is confirmation of satisfactory \ntubal occlusion (e.g. by X Ray hysterosalpingogram). \n4. Timing the operation - Interval preferred          Level B\n \nWherever possible, tubal occlusion should be performed at an appropriate interval following \npregnancy. Sterilisation can be performed in the immedate post-partum period (combined \n\n4.4 Preventing Sterilisation Failure \n337 \n \nwith caesarean section or via minilaparotomy) or post-abortion.  However this period is \nassociated with higher rates of failure and regret by the woman416 417, and this should be \nincorporated into the counselling and documentation prior to the procedure. In terms of post-\npartum sterilisation, salpingectomy and Filshie clip have similar rates of failure (7.5 and 8.8 \nper 1000 respectively).388 416 \n5. Selection of technique- Laparoscopy preferred over laparotomy       Level B  \nEach sterilisation method has specific advantages, disadvantages and individualised failure \nrates according to the sterilisation method and patient characteristics. This information should \nbe conveyed during the counseling process. A meta-analysis 418, and large population study \n419, has shown no significant difference in failure rate or major operative morbidity between \nmini-laparotomy and laparoscopy methods of sterilisation. However, laparoscopic methods \nhave lower minor operative morbidity and are preferred for interval sterilisations as it offers \nobvious advantages in terms of shorter operative time, same day hospital discharge and \nshorter convalescence period. \n6. Selection of technique- Modified Pomeroy at caesarean section       Level B\n \nA modified Pomeroy procedure rather than Filshie clip application may be preferable for \npostpartum sterilisation performed by mini-laparotomy or at the time of caesarean section, as \nthis leads to lower failure rates.388;416;420, although both procedures are equally popular \nchoices with surgeons.421 \n7. Selection of technique- Filshie clip sterilisation is preferred method       Level B\n \nTwo small RCTs 422;423 and observational studies390;424 have shown Filshie clip to have the \nlowest failure rate for interval sterilisation failure and has therefore been recommended by \nthe RCOG 380 as the preferred method at laparoscopic tubal occlusion (Table 4.19). Ring \n\n4.4 Preventing Sterilisation Failure \n338 \n \nmethods have also been recommended by the RCOG, and appear to have equal contraceptive \nefficacy to Filshie clip. However, ring methods tend to be technically more difficult to apply \nto the fallopian tubes and have gradually become less popular in UK clinical practice.  \n8. Operative technique for Filshie clip:                Levels C and GPP  \na) Care should be taken to ensure the Filshie clip is applied to the optimal mid-isthmic tubal \nposition (1cm to 3cm from the uterine cornu) and this structure not be mistaken for an \nadjacent structure e.g. the round ligament or a fold of peritoneum.425 \nb) The Filshie clip should be applied in a manner to completely encapsulate the tube and \nlumen, be fully locked with the upper jaw compressed, completely flattened and its end \nadequately secured under the under the latch which ‗locks‘ the clip jaw (Figure 4.4).  The \nclip should flatten the whole tube portion within the clip without leaving any unflattened \ntubal ‗knuckles‘ without transecting the tube.  Finally, the clip should sit perpendicular to the \nlong axis of the tube. 425 facilitated by stretching the isthmic portion with hinge placed on the \nantimesenteric aspect of the tube.  \nc) Excessive forceful clip applicator overclosure (Figure 4.4) or underclosure may lead to \ntubal transection and subsequent sterilisation failure through luminal regeneration (i.e. tubal \nfistula or re-canalisation) or incomplete tubal occlusion. 425 \nFigure 4.4: Filshie clip under-closure due to operator fault          \nDespite the clip appearing locked, on \ncloser inspection the upper jaw of the \nclip will be noted to be incompletely \ncompressed, rounded rather than \nflattened, and the end insufficiently \nsecured under the under the latch for \nthe upper jaw. Most causes of clip \nunder-closure are due to operator \nfault. \n\n4.4 Preventing Sterilisation Failure \n339 \n \nA predisposing factor to improper closure is a ‗faulty‘ Filshie clip applicator. However, this is \nrare, as it is a legal requirement that device applicators are well maintained and adequately \nchecked to ensure optimum function.  In the case of the Filshie clip, both the manufacturer \n(Femcare, UK \nwww.femcare.co.uk) and MDA strongly recommend that all single Filshie clip \napplicators are serviced and re-adjusted at least once a year or after every 100 procedures.  \nFurthermore, a closing checking gauge should be used prior to every sterilisation procedure \nto ensure the applicator functions correctly.  There is only one published case of failed \nsterilisation, which proposes Filshie clip under-closure as the most likely mechanism of \nsterilisation failure.  Therefore this cause of failure should be considered rare.426  \nd) Applying two mechanical clips adjacent to each other on the tube does not decrease the \nfailure rate, but may even increase it if they are applied too closely together.425;427-429 \ne) Following clip application there should be a systematic checking procedure to ensure the \ncorrect structure and both sides of the tube have been satisfactorily occluded, and this should \nbe documented. Although not a legal requirement in the UK, we recommend: \ne) Taking clinical photographs or operative videos of the sterilised structures \nidentifying them as fallopian tubes. However photographs may be unhelpful in confidently \nexcluding other negligent causes of incomplete tubal occlusion e.g. protruding knuckle of \ntube inadequate locking of clip jaws, clip under-closure, or tubal transection (partial or \ncomplete)  \nf) Presence of second operating surgeon for counter-checking.  A recent study \ninvolving 1094 sterilisations from 1988-1989 showed that Registrars had a 1.3% failure rate, \nConsultants 1.9% and when both a Consultant and Registrar performed the procedure a \n\n4.4 Preventing Sterilisation Failure \n340 \n \nfailure rate of 0.7% was recorded.415  A medical witness to concur the sterilisation procedure \nis a legal requirement in some countries.430 \n9. True method failure. There is evidence that anatomical tubal patency can occur \nfollowing a correctly undertaken sterilisation (true method failure), and has been reported \nfollowing correctly applied Filshie clips in three cases of Filshie clip failure (table 4.19)431 \nand is implied to have occurred in the 29/73 correctly applied clip sterilisation failures \nreported by an observational study. 390However, persisting anatomical tubal patency does not \nnecessarily imply sterilisation failure, as tubal patency rates of 1-2% at three months and 16% \nat five years have been noted following correctly applied tubal ligation, with the actual \npregnancy occurrence of 1-2% over this time period. 400.  Even so, true method failure is rare \nand difficult to prove; nonetheless three possible mechanisms of true method failure are \nsuggested: \n A partially non-occluded segment of tubal lumen has formed within the clip. This tubal \n‗knuckle‘, with a patent lumen, can exist within the completely flattened tube portion \ninside the clip identifiable only at microscopy.  \n Pre-existing utero-tubal structural abnormalities such as accessory fallopian tube, uterine \ndidelphys 432, and utero-tubal fistulas \nMechanical failure of the Filshie clip.  Manufacturers for Filshie clip have not reported \nspontaneous mechanical failure as a possibility for sterilisation failure, and this concurs with \nan absence of such cases in the published literature.  Nevertheless, there remains at least a \ntheoretical possibility of mechanical material failure, and manufacturers like FEMCARE® \noffer an examination of the Filshie clips in failed sterilisation to exclude the possibility of this \nfailure mechanism (Femcare - personal communication). \n\n4.4 Preventing Sterilisation Failure \n341 \n \n10. Operator experience and training                     Levels C and GPP  \nImproper application of tubal occlusive devices by inexperienced surgeons is frequently \nreported in cases of sterilisation failure.394;433;434  Furthermore, operator preference is likely to \nhave an impact on method related failure rates. The CREST study showed failure rates of 7.1 \nto 78.0 per 1000 for the Hulka clip and 0 to 42.5 per 1000 for the silicone ring - dependent on \nthe operating centres being surveyed.388 Higher failure rates were more common in centres \nwith performing fewer annual procedures. RCOG recommends that trainees should perform \nat least 25 supervised laparoscopic tubal occlusions before operating without supervision. 380 \n11. Follow up required if uncertainty in tubal occlusion               Level GPP  \nFollowing a complicated sterilisation good clinical practice (rather than a legal requirement) \ndictates testing of tubal patency.394;435-438  However, a negative dye spill post sterilisation \nHSG does not completely preclude the possibility of pregnancy at a later stage.439  \n12.  Other issues: Clip Migration and dropped ‘lost’ Filshie clips       Level C  \nGood clinical practice dictates that proof of tubal occlusion (X-ray HSG or tubal dye \ninsufflation or histology of salpingectomy) should be undertaken once missing clips are \nidentified, not only when examining failed sterilisation cases, but also at laparoscopy or \nlaparotomy for other reasons.425;440  However, missing clips do not necessarily indicate failed \napplication or imminent pregnancy failure, as over time there is a tendency for clips to \nmigrate and even be expelled without resulting in clinical morbidity.390;423;441-449 There are no \nreports of this leading to sterilisation failures.441 It is estimated that over 25% of women will \nexperience a migration of one or more Filshie clips.441  The tissue between the Filshie clip \njaws normally undergoes avascular necrosis and fibrosis, leaving two healed stumps, which \ntend to separate, permitting clip displacement. Filshie clips may be inadvertently dropped \n\n4.4 Preventing Sterilisation Failure \n342 \n \nduring laparoscopic sterilisation. If possible the clip should be laparoscopically removed \nupon completion of the sterilisation procedure. However, if the clip is irretrievable, either \nopen or closed, it should be left.   Performing a laparotomy would subject the woman to \ngreater operative morbidity risk than leaving the lost clip in the abdomen.  To date, there have \nbeen no reports of any serious morbidity or mortality consequent to a lost clip. Women \nshould be informed of the lost clip and reassured accordingly.425 \nConclusion and Further research \n \nOverall, the level of evidence supporting any screening-preventative measures to reduce the \nrisk of sterilisation failure remains poor. There appears to be a propensity for negligent rather \nthan non-negligent sterilisation failures. However, this can only be verified by establishment \nof a national register of failed sterilisations (as recommended by the RCOG380)  that have \nbeen subjected to systematic enquiry to establish the mechanism of failure. Like other \nConfidential Enquiries, such a registry could identify areas of substandard care that could be \nused as an impetus to improve research and medical training in sterilisation procedures and \nhelp design effective clinical risk prevention strategies. The introduction of an operative \nchecklist or proforma, similar to the pre-operative counseling checklist recommended by the \nRCOG380 and used in another study 450, may result in reduced numbers of negligently \nperformed sterilisations.  \n \n\nChapter 5.Thesis Conclusion \n343 \n \nCHAPTER 5. THESIS CONCLUSION \n \n \nFigure 5.1. Ascension of a research pyramid of research methodologies to benefit \nclinical practice \nAudit\nClinical \nGuidelines\nSystematic \nreviews & RCTs\nCohort studies\nDescriptive studies \nElucidating aetiopathogenesis\nmolecular  in vitro &  in vivo models\nMolecular & epidemiological associations\n \n \nFootnotes \nRCT randomised controlled clinical trials \nAudit refers to clinical audit to assess impact of clinical guidelines \n \n \nThesis Précis \nThe central aim of this PhD thesis was to produce research that could inform and benefit \nclinical practice. Each chapter of this thesis has achieved this aim, within the limits of the \nresearch methodology applied The chapters have been ordered to follow a stepwise ascension \nof a research methodological pyramid (Figure 5.1). \n\nChapter 5.1 Discussion of chapter 1: Experimental investigation of endometriosis \n344 \n \n5.1. Experimental investigation of endometriosis \n \nThe experimental data explored whether there was a causal link for endometriosis as a \nneoplastic precursor to ovarian cancer. The chapter was original in adopting distinctive, yet \ncomplimentary, approaches to testing this hypothesis through clinical epidemiology \n(Bradford Hill causality criteria1), histopathology, immunohistochemistry, genetic and \nmolecular approaches. The chapter reported the largest EAOC series to be subjected to \nclinico-epidemiological and LOH mapping for the entire length of chromosome 9 (20 \nmarkers) and chromosome 11 (27 markers).  Furthermore, this chapter includes the first study \nto apply SNP 100K genome wide genotyping to endometriosis2. \nEpidemiological and causality literature analysis showed: \n No strong evidence to support a causal link between endometriosis and ovarian cancer3-5. \n However, there was moderate quality evidence that endometriosis may display similar \nproperties to a cancer cell (Hanahan‘s Hallmarks of cancer6). There are numerous anti-\ncancer therapies that target the specific molecular properties of the ‗cancer cell‘. Hence it \nis conceivable, that anti-endometriosis therapies may also be designed around such cancer \ncell-like molecular targets (Table 5.1). \nExperimental analysis showed: \n A prognostic association of the LOH (loss of heterozygosity) identified at 9q34.3 and \n11q23.3 chromosomal regions suggesting that these regions may harbour genes that \nimpact on malignant transformation and progression of cancer.  \n\nChapter 5.1 Discussion of chapter 1: Experimental investigation of endometriosis \n345 \n \n Decreased glycodelin immunohistochemical expression in endometriosis adjacent to \novarian cancer compared to endometriosis distant from ovarian cancer- hence a possible \nrole for glycodelin as a TSG responsible for development for ―malignant‖ endometriosis. \n Ovarian endometriosis harbours micro-regions of LOH through Affymetrix genome-wide \n100k SNP microarray. However, to confirm and validate the location of these multiple \nmicro-regions of LOH analysis further customised microsatellite markers analysis is \nrequired. 2;5 \nFuture directions  \nHigh throughput molecular technologies (as used in this chapter) allow parallel genomic, \ntranscriptomic and proteomic evaluation of diseases, at the genome-wide level. Such \napproaches could be used to elucidate the multigene pathways involved in aetiopathogenesis \nof endometriosis (Figure 5.2, Table 5.2), as well as numerous other diseases. Application of \ntechniques in cancer biology may also facilitate the research and development of therapies for \nendometriosis (Tables 5.1 and 5.3).  A key future goal would be the identification of \ncharacteristic endometriotic ‗genetic‘ or ‗proteomic‘ signatures that could form the basis of \nan early screening-preventative testing strategy from women‘s urine, menstrual endometrium \nor blood. Furthermore, a similar genomic/transcriptomic/proteomic approach may be \nconsidered for the investigation of analogous gynaecological disorders (such as adenomyosis \nand fibroids) and explore whether genetic alterations are focal or widespread in a diseased \nreproductive tract. \n\nChapter 5.1 Discussion of chapter 1: Experimental investigation of endometriosis \n346 \n \nTable. 5.1. Individualise therapeutic approach to endometriosis according to cancer cell \nhallmarks 6 \n CANCER \nCELL \nHALLMARKS \nTARGETS UNDER \nEVALUATION IN \nENDOMETRIOSIS BASED ON \nAGENTS USED IN CANCER \nTRIALS \nTARGETS/ PUTATIVE TARGETS UNDER \nINVESTIGATION IN CANCER  \n1 Self-sufficiency in \ngrowth signals \nAromatase inhibitors, Selective \noestrogen (e.g. Arzoxifene) and \nprogesterone receptor modulators, \nMirena Coil, Gonadotrophin releasing \nhormone antagonists (e.g. Cetrorelix) \nInhibitors of: Mitogen-activated protein kinase inhibitors, \nHER-2 receptor (trastuzumab), IGF-1 receptor, EGFR \n(erbitux), EGFR tyrosine kinase (gefitinib) farneysl \ntransferase, Bcr-Abl tyrosine kinase (imatinib mesylate) \n2 Insensitivity to \nantiproliferative \nsignals \n Proteasome inhibitors (bortezomib), cdk inhibitors \n(flavopiridol) \n3 Resistance to \nApoptosis \nAngiostatin gene transfer; transfection \nwith pro-apoptotic (e.g. BAX) gene \nCOX-2 inhibitors \n \n \nCOX-2 inhibitor (celecoxib), thalidomide, apoptosis inducers \n(exisulind inhibits cGMP). \n \nImmunotherapy by genetically modified tumour vaccines \n(e.g. HER-2 peptide vaccination) or humoral factors (e.g. \nimmunokines like IL-12, TNF antagonists; monoclonal \nantibody to CA-125 [ovarex], recombinant immunotoxin to \nmesothelin) \n4 Limitless \nreplicative \npotential \n Telomerase modifiers \n5 Sustained \nangiogenesis \nAnti-VEGF monoclonal antibody \n(bevacizumab), VEGF receptor \ntyrosine kinase inhibitors \nAngiozyme (cleaves mRNA for Flt-1, the main receptor for \nVEGF), Protein kinase C-beta inhibitor (LY317615), COX-2 \ninhibitor, thalidomide, lysophosphatidic acid inhibitors \n \n6 Tissue invasion \nand metastasis \n Inhibit/modify Catenin/Cadherin signaling, Selective MMP \ninhibitors \n7 Genomic instability  Gene therapy to deliver therapeutic or corrective gene to alter \noncogenes/TSG balance. Genes may be delivered by \ninfectious (adenovirus) or non-infectious (liposome) vectors. \nExamples: \nAdenoviral E1A (oppose HE-2/neu oncogene), \nAdenovirus transfection of wild-type p53 (restore TSG) \nTransfect viral suicide genes like HSV-thymidine kinase \n(sensitizes to ganciclovir cytotoxicity) \nAntisense oligonucletoides (targeting proto-oncogenes, \noncogenes like c-myc, protein kinase C-alpha [affinitak]) \nTribozymes (cleave oncogenes transcripts) \n \n\nChapter 5.1 Discussion of chapter 1: Experimental investigation of endometriosis \n347 \n \nFootnotes  Consequent to endometriosis heterogeneity, the exact contribution of each \nhallmark component may vary between individuals and clinical symptoms. Nevertheless, \ntherapies can be designed to target predominant categories following endometriosis \nmolecular classification (expression signature). \n \nFigure 5.2. Evaluating, in parallel, differences between genomic, transcriptomic and \nproteomic array platforms to identify candidate molecular pathways \n \nIntegrating microarrays \ninto endometriosis gene \nidentification strategies\n \n\nChapter 5.1 Discussion of chapter 1: Experimental investigation of endometriosis \n348 \n \nTable 5.2. Summary of studies comparing genomic, transcriptomic and proteomic \nprofiling of endometriosis using high-through put microarray technology \n \nGENOMIC TRANSCRIPTOMIC PROTEOMIC \nComparative Genomic \nHybridisation (CGH)  7-9  \n \nHuman endometriosis  \n(ectopic vs. eutopic \nendometrium) 10-24 \n \nDNA originating from blood \n(lymphocyte sourced ) of \nwomen with endometriosis 25  \n \nAnimal model endometriosis \n26;27  \nEndometriosis: protein tissue \nmicroarrays 28;29 \n \nEndometriosis: surface-\nenhanced laser \ndesorption/ionization time-\nof-flight mass spectrometry \nprotein chip array 30 \n \n \nSingle nucleotide \npolymorphisms (SNP) \nmicroarray based chips \n \nNone identified for \nendometriosis (apart from \ndata presented in this thesis) \n \nSNP microarrays in other \ndisorders \nAdenomyosis 31 \nBladder cancer 32 \nProstate cancer 33 \nExpression microarrays in \nother disorders \nEndometrial cancer  \nNormal endometrium34 \nProteomics in other related \ndisorders \nAdenomyosis and \nleiomyomas: protein tissue \nmicroarrays 35 \n \n \n \nFootnotes \nMost studies have compared endometriosis (ectopic endometrium) with matched eutopic \nendometrium. \nCGH) studies have been undertaken in endometriosis, but could only detect relatively large-\nscale deletions or duplications. \n\nChapter 5.1 Discussion of chapter 1: Experimental investigation of endometriosis \n349 \n \nTable 5.3. Implications of thesis findings and future research directions for \nendometriosis \nKey overall themes \nEstablish biological tissue bank- biological samples (e.g. endometriosis, endometriosis associated \novarian cancer, matched eutopic endometrium and ovarian surface epithelium and blood) should be \nconsidered in conjunction with originating patient clinical epidemiological outcome data \nEstablish international coordinating body to pursue biological tissue bank, basic science and clinical \nendometriosis research \nImplications for genetic epidemiology  \nMeta-analysis of genetic association studies \nMeta-analysis of multiple gene expression analyses to validate candidate genes \nInterrogate accessible bioinformatic tissue expression databases to cross tabulate and identify \ncandidate genes-combine this with above meta-analyses. \nImplications for basic science research \nCompare and contrast molecular profiles from endometriosis and matched eutopic \nendometrium/ovarian surface/peritoneum/blood using genomic, transcriptomic and proteomic \ntechnology IN parallel. \nEvaluate clinical prognostic value of nuclear morphometry in evaluation of endometriosis \nUtilise animal models of endometriosis (e.g. baboon) to study molecular profiling (compare with \nhuman studies) and evaluate novel medical therapies \nImplications for clinical trials \nStandardize methodology (e.g. inclusion, exclusion and diagnostic criteria, fertility and \nendometriosis-specific quality of life outcomes)  \nIntroduce temporality in studies (i.e. long term follow up) as cost-effectiveness between medical and \nsurgical treatments may depend on rates of re-treatment or persistence of impaired quality of life \noutcomes. \nConsider assessing from duration of onset of symptoms rather than from time of diagnosis of \nendometriosis (delayed onset of presentation)-factor this in to epidemiological associations and \nquality of life outcomes of new endometriosis cases or individualised retrospective analysis of cases \nwith known endometriosis onset and duration  \nAim to identify early biomarkers (urine, blood, menstrual flow) which correlate to disease onset to \nenable early therapy (medical or surgical).  \nRCTs to evaluate medical vs. surgical treatments for specific anatomical or clinical presentations of \nendometriosis i.e. consider endometriosis as multiple differing clinical entities \nConsider (or evaluate the need) for long term follow up of early onset severe  or atypical \nendometriosis  in view of increased risk of ovarian cancer \n\nChapter 5.2 Discussion of chapter 2: Observational analytical studies \n350 \n \n5.2. Observational Analytical Studies \n \nThe observational analytical cohort studies have ascertained the incidence, natural history and \ntreatment outcomes of common encountered gynaecological disorders such as heavy \nmenstrual bleeding and endometrial hyperplasia 36-39 (Table 5.4). Furthermore, the cohort \ndesign has been employed to identify prognostic factors associated with the rare outcome of \nfailed female sterilisation40.  \nThe results of each study are immediately transferable to clinical practice and are likely to \nimprove health care outcomes as outlined in Table 5.4 Improvements will mainly arise \nthrough improved pre-operative patient counselling, better patient selection, consideration of \noutpatient rather than inpatient treatment modalities and adopting treatments that would \nreduce rates of hysterectomy for menorrhagia. \nAlthough observational analytical studies (for example, cohort and case-control studies) start \nwith a \"low quality\" rating of evidence (Table 5.5 Level of Evidence; Table 5.6 GRADE \nquality of evidence), grading upwards may be warranted if the magnitude of the treatment \neffect is very large, if there is evidence of a dose-response relation or if all plausible biases \nwould decrease the magnitude of an apparent treatment effect. Further confirmatory \nrandomised controlled trials (RCTs) would be needed to validate the observations noted in \nthe menorrhagia studies cited in this chapter. However, a larger prospective data set( perhaps \nmulticentre or national), with defined inclusion/exclusion criteria, may be provide a study \npopulation that minimises biases, and be of sufficient power to generate data of a standard \nthat approaches a RCT. Hence a future goal would be the creation of large linked \nprospectively collected patient datasets that could be flexibly used by both clinical and \nresearch organisations e.g. NHS electronic patient record.  \n\nChapter 5.2 Discussion of chapter 2: Observational analytical studies \n351 \n \nTable 5.4. Improved health care resulting from analytical observational studies. \nChapter  \nand \nReference \nStudy Beneficial health care outcome \n2.1 \n40 \nPredicting \nnegligence in female \nsterilization failure  \nEarly sterilisation failure is suggestive of a negligent rather \nthan non-negligent mechanism of failure.  \nResult increases awareness of need for adequate surgical \ntraining in sterilisation procedure \nResult also has medico-legal implications \n2.2 \n37 \nEffectiveness of a \nMirena in the \ntreatment of \nendometrial \nhyperplasia \nMirena is highly effective in treating non-atypical endometrial \nhyperplasia \nUse of Mirena will reduce the rate of hysterectomy for women \nwith non-atypical endometrial hyperplasia \n2.3 \n \n38 \nOutpatient vs. \nDaycase \nThermachoice \nablation \nThermachoice balloon ablation may be successfully carried out \nin outpatient local anaesthetic setting \nOutpatient Thermachoice population utilise less analgesia than \nday case Thermachoice population \nDuration of hospital stay is not entirely dependent on whether \noutpatient or daycase endometrial ablation is considered \nConsider patient suitability criteria (e.g. pain thresholds) when \noffering outpatient vs daycase ablation \n2.4 \n39 \nLong term outcome \nof outpatient \nThermachoice \nendometrial balloon \nablation \nThermachoice balloon ablation may be successfully carried out \nin outpatient local anaesthetic setting \nAblation reduces the rate of hysterectomy for women with \nmenorrhagia that is unresponsive to medical therapy \nHigher intrauterine ablation pressures correlate to better long \nterm outcome \n2.5 \n36 \nLong term outcomes \nfollowing \nhysteroscopic \nmyomectomy  \nRemoval of the intracavity component of the fibroid is \neffective in reducing abnormal uterine bleeding. \nThis effect is independent of the size of the fibroid removed, \nuterine cavity size and presence of other intramural/subserous \nfibroids \nWidespread adoption of this minimally invasive surgical \ntechnique will improve patient quality of life and reduce the \nneed for hysterectomy \n \n\nChapter 5.2 Discussion of chapter 2: Observational analytical studies \n352 \n \nTable 5.5.  Classification of evidence used by RCOG Guideline development (originate \nfrom US Agency for Health Care Research and Quality) 41 \nClassification of Evidence Levels  \nIa Evidence obtained from meta-analysis of randomised controlled trials.  \nIb Evidence obtained from at least one randomised controlled trial.  \nIIa Evidence obtained from at least one well-designed controlled study without \nrandomisation.  \nIIb Evidence obtained from at least one other type of well-designed quasi-experimental \nstudy.  \nIII Evidence obtained from well-designed non-experimental descriptive studies, such as \ncomparative studies, correlation studies and case studies.  \nIV Evidence obtained from expert committee reports or opinions and/or clinical \nexperience of respected authorities.  \nGrades of Recommendations  \n \nRequires at least one randomised controlled trial as part of a body of literature of \noverall good quality and consistency addressing the specific recommendation. \n(Evidence levels Ia, Ib) \n \nRequires the availability of well controlled clinical studies but no randomised \nclinical trials on the topic of recommendations. (Evidence levels IIa, IIb, III) \n \nRequires evidence obtained from expert committee reports or opinions and/or \nclinical experiences of respected authorities. Indicates an absence of directly \napplicable clinical studies of good quality. (Evidence level IV) \nGood Practice Point    \n \nRecommended best practice based on the clinical experience of the guideline  \ndevelopment group \n \n \n\nChapter 5.2 Discussion of chapter 2: Observational analytical studies \n353 \n \nTable 5.6. GRADE approach 42 (http://www.gradeworkinggroup.org/index.htm)  \nThe Grading of Recommendations Assessment, Development and Evaluation (GRADE) \nGRADE: Quality of evidence \nThe GRADE system classifies the quality of evidence in one of four levels: \nHigh quality— Further research is very unlikely to change our confidence in the estimate of \neffect \nModerate quality— Further research is likely to have an important impact on our confidence in \nthe estimate of effect and may change the estimate \nLow quality— Further research is very likely to have an important impact on our confidence in \nthe estimate of effect and is likely to change the estimate \nVery low quality— Any estimate of effect is very uncertain \nEvidence based on randomised controlled trials begins as high quality evidence, but our \nconfidence in the evidence may be decreased for several reasons, including:  \n Study limitations  \n Inconsistency of results  \n Indirectness of evidence  \n Imprecision  \n Reporting bias.  \nAlthough observational studies (for example, cohort and case-control studies) start with a \"low \nquality\" rating, grading upwards may be warranted if the magnitude of the treatment effect is \nvery large, if there is evidence of a dose-response relation or if all plausible biases would \ndecrease the magnitude of an apparent treatment effect.  \nGRADE: Strength of recommendation \nThe GRADE system offers two grades of recommendations: \"strong\" and \"weak\" depending \non whether effects of intervention clearly outweigh the undesirable effects, or clearly do not. If \ntrade-offs are less certain—either because of low quality evidence or because evidence \nsuggests that desirable and undesirable effects are closely balanced—weak recommendations \nbecome mandatory. \nFactors that affect the strength of a recommendation  \nFactor Examples of strong \nrecommendations \nExamples of weak \nrecommendations \nQuality of \nevidence \nMany high quality randomised \ntrials have shown the benefit of \ninhaled steroids in asthma \nOnly case series have examined the \nutility of pleurodesis in \npneumothorax \n\nChapter 5.2 Discussion of chapter 2: Observational analytical studies \n354 \n \nUncertainty about \nthe balance \nbetween desirable \nand undesirable \neffects \nAspirin in myocardial infarction \nreduces mortality with minimal \ntoxicity, inconvenience, and cost \nWarfarin in low risk patients with \natrial fibrillation results in small \nstroke reduction but increased \nbleeding risk and substantial \ninconvenience \nUncertainty or \nvariability in \nvalues and \npreferences \nYoung patients with lymphoma \nwill invariably place a higher \nvalue on the life prolonging effects \nof chemotherapy than on treatment \ntoxicity \nOlder patients with lymphoma may \nnot place a higher value on the life \nprolonging effects of chemotherapy \nthan on treatment toxicity \nUncertainty about \nwhether the \nintervention \nrepresents a wise \nuse of resources \nThe low cost of aspirin as \nprophylaxis against stroke in \npatients with transient ischemic \nattacks \nThe high cost of clopidogrel and of \ncombination dipyridamole and \naspirin as prophylaxis against stroke \nin patients with transient ischaemic \nattacks \n \n\nChapter 5.3 Discussion of chapter 3: Systematic reviews \n355 \n \n5.3. Systematic reviews \n \nThe systematic review in chapter 3 has utilised robust methodology (systematic search, meta-\nanalysis, grading of evidence) to assimilate the published literature relating to the screening \nand prevention of preterm labour43;44. A similar approach has been applied to the systematic \nreview of the clinical use of levonorgestrel-releasing intrauterine system (LNG-IUS) 45. \nHowever, for the LNG-IUS study, meta-analysis was precluded due to extensive study \nheterogeneity and paucity of suitable clinical trials.  \nIncluded in each review is an appraisal of the quality of evidence for each therapeutic \nintervention according to standardised criteria (RCOG, GRADE; Table 5.5, Table 5.6). In \nrelation to the screening-prevention of preterm labour, tables listing the evidence appraisal \n(Table 5.7) and resulting care algorithm (Table 5.8) are shown below. The algorithm for \nmanaging a women at high risk of preterm labour exemplifies how this evaluated research \nevidence may be effectively applied in the clinical setting.  \nFurthermore, both reviews have identified areas where future research is likely to be \nclinically advantageous, but where the evidence is currently lacking. The generation of \nhypotheses that require further confirmatory validation is another important end-product of \nsystematic reviews, and has been considered an essential pre-requisite by most research \nfunding councils when seeking funding to conduct the confirmatory clinical trials 46.  \n\nChapter 5.3 Discussion of chapter 3: Systematic reviews \n356 \n \nTable 5.7  Screening and preventative strategies that may reduce the risk of preterm \ndelivery \nStrategy for preventing preterm \ndelivery \nRCOG \nLevel of \nEvidence \nGRADE \nQuality \nof \nEvidence \nGRADE \nStrength of \nRecommendation \nAsymptomatic bacteriuria in all women Ia High Strong \nBacterial vaginosis in low-risk population \ngroups \nIa, Ib Moderate Weak \nElective cervical cerclage in high-risk \npregnancies \nIb, IIa, \nIIb \nModerate Strong \nIndicated cervical cerclage in women with \nshort cervical length on ultrasound \nIb, IIa, \nIIb \nModerate Strong \nProphylactic progesterone \nsupplementation in high-risk pregnancies \nIa, Ib High Strong \nSmoking cessation in all women IIb, III Very \nLow \nWeak \n \n\nChapter 5.3 Discussion of chapter 3: Systematic reviews \n357 \n \n \nANTENATAL \nVISIT AND \nPURPOSE \nInfection \n(Screen and \ntreat  \nBV, UTI) \nCervico-\nvaginal \n fFN \n \nUltrasound \nAbdominal \nand \nTransvaginal \n \nOther interventions to be considered \nPre-pregnancy \nCounselling on recurrence risk \nand any modifiable predisposing \nfactors \nYes \n \nNo No Cessation smoking and illicit drugs \nImprove BMI>25 \nThrombophilia screen if history suggests \nOptimise control of diabetes, high BP \nChange anticoagulation or \nantihypertensive drugs \n8 weeks’ \nRoutine booking bloods \n  \n \nYes  No Dating pregnancy \n \n  \n \nThrombophilia screen and commence \naspirin & LMWH if positive. \nLow dose aspirin if previous pre-\neclampsia (consider use if previous \nstillbirth, abruption, severe IUGR) \nProphylactic progesterone \nGeneral preterm birth education, support, \nand risk factor avoidance. \nScreen and treat BV, UTIs \nLow threshold for GTT testing \n12, 16, 20, 24, 28 weeks’ \nNuchal Translucency(12w) \nand/or  Triple Test or msAFP \n(15-18w) \nNo No Serial Cervical \nassessments in \nwomen at high \nrisk of PTD \nEmergency or elective (12-16w) cervical \ncerclage based on ultrasound findings \nand/or reproductive history \nEmergency cervical cerclage is not \nindicated if above 32 weeks‘ \nLow threshold for GTT testing \n22 weeks’ Yes  No Detailed fetal \nsurvey \nUterine artery \nDoppler \nLow dose aspirin if suspect pre-\neclampsia or IUGR due to uterine artery \nnotching and/or previous history \nScreen and treat BV and UTIs \n24, 28, 32, 36 weeks’ \nGTT at 28 weeks‘ \nNo Only if \nsymptomatic \nFetal growth and \numbilical artery \nDoppler \n \nProphylactic corticosteroids, antibiotics \nif symptomatic of PTL or PPROM. \nIn utero transfer to unit with NICU if \nsymptomatic with positive fFN \n \nLabour \nSpontaneous or induced \nYes  Helps \nconfirm \nLikelihood \nof \nPTL, PPROM \nAsses fetal well-\nbeing, and \npresentation \nProphylactic corticosteroids, antibiotics \n(especially GBS prophylaxis).    \nTocolytics if in utero transfer to unit with \nNICU is needed. \nPost-partum \n6 week antenatal check \nNo No No Review antenatal events and delivery \nIdentify modifiable factors for future \nprevention of PTD \nTable 5.8    Suggested antenatal strategy to prevent preterm delivery \n \nFootNote: Bacterial vaginosis, BV; BMI, body mass index; BP, blood pressure; fFN, fetal \nfibronectin; GBS, Group B streptococcus; GTT, glucose tolerance test; LMWH, IUGR, \nintrauterine growth restriction; low-molecular weight heparin; msAFP, maternal serum alpha-\nfetoprotein; NICU, neonatal intensive care unit; PPROM, preterm premature rupture of \nmembranes; PTL, preterm labour; UA, uterine artery; UTI, urinary tract infection.\n\nChapter 5.4 Discussion of chapter 4: Clinical Guideline Development \n358 \n \n5.4. Clinical guideline development \n \nThere are established methodologies utilised in the production of clinical guidelines (depicted \nin Figure 5.3). Four essential criteria have been defined by the Appraisal of Guidelines for \nResearch and Evaluation in Europe (AGREE) guidelines 41  and include: \n Systematic review of the literature  \n Graded recommendations with explicit links to the evidence (Tables 4v in Chapter 4) \n Input of a multidisciplinary working group  \n Quality control; for example, input by an independent advisory board or by independent \npeer review.  \nAll guidelines developed in this chapter comply with the four essential AGREE criteria 41.   \nFigure 5.3 Derivation of clinical guidelines \nQuality Control (Peer review) \nPublished clinical guideline\n \n\nChapter 5.4 Discussion of chapter 4: Clinical Guideline Development \n359 \n \nThe publications of the clinical guidelines in chapter 4 are likely to have immediate and \nmaximal benefit on clinical practice. Hospital clinical guidelines are normally developed \nfrom guidelines published by professional bodies (e.g. RCOG47; chapter 4), specialist \nevidence-based resources (e.g. BMJ Clinical evidence48; chapter 4) and peer-reviewed \npublications (e.g. Surgical Endoscopy49; chapter 4). Clinical governance demands the \nutilisation of best up-to-date evidence, as indicated by such clinical guidelines, to ensure \npatient receive excellence in their clinical care50. \nThe work performed in the production of this research thesis has also identified potential \ndrawbacks in the research methodologies utilised. Given that the aim of research is to inform \nand benefit clinical practice, due consideration should be given to strategies that may \naugment the research methodological approaches considered in this thesis in order to achieve \nthis goal. These include: \nQuality control. For example, input by an independent advisory board or by independent \npeer review. In relation to the VBAC guideline 47, the RCOG Guidelines Development Group \ninvited peer review from British Maternal Fetal Medicine Society, National Childbirth Trust, \nObstetric anaesthetists, Midwifery supervisors, Obstetricians, and Neonatologists; this served \nas a the multidisciplinary component of the guideline development process. No specific \nmultidisciplinary working groups were employed by the other chapter guidelines48;49;51, \nalthough all had at least three external peer reviewers in addition to the journal‘s editor in \nchief. \nGrading the evidence base of the recommendation. The guidelines produced in this \nchapter have highlighted the significant change in how evidence is now valued for guideline \ndevelopment. The traditional and most common approach has been to value the research \nstudy evidence alone using the SIGN classification 52 or amendments from this (Table 5.5); \n\nChapter 5.4 Discussion of chapter 4: Clinical Guideline Development \n360 \n \nthe SIGN classification had been derived from earlier work by U.S. Preventive Services Task \nForce 53;54. This was used by RCOG VBAC, safe laparoscopic entry and failed sterilisation \nguidelines. However, this system neglects: the study‘s relevance for particular patients and \nsettings; inconsistencies amongst studies examining the same interventions; potential impact \nof health care resource limitations. The newer approach, using the GRADE system42, applied \nin the ectopic pregnancy guideline48, considers all these components when evaluating the \nstudy by using a strict framework  (Table 5.6). Grading judgments are expressed in a clear \nand simple manner as either high, moderate, low, or very low levels of supporting evidence \nand these are incorporated in the allocation of either strong or weak recommendations for \neach intervention (Table 5.9 provides an example for how ectopic pregnancy guideline was \nappraised). However, it has not been proven that this grading system is significantly superior \nto the traditional grading system in pragmatic clinical decision-making. Nonetheless, the \nGRADE system42 is advantageous in clearly identifying areas where there is uncertainty in \nthe level of evidence.  \n \nEvaluating the quality of clinical guidelines. Concern has been raised that guidelines \npublished by both professional medical bodies and peer-reviewed journals :lack sufficient \nreliability; are scientifically inaccurate;  fail to clarify what influence the level of health care \nresources may have on guideline practice; and should be critically reviewed by ‗experts‘ in \nguideline development prior to publication if not produced by ‗experts ‗(such as SIGN or \nNICE)55-59. Such criticism is harsh, but it is true that no guideline can ever achieve complete \ncoverage and applicability in every health care setting. \n\nChapter 5.4 Discussion of chapter 4: Clinical Guideline Development \n361 \n \nTable 5.9.Ectopic pregnancy evidence appraised using RCOG and GRADE criteria \nChapter 4.2 Ectopic pregnancy RCOG \nLevel of \nEvidence \nGRADE \nQuality of \nEvidence \nGRADE  \nStrength of \nRecommendation \nSalpingectomy in women not desiring \nfuture fertility is beneficial compared to \nsalpingotomy or methotrexate in achieving \nprimary treatment success \nIIa, IIb Moderate Strong \nProphylactic methotrexate (systemic) \nfollowing salpingotomy compared to \nsalpingotomy alone is beneficial in \nreducing the risk of persistent trophoblast \nIb, IIa Moderate Strong \nIn women desiring future fertility, \nsystemic methotrexate (single or multiple \ndose) and salpingotomy achieve similar \nprimary treatment success and subsequent \nfertility outcomes \nIa, Ib, IIa Moderate Strong \nIn women desiring future fertility, there is \nmarginally improved subsequent fertility \nrate by performing salpingotomy \ncompared to salpingectomy \nIII Very Low Weak \nSingle dose methotrexate may result in \nhigher rates of treatment failure in women \nwith ectopic pregnancies compared with \nmultiple dose regimens.  \nIa, Ib, IIb Low Weak \nIn selected cases, expectant management \nhas similar primary treatment success and \nfuture fertility outcomes to salpingectomy \nor salpingotomy \nIII Very Low Weak \nMethotrexate plus mifepristone is no more \neffective at increasing treatment success \nrates compared with methotrexate alone \nbut it seems this combination may be more \neffective in increasing treatment success \nrates in women with high levels of \nprogesterone.   \nIb Moderate Weak \n \n \n\nChapter 5.4 Discussion of chapter 4: Clinical Guideline Development \n362 \n \nFuture directions in guideline development \nIt is clear that strategies are needed, above and beyond new developments in grading \nevidence, in order to maintain guideline quality, consistency and applicability across all \nhealth care settings for all health care interventions. Apart from adopting the GRADE criteria \n42 to grade the level of evidence, one possibility is that guidelines should go through a formal \nappraisal process using validated tools, such as AGREE tool ( Appraisal of Guidelines, \nResearch, and Evaluation in Europe (AGREE)60 based on the Cluzeau instrument 59prior to \npublication. Alternatively, a ―second‖ expert consensus panel peer review should be \nundertaken prior to publication; these could be representatives of SIGN, NICE or Cochrane \ngroup. Importantly, consideration should be given to producing a more time efficient means \nof guideline development, with fewer group meetings and a shorter time frame; the RCOG \nVBAC guideline took the author nine months to complete from commencement to final peer-\nreviewed amended draft submission. Finally, there needs to be periodic review of the clinical \nguideline (such as every 2 years) to ensure the guideline remains valid and up-to-date with \nthe latest research evidence.  \n\nChapter 5.5 Future research themes arising from Thesis \n363 \n \n5.5. Future research themes arising from Thesis \n \nThe remainder of this chapter suggests future research themes that may augment the research \nmethodological approaches evaluated in this thesis in order to benefit clinical practice. \nIntegrating genomic, transcriptomic and proteomic high throughput technology \nThe Affymetrix Single Nucleotide Polymorphism (SNP) DNA microarray chip technology \nhas been validated 61and its raison d‘etre of obtaining genome wide association data are now \nbecoming fulfilled across many health conditions62. The Affymetrix SNP microarray chip \ntechnology has been successfully applied to the investigation of endometriosis (in this thesis), \nbladder cancer 32, and prostate cancer 33. The integration of genomic, transcriptomic and \nproteomic microarray technology is likely to yield greater discovery of genetic and molecular \naetiopathology 63-65.  Table 5.2 lists the published studies that utilised either genomic, \ntranscriptomic or proteomic approaches to investigate endometriosis (and related \nadenomyosis). Given this consensus opinion, and the established knowledge relating to the \ndifficulty of identifying endometriosis polymorphisms (that are likely to be multiple and \nhighly variably expressed across the population)  using traditional candidate gene \ninvestigative approaches66, future research should now be focused on adopting this high \nthroughput combinatorial approach (Table 5.2, Figure 5.2). \nAnatomical and molecular re-classification of endometriosis Genomic, \ntranscriptomic and proteomic profiling (Table 5.2, Figure 5.2) should provide a better \nunderstanding of the temporo-spatial relationship of endometriosis in relation to location \nwithin the female reproductive tract, nature during menstrual cycle phase, nature during \npelvic pain and infertility. Novel therapies should target the aberrant process at the molecular \nlevel, rather than focussing solely on endometriotic lesion eradication. Examples of strategies \n\nChapter 5.5 Future research themes arising from Thesis \n364 \n \nusing the cancer hallmark model (Table 5.1) are other approaches (Table 5.3) are discussed \nearlier.  \nTissue banks for endometriosis (and other important diseases) Our host institution \nprovided a unique tissue bank of endometriosis-associated ovarian cancer, but shared \nbiological tissue banks, matched with clinical epidemiological outcome data, would enable \ngreater research throughput and encourage collaboration. These aims are postulated bthey UK  \n67;68 and European bodies69, but various restrictions (particularly  Human Tissue Act 2004, \nimplemented September 2006) have hindered the creation of biological libraries. Given the \nsignificant burden of endometriosis on worldwide health, a central international body to \ncoordinate endometriosis tissue banking and scientific research is justified and urgently \nneeded. \nDeveloping and utilising animal models of disease There may be a significant \ndifference between a pharmacological agent that is effective in vtiro on a fundamental \ncellular process and one that is disease-specific and safe in vivo. Animal models could be \nuseful experimental tools, although opinion is divided as to their correlation to human \ndisease70. Nonetheless, several animal models for endometriosis exist (mouse71;72, rat73, \nbaboon74) , and these have been successfully used to test pathophysiology (such as molecular \nprofiling 26 and capacity for malignant transformation75) and novel pharmacological \ntherapies76.  \nImproved basic science and clinical science collaboration There appears to be \nvirtually no collaboration between clinical trialists and basic scientists. This is undoubtedly a \nmissed opportunity, as clinical trials could be designed to recover both biological specimens \nand clinical outcome measures. The effort and expense in generating a specific patient \npopulation may thus be used for greater benefit, particularly as quality biological sample \n\nChapter 5.5 Future research themes arising from Thesis \n365 \n \nresources are scarce. However, when designing the clinical trial, due consideration should be \ngiven to ensure that the trial is adequately powered for both primary clinical and translational \nresearch endpoints. \nThe need to increase translational potential of basic science research Translational \nresearch has been defined as research that considers and supports the transition of findings \nfrom research setting to benefit clinical practice. Importantly, the concept also includes \nfeedback of findings from the practice to research setting to ensure that the originating \nresearch remains applicable and appropriate; a concept that is frequently neglected. There has \nbeen concern that academic basic science research has not achieved its full potential. \nConsequently, there has been renewed impetus to support and create translational research \npartnerships, with significant financial grants tendered by National Institute for Health \nResearch (NIHR) and Medical Research Council (joint strategy), Wellcome Trust and Cancer \nResearch UK. \nCo-ordinated research programmes and commitment from Government funded \nresearch bodies Endometriosis research has been relatively neglected despite having a \nmajor burden of disease worldwide. It is entirely justified that a central worldwide \ncoordinating body of endometriosis research should be created given the importance of the \ncondition. Funding should be at government, European Parliament 69or international (such as \nWHO) level, particularly given the expense involved in high throughput molecular \ntechnology . In contrast, integrated clinical care and a common research strategy are \nfundamental components of the successful National Cancer Institute (NCI), the Gynecologic \nOncology Group (GOG) and the Specialized Programs of Research Excellence (SPOREs) \ncollaboration in the North American approach to ovarian cancer77. \n\nChapter 5.5 Future research themes arising from Thesis \n366 \n \nUtilising and developing high quality clinical datasets The ―gold standard‖ research \nmethodology that provides the highest level of evidence on interventions is the randomised \ncontrolled clinical trial (RCT). However, it is unrealistic to assume the RCT design may be \nachievable for all disorders. Particularly, as many of the main outcomes of interest are \nrelatively uncommon or tend to occur over several years. Furthermore, randomisation may be \nconsidered unethical in certain conditions, such as randomising between planned VBAC and \nelective caesarean as discussed in chapter 478.  \nThe alternative is to utilise a lower level of evidence in the form of the non-randomised \ncohort study. The optimum components of such a cohort study would be large sample size, \nlong term follow up, standardised definitions for recruiting and assessing outcome and ability \nto compare against matched non-intervention participants.  The Scottish Linked Maternity \ndataset represents such a cohort dataset based on routinely collected obstetric and neonatal \ndata; the interrogation of the dataset yielded publications 79;80 utilised by the VBAC clinical \nguideline (Chapter 4). However, there is considerable heterogeneity in which disorders are \nbeing monitored through such data repositories 81. Inevitably, most clinical guidelines \nconclude a need to establish national database registries for important outcomes (such as \nhypoxic ischaemic encephalopathy by NICE Intrapartum guideline82), but there is no \nagreement on which body should coordinate the data collection. \nFortunately, there appears to be a realisation that such datasets are urgently needed, both by \nDepartment  of Health (NHS Connecting for Health-NHS Care record service is one of many \ncomponents83) and medical research organisations (e.g. MRC67 ). However, there needs to be \nstringent quality assurance to ensure that the electronic health record contains accurate data, a \nconcern that has been raised by those establishing the analogous United States National \nHealth Information Infrastructure (NHII) project84. \n\nChapter 5.5 Future research themes arising from Thesis \n367 \n \nNevertheless, there are immense clinical and biomedical research opportunities by creating \nsuch datasets and sharing these amongst interested public and commercial stakeholders, that \nhave thus far, been relatively underplayed by the UK National Programme for Information \nTechnology85.  It is feasible that given the large sample size within the dataset and utilisation \nof individual patient data meta-analysis techniques, the clinical results generated are at least \nas reliable (if not more robust) than RCT derived data, even though the primary data are non-\nrandomised. Furthermore, the exchange of biological clinical samples, linked with patient \nepidemiological outcome, facilitates translational basic science research. \nCaution with over-reliance on meta-analyses: value according to the quality and \nmethodology of the RCTs included Despite adherence to rigorous quality standards for \nRCTs86, systematic reviews and meta-analyses may provide misleading results or be of poor \nquality87. Greater bias is likely in meta-analyses that pool underpowered trials, fewer trials, or \nexhibit marked heterogeneity86;88;89. The latter factor was identified by the meta-analysis in \nChapter 3 which demonstrated a statistically significant pooled result if the study population \nwere divided into high or low risk of preterm delivery43. Methods for assessment of \nmethodological quality by systematic reviews are still in their infancy, but there is likely to be \nsubstantial room for improvement86. When interpreting meta-analysis, careful consideration \nshould be given to the relevance, quality and robustness of the individual primary study in \naddition to the methodology used by the systematic review90. \n\nChapter 5.5 Future research themes arising from Thesis \n368 \n \nEnsure that clinical guidelines and their utilisation adds value to clinical practice \nThere is a concerted development to expand the programme of guideline development though \nprofessional medical bodies (such as RCOG) and national policy drivers (principally National \nInstitute of Clinical Excellence, UK) to fulfil the remit of delivering excellence in clinical \ncare. However, clinical guidelines are not necessarily transferable to current clinical \npractice91;92. Guidelines are mainly based on data derived from the ―perfect‖ population and \nhealth care resource setting because of the weighting it affords to RCTs. In contrast, the \nactual clinical population and environment tends to be heterogeneous and ―imperfect‖. \nFurthermore, such evidence-based health care policy will be practiced in a mainly non-\nevidence-based health care system.  \nImportantly, there is a dearth of evidence that has conclusively shown that the introduction of \nclinical guidelines has actually improved clinical practice93;94.  This research is vitally \nimportant, as it not only justifies continuing with the guideline but also provides feedback to \nguideline developers to enable improvements to content to maintain guideline validity, \napplicability and overall effectiveness95;96. A national audit represents the best study design \nfor evaluation and feedback of clinical guideline-led practice; examples of such audits are the \nlong-established perinatal and maternal mortality reports produced by CEMACH \n(Confidential Enquiry into Maternal and Child Health, \nwww.cemach.org.uk ). Alternatively, \ninstead of establishing numerous discrete national audits, data on multiple health outcomes \ncould be retrieved and processed from the proposed national electronic linked patient records \nsystem83. Again, the problem lies in creating a national coordinating audit body that could \nvalidate the effectiveness of guidelines, perhaps allied to the professional medical colleges. \n \n\nREFERENCES   Chapter: Introduction & Preliminaries \n369 \n \nREFERENCES \nIntroduction & Preliminaries  \n(1)  Department of Health. Research Governance Framework for Health and Social Care. Second \nedition, 2005. \nhttp://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4108962.  \n2005.  \n (2)  Medical Research Council. Medical Research Council of United Kingdom. Royal Chart er. \nAwarded October 1966. Amended July \n2003.http://www.mrc.ac.uk/Utilities/Documentrecord/index.htm?d=MRC002423.  2003.  \n (3)  Varma R, Gupta JK. Royal College of Obstetricians and Gynaecologists. Abnornal Uterine \nBleeding. Module 13. StratOG.net. Stuctured Training Resource to Assist Trainees in Obstetrics and \nGynaecology. June 2007. http://www stratog net/ [ 2007  \n (4)  Prowse AH, Manek S, Varma R, Liu J, Godwin AK, Maher ER et al. Molecular genetic \nevidence that endometriosis is a precursor of ovarian cancer. Int J Cancer 2006; 119(3):556-562. \n (5)  Varma R, Rollason T, Gupta JK, Maher ER. Endometriosis and the neoplastic process. \nReproduction 2004; 127(3):293-304. \n (6)   Genome wide SNP 100K analysis of endometriosis: demonstration of genomic imbalance. \n6th July 2007, EXCEL Conference centre, London. Oral Presentation. British Congress of Obstetrics and \nGynaecology (BCOG); Royal College of Obstetricians and Gynaecologists; 2007.  \n (7)   Genetic evidence for malignant transformation of endometriosis. Oral Presentation. 23rd \nAnnual Meeting of European Society of Human Reproduction and Embryology (ESHRE), Tête d'Or , Lyon \nCongress Centre, France; 1st July 2007. 2007. \n (8)  Varma R, Gupta JK. Predicting negligence in female sterilization failure using time inte rval to \nsterilization failure: analysis of 131 cases. Hum Reprod 2007; 22(9):2437 -2443. \n\nREFERENCES   Chapter: Introduction & Preliminaries \n370 \n \n (9)  Varma R, Soneja H, Bhatia K, Ganesan R, Rollason T, Clark TJ et al. The effectiveness of a \nlevonorgestrel-releasing intrauterine system (LNG-IUS) in the treatment of endometrial hyperplasia --a long-\nterm follow-up study. Eur J Obstet Gynecol Reprod Biol 2008; 139(2):169-175. \n (10)  Varma R, Soneja H, Samuel N, Sangha E, Clark TJ, Gupta JK. Hospital recovery following \nThermachoice ablation is not dependent on setting (outpatient or daycase) or rescue analgesia: unexpected \nresult. Eur J Obstet Gynecol Reprod Biol 2008; 140(1):76-81. \n (11)  Varma R, Soneja H, Clark TJ, Gupta JK. Outpatient Thermachoice endometrial balloon \nablation: long term, prognostic and quality of life measures. In Submission 2008. Eur J Obstet Gynecol Reprod \nBiol 2008. \n (12)  Varma R, Soneja H, Clark TJ, Gupta JK. Hysteroscopic myomectomy for menorrhagia using \nVersascopetrade mark bipolar system: Efficacy and prognostic factors at a minimum of one year follow up. Eur \nJ Obstet Gynecol Reprod Biol 2008. \n (13)  Varma R, Gupta JK, James DK, Kilby MD. Do screening-preventative interventions in \nasymptomatic pregnancies reduce the risk of preterm delivery--a critical appraisal of the literature. Eur J Obstet \nGynecol Reprod Biol 2006; 127(2):145-159. \n (14)  Varma R, Gupta JK. Antibiotic treatment of bacterial vaginosis in pregnancy: multiple meta-\nanalyses and dilemmas in interpretation. Eur J Obstet Gynecol Reprod Biol 2006; 124(1):10- 14. \n (15)  Varma R, Sinha D, Gupta JK. Non-contraceptive uses of levonorgestrel-releasing hormone \nsystem (LNG-IUS)--a systematic enquiry and overview. Eur J Obstet Gynecol Reprod Biol 2006; 125(1):9-28. \n (16)  Varma R, Gupta JK, Smith GC. Birth after previous caesarean section. Royal College of \nObstetricians and Gynaecologists Clinical Green top guideline No.45. February 2007. \nhttp://www.rcog.org.uk/index.asp?PageID=1913.  2008.  RCOG.  \n (17)  Varma R, Smith GC. Management of women with previous caesarean section. In Press. In: \nWarren R, Arulkumaran S, editors. Best Practice in Labour and Delivery. Cambridge University Press, \nCambridge, UK.; 2008. \n\nREFERENCES   Chapter: Introduction & Preliminaries \n371 \n \n (18)  Varma R, Gupta JK. Ectopic Pregnancy. \nhttp://clinicalevidence.bmj.com/ceweb/conditions/pac/1406/1406_background.jsp. BMJ Clinical Evidence . \n2006.  \n (19)  Varma R, Gupta JK. Laparoscopic entry techniques: clinical guideline, national survey, and \nmedicolegal ramifications. Surg Endosc 2008; 22(12):2686-2697. \n (20)  Varma R, Gupta JK. Failed sterilisation: evidence-based review and medico-legal \nramifications. BJOG 2004; 111(12):1322-1332. \n (21)  Varma R, Gupta JK. Minimising the risk of sterilisation failure: an evidence based approach. \nIn: O'Donovan P, editor. Complications in Gynaecological Surgery. Springer-Verlag, London; 2008. 106-126. \n \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n372 \n \nChapter 1 \n (1)  Lapp T. ACOG issues recommendations for the management of endometriosis. American \nCollege of Obstetricians and Gynecologists. American Family Physician 2000; 62(6):1431. \n (2)  Roberts CP, Rock JA. The current staging system for endometriosis: does it help?. Obstetrics \n& Gynecology Clinics of North America 2003; 30(1):115-132. \n (3)  Donnez J, Squifflet J, Pirard C, Jadoul P, Wyns C, Smets M. The efficacy of medical and \nsurgical treatment of endometriosis-associated infertility and pelvic pain. [Review] [46 refs]. Gynecologic & \nObstetric Investigation 2002; 54:Suppl-7. \n (4)  Sharpe-Timms KL. Endometrial anomalies in women with endometriosis. Annals of the New \nYork Academy of Sciences 2001; 943:131-147. \n (5)  Sampson JA. Endometrial carcinoma of the ovary arising in endometrial tissue in that organ. \nArch Surg 1925; 10:1-72. \n (6)  Hanahan D, Weinberg RA. The hallmarks of cancer. [Review] [94 refs]. Cell 2000; 100(1):57-\n70. \n (7)  Hill AB. The environment and disease:association or causation. Proc R Soc Med 1965; \n58:295-300. \n (8)  Bulun SE, Yang S, Fang Z, Gurates B, Tamura M, Sebastian S. Estrogen production and \nmetabolism in endometriosis. Annals of the New York Academy of Sciences 2002; 955:75- 85. \n (9)  Arvanitis DA, Koumantakis GE, Goumenou AG, Matalliotakis IM, Koumantakis EE, \nSpandidos DA. CYP1A1, CYP19, and GSTM1 polymorphisms increase the risk of endometriosis. Fertility & \nSterility 2003; 79:Suppl-9. \n (10)  Baxter SW, Thomas EJ, Campbell IG. GSTM1 null polymorphism and susceptibility t o \nendometriosis and ovarian cancer.Carcinogenesis 2001; 22(1):63-65. \n (11)  Attia GR, Zeitoun K, Edwards D, Johns A, Carr BR, Bulun SE. Progesterone receptor isoform \nA but not B is expressed in endometriosis. Journal of Clinical Endocrinology & Metabolism 2000; 85(8):2897-\n2902. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n373 \n \n (12)  Fauvet R, Poncelet C, Hugol D, Lavaur A, Feldmann G, Darai E. Expression of apoptosis-\nrelated proteins in endometriomas and benign and malignant ovarian tumours. Virchows Archiv 2003; \n443(1):38-43. \n (13)  Matsuzaki S, Canis M, Murakami T, Dechelotte P, Bruhat MA, Okamura K. Expression of the \ncyclin-dependent kinase inhibitor p27Kip1 in eutopic endometrium and peritoneal endometriosis. Fertility & \nSterility 2001; 75(5):956-960. \n (14)  Garcia-Velasco JA, Mulayim N, Kayisli UA, Arici A. Elevated soluble Fas ligand levels may \nsuggest a role for apoptosis in women with endometriosis. Fertility & Sterility 2002; 78(4):855- 859. \n (15)  Bischoff FZ, Heard M, Simpson JL. Somatic DNA alterations in endometriosis: high \nfrequency of chromosome 17 and p53 loss in late-stage endometriosis. Journal of Reproductive Immunology \n2002; 55(1-2):49-64. \n (16)  Chang CC, Hsieh YY, Tsai FJ, Tsai CH, Tsai HD, Lin CC. The proline form of p53 codon 72 \npolymorphism is associated with endometriosis. Fertility & Sterility 2002; 77(1):43- 45. \n (17)  Ria R, Loverro G, Vacca A, Ribatti D, Cormio G, Roccaro AM et al. Angiogenesis extent and \nexpression of matrix metalloproteinase-2 and -9 agree with progression of ovarian endometriomas. European \nJournal of Clinical Investigation 2002; 32(3):199-206. \n (18)  Ueda M, Yamashita Y, Takehara M, Terai Y, Kumagai K, Ueki K et al. Gene expression of \nadhesion molecules and matrix metalloproteinases in endometriosis. Gynecological Endocrinology 2002; \n16(5):391-402. \n (19)  Ueda M, Yamashita Y, Takehara M, Terai Y, Kumagai K, Ueki K et al. Survivin gene \nexpression in endometriosis. Journal of Clinical Endocrinology & Metabolism 2002; 87(7):3452-3459. \n (20)  Vidal JD, Register TC, Gupta M, Cline JM. Estrogen replacement therapy induces telomerase \nRNA expression in the macaque endometrium. Fertility & Sterility 2002; 77(3):601- 608. \n (21)  Wang Z, Kyo S, Maida Y, Takakura M, Tanaka M, Yatabe N et al. Tamoxifen regulates \nhuman telomerase reverse transcriptase (hTERT) gene expression differently in breast and endometrial cancer \ncells. Oncogene 2002; 21(22):3517-3524. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n374 \n \n (22)  Gazvani R, Templeton A. Peritoneal environment, cytokines and angiogenesis in the \npathophysiology of endometriosis. [Review] [99 refs]. Reproduction 2002; 123(2):217- 226. \n (23)  Folkman J. Role of angiogenesis in tumor growth and metastasis. [Review] [30 refs]. \nSeminars in Oncology 2002; 29(6:Suppl 16):Suppl-8. \n (24)  Kitawaki J, Obayashi H, Ohta M, Kado N, Ishihara H, Koshiba H et al. Genetic contribution \nof the interleukin-10 promoter polymorphism in endometriosis susceptibility. American Journal of Reproductive \nImmunology (Copenhagen) 2002; 47(1):12-18. \n (25)  Vigano P, Infantino M, Lattuada D, Lauletta R, Ponti E, Somigliana E et al. Intercellular \nadhesion molecule-1 (ICAM-1) gene polymorphisms in endometriosis. Molecular Human Reproduction 2003; \n9(1):47-52. \n (26)  Wieser F, Schneeberger C, Tong D, Tempfer C, Huber JC, Wenzl R. PROGINS receptor gene \npolymorphism is associated with endometriosis. Fertility & Sterility 2002; 77(2):309-312. \n (27)  Sehouli J, Mustea A, Koensgen D, Chen FC, Lichtenegger W. Interleukin-1 receptor \nantagonist gene polymorphism is associated with increased risk of epithelial ovarian cancer. Ann Oncol 2003; \n14(10):1501-1504. \n (28)  Kaklamani VG, Hou N, Bian Y, Reich J, Offit K, Michel LS et al. TGFBR1*6A and cancer \nrisk: a meta-analysis of seven case-control studies \n9. J Clin Oncol 2003; 21(17):3236-3243. \n (29)  Bojesen SE, Tybjaerg-Hansen A, Nordestgaard BG. Integrin beta3 Leu33Pro homozygosity \nand risk of cancer. J Natl Cancer Inst 2003; 95(15):1150-1157. \n (30)  Hefler LA, Grimm C, Ackermann S, Malur S, Radjabi-Rahat AR, Leodolter S et al. An \ninterleukin-6 gene promoter polymorphism influences the biological phenotype of ovarian cancer. Cancer Re s \n2003; 63(12):3066-3068. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n375 \n \n (31)  Kanamori Y, Matsushima M, Minaguchi T, Kobayashi K, Sagae S, Kudo R et al. Correlation \nbetween expression of the matrix metalloproteinase-1 gene in ovarian cancers and an insertion/deletion \npolymorphism in its promoter region. Cancer Res 1999; 59(17):4225- 4227. \n (32)  Chen GT, Tai CT, Yeh LS, Yang TC, Tsai HD. Identification of the cadherin subtypes present \nin the human peritoneum and endometriotic lesions: potential role for P-cadherin in the development of \nendometriosis. Molecular Reproduction & Development 2002; 62(3):289- 294. \n (33)  Scotti S, Regidor PA, Schindler AE, Winterhager E. Reduced proliferation and cell adhesion \nin endometriosis. Molecular Human Reproduction 2000; 6(7):610-617. \n (34)  Witz CA, Takahashi A, Monto ya-Rodriguez IA, Cho S, Schenken RS. Expression of the \nalpha2beta1 and alpha3beta1 integrins at the surface of mesothelial cells: a potential attachment site of \nendometrial cells. Fertility & Sterility 2000; 74(3):579-584. \n (35)  Moreno-Bueno G, Gamallo C, Perez-Gallego L, de Mora JC, Suarez A, Palacios J. beta-\nCatenin expression pattern, beta-catenin gene mutations, and microsatellite instability in endometrioid ovarian \ncarcinomas and synchronous endometrial carcinomas. Diagnostic Molecular Pathology 2001; 10(2):116- 122. \n (36)  Palacios J, Gamallo C. Mutations in the beta-catenin gene (CTNNB1) in endometrioid ovarian \ncarcinomas. Cancer Research 1998; 58(7):1344-1347. \n (37)  Wu Y, Basir Z, Kajdacsy-Balla A, Strawn E, Macias V, Montgomery K et al. Resolution of \nclonal origins for endometriotic lesions using laser capture microdissection and the human androgen receptor \n(HUMARA) assay. Fertility & Sterility 2003; 79:Suppl-7. \n (38)  Bischoff FZ, Simpson JL. Heritability and molecular genetic studies of endometriosis. Human \nReproduction Update 2000; 6(1):37-44. \n (39)  Bischoff FZ, Heard M, Simpson JL. Somatic DNA alterations in endometriosis: high \nfrequency of chromosome 17 and p53 loss in late-stage endometriosis. Journal of Reproductive Immunology \n2002; 55(1-2):49-64. \n (40)  Sato N, Tsunoda H, Nishida M, Morishita Y, Takimoto Y, Kubo T et al. Loss of \nheterozygosity on 10q23.3 and mutation of the tumor suppressor gene PTEN in benign endometrial cyst of the \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n376 \n \novary: possible sequence progression from benign endometrial cyst to endometrioid carcinoma and clear cell \ncarcinoma of the ovary. Cancer Research 2000; 60(24):7052-7056. \n (41)  Jiang X, Morland SJ, Hitchcock A, Thomas EJ, Campbell IG. Allelotyping of endometriosis \nwith adjacent ovarian carcinoma reveals evidence of a common lineage \n55. Cancer Research 1998; 58(8):1707-1712. \n (42)  LaGrenade A, Silverberg SG. Ovarian tumors associated with atypical endometriosis. Human \nPathology 1988; 19(9):1080-1084. \n (43)  Nishida M, Watanabe K, Sato N, Ichikawa Y. Malignant transformation of ovarian \nendometriosis. Gynecologic & Obstetric Investigation 2000; 50:Suppl-25. \n (44)  Seidman JD. Prognostic importance of hyperplasia and atypia in endometriosis. International \nJournal of Gynecological Pathology 1996; 15(1):1-9. \n (45)  Bayramoglu H, Duzcan E. Atypical epithelial changes and mutant p53 gene expression in \novarian endometriosis. Pathology Oncology Research 2001; 7(1):33-38. \n (46)  Ogawa S, Kaku T, Amada S, Kobayashi H, Hirakawa T, Ariyoshi K et al. Ovarian \nendometriosis associated with ovarian carcinoma: a clinicopathological and immunohistochemical study. \nGynecologic Oncology 2000; 77(2):298-304. \n (47)  Oral E, Ilvan S, Tustas E, Korbeyli B, Bese T, Demirkiran F et al. Prevalence of \nendometriosis in malignant epithelial ovary tumours. European Journal of Obstetrics, Gynecology, & \nReproductive Biology 2003; 109(1):97-101. \n (48)  Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypical endometriosis: its close \nassociation with malignant epithelial tumours.[comment]. Histopathology 1997; 30(3):249- 255. \n (49)  Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypical endometriosis: its close \nassociation with malignant epithelial tumours.[comment]. Histopathology 1997; 30(3):249- 255. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n377 \n \n (50)  Baak JP, ten Kate FJ, Offerhaus GJ, van Lanschot JJ, Meijer GA. Routine morphometrical \nanalysis can improve reproducibility of dysplasia grade in Barrett's oesophagus surveillance biopsies. Journal of \nClinical Pathology 2002; 55(12):910-916. \n (51)  Kronqvist P, Kuopio T, Jalava P, Collan Y. Morphometrical malignancy grading is a valuable \nprognostic factor in invasive ductal breast cancer. British Journal of Cancer 2002; 87(11):1275-1280. \n (52)  Defrere S, Van LA, Gonzalez RR, Jouret M, Mettlen M, Donnez J. Quantification of \nendometriotic lesions in a murine model by fluorimetric and morphometric analyses. Hum Reprod 2006; \n21(3):810-817. \n (53)  Donnez J, Nisolle M, Smoes P, Gillet N, Beguin S, Casanas-Roux F. Peritoneal endometriosis \nand \"endometriotic\" nodules of the rectovaginal septum are two different entities. Fertil Steril 1996; 66(3):362-\n368. \n (54)  Fedele L, Marchini M, Bianchi S, Dorta M, Arcaini L, Fontana PE. Structural and \nultrastructural defects in preovulatory endometrium of normo-ovulating infertile women with minimal or mild \nendometriosis. Fertil Steril 1990; 53(6):989-993. \n (55)  Nisolle M, Casanas-Roux F, Anaf V, Mine JM, Donnez J. Morphometric study of the stromal \nvascularization in peritoneal endometriosis. Fertil Steril 1993; 59(3):681-684. \n (56)  Nisolle M, Donnez J. Peritoneal endometriosis, ovarian endometriosis, and adenomyotic \nnodules of the rectovaginal septum are three different entities. Fertil Steril 1997; 68(4):585- 596. \n (57)  Ota H, Igarashi S, Tanaka T. Morphometric evaluation of stromal vascularization in the \nendometrium in adenomyosis. Hum Reprod 1998; 13(3):715-719. \n (58)  Ota H, Tanaka T. Stromal vascularization in the endometrium during adenomyosis. Microsc \nRes Tech 2003; 60(4):445-449. \n (59)  Palatynski A. Morphometric investigations in experimental ovarian endometriosis. Ginekol \nPol 1988; 59(12):713-718. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n378 \n \n (60)  Regidor PA, Wagner I, Ruwe M, Regidor M, Schindler AE. Morphometric analyses of \nendometriotic tissues to determine their grade of activity. Gynecol Endocrinol 2002; 16(3):235- 243. \n (61)  Ruwe M, Donhuijsen K, Regidor PA, Leder LD, Schindler AE. Endometriosis: clinical, \nhistological and morphometric findings before and after Gn-RH agonist therapy. Zentralbl Gynakol 1998; \n120(8):391-398. \n (62)  Regidor PA, Wagner I, Ruwe M, Regidor M, Schindler AE. Morphometric analyses of \nendometriotic tissues to determine their grade of activity. Gynecological Endocrinology 2002; 16(3):235-243. \n (63)  Ballouk F, Ross JS, Wolf BC. Ovarian endometriotic cysts. An analysis of cytologic atypia \nand DNA ploidy patterns. American Journal of Clinical Pathology 1994; 102(4):415-419. \n (64)  Modesitt SC, Tortolero-Luna G, Robinson JB, Gershenson DM, Wolf JK. Ovarian and \nextraovarian endometriosis-associated cancer. Obstetrics & Gynecology 2002; 100(4):788- 795. \n (65)  Erzen M, Kovacic J. Relationship between endometriosis and ovarian cancer. European \nJournal of Gynaecological Oncology 1998; 19(6):553-555. \n (66)  Erzen M, Rakar S, Klancnik B, Syrjanen K, Klancar B. Endometriosis-associated ovarian \ncarcinoma (EAOC): an entity distinct from other ovarian carcinomas as suggested by a nested case-control \nstudy.[erratum appears in Gynecol Oncol 2001 Dec;83(3):617 Note: Klancar B [corrected to Klancnik B]]. \nGynecologic Oncology 2001; 83(1):100-108. \n (67)  Stern RC, Dash R, Bentley RC, Snyder MJ, Haney AF, Robboy SJ. Malignancy in \nendometriosis: frequency and comparison of ovarian and extraovarian types. International Journal of \nGynecological Pathology 2001; 20(2):133-139. \n (68)  Zaino R, Whitney C, Brady MF, DeGeest K, Burger RA, Buller RE. Simultaneously detected \nendometrial and ovarian carcinomas--a prospective clinicopathologic study of 74 cases: a gynecologic oncology \ngroup study. Gynecologic Oncology 2001; 83(2):355-362. \n (69)  Stern RC, Dash R, Bentley RC, Snyder MJ, Haney AF, Robboy SJ. Malignancy in \nendometriosis: frequency and comparison of ovarian and extraovarian types. International Journal of \nGynecological Pathology 2001; 20(2):133-139. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n379 \n \n (70)  Erzen M, Kovacic J. Relationship between endometriosis and ovarian cancer. European \nJournal of Gynaecological Oncology 1998; 19(6):553-555. \n (71)  Vercellini P, Scarfone G, Bolis G, Stellato G, Carinelli S, Crosignani PG. Site of origin of \nepithelial ovarian cancer: the endometriosis connection. BJOG: an International Journal of Obstetrics & \nGynaecology 2000; 107(9):1155-1157. \n (72)  Al Fozan H, Tulandi T. Left lateral predisposition of endometriosis and endometrioma. \nObstetrics & Gynecology 2003; 101(1):164-166. \n (73)  National Statistics. Cancer trends in England and Wales 1950-1999.  2003.  UK Government.  \n (74)  Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypical endometriosis: its close \nassociation with malignant epithelial tumours. Histopathology 1997; 30(3):249-255. \n (75)  Oral E, Ilvan S, Tustas E, Korbeyli B, Bese T, Kosebay D. Prevalence of endometriosis in \nmalignant epithelial ovary tumours. Fertility & Sterility 77 Suppl 1:S33-4, 2002. \n (76)  Brinton LA, Gridley G, Persson I, Baron J, Bergqvist A. Cancer risk after a hospital discharge \ndiagnosis of endometriosis.American Journal of Obstetrics & Gynecology 1997; 176(3):572- 579. \n (77)  Yoshikawa H, Jimbo H, Okada S, Matsumoto K, Onda T, Yasugi T et al. Prevalence of \nendometriosis in ovarian cancer. Gynecologic & Obstetric Investigation 2000; 50:Suppl- 7. \n (78)  Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypical endometriosis: its close \nassociation with malignant epithelial tumours.  Histopathology 1997; 30(3):249-255. \n (79)  Modesitt SC, Tortolero-Luna G, Robinson JB, Gershenson DM, Wolf JK. Ovarian and \nextraovarian endometriosis-associated cancer. Obstetrics & Gynecology 2002; 100(4):788- 795. \n (80)  Pecorelli S, Odicino F, Maisonneuve P, Creasman W, Shepherd J, Sideri M et al. Carcinoma \nof the ovary.Annual report on the results of treatment in gynaecological cancer. J Epidemiol Biostat 1998; 3:75-\n102. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n380 \n \n (81)  Stern RC, Dash R, Bentley RC, Snyder MJ, Haney AF, Robboy SJ. Malignancy in \nendometriosis: frequency and comparison of ovarian and extraovarian types. International Journal of \nGynecological Pathology 2001; 20(2):133-139. \n (82)  Erzen M, Rakar S, Klancar B, Syrjanen K. Endometriosis-associated ovarian carcinoma \n(EAOC): an entity distinct from other ovarian carcinomas as suggested by a nested case-control study. \nGynecologic Oncology 2001; 83(1):100-108. \n (83)  Modesitt SC, Tortolero-Luna G, Robinson JB, Gershenson DM, Wolf JK. Ovarian and \nextraovarian endometriosis-associated cancer. Obstetrics & Gynecology 2002; 100(4):788- 795. \n (84)  Modesitt SC, Tortolero-Luna G, Robinson JB, Gershenson DM, Wolf JK. Ovarian and \nextraovarian endometriosis-associated cancer. Obstetrics & Gynecology 2002; 100(4):788- 795. \n (85)  Brinton LA, Gridley G, Persson I, Baron J, Bergqvist A. Cancer risk after a hospital discharge \ndiagnosis of endometriosis.[comment]. American Journal of Obstetrics & Gynecology 1997; 176(3):572- 579. \n (86)  Bulun SE, Zeitoun KM, Takayama K, Sasano H. Estrogen biosynthesis in endometriosis: \nmolecular basis and clinical relevance. [Review] [31 refs]. Journal of Molecular Endocrinology 2000; 25(1):35-\n42. \n (87)  Matsuzaki S, Murakami T, Uehara S, Canis M, Sasano H, Okamura K. Expression of estrogen \nreceptor alpha and beta in peritoneal and ovarian endometriosis. Fertility & Sterility 2001; 75(6):1198- 1205. \n (88)  Wieser F, Schneeberger C, Tong D, Tempfer C, Huber JC, Wenzl R. PROGINS receptor gene \npolymorphism is associated with endometriosis. Fertil Steril 2002; 77(2):309- 312. \n (89)  Kitawaki J, Obayashi H, Ishihara H, Koshiba H, Kusuki I, Kado N et al. Oestrogen receptor-\nalpha gene polymorphism is associated with endometriosis, adenomyosis and leiomyomata. Human \nReproduction 2001; 16(1):51-55. \n (90)  Baranova H, Canis M, Ivaschenko T, Albuisson E, Bothorishvilli R, Baranov V et al. Possible \ninvolvement of arylamine N-acetyltransferase 2, glutathione S-transferases M1 and T1 genes in the development \nof endometriosis. Molecular Human Reproduction 1999; 5(7):636-641. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n381 \n \n (91)  Hadfield RM, Manek S, Weeks DE, Mardon HJ, Barlow DH, Kennedy SH et al. Linkage and \nassociation studies of the relationship between endometriosis and genes encoding the detoxification enzymes \nGSTM1, GSTT1 and CYP1A1. Molecular Human Reproduction 2001; 7(11):1073- 1078. \n (92)  Nakago S, Hadfield RM, Zondervan KT, Mardon H, Manek S, Weeks DE et al. Association \nbetween endometriosis and N-acetyl transferase 2 polymorphisms in a UK population. Molecular Human \nReproduction 2001; 7(11):1079-1083. \n (93)  Baxter SW, Thomas EJ, Campbell IG. GSTM1 null polymorphism and susceptibility to \nendometriosis and ovarian cancer. Carcinogenesis 2001; 22(1):63-65. \n (94)  Birnbaum LS, Cummings AM. Dioxins and endometriosis: a plausible hypothesis. \nEnvironmental Health Perspectives 2002; 110(1):15-21. \n (95)  Eskenazi B, Mocarelli P, Warner M, Samuels S, Vercellini P, Olive D et al. Serum dioxin \nconcentrations and endometriosis: a cohort study in Seveso, Italy. Environmental Health Perspectives 2002; \n110(7):629-634. \n (96)  Bulun SE, Zeitoun KM, Kilic G. Expression of dioxin-related transactivating factors and \ntarget genes in human eutopic endometrial and endometriotic tissues. American Journal of Obstetrics & \nGynecology 2000; 182(4):767-775. \n (97)  Ohtake F, Takeyama K, Matsumoto T, Kitagawa H, Yamamoto Y, Nohara K et al. \nModulation of oestrogen receptor signalling by association with the activated dioxin receptor.[comment]. Nature \n2003; 423(6939):545-550. \n (98)  Lebovic DI, Mueller MD, Taylor RN. Immunobiology of endometriosis. [Review] [100 refs]. \nFertility & Sterility 2001; 75(1):1-10. \n (99)  Druckmann R, Rohr UD. IGF-1 in gynaecology and obstetrics: update 2002. [Review] [172 \nrefs]. Maturitas 2002; 41:Suppl-83. \n (100)  Meresman GF, Vighi S, Buquet RA, Contreras-Ortiz O, Tesone M, Rumi LS. Apoptosis and \nexpression of Bcl-2 and Bax in eutopic endometrium from women with endometriosis. Fertility & Sterility \n2000; 74(4):760-766. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n382 \n \n (101)  Bischoff FZ, Heard M, Simpson JL. Somatic DNA alterations in endometriosis: high \nfrequency of chromosome 17 and p53 loss in late-stage endometriosis. Journal of Reproductive Immunology \n2002; 55(1-2):49-64. \n (102)  Vigano P, Infantino M, Lattuada D, Lauletta R, Ponti E, Somigliana E et al. Intercellular \nadhesion molecule-1 (ICAM-1) gene polymorphisms in endometriosis. Molecular Human Reproduction 2003; \n9(1):47-52. \n (103)  Wieser F, Fabjani G, Tempfer C, Schneeberger C, Sator M, Huber J et al. Analysis of an \ninterleukin-6 gene promoter polymorphism in women with endometriosis by pyrosequencing. Journal of the \nSociety for Gynecologic Investigation 2003; 10(1):32-36. \n (104)  Kitawaki J, Obayashi H, Ohta M, Kado N, Ishihara H, Koshiba H et al. Genetic contribution \nof the interleukin-10 promoter polymorphism in endometriosis susceptibility. American Journal of Reproductive \nImmunology (Copenhagen) 2002; 47(1):12-18. \n (105)  Landi S, Moreno V, Gioia-Patricola L, Guino E, Navarro M, de Oca J et al. Association of \ncommon polymorphisms in inflammatory genes interleukin (IL)6, IL8, tumor necrosis factor alpha, NFKB1, and \nperoxisome proliferator-activated receptor gamma with colorectal cancer. Cancer Research 2003; 63(13):3560-\n3566. \n (106)  Baxter SW, Thomas EJ, Campbell IG. GSTM1 null polymorphism and susceptibility to \nendometriosis and ovarian cancer. Carcinogenesis 2001; 22(1):63-65. \n (107)  Hadfield RM, Manek S, Weeks DE, Mardon HJ, Barlow DH, Kennedy SH et al. Linkage and \nassociation studies of the relationship between endometriosis and genes encoding the detoxification enzymes \nGSTM1, GSTT1 and CYP1A1. Molecular Human Reproduction 2001; 7(11):1073- 1078. \n (108)  Nakago S, Hadfield RM, Zondervan KT, Mardon H, Manek S, Weeks DE et al. Association \nbetween endometriosis and N-acetyl transferase 2 polymorphisms in a UK population. Molecular Human \nReproduction 2001; 7(11):1079-1083. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n383 \n \n (109)  Baranova H, Canis M, Ivaschenko T, Albuisson E, Bothorishvilli R, Baranov V et al. Possible \ninvolvement of arylamine N-acetyltransferase 2, glutathione S-transferases M1 and T1 genes in the development \nof endometriosis. Molecular Human Reproduction 1999; 5(7):636-641. \n (110)  Hull ML, Charnock-Jones DS, Chan CL, Bruner-Tran KL, Osteen KG, Tom BD et al. \nAntiangiogenic agents are effective inhibitors of endometriosis. Journal of Clinical Endocrinology & \nMetabolism 2003; 88(6):2889-2899. \n (111)  Dabrosin C, Gyorffy S, Margetts P, Ross C, Gauldie J. Therapeutic effect of angiostatin gene \ntransfer in a murine model of endometriosis. American Journal of Pathology 2002; 161(3):909- 918. \n (112)  Mizumoto H, Saito T, Ashihara K, Nishimura M, Takehara M, Tanaka R et al. Expression of \nmatrix metalloproteinases in ovarian endometriomas: immunohistochemical study and enzyme immunoassay. \nLife Sciences 2002; 71(3):259-273. \n (113)  Morin PJ. beta-catenin signaling and cancer. Bioessays 1999; 21(12):1021-1030. \n (114)  Witz CA, Takahashi A, Montoya-Rodriguez IA, Cho S, Schenken RS. Expression of the \nalpha2beta1 and alpha3beta1 integrins at the surface of mesothelial cells: a potential attachment site of \nendometrial cells. Fertility & Sterility 2000; 74(3):579-584. \n (115)  Starzinski-Powitz A, Handrow-Metzmacher H, Kotzian S. The putative role of cell adhesion \nmolecules in endometriosis: can we learn from tumour metastasis?. [Review] [46 refs]. Molecular Medicine \nToday 1999; 5(7):304-309. \n (116)  Lengauer C, Kinzler KW, Vogelstein B. Genetic instability in colorectal cancers. Nature \n1997; 386(6625):623-627. \n (117)  Lengauer C, Kinzler KW, Vogelstein B. Genetic instabilities in human cancers. Nature 1998; \n396(6712):643-649. \n (118)  Mills GB, Lu Y, Fang X, Wang H, Eder A, Mao M et al. The role of genetic abnormalities of \nPTEN and the phosphatidylinositol 3-kinase pathway in breast and ovarian tumorigenesis, prognosis, and \ntherapy. [Review] [110 refs]. Seminars in Oncology 2001; 28(5 Suppl 16):125-141. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n384 \n \n (119)  Lalloo F, Evans G. Molecular genetics and endometrial cancer. [Review] [38 refs]. Best \nPractice & Research in Clinical Obstetrics & Gynaecology 2001; 15(3):355-363. \n (120)  Jimbo H, Hitomi Y, Yoshikawa H, Yano T, Momoeda M, Sakamoto A et al. Evidence for \nmonoclonal expansion of epithelial cells in ovarian endometrial cysts. American Journal of Pathology 1997; \n150(4):1173-1178. \n (121)  Wu Y, Basir Z, Kajdacsy-Balla A, Strawn E, Macias V, Montgomery K et al. Resolution of \nclonal origins for endometriotic lesions using laser capture microdissection and the human androgen receptor \n(HUMARA) assay. Fertility & Sterility 2003; 79:Suppl-7. \n (122)  Tamura M, Fukaya T, Murakami T, Uehara S, Yajima A. Analysis of clonality in human \nendometriotic cysts based on evaluation of X chromosome inactivation in archival formalin-fixed, paraffin-\nembedded tissue. [Review] [36 refs]. Laboratory Investigation 1998; 78(2):213-218. \n (123)  Mayr D, Amann G, Siefert C, Diebold J, Anderegg B. Does endometriosis really have \npremalignant potential? A clonal analysis of laser-microdissected tissue. FASEB Journal 2003; 17(6):693-695. \n (124)  Gogusev J, Bouquet dJ, Telvi L, Doussau M, du MS, Stojkoski A et al. Genetic abnormalities \ndetected by comparative genomic hybridization in a human endometriosis-derived cell line. Mol Hum Reprod \n2000; 6(9):821-827. \n (125)  Gogusev J, Bouquet dJ, Telvi L, Doussau M, du MS, Stojkoski A et al. Detection of DNA \ncopy number changes in human endometriosis by comparative genomic hybridization. Human Genetics 1999; \n105(5):444-451. \n (126)  Bischoff FZ, Heard M, Simpson JL. Somatic DNA alterations in endometriosis: high \nfrequency of chromosome 17 and p53 loss in late-stage endometriosis. Journal of Reproductive Immunology \n2002; 55(1-2):49-64. \n (127)  Kosugi Y, Elias S, Malinak LR, Nagata J, Isaka K, Takayama M et al. Increased heterogeneity \nof chromosome 17 aneuploidy in endometriosis. American Journal of Obstetrics & Gynecology 1999; \n180(4):792-797. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n385 \n \n (128)  Goumenou AG, Arvanitis DA, Matalliotakis IM, Koumantakis EE, Spandidos DA. \nMicrosatellite DNA assays reveal an allelic imbalance in p16(Ink4), GALT, p53, and APOA2 loci in patients \nwith endometriosis. Fertility & Sterility 2001; 75(1):160-165. \n (129)  Obata K, Hoshiai H. Common genetic changes between endometriosis and ovarian cancer. \n[Review] [19 refs]. Gynecologic & Obstetric Investigation 2000; 50:Suppl-43. \n (130)  Thomas EJ, Campbell IG. Molecular genetic defects in endometriosis. [Review] [9 refs]. \nGynecologic & Obstetric Investigation 2000; 50:Suppl-50. \n (131)  Chen H-W, Li H-N, Lee Y-T, Yang P-C, Chen JJW, Tzeng C-R. Global analysis of \ndifferentially expressed genes in endometrium with or without endometriosis using human cDNA microarray. \nFertility and Sterility 2002; 77(1):S16. \n (132)  Tamura M, Fukaya T, Murakami T, Uehara S, Yajima A. Analysis of clonality in human \nendometriotic cysts based on evaluation of X chromosome inactivation in archival formalin-fixed, paraffin-\nembedded tissue. [Review] [36 refs]. Laboratory Investigation 1998; 78(2):213-218. \n (133)  Jiang X, Hitchcock A, Bryan EJ, Watson RH, Englefield P, Thomas EJ et al. Microsatellite \nanalysis of endometriosis reveals loss of heterozygosity at candidate ovarian tumor suppressor gene loci. Cancer \nRes 1996; 56(15):3534-3539. \n (134)  Campbell IG, Thomas EJ. Endometriosis: candidate genes. [Review] [41 refs]. Human \nReproduction Update 2001; 7(1):15-20. \n (135)  Jiang X, Morland SJ, Hitchcock A, Thomas EJ, Campbell IG. Allelotyping of endometriosis \nwith adjacent ovarian carcinoma reveals evidence of a common lineage. Cancer Research 1998; 58(8):1707-\n1712. \n (136)  Martini M, Ciccarone M, Garganese G, Maggiore C, Evangelista A, Rahimi S et al. Possible \ninvolvement of hMLH1, p16(INK4a) and PTEN in the malignant transformation of endometriosis. International \nJournal of Cancer 2002; 102(4):398-406. \n (137)  Sato N, Tsunoda H, Nishida M, Morishita Y, Takimoto Y, Kubo T et al. Loss of \nheterozygosity on 10q23.3 and mutation of the tumor suppressor gene PTEN in benign endometrial cyst of the \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n386 \n \novary: possible sequence progression from benign endometrial cyst to endometrioid carcinoma and clear cell \ncarcinoma of the ovary. Cancer Research 2000; 60(24):7052-7056. \n (138)  Bischoff FZ, Heard M, Simpson JL. Somatic DNA alterations in endometriosis: high \nfrequency of chromosome 17 and p53 loss in late-stage endometriosis. Journal of Reproductive Immunology \n2002; 55(1-2):49-64. \n (139)  Otsuka J, Okuda T, Sekizawa A, Amemiya S, Saito H, Okai T et al. K-ras mutation may \npromote carcinogenesis of endometriosis leading to ovarian clear cell carcinoma. Med Electron Microsc 2004; \n37(3):188-192. \n (140)  Sato N, Tsunoda H, Nishida M, Morishita Y, Takimoto Y, Kubo T et al. Loss of \nheterozygosity on 10q23.3 and mutation of the tumor suppressor gene PTEN in benign endometrial cyst of the \novary: possible sequence progression from benign endometrial cyst to endometrioid carcinoma and clear cell \ncarcinoma of the ovary. Cancer Res 2000; 60(24):7052-7056. \n (141)  Dinulescu DM, Ince TA, Quade BJ, Shafer SA, Crowley D, Jacks T. Role of K-ras and Pten in \nthe development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005; 11(1):63-\n70. \n (142)  Prefumo F, Venturini PL, Fulcheri E. Analysis of p53 and c-erbB-2 expression in ovarian \nendometrioid carcinomas arising in endometriosis. International Journal of Gynecological Pathology 2003; \n22(1):83-88. \n (143)  Modesitt SC, Tortolero-Luna G, Robinson JB, Gershenson DM, Wolf JK. Ovarian and \nextraovarian endometriosis-associated cancer. Obstet Gynecol 2002; 100(4):788-795. \n (144)  Missmer SA, Cramer DW. The epidemiology of endometriosis. [Review] [87 refs]. Obstetrics \n& Gynecology Clinics of North America 2003; 30(1):1-19. \n (145)  Hadfield R, Mardon H, Barlow D, Kennedy S. Delay in the diagnosis of endometriosis: a \nsurvey of women from the USA and the UK. Human Reproduction 1996; 11(4):878- 880. \n (146)  Arruda MS, Petta CA, Abrao MS, Benetti-Pinto CL. Time elapsed from onset of symptoms to \ndiagnosis of endometriosis in a cohort study of Brazilian women. Hum Reprod 2003; 18(4):756- 759. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n387 \n \n (147)  Kirwan JM, Tincello DG, Herod JJ, Frost O, Kingston RE. Effect of delays in primary care \nreferral on survival of women with epithelial ovarian cancer: retrospective audit. BMJ 2002; 324(7330):148-\n151. \n (148)  Horiuchi A, Itoh K, Shimizu M, Nakai I, Yamazaki T, Kimura K et al. Toward understanding \nthe natural history of ovarian carcinoma development: a clinicopathological approach. Gynecologic Oncology \n2003; 88(3):309-317. \n (149)  Moll UM, Chumas JC, Chalas E, Mann WJ. Ovarian carcinoma arising in atypical \nendometriosis. Obstetrics & Gynecology 1990; 75(3:Pt 2):t-9. \n (150)  Daya S. Characteristics of good causation studies. Semin Reprod Med 2003; 21(1):73- 83. \n (151)  Erzen M, Kovacic J. Relationship between endometriosis and ovarian cancer. Eur J Gynaecol \nOncol 1998; 19(6):553-555. \n (152)  Heaps JM, Nieberg RK, Berek JS. Malignant neoplasms arising in endometriosis.Obstet \nGynecol 1990; 75(6):1023-1028. \n (153)  Rusell P. The pathological assessment of ovarian neoplasms: Introduction to the common \n'epithelial' tumors and analysis of benign 'epithelial' tumors. Pathology 1979; 11:5- 26. \n (154)  Sainz dlC, Eichhorn JH, Rice LW, Fuller AF, Jr., Nikrui N, Goff BA. Histologic \ntransformation of benign endometriosis to early epithelial ovarian cancer. Gynecol Oncol 1996; 60(2):238-244. \n (155)  Takahashi K, Kurioka H, Irikoma M, Ozaki T, Kanasaki H, Miyazaki K. Benign or malignant \novarian neoplasms and ovarian endometriomas. Journal of the American Association of Gynecologic \nLaparoscopists 2001; 8(2):278-284. \n (156)  Vercellini P, Parazzini F, Bolis G, Carinelli S, Dindelli M, Vendola N et al. Endometriosis and \novarian cancer. Am J Obstet Gynecol 1993; 169(1):181-182. \n (157)  Yoshikawa H, Jimbo H, Okada S, Matsumoto K, Onda T, Yasugi T et al. Prevalence of \nendometriosis in ovarian cancer. Gynecologic & Obstetric Investigation 2000; 50 Suppl 1:11- 17. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n388 \n \n (158)  Modesitt SC, Tortolero-Luna G, Robinson JB, Gershenson DM, Wolf JK. Ovarian and \nextraovarian endometriosis-associated cancer. Obstetrics & Gynecology 2002; 100(4):788- 795. \n (159)  Aure JC, Hoeg K, Kolstad P. Carcinoma of the ovary and endometriosis. Acta Obstet Gynecol \nScand 1971; 50(1):63-67. \n (160)  Crozier MA, Copeland LJ, Silva EG, Gershenson DM, Stringer CA. Clear cell carcinoma of \nthe ovary: a study of 59 cases. Gynecol Oncol 1989; 35(2):199-203. \n (161)  Komiyama S, Aoki D, Tominaga E, Susumu N, Udagawa Y, Nozawa S. Prognosis of \nJapanese patients with ovarian clear cell carcinoma associated with pelvic endometriosis: clinicopathologic \nevaluation. Gynecol Oncol 1999; 72(3):342-346. \n (162)  Borgfeldt C, Andolf E. Cancer risk after hospital discharge diagnosis of benign ovarian cysts \nand endometriosis. Acta Obstetricia et Gynecologica Scandinavica 83(4):395-400, 2004. \n (163)  Ness RB, Cramer DW, Goodman MT, Kjaer SK, Mallin K, Mosgaard BJ et al. Infertility, \nfertility drugs, and ovarian cancer: a pooled analysis of case-control studies. American Journal of Epidemiology \n2002; 155(3):217-224. \n (164)  Olson JE, Cerhan JR, Janney CA, Anderson KE, Vachon CM, Sellers TA. Postmenopausal \ncancer risk after self-reported endometriosis diagnosis in the Iowa Women's Health Study. Cancer 2002; \n94(5):1612-1618. \n (165)  Brinton LA, Gridley G, Persson I, Baron J, Bergqvist A. Cancer risk after a hospital discharge \ndiagnosis of endometriosis. Am J Obstet Gynecol 1997; 176(3):572-579. \n (166)  Jimbo H, Yoshikawa H, Onda T, Yasugi T, Sakamoto A, Taketani Y. Prevalence of ovarian \nendometriosis in epithelial ovarian cancer. Int J Gynaecol Obstet 1997; 59(3):245-250. \n (167)  Vercellini P, Scarfone G, Bolis G, Stellato G, Carinelli S, Crosignani PG. Site of origin of \nepithelial ovarian cancer: the endometriosis connection. BJOG 2000; 107(9):1155- 1157. \n (168)  Varma R, Rollason T, Gupta JK, Maher ER. Endometriosis and the neoplastic process. \nReproduction 2004; 127(3):293-304. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n389 \n \n (169)  Ness RB, Cramer DW, Goodman MT, Kjaer SK, Mallin K, Mosgaard BJ et al. Infertility, \nfertility drugs, and ovarian cancer: a pooled analysis of case-control studies. Am J Epidemiol 2002; 155(3):217-\n224. \n (170)  Lavery S, Gillmer M. Malignant transformation of residual endometriosis in women on \nunopposed oestrogen hormone replacement therapy. BJOG: an International Journal of Obstetrics & \nGynaecology 2001; 108(10):1106-1107. \n (171)  Reimnitz C, Brand E, Nieberg RK, Hacker NF. Malignancy arising in endometriosis \nassociated with unopposed estrogen replacement. Obstet Gynecol 1988; 71(3 Pt 2):444-447. \n (172)  Purdie DM, Bain CJ, Siskind V, Russell P, Hacker NF, Ward BG et al. Hormone replacement \ntherapy and risk of epithelial ovarian cancer. Br J Cancer 1999; 81(3):559- 563. \n (173)  Riman T, Dickman PW, Nilsson S, Correia N, Nordlinder H, Magnusson CM et al. Hormone \nreplacement therapy and the risk of invasive epithelial ovarian cancer in Swedish women. J Natl Cancer Inst \n2002; 94(7):497-504. \n (174)  Purdie DM, Bain CJ, Siskind V, Webb PM, Green AC. Ovulation and risk of epithelial \novarian cancer. Int J Cancer 2003; 104(2):228-232. \n (175)  Deligeoroglou E, Michailidis E, Creatsas G. Oral contraceptives and reproductive system \ncancer. Ann N Y Acad Sci 2003; 997:199-208. \n (176)  Bosetti C, Negri E, Trichopoulos D, Franceschi S, Beral V, Tzonou A et al. Long-term effects \nof oral contraceptives on ovarian cancer risk. Int J Cancer 2002; 102(3):262-265. \n (177)  Hankinson SE, Hunter DJ, Colditz GA, Willett WC, Stampfer MJ, Rosner B et al. Tubal \nligation, hysterectomy, and risk of ovarian cancer. A prospective study. JAMA 1993; 270(23):2813- 2818. \n (178)  Rosenblatt KA, Thomas DB. Reduced risk of ovarian cancer in women with a tubal ligation or \nhysterectomy. The World Health Organization Collaborative Study of Neoplasia and Steroid Contraceptives. \nCancer Epidemiol Biomarkers Prev 1996; 5(11):933-935. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n390 \n \n (179)  Tung KH, Goodman MT, Wu AH, McDuffie K, Wilkens LR, Kolonel LN et al. Reproductive \nfactors and epithelial ovarian cancer risk by histologic type: a multiethnic case-control study. Am J Epidemiol \n2003; 158(7):629-638. \n (180)  Konno R, Fujiwara H, Netsu S, Odagiri K, Shimane M, Nomura H et al. Gene expression \nprofiling of the rat endometriosis model. American Journal of Reproductive Immunology 2007; 58(4):330-343. \n (181)  Lebovic DI, Mwenda JM, Chai DC, Mueller MD, Santi A, Fisseha S et al. PPAR-gamma \nreceptor ligand induces regression of endometrial explants in baboons: a prospective, randomized, placebo- and \ndrug-controlled study. Fertility & Sterility 2007; 88(4:Suppl):Suppl-19. \n (182)  Nyachieo A, Chai DC, Deprest J, Mwenda JM, D'Hooghe TM. The baboon as a research \nmodel for the study of endometrial biology, uterine receptivity and embryo implantation. Gynecologic & \nObstetric Investigation 2007; 64(3):149-155. \n (183)  Becker CM, Sampson DA, Short SM, Javaherian K, Folkman J, D'Amato RJ. Short synthetic \nendostatin peptides inhibit endothelial migration in vitro and endometriosis in a mouse model. Fertil Steril 2006; \n85(1):71-77. \n (184)  Einspanier A, Lieder K, Bruns A, Husen B, Thole H, Simon C. Induction of endometriosis in \nthe marmoset monkey (Callithrix jacchus). Molecular Human Reproduction 2006; 12(5):291- 299. \n (185)  Gashaw I, Hastings JM, Jackson KS, Winterhager E, Fazleabas AT. Induced endometriosis in \nthe baboon (Papio anubis) increases the expression of the proangiogenic factor CYR61 (CCN1) in eutopic and \nectopic endometria. Biology of Reproduction 2006; 74(6):1060-1066. \n (186)  Hirata T, Osuga Y, Yoshino O, Hirota Y, Harada M, Takemura Y et al. Development of an \nexperimental model of endometriosis using mice that ubiquitously express green fluorescent protein. Human \nReproduction 2005; 20(8):2092-2096. \n (187)  Leiserowitz GS, Gumbs JL, Oi R, Dalrymple JL, Smith LH, Ryu J et al. Endometriosis-related \nmalignancies. International Journal of Gynecological Cancer 2003; 13(4):466- 471. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n391 \n \n (188)  Rubod C, Narducci F, Delattre C, Decocq J, Verbert A, Delahousse G. [Endometrioid \nadenocarcinoma arising from adenomyosis: a case report and literature review]. Journal de Gynecologie, \nObstetrique et Biologie de la Reproduction 33(2):140-4, 2004. \n (189)  Balmain A, Gray J, Ponder B. The genetics and genomics of cancer.Nature Genetics 2003; \n33:Suppl-44. \n (190)  Zondervan KT, Cardon LR, Kennedy SH. The genetic basis of endometriosis. [Review] [57 \nrefs]. Current Opinion in Obstetrics & Gynecology 2001; 13(3):309-314. \n (191)  Hadfield RM, Manek S, Weeks DE, Mardon HJ, Barlow DH, Kennedy SH et al. Linkage and \nassociation studies of the relationship between endometriosis and genes encoding the detoxification enzymes \nGSTM1, GSTT1 and CYP1A1. Molecular Human Reproduction 2001; 7(11):1073- 1078. \n (192)  Campbell IG, Thomas EJ. Endometriosis: candidate genes. [Review] [41 refs]. Human \nReproduction Update 2001; 7(1):15-20. \n (193)  Holt VL, Weiss NS. Recommendations for the design of epidemiologic studies of \nendometriosis. Epidemiology 2000; 11(6):654-659. \n (194)  Zondervan KT, Cardon LR, Kennedy SH. What makes a good case-control study? Design \nissues for complex traits such as endometriosis. Hum Reprod 2002; 17(6):1415-1423. \n (195)  Vigano P, Somigliana E, Parazzini F, Vercellini P. Bias versus causality: interpreting recent \nevidence of association between endometriosis and ovarian cancer. Fertility & Sterility 2007; 88(3):588- 593. \n (196)   Genome wide SNP 100K analysis of endometriosis: demonstration of genomic imbalance. \n6th July 2007, EXCEL Conference centre, London. Oral Presentation. British Congress of Obstetrics and \nGynaecology (BCOG); Royal College of Obstetricians and Gynaecologists; 2007. \n (197)   Genetic evidence for malignant transformation of endometriosis. Oral Presentation. 23rd \nAnnual Meeting of European Society of Human Reproduction and Embryology (ESHRE), Tête d'Or , Lyon \nCongress Centre, France; 1st July 2007. 2007. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n392 \n \n (198)  Somigliana E, Vigano' P, Parazzini F, Stoppelli S, Giambattista E, Vercellini P. Association \nbetween endometriosis and cancer: a comprehensive review and a critical analysis of clinical and \nepidemiological evidence. [Review] [91 refs]. Gynecologic Oncology 2006; 101(2):331-341. \n (199)  Vigano P, Somigliana E, Chiodo I, Abbiati A, Vercellini P. Molecular mechanisms and \nbiological plausibility underlying the malignant transformation of endometriosis: a critical analysis. [Review] \n[112 refs]. Human Reproduction Update 2006; 12(1):77-89. \n (200)  Jiang X, Hitchcock A, Bryan EJ, Watson RH, Englefield P, Thomas EJ et al. Microsatellite \nanalysis of endometriosis reveals loss of heterozygosity at candidate ovarian tumor suppressor gene loci. Cancer \nRes 1996; 56(15):3534-3539. \n (201)  Bischoff F, Simpson JL. Genetics of endometriosis: heritability and candidate genes. Best \nPractice & Research in Clinical Obstetrics & Gynaecology 2004; 18(2):219-232. \n (202)  Dinulescu DM, Ince TA, Quade BJ, Shafer SA, Crowley D, Jacks T. Role of K-ras and Pten in \nthe development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005; 11(1):63-\n70. \n (203)  Modesitt SC, Tortolero-Luna G, Robinson JB, Gershenson DM, Wolf JK. Ovarian and \nextraovarian endometriosis-associated cancer. Obstet Gynecol 2002; 100(4):788-795. \n (204)  Yoshikawa H, Jimbo H, Okada S, Matsumoto K, Onda T, Yasugi T et al. Prevalence of \nendometriosis in ovarian cancer. Gynecologic & Obstetric Investigation 2000; 50 Suppl 1:11- 17. \n (205)  Jimbo H, Yoshikawa H, Onda T, Yasugi T, Sakamoto A, Taketani Y. Prevalence of ovarian \nendometriosis in epithelial ovarian cancer. International Journal of Gynaecology & Obstetrics 1997; 59(3):245-\n250. \n (206)  Jiang X, Morland SJ, Hitchcock A, Thomas EJ, Campbell IG. Allelotyping of endometriosis \nwith adjacent ovarian carcinoma reveals evidence of a common lineage. Cancer Research 1998; 58(8):1707-\n1712. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n393 \n \n (207)  Noack F, Schmidt H, Buchweitz O, Malik E, Horny HP. Genomic imbalance and onco-protein \nexpression of ovarian endometrioid adenocarcinoma arisen in an endometriotic cyst. Anticancer Res 2004; \n24(1):151-154. \n (208)  Mhawech P, Kinkel K, Vlastos G, Pelte MF. Ovarian carcinomas in endometriosis: an \nimmunohistochemical and comparative genomic hybridization study. Int J Gynecol Pathol 2002; 21(4):401- 406. \n (209)  Bell DA. Origins and molecular pathology of ovarian cancer. Mod Pathol 2005; 18 Suppl \n2:S19-S32. \n (210)  Banerjee SK, Makdisi WF, Weston AP, Mitchell SM, Campbell DR. Microwave-based DNA \nextraction from paraffin-embedded tissue for PCR amplification. Biotechniques 1995; 18(5):768- 3. \n (211)  Liu Y, Ganesan TS. Tumour suppressor genes in sporadic epithelial ovarian cancer. \nReproduction 2002; 123(3):341-353. \n (212)  Esfandiari N, Ai J, Nazemian Z, Javed MH, Gotlieb L, Casper RF. Expression of glycodelin \nand cyclooxygenase-2 in human endometrial tissue following three-dimensional culture. American Journal of \nReproductive Immunology 2007; 57(1):49-54. \n (213)  Richter C, Baetje M, Bischof A, Makovitzky J, Richter DU, Gerber B et al. Expression of the \nglycodelin A gene and the detection of its protein in tissues and serum of ovarian carcinoma patients. Anticancer \nResearch 2007; 27(4A):2023-2025. \n (214)  Bersinger NA, von RS, Wunder DM, Raio L, Dreher E, Mueller MD. PAPP-A and \nosteoprotegerin, together with interleukin-8 and RANTES, are elevated in the peritoneal fluid of women with \nendometriosis. American Journal of Obstetrics & Gynecology 2006; 195(1):103-108. \n (215)  Kao LC, Germeyer A, Tulac S, Lobo S, Yang JP, Taylor RN et al. Expression profiling of \nendometrium from women with endometriosis reveals candidate genes for disease-based implantation failure \nand infertility. Endocrinology 2003; 144(7):2870-2881. \n (216)  Del Carmen MG, Smith Sehdev AE, Fader AN, Zahurak ML, Richardson M, Fruehauf JP et \nal. Endometriosis-associated ovarian carcinoma: differential expression of vascular endothelial growth factor \nand estrogen/progesterone receptors. Cancer 2003; 98(8):1658-1663. \n\nREFERENCES  Chapter 1. Basic science investigations of endometriosis \n394 \n \n (217)  Fazleabas AT, Brudney A, Chai D, Langoi D, Bulun SE. Steroid receptor and aromatase \nexpression in baboon endometriotic lesions. Fertility & Sterility 2003; 80:Suppl-7. \n (218)  Nisolle M, Casanas-Roux F, Donnez J. Immunohistochemical analysis of proliferative activity \nand steroid receptor expression in peritoneal and ovarian endometriosis. Fertility & Sterility 1997; 68(5):912-\n919. \n (219)  Wieser F, Schneeberger C, Tong D, Tempfer C, Huber JC, Wenzl R. PROGINS receptor gene \npolymorphism is associated with endometriosis. Fertility & Sterility 2002; 77(2):309-312. \n (220)  Prowse AH, Manek S, Varma R, Liu J, Godwin AK, Maher ER et al. Molecular genetic \nevidence that endometriosis is a precursor of ovarian cancer. Int J Cancer 2006; 119(3):556- 562. \n (221)  Potlog-Nahari C, Feldman AL, Stratton P, Koziol DE, Segars J, Merino MJ et al. CD10 \nimmunohistochemical staining enhances the histological detection of endometriosis. Fertility & Sterility 2004; \n82(1):86-92. \n \n\nREFERENCES   Chapter 2. Analytical observational studies \n395 \n \nChapter 2  \n \n(1)  RCOG. Royal College of Obstetricians and Gynaecologists. Male and female sterilisation. \nNational Evidence-Based Clinical Guideline Number 4.  2004.  RCOG Press, London.  \n (2)  Peterson HB, Xia Z, Hughes JM, Wilcox LS, Tylor LR, Trussell J. The risk of pregnancy after \ntubal sterilization: findings from the U.S. Collaborative Review of Sterilization. Am J Obstet Gynecol 1996; \n174(4):1161-1168. \n (3)  Trussell J, Guilbert E, Hedley A. Sterilization failure, sterilization reversal, and pregnancy \nafter sterilization reversal in Quebec. Obstet Gynecol 2003; 101(4):677-684. \n (4)  Varma R, Gupta JK. Failed sterilisation: evidence-based review and medico-legal \nramifications. BJOG 2004; 111(12):1322-1332. \n (5)  Cattanach v Melchior [2003] HCA 38. The costs of raising a child: Cattanach v Melchior and \nthe Justice and other legislation amendement bill 2003 (Qld). Cattanach v Melchior [2003] HCA 38 (16 July \n2003).  2003.  \n (6)  Kovacs GT, Krins AJ. Female sterilisations with Filshie clips: what is the risk failure? A \nretrospective survey of 30,000 applications. J Fam Plann Reprod Health Care 2002; 28(1):34- 35. \n (7)  Filshie GM, Helson K, Teper S. Day case sterilization with the Filshie Clip in Nottingham. \n10-year follow up study: the first 200 cases. In: Kruger T, Gome V, Van der Wat J, editors. 7th Annual Meeting \nof the International Society for Gynecologic Endoscopy. Bologna: Monduzzi Editore International Proceedings \nDivision; 1998. p.145-58.  1998.  \n (8)  Birdsall MA, Pattison NS, Wilson P. Female sterilisation: National Women's Hospital 1988-9. \nN Z Med J 1994; 107(990):473-475. \n (9)  Sokal D, Gates D, Amatya R, Dominik R. Two randomized controlled trials comparing the \ntubal ring and filshie clip for tubal sterilization. Fertil Steril 2000; 74(3):525-533. \n\nREFERENCES   Chapter 2. Analytical observational studies \n396 \n \n (10)  Dominik R, Gates D, Sokal D, Cordero M, Lasso dl, V, Remes RA et al. Two randomized \ncontrolled trials comparing the Hulka and Filshie Clips for tubal sterilization. Contraception 2000; 62(4):169-\n175. \n (11)  Femcare. Series of 32 Filshie Clip sterilisation failures litigated in Australia 1990-2000.  2004.  \n (12)  Nardin JM, Kulier R, Boulvain M. Techniques for the interruption of tubal patency for female \nsterilisation. Cochrane Database Syst Rev 2003;(1):CD003034. \n (13)  International Society of Gynaecological Pathologists (ISGP), World Health Organisation \n(WHO). Classification of Endometrial Hyperplasia. Blaustein's Pathology of the Female Genital Tract. 5 ed. \nSpringer; 2002. 467-500. \n (14)  Fox H, Buckley CH. The endometrial hyperplasias and their relationship to endometrial \nneoplasia. Histopathology 1982; 6(5):493-510. \n (15)  Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia. A long-term \nstudy of \"untreated\" hyperplasia in 170 patients. Cancer 1985; 56(2):403-412. \n (16)  Tabata T, Yamawaki T, Yabana T, Ida M, Nishimura K, Nose Y. Natural history of \nendometrial hyperplasia. Study of 77 patients. Arch Gynecol Obstet 2001; 265(2):85- 88. \n (17)  Terakawa N, Kigawa J, Taketani Y, Yoshikawa H, Yajima A, Noda K et al. The behavior of \nendometrial hyperplasia: a prospective study. Endometrial Hyperplasia Study Group. J Obstet Gynaecol Res \n1997; 23(3):223-230. \n (18)  Clark TJ, Neelakantan D, Gupta JK. The management of endometrial hyperplasia: An \nevaluation of current practice. Eur J Obstet Gynecol Reprod Biol 2005. \n (19)  Montgomery BE, Daum GS, Dunton CJ. Endometrial hyperplasia: a review. Obstetrical & \nGynecological Survey 59(5):368-78, 2004. \n (20)  Marsden DE, Hacker NF. Optimal management of endometrial hyperplasia. Best Practice & \nResearch in Clinical Obstetrics & Gynaecology 15(3):393-405, 2001. \n\nREFERENCES   Chapter 2. Analytical observational studies \n397 \n \n (21)  Marsden DE, Hacker NF. Optimal management of endometrial hyperplasia. Best Practice & \nResearch in Clinical Obstetrics & Gynaecology 15(3):393-405, 2001. \n (22)  Wildemeersch D, Janssens D, Pylyser K, De Wever N, Verbeeck G, Dhont M et al. \nManagement of patients with non-atypical and atypical endometrial hyperplasia with a levonorgestrel-releasing \nintrauterine system: long-term follow-up. Maturitas 2007; 57(2):210-213. \n (23)  Wheeler DT, Bristow RE, Kurman RJ. Histologic alterations in endometrial hyperplasia and \nwell-differentiated carcinoma treated with progestins. Am J Surg Pathol 2007; 31(7):988- 998. \n (24 )  Varma R, Sinha D, Gupta JK. Non-contraceptive uses of levonorgestrel-releasing hormone \nsystem (LNG-IUS)--a systematic enquiry and overview. Eur J Obstet Gynecol Reprod Biol 2006; 125(1):9-28. \n (25)  Perino A, Quartararo P, Catinella E, Genova G, Cittadini E. Treatment of endometrial \nhyperplasia with levonorgestrel releasing intrauterine devices. Acta Eur Fertil 1987; 18(2):137- 140. \n (26)  Scarselli G, Tantini C, Colafranceschi M, Taddei GL, Bargelli G, Venturini N et al. Levo-\nnorgestrel-nova-T and precancerous lesions of the endometrium. Eur J Gynaecol Oncol 1988; 9(4):284- 286. \n (27)  Vereide AB, Kaino T, Sager G, Arnes M, Orbo A. Effect of levonorgestrel IUD and oral \nmedroxyprogesterone acetate on glandular and stromal progesterone receptors (PRA and PRB), and estrogen \nreceptors (ER-alpha and ER-beta) in human endometrial hyperplasia. Gynecol Oncol 2006; 101(2):214- 223. \n (28)  Wildemeersch D, Dhont M. Treatment of nonatypical and atypical endometrial hyperplasia \nwith a levonorgestrel-releasing intrauterine system. American Journal of Obstetrics & Gynecology 2003; \n188(5):1297-1298. \n (29)  Hejmadi RK, Chaudhri S, Ganesan R, Rollason TP. Morphologic changes in the endometrium \nassociated with the use of the mirena coil: a retrospective study of 106 cases. Int J Surg Pathol 2007; 15(2):148-\n154. \n (30)  Horn LC, Meinel A, Handzel R, Einenkel J. Histopathology of endometrial hyperplasia and \nendometrial carcinoma: an update. Ann Diagn Pathol 2007; 11(4):297-311. \n\nREFERENCES   Chapter 2. Analytical observational studies \n398 \n \n (31)  Wheeler DT, Bristow RE, Kurman RJ. Histologic alterations in endometrial hyperplasia and \nwell-differentiated carcinoma treated with progestins. Am J Surg Pathol 2007; 31(7):988- 998. \n (32)  Phillips V, Graham CT, Manek S, McCluggage WG. The effects of the levonorgestrel \nintrauterine system (Mirena coil) on endometrial morphology. J Clin Pathol 2003; 56(4):305- 307. \n (33)  Kelly P, Dobbs SP, McCluggage WG. Endometrial hyperplasia involving endometrial polyps: \nreport of a series and discussion of the significance in an endometrial biopsy specimen. BJOG 2007 ; \n114(8):944-950. \n (34)  Altman DG, Bland JM. Diagnostic tests. 1: Sensitivity and specificity. BMJ 1994; \n308(6943):1552. \n (35)  Wildemeersch D, Janssens D, Pylyser K, De Wever N, Verbeeck G, Dhont M et al. \nManagement of patients with non-atypical and atypical endometrial hyperplasia with a levonorgestrel-releasing \nintrauterine system: long-term follow-up. Maturitas 2007; 57(2):210-213. \n (36)  Wheeler DT, Bristow RE, Kurman RJ. Histologic alterations in endometrial hyperplasia and \nwell-differentiated carcinoma treated with progestins. Am J Surg Pathol 2007; 31(7):988- 998. \n (37)  Vereide AB, Kaino T, Sager G, Arnes M, Orbo A. Effect of levonorgestrel IUD and oral \nmedroxyprogesterone acetate on glandular and stromal progesterone receptors (PRA and PRB), and estrogen \nreceptors (ER-alpha and ER-beta) in human endometrial hyperplasia. Gynecol Oncol 2006; 101(2):214- 223. \n (38)  Perino A, Quartararo P, Catinella E, Genova G, Cittadini E. Treatment of endometrial \nhyperplasia with levonorgestrel releasing intrauterine devices. Acta Eur Fertil 1987; 18(2):137- 140. \n (39)  Clark TJ, Mann CH, Shah N, Khan KS, Song F, Gupta JK. Accuracy of outpatient \nendometrial biopsy in the diagnosis of endometrial hyperplasia. Acta Obstet Gynecol Scand 2001; 80(9):784-\n793. \n (40)  Garuti G, Cellani F, Garzia D, Colonnelli M, Luerti M. Accuracy of hysteroscopic diagnosis \nof endometrial hyperplasia: a retrospective study of 323 patients. J Minim Invasive Gynecol 2005; 12(3):247-\n253. \n\nREFERENCES   Chapter 2. Analytical observational studies \n399 \n \n (41)  Gundem G, Sendag F, Kazandi M, Akercan F, Mgoyi L, Terek MC et al. Preoperative and \npostoperative correlation of histopathological findings in cases of endometrial hyperplasia. Eur J Gynaecol \nOncol 2003; 24(3-4):330-333. \n (42)  Lambert B, Muteganya D, Lepage Y, Boivin Y. Complex hyperplasia of the endometrium. \nPredictive value of curettage vs. hysterectomy specimens. J Reprod Med 1994; 39(8):639- 642. \n (43)  Jesadapatrakul S, Tangjitgamol S, Manusirivitaya S. Histopathologic consistency between \nendometrial hyperplasia diagnosis from endometrial curettage and pathologic diagnoses from hysterectomy \nspecimens. J Med Assoc Thai 2005; 88 Suppl 2:S16-S21. \n (44)  Wildemeersch D, Janssens D, Pylyser K, De Wever N, Verbeeck G, Dhont M et al. \nManagement of patients with non-atypical and atypical endometrial hyperplasia with a levonorgestrel-releasing \nintrauterine system: long-term follow-up. Maturitas 2007; 57(2):210-213. \n (45)  Wildemeersch D, Janssens D, Pylyser K, De Wever N, Verbeeck G, Dhont M et al. \nManagement of patients with non-atypical and atypical endometrial hyperplasia with a levonorgestrel-releasing \nintrauterine system: long-term follow-up. Maturitas 2007; 57(2):210-213. \n (46)  Rattanachaiyanont M, Angsuwathana S, Techatrisak K, Tanmahasamut P, Indhavivadhana S, \nLeerasiri P. Clinical and pathological responses of progestin therapy for non-atypical endometrial hyperplasia: a \nprospective study. J Obstet Gynaecol Res 2005; 31(2):98-106. \n (47)  Horn LC, Schnurrbusch U, Bilek K, Hentschel B, Einenkel J. Risk of progression in complex \nand atypical endometrial hyperplasia: clinicopathologic analysis in cases with and without progestogen \ntreatment. Int J Gynecol Cancer 2004; 14(2):348-353. \n (48)  Ferenczy A, Gelfand M. The biologic significance of cytologic atypia in progestogen-treated \nendometrial hyperplasia. Am J Obstet Gynecol 1989; 160(1):126-131. \n (49)  Randall TC, Kurman RJ. Progestin treatment of atypical hyperplasia and well-differentiated \ncarcinoma of the endometrium in women under age 40. Obstet Gynecol 1997; 90(3):434- 440. \n\nREFERENCES   Chapter 2. Analytical observational studies \n400 \n \n (50)  Perez-Medina T, Bajo J, Folgueira G, Haya J, Ortega P. Atypical endometrial hyperplasia \ntreatment with progestogens and gonadotropin-releasing hormone analogues: long-term follow-up. Gynecol \nOncol 1999; 73(2):299-304. \n (51)  Wildemeersch D, Janssens D, Pylyser K, De Wever N, Verbeeck G, Dhont M et al. \nManagement of patients with non-atypical and atypical endometrial hyperplasia with a levonorgestrel-releasing \nintrauterine system: long-term follow-up. Maturitas 2007; 57(2):210-213. \n (52)  Marsden DE, Hacker NF. Optimal management of endometrial hyperplasia. Best Practice & \nResearch in Clinical Obstetrics & Gynaecology 15(3):393-405, 2001. \n (53)  Jarvela IY, Santala M. Treatment of non-atypic endometrial hyperplasia using thermal balloon \nendometrial ablation therapy. Obstet Gynecol Surv 2005; 60(9):580-582. \n (54)  Agorastos T, Vaitsi V, Pantazis K, Efstathiadis E, Vavilis D, Bontis JN. Aromatase inhibitor \nanastrozole for treating endometrial hyperplasia in obese postmenopausal women. Eur J Obstet Gynecol Reprod \nBiol 2005; 118(2):239-240. \n (55)  NICE. CG44. Heavy Menstrual Bleeding. Clinical Guideline No 44. National Collaborating \nCentre for Women's and Children's Health Commissioned by the National Institute for Health and Clinical \nExcellence.  24-1-2007.  RCOG Press.  \n (56)  Reid PC. Endometrial ablation in England-coming of age? An examination of hospital episode \nstatistics 1989/1990 to 2004/2005. Eur J Obstet Gynecol Reprod Biol 2006. \n (57)  Amso NN. Clinical and health service implications of second generation endometrial ablation \ndevices. Current Opinion in Obstetrics & Gynecology 18(4):457-63, 2006. \n (58)  Lethaby A, Hickey M, Garry R. Endometrial destruction techniques for heavy menstrual \nbleeding. Cochrane Database Syst Rev 2007;(4):CD001501. \n (59)  Garside R, Stein K, Wyatt K, Round A. Microwave and thermal balloon ablation for heavy \nmenstrual bleeding: a systematic review. BJOG 2005; 112(1):12-23. \n\nREFERENCES   Chapter 2. Analytical observational studies \n401 \n \n (60)  Hurskainen R, Grenman S, Komi I, Kujansuu E, Luoto R, Orrainen M et al. Diagnosis and \ntreatment of menorrhagia. Acta Obstet Gynecol Scand 2007; 86(6):749-757. \n (61)  Practice Committee of the American Society for Reproductive Medicine. Indications and \noptions for endometrial ablation. [Review] [37 refs]. Fertility & Sterility 86(5 Suppl):S6-10, 2006. \n (62)  Aletebi FA, Vilos GA, Eskandar MA. Thermal balloon endometrial ablation to treat \nmenorrhagia in high-risk surgical candidates. J Am Assoc Gynecol Laparosc 1999; 6(4):435- 439. \n (63)  Soysal M, Soysal SK. Endometrial thermal balloon ablation under local anesthesia in patients \nwith prosthetic heart valves: a pilot study. Zentralbl Gynakol 2000; 122(11):556-560. \n (64)  Clark TJ, Gupta JK. Outpatient thermal balloon ablation of the endometrium. Fertil Steril \n2004; 82(5):1395-1401. \n (65)  Marsh F, Thewlis J, Duffy S. Randomized controlled trial comparing Thermachoice III* in the \noutpatient versus daycase setting. Fertil Steril 2007; 87(3):642-650. \n (66)  Olah KS, Alliston J, Jones J, Stewart G, Mavrommatis R. Thermal ablation performed in a \nprimary care setting: the South Warwickshire Experience. BJOG 2005; 112(8):1117- 1120. \n (67)  Amso NN, Fernandez H, Vilos G, Fortin C, McFaul P, Schaffer M et al. Uterine endometrial \nthermal balloon therapy for the treatment of menorrhagia: long-term multicentre follow-up study. Hum Reprod \n2003; 18(5):1082-1087. \n (68)  Bongers MY, Mol BW, Brolmann HA. Prognostic factors for the success of thermal balloon \nablation in the treatment of menorrhagia. Obstet Gynecol 2002; 99(6):1060-1066. \n (69)  Amso NN. Clinical and health service implications of second generation endometrial ablation \ndevices. Curr Opin Obstet Gynecol 2006; 18(4):457-463. \n (70)  Munro MG. Endometrial ablation: where have we been? Where are we going? Clinical \nObstetrics & Gynecology 49(4):736-66, 2006. \n\nREFERENCES   Chapter 2. Analytical observational studies \n402 \n \n (71)  Regional Palliative Care Program. In Edmonton Alberta.University of Alberta. Canada. \nMorphine Mean Equivalent Dose Calculation. \nwww.palliative.org/PC/ClinicalInfo/AssessmentTools/AssessmentToolsIDX.html.  2007.  \n (72)  Clark TJ, Gupta JK. Outpatient Hysteroscopy. The Obstetrician and Gynaecologist RCOG \nPress 2002; 4(4):217-221. \n (73)  Garside R, Stein K, Wyatt K, Round A, Price A. The effectiveness and cost-effectiveness of \nmicrowave and thermal balloon endometrial ablation for heavy menstrual bleeding: a systematic review and \neconomic modelling. Health Technol Assess 2004; 8(3):iii, 1-iii155. \n (74)  Lethaby A, Hickey M, Garry R. Endometrial destruction techniques for heavy menstrual \nbleeding. Cochrane Database Syst Rev 2006;(4):CD001501. \n (75)  Owusu-Ansah R, Gatongi D, Chien PF. Health technology assessment of surgical therapies for \nbenign gynaecological disease. Best Pract Res Clin Obstet Gynaecol 2006; 20(6):841- 879. \n (76)  Shaamash AH, Sayed EH. Prediction of successful menorrhagia treatment after thermal \nballoon endometrial ablation. J Obstet Gynaecol Res 2004; 30(3):210-216. \n (77)  Lok IH, Leung PL, Ng PS, Yuen PM. Life-table analysis of the success of thermal balloon \nendometrial ablation in the treatment of menorrhagia. Fertil Steril 2003; 80(5):1255- 1259. \n (78)  Vilos GA, Aletebi FA, Eskandar MA. Endometrial thermal balloon ablation with the \nThermaChoice system: effect of intrauterine pressure and duration of treatment. J Am Assoc Gynecol Laparosc \n2000; 7(3):325-329. \n (79)  Amso NN, Stabinsky SA, McFaul P, Blanc B, Pendley L, Neuwirth R. Uterine thermal \nballoon therapy for the treatment of menorrhagia: the first 300 patients from a multi-centre study. International \nCollaborative Uterine Thermal Balloon Working Group. Br J Obstet Gynaecol 1998; 105(5):517-523. \n (80)  Edwards P, Roberts I, Clarke M, DiGuiseppi C, Pratap S, Wentz R et al. Increasing response \nrates to postal questionnaires: systematic review. BMJ 2002; 324(7347):1183. \n\nREFERENCES   Chapter 2. Analytical observational studies \n403 \n \n (81)  Shaw RW, Brickley MR, Evans L, Edwards MJ. Perceptions of women on the impact of \nmenorrhagia on their health using multi-attribute utility assessment. Br J Obstet Gynaecol 1998; 105(11):1155-\n1159. \n (82)  Badia X, Roset M, Herdman M, Kind P. A comparison of United Kingdom and Spanish \ngeneral population time trade-off values for EQ-5D health states. Med Decis Making 2001; 21(1):7- 16. \n (83)  Clark TJ, Khan KS, Foon R, Pattison H, Bryan S, Gupta JK. Quality of life instruments in \nstudies of menorrhagia: a systematic review. Eur J Obstet Gynecol Reprod Biol 2002; 104(2):96-104. \n (84)  Jones GL, Kennedy SH, Jenkinson C. Health-related quality of life measurement in women \nwith common benign gynecologic conditions: a systematic review. Am J Obstet Gynecol 2002; 187(2):501- 511. \n (85)  Perneger TV. What's wrong with Bonferroni adjustments. BMJ 1998; 316(7139):1236-1238. \n (86)  Bender R, Lange S. Multiple test procedures other than Bonferroni's deserve wider use. BMJ \n1999; 318(7183):600-601. \n (87)  Marsh F, Thewlis J, Duffy S. Thermachoice endometrial ablation in the outpatient setting, \nwithout local anesthesia or intravenous sedation: a prospective cohort study. Fertil Steril 2005; 83(3):715-720. \n (88)  Loffer FD, Grainger D. Five-year follow-up of patients participating in a randomized trial of \nuterine balloon therapy versus rollerball ablation for treatment of menorrhagia. J Am Assoc Gynecol Laparosc \n2002; 9(4):429-435. \n (89)  NICE. CG44. Heavy Menstrual Bleeding. Clinical Guideline No 44. National Collaborating \nCentre for Women's and Children's Health Commissioned by the National Institute for Health and Clinical \nExcellence.  24-1-2007.  RCOG Press.  \n (90)  ACOG Committee on Practice Bulletins-Gynecology. ACOG practice bulletin. Surgical \nalternatives to hysterectomy in the management of leiomyomas. Number 16, May 2000 (replaces educational \nbulletin number 192, May 1994). International Journal of Gynaecology & Obstetrics 73(3):285-93, 2001. \n\nREFERENCES   Chapter 2. Analytical observational studies \n404 \n \n (91)  Farquhar C, Arroll B, Ekeroma A, Fentiman G, Lethaby A, Rademaker L et al. An evidence-\nbased guideline for the management of uterine fibroids. [Review] [135 refs]. Australian & New Zealand Journal \nof Obstetrics & Gynaecology 41(2):125-40, 2001. \n (92)  Lefebvre G, Vilos G, Allaire C, Jeffrey J, Arneja J, Birch C et al. The management of uterine \nleiomyomas. SOGC Canada. Journal of Obstetrics & Gynaecology Canada: JOGC 25(5):396- 418; quiz 419-22, \n2003. \n (93)  Indman PD. Hysteroscopic treatment of submucous myomas. [Review] [29 refs]. Clinical \nObstetrics & Gynecology 49(4):811-20, 2006. \n (94)  Clark TJ, Mahajan D, Sunder P, Gupta JK. Hysteroscopic treatment of symptomatic \nsubmucous fibroids using a bipolar intrauterine system: a feasibility study. European Journal of Obstetrics, \nGynecology, & Reproductive Biology 100(2):237-42, 2002. \n (95)  Clark TJ. Outpatient hysteroscopy and ultrasonography in the management of endometrial \ndisease. Current Opinion in Obstetrics & Gynecology 16(4):305-11, 2004. \n (96)  Bettocchi S, Ceci O, Di Venere R, Pansini MV, Pellegrino A, Marello F et al. Advanced \noperative office hysteroscopy without anaesthesia: analysis of 501 cases treated with a 5 Fr. bipolar electrode. \nHuman Reproduction 17(9):2435-8, 2002. \n (97)  Boe E, I, Woie K, Hordnes K. Transcervical endometrial resection: long-term results of 390 \nprocedures. Acta Obstetricia et Gynecologica Scandinavica 85(1):82-7, 2006. \n (98)  Campo S, Campo V, Gambadauro P. Short-term and long-term results of resectoscopic \nmyomectomy with and without pretreatment with GnRH analogs in premenopausal women. Acta Obstetricia et \nGynecologica Scandinavica 84(8):756-60, 2005. \n (99)  Cravello L, Farnarier J, Roger V, D'Ercole C, Blanc B. Hysteroscopic myomectomy. \nFunctional results with an average follow-up of 6 years. Journal de Gynecologie, Obstetrique et Biologie de la \nReproduction 27(6):593-7, 1998. \n (100)  Emanuel MH, Wamsteker K, Hart AA, Metz G, Lammes FB. Long-term results of \nhysteroscopic myomectomy for abnormal uterine bleeding. Obstetrics & Gynecology 93(5 Pt 1):743-8, 1999. \n\nREFERENCES   Chapter 2. Analytical observational studies \n405 \n \n (101)  Hart R, Molnar BG, Magos A. Long term follow up of hysteroscopic myomectomy assessed \nby survival analysis. British Journal of Obstetrics & Gynaecology 106(7):700-5, 1999. \n (102)  Polena V, Mergui JL, Perrot N, Poncelet C, Barranger E, Uzan S. Long-term results of \nhysteroscopic myomectomy in 235 patients. European Journal of Obstetrics, Gynecology, & Reproductive \nBiology 130(2):232-7, 2007. \n (103)  van Dongen H, Emanuel MH, Smeets MJ, Trimbos B, Jansen FW. Follow-up after incomplete \nhysteroscopic removal of uterine fibroids. Acta Obstetricia et Gynecologica Scandinavica 85(12):1463-7, 2006. \n (104)  Vercellini P, Zaina B, Yaylayan L, Pisacreta A, De Giorgi O, Crosignani PG. Hysteroscopic \nmyomectomy: long-term effects on menstrual pattern and fertility. Obstetrics & Gynecology 94(3):341-7, 1999. \n (105)  Dueholm M, Lundorf E. Transvaginal ultrasound or MRI for diagnosis of adenomyosis. Curr \nOpin Obstet Gynecol 2007; 19(6):505-512. \n (106)  Kepkep K, Tuncay YA, Goynumer G, Tutal E. Transvaginal sonography in the diagnosis of \nadenomyosis: which findings are most accurate? Ultrasound Obstet Gynecol 2007; 30(3):341- 345. \n \n \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n406 \n \nChapter 3.1 \n \n(1)  Varma R, Gupta JK, James DK, Kilby MD. Do screening-preventative interventions in \nasymptomatic pregnancies reduce the risk of preterm delivery--a critical appraisal of the literature. Eur J Obstet \nGynecol Reprod Biol 2006; 127(2):145-159. \n (2)  Varma R, Gupta JK. Antibiotic treatment of bacterial vaginosis in pregnancy: multiple meta-\nanalyses and dilemmas in interpretation. Eur J Obstet Gynecol Reprod Biol 2006; 124(1):10- 14. \n (3)  Varma R, Sinha D, Gupta JK. Non-contraceptive uses of levonorgestrel-releasing hormon e \nsystem (LNG-IUS)--a systematic enquiry and overview. Eur J Obstet Gynecol Reprod Biol 2006; 125(1):9-28. \n (4)  Higgins JPT, Green S. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic \nReviews of Interventions 4.2.6 [updated September 2006]. \nhttp://www.cochrane.org/resources/handbook/hbook.htm.  2006.  The Cochrane Collaboration.  \n (5)  Centre for Reviews and Dissemination. Undertaking systematic reviews of research on \neffectiveness: CRD's guidance for those carrying out or commissioning reviews. CRD Report 4 (2nd edition). \nhttp://www.york.ac.uk/inst/crd/report4.htm.  2001.  \n (6)  RCOG. Clinical Governance Advice No 1.Guidance for the Development of RCOG Green-top \nGuidelines.  2000.  Royal College of Obstetricians and Gynaecologists, London, UK.  \n (7)  Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S et al. Grading quality of \nevidence and strength of recommendations. BMJ 2004; 328(7454):1490. \n (8)  Moutquin JM. Classification and heterogeneity of preterm birth. BJOG: an International \nJournal of Obstetrics & Gynaecology 2003; 110:Suppl-3. \n (9)  Slattery MM, Morrison JJ. Preterm delivery. Lancet 2002; 360(9344):1489- 1497. \n (10)  Ward RM, Beachy JC. Neonatal complications following preterm birth. BJOG 2003; 110 \nSuppl 20:8-16. \n (11)  Institute for Clinical Systems Improvement (ICSI). Preterm Birth Prevention. ICSI Health \nCare Guideline.  2003.  Institute for Clinical Systems Improvement (ICSI). Bloomington (MN).  \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n407 \n \n (12)  American College of Obstetricians and Gynecologists. ACOG Practice Bulletin. Assessment \nof risk factors for preterm birth. Clinical management guidelines for obstetrician-gynecologists. Number 31, \nOctober 2001. Obstetrics & Gynecology 2001; 98(4):709-716. \n (13)  Iams JD. Prediction and early detection of preterm labor. Obstetrics & Gynecology 2003; \n101(2):402-412. \n (14)  El Bastawissi AY, Sorensen TK, Akafomo CK, Frederick IO, Xiao R, Williams MA. History \nof fetal loss and other adverse pregnancy outcomes in relation to subsequent risk of preterm delivery. Matern \nChild Health J 2003; 7(1):53-58. \n (15)  Carlini L, Somigliana E, Rossi G, Veglia F, Busacca M, Vignali M. Risk factors for \nspontaneous preterm birth: a Northern Italian multicenter case-control study. Gynecologic & Obstetric \nInvestigation 2002; 53(3):174-180. \n (16)  Adams MM, Elam-Evans LD, Wilson HG, Gilbertz DA. Rates of and factors associated with \nrecurrence of preterm delivery. JAMA 2000; 283(12):1591-1596. \n (17)  Airoldi J, Berghella V, Sehdev H, Ludmir J. Transvaginal ultrasonography of the cervix to \npredict preterm birth in women with uterine anomalies. Obstet Gynecol 2005; 106(3):553- 556. \n (18)  Jolly MC, Sebire N, Harris J, Robinson S, Regan L. Obstetric risks of pregnancy in women \nless than 18 years old. Obstetrics & Gynecology 2000; 96(6):962-966. \n (19)  Smith GC, Pell JP. Teenage pregnancy and risk of adverse perinatal outcomes associated with \nfirst and second births: population based retrospective cohort study. BMJ 2001; 323(7311):476. \n (20)  Bloom SL, Yost NP, McIntire DD, Leveno KJ. Recurrence of preterm birth in singleton and \ntwin pregnancies. Obstetrics & Gynecology 2001; 98(3):379-385. \n (21)  Moutquin JM. Socio-economic and psychosocial factors in the management and prevention of \npreterm labour. BJOG: an International Journal of Obstetrics & Gynaecology 2003; 110:Suppl- 60. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n408 \n \n (22)  Goldenberg RL, Iams JD, Mercer BM, Meis PJ, Moawad A, Das A et al. The Preterm \nPrediction Study: toward a multiple-marker test for spontaneous preterm birth. Am J Obstet Gynecol 2001; \n185(3):643-651. \n (23)  Jackson RA, Gibson KA, Wu YW, Croughan MS. Perinatal outcomes in singletons following \nin vitro fertilization: a meta-analysis. Obstet Gynecol 2004; 103(3):551-563. \n (24)  Ananth CV, Demissie K, Smulian JC, Vintzileos AM. Relationship among placenta previa, \nfetal growth restriction, and preterm delivery: a population-based study. Obstetrics & Gynecology 2001; \n98(2):299-306. \n (25)  Crane JM. Pregnancy outcome after loop electrosurgical excision procedure: a systematic \nreview. Obstet Gynecol 2003; 102(5 Pt 1):1058-1062. \n (26)  Smith GC, Pell JP, Dobbie R. Interpregnancy interval and risk of preterm birth and neonatal \ndeath: retrospective cohort study. BMJ 2003; 327(7410):313. \n (27)  Patel RR, Steer P, Doyle P, Little MP, Elliott P. Does gestation vary by ethnic group? A \nLondon-based study of over 122,000 pregnancies with spontaneous onset of labour. Int J Epidemiol 2004; \n33(1):107-113. \n (28)  Albertsen K, Andersen AM, Olsen J, Gronbaek M. Alcohol consumption during pregnancy \nand the risk of preterm delivery. Am J Epidemiol 2004; 159(2):155-161. \n (29)  Burguet A, Kaminski M, Abraham-Lerat L, Schaal JP, Cambonie G, Fresson J et al. The \ncomplex relationship between smoking in pregnancy and very preterm delivery. Results of the Epipage study. \nBJOG 2004; 111(3):258-265. \n (30)  Leitich H, Bodner-Adler B, Brunbauer M, Kaider A, Egarter C, Husslein P. Bacterial \nvaginosis as a risk factor for preterm delivery: a meta-analysis. Am J Obstet Gynecol 2003; 189(1):139- 147. \n (31)  Harrington K, Carpenter RG, Goldfrad C, Campbell S. Transvaginal Doppler ultrasound of \nthe uteroplacental circulation in the early prediction of pre-eclampsia and intrauterine growth retardation. Br J \nObstet Gynaecol 1997; 104(6):674-681. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n409 \n \n (32)  National Institute of Clinical Excellence U. NICE Clinical Guideline - Antenatal care - routine \ncare for healthy pregnant women.  2003.  \n (33)  Ehrenberg HM, Dierker L, Milluzzi C, Mercer BM. Low maternal weight, failure to thrive in \npregnancy, and adverse pregnancy outcomes. Am J Obstet Gynecol 2003; 189(6):1726-1730. \n (34)  Berkman ND, Thorp JM, Hartmann KE, Idicula AE, McPheeters M, Gavin NI et al. \nManagement of preterm labor.Evidence Report/Technology Assessment. Number 18 .AHRQ Publication No. \n01-E021.  2000.  Agency for Healthcare Research and Quality , Rockville, USA.  \n (35)  Lu MC, Tache V, Alexander GR, Kotelchuck M, Halfon N. Preventing low birth weight: is \nprenatal care the answer? Journal of Maternal-Fetal & Neonatal Medicine 2003; 13(6):362- 380. \n (36)  Lumley J, Oliver S, Waters E. Interventions for promoting smoking cessation during \npregnancy. Cochrane Database Syst Rev 2000;(2):CD001055. \n (37)  Armstrong MA, Gonzales O, V, Lieberman L, Carpenter DM, Pantoja PM, Escobar GJ. \nPerinatal substance abuse intervention in obstetric clinics decreases adverse neonatal outcomes. J Perinatol \n2003; 23(1):3-9. \n (38)  Little BB, Snell LM, Van Beveren TT, Crowell RB, Trayler S, Johnston WL. Treatment of \nsubstance abuse during pregnancy and infant outcome. Am J Perinatol 2003; 20(5):255-262. \n (39)  Sweeney PJ, Schwartz RM, Mattis NG, Vohr B. The effect of integrating substance abuse \ntreatment with prenatal care on birth outcome. J Perinatol 2000; 20(4):219-224. \n (40)  Xiong X, Buekens P, Alexander S, Demianczuk N, Wollast E. Anemia during pregnancy and \nbirth outcome: a meta-analysis. Am J Perinatol 2000; 17(3):137-146. \n (41)  Rasmussen K. Is There a Causal Relationship between Iron Deficiency or Iron-Deficiency \nAnemia and Weight at Birth, Length of Gestation and Perinatal Mortality? J Nutr 2001; 131(2S-2):590S-601S. \n (42)  Villar J, Merialdi M, Gulmezoglu AM, Abalos E, Carroli G, Kulier R et al. Nutritional \ninterventions during pregnancy for the prevention or treatment of maternal morbidity and preterm delivery: an \noverview of randomized controlled trials. J Nutr 2003; 133(5 Suppl 2):1606S-1625S. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n410 \n \n (43)  Cogswell ME, Parvanta I, Ickes L, Yip R, Brittenham GM. Iron supplementation during \npregnancy, anemia, and birth weight: a randomized controlled trial. Am J Clin Nutr 2003; 78(4):773-781. \n (44)  Friis H, Gomo E, Nyazema N, Ndhlovu P, Krarup H, Kaestel P et al. Effect of \nmultimicronutrient supplementation on gestational length and birth size: a randomized, placebo-controlled, \ndouble-blind effectiveness trial in Zimbabwe. Am J Clin Nutr 2004; 80(1):178- 184. \n (45)  Smaill F. Antibiotics for asymptomatic bacteriuria in pregnancy. Cochrane Database of \nSystematic Reviews 2003;(3). \n (46)  Bachmann LM, Coomarasamy A, Honest H, Khan KS. Elective cervical cerclage for \nprevention of preterm birth: a systematic review. Acta Obstetricia et Gynecologica Scandinavica 2003; \n82(5):398-404. \n (47)  Drakeley AJ, Roberts D, Alfirevic Z. Cervical stitch (cerclage) for preventing pregnancy loss \nin women. Cochrane Database of Systematic Reviews 2003;(1):CD003253. \n (48)  Odibo AO, Elkousy M, Ural SH, Macones GA. Prevention of preterm birth by cervical \ncerclage compared with expectant management: a systematic review. Obstetrical & Gynecological Survey 2003; \n58(2):130-136. \n (49)  MRC RCOG. Final report of the Medical Research Council/Royal College of Obstetricians \nand Gynaecologists multicentre randomised trial of cervical cerclage. MRC/RCOG Working Party on Cervical \nCerclage. British Journal of Obstetrics & Gynaecology 1993; 100(6):516-523. \n (50)  Guise JM, Mahon SM, Aickin M, Helfand M, Peipert JF, Westhoff C. Screening for bacterial \nvaginosis in pregnancy. Am J Prev Med 2001; 20(3 Suppl):62-72. \n (51)  Honest H, Bachmann LM, Knox EM, Gupta JK, Kleijnen J, Khan KS. The accuracy of \nvarious tests for bacterial vaginosis in predicting preterm birth: a systematic review. BJOG 2004; 111(5):409-\n422. \n (52)  Lamont RF. Infection in the prediction and antibiotics in the prevention of spontaneous \npreterm labour and preterm birth. BJOG: an International Journal of Obstetrics & Gynaecology 2003; \n110:Suppl-5. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n411 \n \n (53)  McDonald H, Brocklehurst P, Parsons J, Vigneswaran R. Antibiotics for treating bacterial \nvaginosis in pregnancy. Cochrane Database of Systematic Reviews 2004;(3). \n (54)  Riggs MA, Klebanoff MA. Treatment of vaginal infections to prevent preterm birth: a meta-\nanalysis. Clinical Obstetrics & Gynecology 47(4):796-807 2004. \n (55)  Leitich H, Brunbauer M, Bodner-Adler B, Kaider A, Egarter C, Husslein P. Antibiotic \ntreatment of bacterial vaginosis in pregnancy: a meta-analysis. American Journal of Obstetrics & Gynecology \n2003; 188(3):752-758. \n (56)  Guise JM, Mahon SM, Aickin M, Helfand M, Peipert JF, Westhoff C. Screening for bacterial \nvaginosis in pregnancy. American Journal of Preventive Medicine 2001; 20(3:Suppl):Suppl- 72. \n (57)  Okun N, Gronau KA, Hannah ME. Antibiotics for Bacterial Vaginosis or Trichomonas \nvaginalis in Pregnancy: A Systematic Review. Obstet Gynecol 2005; 105(4):857-868. \n (58)  Varma R, Gupta JK. Antibiotic treatment of bacterial vaginosis in pregnancy: Multiple meta-\nanalyses and dilemmas in interpretation. Eur J Obstet Gynecol Reprod Biol 2006; 124(1):10- 14. \n (59)  Lamont RF. Can antibiotics prevent preterm birth --the pro and con debate. BJOG 2005; 112 \nSuppl 1:67-73. \n (60)  Carey JC, Klebanoff MA, Hauth JC, Hillier SL, Thom EA, Ernest JM et al. Metronidazole to \nprevent preterm delivery in pregnant women with asymptomatic bacterial vaginosis. National Institute of Child \nHealth and Human Development Network of Maternal-Fetal Medicine Units. N Engl J Med 2000; 342(8):534-\n540. \n (61)  Kiss H, Petricevic L, Husslein P. Prospective randomised controlled trial of an infection \nscreening programme to reduce the rate of preterm delivery. BMJ 2004; 329(7462):371. \n (62)  Klebanoff MA, Carey JC, Hauth JC, Hillier SL, Nugent RP, Thom EA et al. Failure of \nmetronidazole to prevent preterm delivery among pregnant women with asymptomatic Trichomonas vaginalis \ninfection. New England Journal of Medicine 2001; 345(7):487-493. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n412 \n \n (63)  Gulmezoglu AM. Interventions for trichomoniasis in pregnancy.Cochrane Pregnancy and \nChildbirth Group. Cochrane Database of Systematic Reviews 2004;(2). \n (64)  Andrews WW, Goldenberg RL, Mercer B, Iams J, Meis P, Moawad A et al. The Preterm \nPrediction Study: association of second-trimester genitourinary chlamydia infection with subsequent \nspontaneous preterm birth. Am J Obstet Gynecol 2000; 183(3):662-668. \n (65)  Martin DH, Eschenbach DA, Cotch FA. Double-blind placebo controlled treatment trial of \nchlamdyia trachomatis endocervical infections in pregnant women. Inf Dis Ob Gynecol 1997; 5:10- 17. \n (66)  Raynes- Greenow CH, Roberts CL, Bell JC, Peat B, Gilbert GL. Antibiotics for ureaplasma in \nthe vagina in pregnancy. Cochrane Pregnancy and Childbirth Group. Cochrane Database of Systematic Reviews \n2004;(2). \n (67)  RCOG. Prevention of Early Onset Neonatal Group B Streptococcal Disease (36) - November \n2003 .  2003.  RCOG Press, London.  \n (68)  Klebanoff MA, Regan JA, Rao AV, Nugent RP, Blackwelder WC, Eschenbach DA et al. \nOutcome of the Vaginal Infections and Prematurity Study: results of a clinical trial of erythromycin among \npregnant women colonized with group B streptococci. Am J Obstet Gynecol 1995; 172(5):1540-1545. \n (69)  Schrag S, Gorwitz R, Fultz-Butts K, Schuchat A. Prevention of perinatal group B \nstreptococcal disease. Revised guidelines from CDC. MMWR Recomm Rep 2002; 51(RR-11):1- 22. \n (70)  Canadian Task Force on Preventive Health Care. Prevention of group B streptococcal \ninfection in newborns. Recommendation statement from the Canadian Task Force on Preventive Health Care. \nCan Fam Physician 2002; 48:934-936. \n (71)  Goldenberg RL, Iams JD, Das A, Mercer BM, Meis PJ, Moawad AH et al. The Preterm \nPrediction Study: sequential cervical length and fetal fibronectin testing for the prediction of spontaneous \npreterm birth. National Institute of Child Health and Human Development Maternal-Fetal Medicine Units \nNetwork. Am J Obstet Gynecol 2000; 182(3):636-643. \n (72)  Hibbard JU, Tart M, Moawad AH. Cervical length at 16-22 weeks' gestation and risk for \npreterm delivery. Obstetrics & Gynecology 2000; 96(6):972-978. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n413 \n \n (73)  Heath VC, Southall TR, Souka AP, Elisseou A, Nicolaides KH. Cervical length at 23 weeks \nof gestation: prediction of spontaneous preterm delivery. Ultrasound Obstet Gynecol 1998; 12(5):312-317. \n (74)  Taipale P, Hiilesmaa V. Sonographic measurement of uterine cervix at 18-22 weeks' gestation \nand the risk of preterm delivery. Obstet Gynecol 1998; 92(6):902-907. \n (75)  Guzman ER, Walters C, Ananth CV, O'Reilly-Green C, Benito CW, Palermo A et al. A \ncomparison of sonographic cervical parameters in predicting spontaneous preterm birth in high-risk singleton \ngestations. Ultrasound in Obstetrics & Gynecology 2001; 18(3):204-210. \n (76)  Honest H, Bachmann LM, Coomarasamy A, Gupta JK, Kleijnen J, Khan KS. Accuracy of \ncervical transvaginal sonography in predicting preterm birth: a systematic review. Ultrasound in Obstetrics & \nGynecology 2003; 22(3):305-322. \n (77)  Owen J, Yost N, Berghella V, Thom E, Swain M, Dildy GA, III et al. Mid-trimester \nendovaginal sonography in women at high risk for spontaneous preterm birth. JAMA 2001; 286(11):1340- 1348. \n (78)  To MS, Skentou C, Liao AW, Cacho A, Nicolaides KH. Cervical length and funneling at 23 \nweeks of gestation in the prediction of spontaneous early preterm delivery. Ultrasound in Obstetrics & \nGynecology 2001; 18(3):200-203. \n (79)  Belej-Rak T, Okun N, Windrim R, Ross S, Hannah ME. Effectiveness of cervical cerclage for \na sonographically shortened cervix: a systematic review and meta-analysis. Am J Obstet Gynecol 2003; \n189(6):1679-1687. \n (80)  Berghella V, Odibo AO, To MS, Rust OA, Althuisius SM. Cerclage for short cervix on \nultrasonography: meta-analysis of trials using individual patient-level data. Obstet Gynecol 2005; 106(1):181-\n189. \n (81)  To MS, Alfirevic Z, Heath VC, Cicero S, Cacho AM, Williamson PR et al. Cervical cerclage \nfor prevention of preterm delivery in women with short cervix: randomised controlled trial. Lancet 2004; \n363(9424):1849-1853. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n414 \n \n (82)  Berghella V, Odibo AO, Tolosa JE. Cerclage for prevention of preterm birth in women with a \nshort cervix found on transvaginal ultrasound examination: a randomized trial. Am J Obstet Gynecol 2004; \n191(4):1311-1317. \n (83)  Althuisius SM, Dekker GA, Hummel P, Bekedam DJ, van Geijn HP. Final results of the \nCervical Incompetence Prevention Randomized Cerclage Trial (CIPRACT): therapeutic cerclage with bed rest \nversus bed rest alone. American Journal of Obstetrics & Gynecology 2001; 185(5):1106-1112. \n (84)  Althuisius SM, Dekker GA, Hummel P, van Geijn HP. Cervical incompetence prevention \nrandomized cerclage trial: emergency cerclage with bed rest versus bed rest alone. Am J Obstet Gynecol 2003; \n189(4):907-910. \n (85)  Guzman ER, Forster JK, Vintzileos AM, Ananth CV, Walters C, Gipson K. Pregnancy \noutcomes in women treated with elective versus ultrasound-indicated cervical cerclage. Ultrasound in Obstetrics \n& Gynecology 1998; 12:323-327. \n (86)  Groom KM, Bennett PR, Golara M, Thalon A, Shennan AH. Elective cervical cerclage versus \nserial ultrasound surveillance of cervical length in a population at high risk for preterm delivery. Eur J Obstet \nGynecol Reprod Biol 2004; 112(2):158-161. \n (87)  To MS, Palaniappan V, Skentou C, Gibb D, Nicolaides KH. Elective cerclage vs. ultrasound-\nindicated cerclage in high-risk pregnancies. Ultrasound in Obstetrics & Gynecology 2002; 19(5):475-477. \n (88)  Berghella V, Haas S, Chervoneva I, Hyslop T. Patients with prior second-trimester loss: \nprophylactic cerclage or serial transvaginal sonograms? American Journal of Obstetrics & Gynecology 2002; \n187(3):747-751. \n (89)  Higgins SP, Kornman LH, Bell RJ, Brennecke SP. Cervical surveillance as an alternative to \nelective cervical cerclage for pregnancy management of suspected cervical incompetence. Aust N Z J Obstet \nGynaecol 2004; 44(3):228-232. \n (90)  Odibo AO, Farrell C, Macones GA, Berghella V. Development of a scoring system for \npredicting the risk of preterm birth in women receiving cervical cerclage. J Perinatol 2003; 23(8):664-667. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n415 \n \n (91)  Honest H, Bachmann LM, Gupta JK, Kleijnen J, Khan KS. Accuracy of cervicovaginal fetal \nfibronectin test in predicting risk of spontaneous preterm birth: systematic review. BMJ 2002; 325(7359):301. \n (92)  Leitich H, Kaider A. Fetal fibronectin --how useful is it in the prediction of preterm birth? \nBJOG: an International Journal of Obstetrics & Gynaecology 2003; 110:Sup pl-70. \n (93)  Andrews WW, Sibai BM, Thom EA, Dudley D, Ernest JM, McNellis D et al. Randomized \nclinical trial of metronidazole plus erythromycin to prevent spontaneous preterm delivery in fetal fibronectin-\npositive women. Obstetrics & Gynecology 2003; 101(5:Pt 1):847-855. \n (94)  Goldenberg RL, Klebanoff M, Carey JC, MacPherson C. Metronidazole treatment of women \nwith a positive fetal fibronectin test result. American Journal of Obstetrics & Gynecology 2001; 185(2):485-\n486. \n (95)  Dirnberger DR, Yoder BA, Gordon MC. Single versus repeated-course antenatal \ncorticosteroids: outcomes in singleton and multiple-gestation pregnancies. American Journal of Perinatology \n2001; 18(5):267. \n (96)  Murphy DJ, Caukwell S, Joels LA, Wardle P. Cohort study of the neonatal outcome of twin \npregnancies that were treated with prophylactic or rescue antenatal corticosteroids. American Journal of \nObstetrics & Gynecology 2002; 187(2):483-488. \n (97)  Gibson JL, Macara LM, Owen P, Young D, Macauley J, Mackenzie F. Prediction of preterm \ndelivery in twin pregnancy: a prospective, observational study of cervical length and fetal fibronectin testing. \nUltrasound Obstet Gynecol 2004; 23(6):561-566. \n (98)  Yang JH, Kuhlman K, Daly S, Berghella V. Prediction of preterm birth by second trimester \ncervical sonography in twin pregnancies. Ultrasound in Obstetrics & Gynecology 2000; 15(4):288- 291. \n (99)  Fujita MM, Brizot ML, Liao AW, Bernath T, Cury L, Neto JD et al. Reference range for \ncervical length in twin pregnancies. Acta Obstetricia et Gynecologica Scandinavica 2002; 81(9):856-859. \n (100)  Souka AP, Heath V, Flint S, Sevastopoulou I, Nicolaides KH. Cervical length at 23 weeks in \ntwins in predicting spontaneous preterm delivery. Obstet Gynecol 1999; 94(3):450- 454. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n416 \n \n (101)  Newman RB, Krombach RS, Myers MC, McGee DL. Effect of cerclage on obstetrical \noutcome in twin gestations with a shortened cervical length. American Journal of Obstetrics & Gynecology \n2002; 186(4):634-640. \n (102)  Strauss A, Heer IM, Janssen U, Dannecker C, Hillemanns P, Muller-Egloff S. Routine \ncervical cerclage in higher order multiple gestation -- does it prolong the pregnancy? Twin Research 2002; \n5(2):67-70. \n (103)  Elimian A, Figueroa R, Nigam S, Verma U, Tejani N, Kirshenbaum N. Perinatal outcome of \ntriplet gestation: does prophylactic cerclage make a difference? Journal of Maternal-Fetal Medicine 1999; \n8(3):119-122. \n (104)  Alfirevic Z, Roberts D, Martlew V. How strong is the association between maternal \nthrombophilia and adverse pregnancy outcome? A systematic review. Eur J Obstet Gynecol Reprod Biol 2002; \n101(1):6-14. \n (105)  Kujovich JL. Thrombophilia and pregnancy complications. Am J Obstet Gynecol 2004; \n191(2):412-424. \n (106)  Rey E, Kahn SR, David M, Shrier I. Thrombophilic disorders and fetal loss: a meta-analysis. \nLancet 2003; 361(9361):901-908. \n (107)  Clark CA, Spitzer KA, Nadler JN, Laskin CA. Preterm deliveries in women with systemic \nlupus erythematosus. J Rheumatol 2003; 30(10):2127-2132. \n (108)  Empson M, Lassere M, Craig JC, Scott JR. Recurrent pregnancy loss with antiphospholipid \nantibody: a systematic review of therapeutic trials. Obstet Gynecol 2002; 99(1):135- 144. \n (109)  Lassere M, Empson M. Treatment of antiphospholipid syndrome in pregnancy --a systematic \nreview of randomized therapeutic trials. Thromb Res 2004; 114(5-6):419-426. \n (110)  Gris JC, Mercier E, Quere I, Lavigne-Lissalde G, Cochery-Nouvellon E, Hoffet M et al. Low-\nmolecular-weight heparin versus low-dose aspirin in women with one fetal loss and a constitutional \nthrombophilic disorder. Blood 2004; 103(10):3695-3699. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n417 \n \n (111)  Stephenson MD, Ballem PJ, Tsang P, Purkiss S, Ensworth S, Houlihan E et al. Treatment of \nantiphospholipid antibody syndrome (APS) in pregnancy: a randomized pilot trial comparing low molecular \nweight heparin to unfractionated heparin. J Obstet Gynaecol Can 2004; 26(8):729- 734. \n (112)  Brenner B, Grabowski EF, Hellgren M, Kenet G, Massicotte P, Manco-Johnson M et al. \nThrombophilia and pregnancy complications. Thromb Haemost 2004; 92(4):678- 681. \n (113)  Koike T, Minakami H, Izumi A, Watanabe T, Matsubara S, Sato I. Recurrence risk of preterm \nbirth due to preeclampsia. Gynecologic & Obstetric Investigation 2002; 53(1):22- 27. \n (114)  Hnat MD, Sibai BM, Caritis S, Hauth J, Lindheimer MD, MacPherson C et al. Perinatal \noutcome in women with recurrent preeclampsia compared with women who develop preeclampsia as nulliparas. \nAmerican Journal of Obstetrics & Gynecology 2002; 186(3):422-426. \n (115)  Coomarasamy A, Honest H, Papaioannou S, Gee H, Khan KS. Aspirin for prevention of \npreeclampsia in women with historical risk factors: a systematic review. Obstetrics & Gynecology 2003; \n101(6):1319-1332. \n (116)  Papageorghiou AT, Yu CK, Cicero S, Bower S, Nicolaides KH. Second-trimester uterine \nartery Doppler screening in unselected populations: a review. J Matern Fetal Neonatal Med 2002; 12(2):78-88. \n (117)  Coomarasamy A, Papaioannou S, Gee H, Khan KS. Aspirin for the prevention of \npreeclampsia in women with abnormal uterine artery Doppler: a meta-analysis. Obstetrics & Gynecology 2001; \n98(5:Pt 1):t-6. \n (118)  Vainio M, Kujansuu E, Iso-Mustajarvi M, Maenpaa J. Low dose acetylsalicylic acid in \nprevention of pregnancy-induced hypertension and intrauterine growth retardation in women with bilateral \nuterine artery notches. BJOG 2002; 109(2):161-167.\n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n418 \n \n \n (119)  Yu CK, Papageorghiou AT, Parra M, Palma DR, Nicolaides KH, Fetal Medicine Foundation \nSecond Trimester Screening Group. Randomized controlled trial using low-dose aspirin in the prevention of pre-\neclampsia in women with abnormal uterine artery Doppler at 23 weeks' gestation. Ultrasound in Obstetrics & \nGynecology 2003; 22(3):233-239. \n (120)  Subtil D, Goeusse P, Houfflin-Debarge V, Puech F, Lequien P, Breart G et al. Randomised \ncomparison of uterine artery Doppler and aspirin (100 mg) with placebo in nulliparous women: the Essai \nRegional Aspirine Mere-Enfant study (Part 2). BJOG 2003; 110(5):485-491. \n (121)  Goffinet F, Aboulker D, Paris-Llado J, Bucourt M, Uzan M, Papiernik E et al. Screening with \na uterine Doppler in low risk pregnant women followed by low dose aspirin in women with abnormal results: a \nmulticenter randomised controlled trial. BJOG 2001; 108(5):510-518. \n (122)  Tuffnell DJ, West J, Walkinshaw SA. Treatments for gestational diabetes and impaired \nglucose tolerance in pregnancy. Cochrane Database Syst Rev 2003;(3):CD003395. \n (123)  Walkinshaw SA. Very tight versus tight control for diabetes in pregnancy. Cochrane Database \nSyst Rev 2000;(2):CD000226. \n (124)  Walkinshaw SA. Dietary regulation for 'gestational diabetes'. Cochrane Database Syst Rev \n2000;(2):CD000070. \n (125)  Bartha JL, Martinez- Del-Fresno P, Comino-Delgado R. Early diagnosis of gestational diabetes \nmellitus and prevention of diabetes-related complications. European Journal of Obstetrics, Gynecology, & \nReproductive Biology 2003; 109(1):41-44. \n (126)  Vintzileos AM, Ananth CV, Smulian JC, Scorza WE, Knuppel RA. The impact of prenatal \ncare in the United States on preterm births in the presence and absence of antenatal high-risk conditions. Am J \nObstet Gynecol 2002; 187(5):1254-1257. \n (127)  Heaman MI, Sprague AE, Stewart PJ. Reducing the preterm birth rate: a population health \nstrategy. [Review] [84 refs]. JOGNN - Journal of Obstetric, Gynecologic, & Neonatal Nursing 2001; 30(1):20-\n29. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n419 \n \n (128)  Papiernik E, Goffinet F. Prevention of preterm births, the French experience. Clinical \nObstetrics & Gynecology 47(4):755-67; discussion 881-2, 2004. \n (129)  Armson BA, Dodds L, Haliburton SC, Cervin C, Rinaldo K. Impact of participation in the \nHalifax County Preterm Birth Prevention Project. J Obstet Gynaecol Can 2003; 25(3):209- 217. \n (130)  Carroli G, Villar J, Piaggio G, Khan-Neelofur D, Gulmezoglu M, Mugford M et al. WHO \nsystematic review of randomised controlled trials of routine antenatal care. Lancet 2001; 357(9268):1565- 1570. \n (131)  Villar J, Carroli G, Khan-Neelofur D, Piaggio G, Gulmezoglu M. Patterns of routine antenatal \ncare for low-risk pregnancy. Cochrane Database Syst Rev 2001;(4):CD000934. \n (132)  Makrides M, Duley L, Olsen SF. Fish oil and other prostaglandin precursor supplementation \nduring pregnancy for reducing pre-eclampsia, preterm birth, low birth weight and intrauterine growth restriction. \n[Protocol]. Cochrane Database of Systematic Reviews 2004;(2). \n (133)  Smuts CM, Huang M, Mundy D, Plasse T, Major S, Carlson SE. A randomized trial of \ndocosahexaenoic acid supplementation during the third trimester of pregnancy. Obstetrics & Gynecology 2003; \n101(3):469-479. \n (134)  ACOG. ACOG Practice Bulletin. Assessment of risk factors for preterm birth. Clinical \nmanagement guidelines for obstetrician-gynecologists. Number 31, October 2001. Obstet Gynecol 2001; \n98(4):709-716. \n (135)  Olsen SF, Secher NJ, Tabor A, Weber T, Walker JJ, Gluud C. Randomised clinical trials of \nfish oil supplementation in high risk pregnancies. Fish Oil Trials In Pregnancy (FOTIP) Team. BJOG: an \nInternational Journal of Obstetrics & Gynaecology 2000; 107(3):382-395. \n (136)  Crowther CA, Hiller JE, Pridmore B, Bryce R, Duggan P, Hague WM et al. Calcium \nsupplementation in nulliparous women for the prevention of pregnancy-induced hypertension, preeclampsia and \npreterm birth: an Australian randomized trial. FRACOG and the ACT Study Group. Australian & New Zealand \nJournal of Obstetrics & Gynaecology 1999; 39(1):12-18. \n (137)  Vahratian A, Siega-Riz AM, Savitz DA, Thorp JM, Jr. Multivitamin use and the risk of \npreterm birth. Am J Epidemiol 2004; 160(9):886-892. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n420 \n \n (138)  Olsen SF. Is supplementation with marine omega-3 fatty acids during pregnancy a useful tool \nin the prevention of preterm birth? Clinical Obstetrics & Gynecology 47(4):768-74; discussion 881-2, 2004. \n (139)  Rumbold A, Crowther CA. Vitamin E supplementation in pregnancy. Cochrane Database Syst \nRev 2005;(2):CD004069. \n (140)  Rumbold A, Crowther CA. Vitamin C supplementation in pregnancy. Cochrane Database Syst \nRev 2005;(2):CD004072. \n (141)  Sanchez-Ramos L, Huddleston JF. The therapeutic value of maintenance tocolysis: an \noverview of the evidence. Clin Perinatol 2003; 30(4):841-854. \n (142)  Meirowitz NB, Ananth CV, Smulian JC, Vintzileos AM. Value of maintenance therapy with \noral tocolytics: a systematic review. J Matern Fetal Med 1999; 8(4):177-183. \n (143)  Berkman ND, Thorp JM, Jr., Lohr KN, Carey TS, Hartmann KE, Gavin NI et al. Tocolytic \ntreatment for the management of preterm labor: a review of the evidence. [Review] [79 refs]. American Journal \nof Obstetrics & Gynecology 2003; 188(6):1648-1659. \n (144)  RCOG. Tocolytic Drugs for women in preterm labour. Guideline no 1B.  2002.  RCOG, \nLondon.  \n (145)  Groom KM, Shennan AH, Jones BA, Seed P, Bennett PR. TOCOX --a randomised, double-\nblind, placebo-controlled trial of rofecoxib (a COX-2-specific prostaglandin inhibitor) for the prevention of \npreterm delivery in women at high risk. BJOG 2005; 112(6):725-730. \n (146)  Crowley P. Prophylactic corticosteroids for preterm birth. Cochrane Database of Systematic \nReviews 2003;(3). \n (147)  RCOG. Antenatal corticosteroids to prevent respiratroy distress syndrome. Guideline number \n7.  2004.  RCOG, London.  \n (148)  Crowther CA, Harding J. Repeat doses of prenatal corticosteroids for women at risk of \npreterm birth for preventing neonatal respiratory disease. Cochrane Database of Systematic Reviews 2003;(3). \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n421 \n \n (149)  Bloom SL, Leveno KJ. Corticosteroid use in special circumstances: preterm ruptured \nmembranes, hypertension, fetal growth restriction, multiple fetuses. Clinical Obstetrics & Gynecology 2003; \n46(1):150-160. \n (150)  Boggess KA, Bailit JL, Singer ME, Parisi VM, Mercer BM. Projected benefits of universal or \nscheduled antepartum corticosteroids to prevent neonatal morbidity: a decision analysis. Am J Obstet Gynecol \n2005; 193(4):1415-1423. \n (151)  NIH Consensus Statement. Antenatal corticosteroids revisited: repeat courses.  2000.  NIH \nConsensus Statement.  \n (152)  Guinn DA AMS. Single vs weekly courses of antenatal corticosteroids for women at risk of \npreterm delivery: A randomized controlled trial. JAMA 286(13):1581-7, 2001. \n (153)  Lee MJ, Davies J, Guinn D, Sullivan L, Atkinson MW, McGregor S et al. Single versus \nweekly courses of antenatal corticosteroids in preterm premature rupture of membranes. Obstetrics & \nGynecology 2004; 103(2):274-281. \n (154)  Thinkhamrop J, Hofmeyr GJ, Adetoro O, Lumbiganon P. Prophylactic antibiotic \nadministration in pregnancy to prevent infectious morbidity and mortality. [Review] [22 refs]. Cochran e \nDatabase of Systematic Reviews 2002;(4):CD002250. \n (155)  Dodd JM, Crowther CA, Cincotta R, Flenady V, Robinson JS. Progesterone supplementation \nfor preventing preterm birth: a systematic review and meta-analysis. Acta Obstet Gynecol Scand 2005; \n84(6):526-533. \n (156)  Sanchez-Ramos L, Kaunitz AM, Delke I. Progestational agents to prevent preterm birth: a \nmeta-analysis of randomized controlled trials. Obstet Gynecol 2005; 105(2):273-279. \n (157)  ACOG. Use of progesterone to reduce preterm birth. Int J Gynaecol Obstet 2004; 84(1):93- 94. \n (158)  da Fonseca EB, Bittar RE, Carvalho MH, Zugaib M. Prophylactic administration of \nprogesterone by vaginal suppository to reduce the incidence of spontaneous preterm birth in women at increased \nrisk: a randomized placebo-controlled double-blind study.[comment]. American Journal of Obstetrics & \nGynecology 2003; 188(2):419-424. \n\nREFERENCES   Chapter 3.1. Prevention of Preterm Delivery \n422 \n \n (159)  Meis PJ, Klebanoff M, Thom E, Dombrowski MP, Sibai B, Moawad AH et al. Prevention of \nrecurrent preterm delivery by 17 alpha-hydroxyprogesterone caproate. New England Journal of Medicine 2003; \n348(24):2379-2385. \n (160)  Spong CY, Meis PJ, Thom EA, Sibai B, Dombrowski MP, Moawad AH et al. Progesterone \nfor prevention of recurrent preterm birth: impact of gestational age at previous delivery. Am J Obstet Gynecol \n2005; 193(3 Pt 2):1127-1131. \n (161)  Jones B, Groom KM, Shennan AH, Simcox R, Seed PG, Teoh TG et al. Experience of a clinic \nfor surveillance of women at high risk of preterm delivery: cerclage rates and outcome. J Soc Gynecol Investig \n2004; 11(2 (Supplement)):311A. \n (162)  Tudur-Smith C, Jorgensen AL, Alfirevic Z, Williamson PR. Individual patient data meta-\nanalysis : Cervical stitch (cerclage) for preventing pregnancy loss in women. BMC Pregnancy Childbirth 2005; \n5(1):5. \n (163)  Antiplatelet agents for prevention of pre-eclampsia and its consequences: a systematic review \nand individual patient data meta-analysis. BMC Pregnancy Childbirth 2005; 5(1):7. \n (164)  Lumley J. Defining the problem: the epidemiology of preterm birth. BJOG 2003; 110 Suppl \n20:3-7. \n (165)  Thompson SG, Pocock SJ. Can meta-analyses be trusted? Lancet 1991; 338(8775):1127- 1130. \n (166)  Egger M, Smith GD, Sterne JA. Uses and abuses of meta-analysis. Clin Med 2001; 1(6):478-\n484\n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n423 \n \nChapter 3.2  \n \n(1)  Lockhat FB, Emembolu JE, Konje JC. Serum and peritoneal fluid levels of levonorgestrel in \nwomen with endometriosis who were treated with an intrauterine contraceptive device containing \nlevonorgestrel. Fertil Steril 2005; 83(2):398-404. \n (2)  Nilsson CG, Lahteenmaki PL, Luukkainen T, Robertson DN. Sustained intrauterine release of \nlevonorgestrel over five years. Fertil Steril 1986; 45(6):805-807. \n (3)  Sturdee D, Rantala ML, Colau JC, Zahradnik HP, Riphagen FE. The acceptability of a small \nintrauterine progestogen-releasing system for continuous combined hormone therapy in early postmenopausal \nwomen. Climacteric 2004; 7(4):404-411. \n (4)  Wildemeersch D, Schacht E, Wildemeersch P, Janssens D, Thiery M. Development of a \nminiature, low-dose, frameless intrauterine levonorgestrel-releasing system for contraception and treatment: a \nreview of initial clinical experience. Reproductive Biomedicine Online 2002; 4(1):71- 82. \n (5)  French R. Hormonally impregnated intrauterine systems (IUSs) versus other forms of \nreversible contraceptives as effective methods of preventing pregnancy. Cochrane Database of Systematic \nReviews 2005; 1, 2005. \n (6)  McGavigan CJ, Cameron IT. The Mirena levonorgestrel system. Drugs of Today 2003; \n39(12):973-984. \n (7)  Ikomi A, Pepra EF. Efficacy of the levonorgestrel intrauterine system in treating menorrhagia: \nactualities and ambiguities. J Fam Plann Reprod Health Care 2002; 28(2):99-100. \n (8)  Pakarinen P, Toivonen J, Luukkainen T. Therapeutic use of the LNG IUS, and counseling. \nSemin Reprod Med 2001; 19(4):365-372. \n (9)  Lahteenmaki P, Rauramo I, Backman T. The levonorgestrel intrauterine system in \ncontraception. Steroids 2000; 65(10-11):693-697. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n424 \n \n (10)  Faculty of Family Planning and Reproductive Health Care Clinical Effectiveness Unit. \nFFPRHC Guidance (April 2004). The levonorgestrel-releasing intrauterine system (LNG-IUS) in contraception \nand reproductive health. J Fam Plann Reprod Health Care 2004; 30(2):99-108. \n (11)  Hubacher D, Grimes DA. Noncontraceptive health benefits of intrauterine devices: a \nsystematic review. [Review] [72 refs]. Obstetrical & Gynecological Survey 2002; 57(2):120- 128. \n (12)  RCOG. Clinical Governance Advice No 1.Guidance for the Development of RCOG Green-top \nGuidelines.  2000.  Royal College of Obstetricians and Gynaecologists, London, UK.  \n (13 )  Hurskainen R, Teperi J, Rissanen P, Aalto AM, Grenman S, Kivela A et al. Quality of life and \ncost-effectiveness of levonorgestrel-releasing intrauterine system versus hysterectomy for treatment of \nmenorrhagia: a randomised trial. Lancet 2001; 357(9252):273-277. \n (14)  Hurskainen R, Teperi J, Rissanen P, Aalto AM, Grenman S, Kivela A et al. Clinical outcomes \nand costs with the levonorgestrel-releasing intrauterine system or hysterectomy for treatment of menorrhagia: \nrandomized trial 5-year follow-up. JAMA 2004; 291(12):1456-1463. \n (15)  Soysal M, Soysal S, Ozer S. A randomized controlled trial of levonorgestrel releasing IUD \nand thermal balloon ablation in the treatment of menorrhagia. Zentralblatt fur Gynakologie 2002; 124(4):213-\n219. \n (16)  Crosignani PG, Vercellini P, Mosconi P, Oldani S, Cortesi I, De Giorgi O. Levonorgestrel-\nreleasing intrauterine device versus hysteroscopic endometrial resection in the treatment of dysfunctional uterine \nbleeding. Obstet Gynecol 1997; 90(2):257-263. \n (17)  Kittelsen N. A randomized study comparing levonorgestrel intrauterine system (LNG IUS) \nand transcervical resection of the endometrium (TCRE) in the treatment of menorrhagia: Preliminary results. \nGynaecological Endoscopy 1998; 7(2):61-65. \n (18)  Istre O, Trolle B. Treatment of menorrhagia with the levonorgestrel intrauterine system versus \nendometrial resection. Fertility & Sterility 2001; 76(2):304-309. \n (19)  Rauramo I, Elo I, Istre O. Long-term treatment of menorrhagia with levonorgestrel \nintrauterine system versus endometrial resection. Obstet Gynecol 2004; 104(6):1314- 1321. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n425 \n \n (20)  Lahteenmaki P, Haukkamaa M, Puolakka J, Riikonen U, Sainio S, Suvisaari J et al. Open \nrandomised study of use of levonorgestrel releasing intrauterine system as alternative to hysterectomy. BMJ \n1998; 316(7138):1122-1126. \n (21)  Reid PC, Virtanen-Kari S. Randomised comparative trial of the levonorgestrel intrauterine \nsystem and mefenamic acid for the treatment of idiopathic menorrhagia: a multiple analysis using total \nmenstrual fluid loss, menstrual blood loss and pictorial blood loss assessment charts. Epub. BJOG: an \nInternational Journal of Obstetrics & Gynaecology 2005; 112:1-5. \n (22)  Barrington JW, Arunkalaivanan AS, Abdel-Fattah M. Comparison between the levonorgestrel \nintrauterine system (LNG-IUS) and thermal balloon ablation in the treatment of menorrhagia. Eur J Obstet \nGynecol Reprod Biol 2003; 108(1):72-74. \n (23)  Irvine GA, Campbell-Brown MB, Lumsden MA, Heikkila A, Walker JJ, Cameron IT. \nRandomised comparative trial of the levonorgestrel intrauterine system and norethisterone for treatment of \nidiopathic menorrhagia. British Journal of Obstetrics & Gynaecology 1998; 105(6):592-598. \n (24)  Milsom I, Andersson K, Andersch B, Rybo G. A comparison of flurbiprofen, tranexamic acid, \nand a levonorgestrel-releasing intrauterine contraceptive device in the treatment of idiopathic menorrhagia. Am \nJ Obstet Gynecol 1991; 164(3):879-883. \n (25)  Romer T. Prospective comparison study of levonorgestrel IUD versus Roller-Ball endometrial \nablation in the management of refractory recurrent hypermenorrhea. European Journal of Obstetrics, \nGynecology, & Reproductive Biology 2000; 90(1):27-29. \n (26)  Henshaw R, Coyle C, Low S, Barry C. A retrospective cohort study comparing microwave \nendometrial ablation with levonorgestrel-releasing intrauterine device in the management of heavy menstrual \nbleeding. Aust N Z J Obstet Gynaecol 2002; 42(2):205-209. \n (27)  Mansour MSA, Mansour DJA. The effectiveness of the levonorgestrel intauterine system \n(LNG IUS) in the treatment of menorrhagia in a district clinic for abnormal uterine bleeding. Abstract 323. \nBJOG: an International Journal of Obstetrics & Gynaecology 1998; 105(Supplement):105. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n426 \n \n (28)  Barrington JW, Bowen-Simpkins P. The levonorgestrel intrauterine system in the \nmanagement of menorrhagia. British Journal of Obstetrics & Gynaecology 1997; 104(5):614- 616. \n (29)  Monteiro I, Bahamondes L, Diaz J, Perrotti M, Petta C. Therapeutic use of levonorgestrel-\nreleasing intrauterine system in women with menorrhagia: a pilot study(1). Contraception 2002; 65(5):325- 328. \n (30)  Wildemeersch D, Rowe PJ. Assessment of menstrual blood loss in women with ideopathic \nmenorrhagia using the frameless levonorgestrel-releasing intrauterine system. Contraception 2004; 70(2):165-\n168. \n (31)  Wildemeersch D, Schacht E. Treatment of menorrhagia with a novel 'frameless' intrauterine \nlevonorgestrel-releasing drug delivery system: a pilot study. Eur J Contracept Reprod Health Care 2001; \n6(2):93-101. \n (32)  Wildemeersch D, Schacht E, Wildemeersch P. Treatment of primary and secondary \ndysmenorrhea with a novel 'frameless' intrauterine levonorgestrel-releasing drug delivery system: a pilot study. \nEuropean Journal of Contraception & Reproductive Health Care 2001; 6(4):192-198. \n (33)  Wildemeersch D, Rowe PJ. Assessment of menstrual blood loss in Belgian users of a new T-\nshaped levonorgestrel-releasing intrauterine system. Contraception 2005; 71(6):470- 473. \n (34)  Xiao B, Wu SC, Chong J, Zeng T, Han LH, Luukkainen T. Therapeutic effects of the \nlevonorgestrel-releasing intrauterine system in the treatment of idiopathic menorrhagia. Fertil Steril 2003; \n79(4):963-969. \n (35)  Sivin I, Stern J. Health during prolonged use of levonorgestrel 20 micrograms/d and the \ncopper TCu 380Ag intrauterine contraceptive devices: a multicenter study. International Committee for \nContraception Research (ICCR). Fertil Steril 1994; 61(1):70-77. \n (36)  Nilsson CG, Allonen H, Diaz J, Luukkainen T. Two years' experience with two \nlevonorgestrel-releasing intrauterine devices and one copper-releasing intrauterine device: a randomized \ncomparative performance study. Fertil Steril 1983; 39(2):187-192. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n427 \n \n (37)  Andrade AT, Souza JP, Andrade GN, Rowe PJ, Wildemeersch D. Assessment of menstrual \nblood loss in Brazilian users of the frameless copper-releasing IUD with copper surface area of 330 mm2 and \nthe frameless levonorgestrel-releasing intrauterine system. Contraception 2004; 70(2):173- 177. \n (38)  Rogerson L, Duffy S, Crocombe W, Stead M, Dassu D. Management of menorrhagia --\nSMART study (Satisfaction with Mirena and Ablation: a Randomised Trial. BJOG: an International Journal of \nObstetrics & Gynaecology 2000; 107(10):1325-1326. \n (39)  Johnson N, Busfield R, Sadler L, Lethaby A, Farquhar C. The management of menorrhagia --\nSMART study (Satisfaction with Mirena and Ablation: a Randomised Trial). BJOG 2001; 108(7):773- 774. \n (40)  Lethaby AE. Progesterone/progestogen releasing intrauterine systems for heavy menstrual \nbleeding. Cochrane Database of Systematic Reviews 2005; 1, 2005. \n (41)  Marjoribanks J. Surgery versus medical therapy for heavy menstrual bleeding. Cochrane \nDatabase of Systematic Reviews 2005; 1, 2005. \n (42)  Stewart A, Cummins C, Gold L, Jordan R, Phillips W. The effectiveness of the levonorgestrel-\nreleasing intrauterine system in menorrhagia: a systematic review. BJOG: an International Journal of Obstetrics \n& Gynaecology 2001; 108(1):74-86. \n (43)  Radesic B, Sharma A. Levonorgestrel-releasing intrauterine system for treating menstrual \ndisorders: a patient satisfaction questionnaire. Aust N Z J Obstet Gynaecol 2004; 44(3):247- 251. \n (44)  Nagrani R, Bowen-Simpkins P, Barrington JW. Can the levonorgestrel intrauterine system \nreplace surgical treatment for the management of menorrhagia? BJOG: an International Journal of Obstetrics & \nGynaecology 2002; 109(3):345-347. \n (45)  Lethaby A, Hickey M. Endometrial destruction techniques for heavy menstrual bleeding: a \nCochrane review. Hum Reprod 2002; 17(11):2795-2806. \n (46)  Kingman CE, Kadir RA, Lee CA, Economides DL. The use of levonorgestrel-releasing \nintrauterine system for treatment of menorrhagia in women with inherited bleeding disorders. BJOG 2004; \n111(12):1425-1428. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n428 \n \n (47)  Soysal S, Soysal ME. The efficacy of levonorgestrel-releasing intrauterine device in selected \ncases of myoma-related menorrhagia: a prospective controlled trial. Gynecol Obstet Invest 2005; 59(1):29-35. \n (48)  Grigorieva V, Chen-Mok M, Tarasova M, Mikhailov A. Use of a levonorgestrel-releasing \nintrauterine system to treat bleeding related to uterine leiomyomas. Fertility & Sterility 2003; 79(5):1194-1198. \n (49)  Mercorio F, De Simone R, Di Spiezio SA, Cerrota G, Bifulco G, Vanacore F et al. The effect \nof a levonorgestrel-releasing intrauterine device in the treatment of myoma-related menorrhagia. Contraception \n2003; 67(4):277-280. \n (50)  Wildemeersch D, Schacht E. The effect on menstrual blood loss in women with uterine \nfibroids of a novel \"frameless\" intrauterine levonorgestrel-releasing drug delivery system: a pilot study. Eur J \nObstet Gynecol Reprod Biol 2002; 102(1):74-79. \n (51)  Starczewski A, Iwanicki M. Intrauterine therapy with levonorgestrel releasing IUD of women \nwith hypermenorrhea secondary to uterine fibroids. Ginekol Pol 2000; 71(9):1221-1225. \n (52)  Singer A, Ikomi A. Successful treatment of uterine fibroids using an intrauterine progesterone \ndevice. International Journal of Obstetrics & Gynecology 1994; 46(Supplement):55. \n (53)  Fong YF, Singh K. Effect of the levonorgestrel-releasing intrauterine system on uterine \nmyomas in a renal transplant patient. Contraception 1999; 60(1):51-53. \n (54)  Gardner FJ, Konje JC, Abrams KR, Brown LJ, Khanna S, Al Azzawi F et al. Endometrial \nprotection from tamoxifen-stimulated changes by a levonorgestrel-releasing intrauterine system: a randomised \ncontrolled trial. Lancet 2000; 356(9243):1711-1717. \n (55)  INKI P, Hurskainen R, PALO P, Ekholme E, Grenman S, Kivela A et al. Comparison of \novarian cyst formation in women using the levonorgestrel-releasing intrauterine system vs. hysterectomy. \nUltrasound in Obstetrics & Gynecology 2002; 20(4):381-385. \n (56)  Petta CA, Ferriani RA, Abrao MS, Hassan D, Rosa E Silva JC, Pdgaec S et al. Randomized \nclinical trial of a levonorgestrel-releasing intrauterine system and a depot GnRH analogue for the treatment of \nchronic pelvic pain in women with endometriosis. Hum Reprod 2005. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n429 \n \n (57)  Vercellini P, Frontino G, De Giorgi O, Aimi G, Zaina B, Crosignani PG. Comparison of a \nlevonorgestrel-releasing intrauterine device versus expectant management after conservative surgery for \nsymptomatic endometriosis: a pilot study. Fertil Steril 2003; 80(2):305-309. \n (58)  Lockhat FB, Emembolu JO, Konje JC. The evaluation of the effectiveness of an intrauterine-\nadministered progestogen (levonorgestrel) in the symptomatic treatment of endometriosis and in the staging o f \nthe disease. Human Reproduction 2004; 19(1):179-184. \n (59)  Lockhat FB, Emembolu JO, Konje JC. The efficacy, side-effects and continuation rates in \nwomen with symptomatic endometriosis undergoing treatment with an intra-uterine administered progestogen \n(levonorgestrel): a 3 year follow-up. Hum Reprod 2005; 20(3):789-793. \n (60)  Vercellini P, Aimi G, Panazza S, De Giorgi O, Pesole A, Crosignani PG. A levonorgestrel-\nreleasing intrauterine system for the treatment of dysmenorrhea associated with endometriosis: a pilot study. \nFertility & Sterility 1999; 72(3):505-508. \n (61)  Fedele L, Bianchi S, Zanconato G, Portuese A, Raffaelli R. Use of a levonorgestrel-releasing \nintrauterine device in the treatment of rectovaginal endometriosis. Fertil Steril 2001; 75(3):4 85-488. \n (62)  Maia H, Jr., Maltez A, Coelho G, Athayde C, Coutinho EM. Insertion of mirena after \nendometrial resection in patients with adenomyosis. J Am Assoc Gynecol Laparosc 2003; 10(4):512-516. \n (63)  Fedele L, Bianchi S, Raffaelli R, Portuese A, Dorta M. Treatment of adenomyosis-associated \nmenorrhagia with a levonorgestrel-releasing intrauterine device. Fertility & Sterility 1997; 68(3):426-429. \n (64)  Fong YF, Singh K. Medical treatment of a grossly enlarged adenomyotic uterus with the \nlevonorgestrel-releasing intrauterine system. Contraception 1999; 60(3):173-175. \n (65)  Turnbull LW, Tetlow RL, Manton DJ, Purdie DW. The effectsof a progestogen on the \nendometrium of long term users of tamoxifen treated breast cancer patients using transvaginal scanning and \nMRI. British Journal of Radiology 1998; 71:45 (S). \n (66)  Boon J, Scholten PC, Oldenhave A, Heintz AP. Continuous intrauterine compared with cyclic \noral progestin administration in perimenopausal HRT. Maturitas 2003; 46(1):69- 77. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n430 \n \n (67)  Wollter-Svensson LO, Stadberg E, Andersson K, Mattsson LA, Odlind V, Persson I. \nIntrauterine administration of levonorgestrel 5 and 10 microg/24 hours in perimenopausal hormone replacement \ntherapy. A randomized clinical study during one year. Acta Obstetricia et Gynecologica Scandinavica 1997; \n76(5):449-454. \n (68)  Raudaskoski T, Tapanainen J, Tomas E, Luotola H, Pekonen F, Ronni-Sivula H et al. \nIntrauterine 10 microg and 20 microg levonorgestrel systems in postmenopausal women receiving oral \noestrogen replacement therapy: clinical, endometrial and metabolic response. BJOG 2002; 109(2):136- 144. \n (69)  Raudaskoski TH, Lahti EI, Kauppila AJ, Apaja-Sarkkinen MA, Laatikainen TJ. Transdermal \nestrogen with a levonorgestrel-releasing intrauterine device for climacteric complaints: clinical and endometrial \nresponses. Am J Obstet Gynecol 1995; 172(1 Pt 1):114-119. \n (70)  Andersson K, Mattsson LA, Rybo G, Stadberg E. Intrauterine release of levonorgestrel --a new \nway of adding progestogen in hormone replacement therapy. Obstet Gynecol 1992; 79(6):963- 967. \n (71)  Suhonen SP, Allonen HO, Lahteenmaki P. Sustained-release estradiol implants and a \nlevonorgestrel-releasing intrauterine device in hormone replacement therapy. Am J Obstet Gynecol 1995; 172(2 \nPt 1):562-567. \n (72)  Suhonen SP, Holmstrom T, Allonen HO, Lahteenmaki P. Intrauterine and subdermal \nprogestin administration in postmenopausal hormone replacement therapy. Fertil Steril 1995; 63(2):336- 342. \n (73)  Suvanto-Luukkonen E, Malinen H, Sundstrom H, Penttinen J, Kauppila A. Endometrial \nmorphology during hormone replacement therapy with estradiol gel combined to levonorgestrel-releasing \nintrauterine device or natural progesterone. Acta Obstet Gynecol Scand 1998; 77(7):758- 763. \n (74)  Suvanto-Luukkonen E, Sundstrom H, Penttinen J, Laara E, Pramila S, Kauppila A. \nPercutaneous estradiol gel with an intrauterine levonorgestrel releasing device or natural progesterone in \nhormone replacement therapy. Maturitas 1997; 26(3):211-217. \n (75)  Suvanto-Luukkonen E, Kauppila A. The levonorgestrel intrauterine system in menopausal \nhormone replacement therapy: five-year experience. Fertility & Sterility 1999; 72(1):161- 163. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n431 \n \n (76)  Antoniou G, Kalogirou D, Karakitsos P, Antoniou D, Kalogirou O, Giannikos L. Transdermal \nestrogen with a levonorgestrel-releasing intrauterine device for climacteric complaints versus estradiol-releasing \nvaginal ring with a vaginal progesterone suppository: clinical and endometrial responses. Maturitas 1997; \n26(2):103-111. \n (77)  Kalogirou D. A comparative study of the effects of an estradiol-releasing vaginal ring \ncombined with an oral gestagen versus transdermal estrogen combined with a levonorgestrel-releasing IUD: \nclinical findings and endometrial response. International Journal of Fertility & Menopausal Studies 1996; \n41(6):522-527. \n (78)  Wildemeersch D, Schacht E, Wildemeersch P. Performance and acceptability of intrauterine \nrelease of levonorgestrel with a miniature delivery system for hormonal substitution therapy, contraception and \ntreatment in peri and postmenopausal women. Maturitas 2003; 44(3):237-245. \n (79)  Hampton NR, Rees MC, Lowe DG, Rauramo I, Barlow D, Guillebaud J. Levonorgestrel \nintrauterine system (LNG-IUS) with conjugated oral equine estrogen: a successful regimen for HRT in \nperimenopausal women. Hum Reprod 2005. \n (80)  Varila E, Wahlstrom T, Rauramo I. A 5-year follow-up study on the use of a levonorgestrel \nintrauterine system in women receiving hormone replacement therapy. Fertil Steril 2001; 76(5):969-973. \n (81)  Wildemeersch D, Schacht E. Endometrial suppression with a new 'frameless' levonorgestrel \nreleasing intrauterine system in perimenopausal and postmenopausal women: a pilot study. Maturitas 2000; \n36(1):63-68. \n (82)  Suhonen S, Holmstrom T, Lahteenmaki P. Three-year follow-up of the use of a \nlevonorgestrel-releasing intrauterine system in hormone replacement therapy. Acta Obstetricia et Gynecologica \nScandinavica 1997; 76(2):145-150. \n (83)  Wildemeersch D, Schacht E, Wildemeersch P, Calleweart K, Pylyser K, De Wever N. \nEndometrial safety with a low-dose intrauterine levonorgestrel-releasing system after 3 years of estrogen \nsubstitution therapy. Maturitas 2004; 48(1):65-70. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n432 \n \n (84)  Vereide AB, Arnes M, Straume B, Maltau JM, Orbo A. Nuclear morphometric changes and \ntherapy monitoring in patients with endometrial hyperplasia: a study comparing effects of intrauterine \nlevonorgestrel and systemic medroxyprogesterone. Gynecol Oncol 2003; 91(3):526-533. \n (85)  Scarselli G, Tantini C, Colafranceschi M, Taddei GL, Bargelli G, Venturini N et al. Levo-\nnorgestrel-nova-T and precancerous lesions of the endometrium. Eur J Gynaecol Oncol 1988; 9(4):284- 286. \n (86)  Perino A, Quartararo P, Catinella E, Genova G, Cittadini E. Treatment of endometrial \nhyperplasia with levonorgestrel releasing intrauterine devices. Acta Eur Fertil 1987; 18(2):137- 140. \n (87)  Wildemeersch D, Dhont M. Treatment of nonatypical and atypical endometrial hyperplasia \nwith a levonorgestrel-releasing intrauterine system. American Journal of Obstetrics & Gynecology 2003; \n188(5):1297-1298. \n (88)  Rose G, Edmonds DK. Levonorgestrel IUS-treatment for endometrial cystic hyperplasia. J \nObstet Gynaecol 2001; 21:642-643. \n (89)  Bahamondes L, Ribeiro-Huguet P, de Andrade KC, Leon-Martins O, Petta CA. \nLevonorgestrel-releasing intrauterine system (Mirena) as a therapy for endometrial hyperplasia and carcinoma. \nActa Obstetricia et Gynecologica Scandinavica 2003; 82(6):580-582. \n (90)  Ramirez PT, Frumovitz M, Bodurka DC, Sun CC, Levenback C. Hormonal therapy for the \nmanagement of grade 1 endometrial adenocarcinoma: a literature review. Gynecol Oncol 2004; 95(1):133- 138. \n (91)  Giannopoulos T, Butler-Manuel S, Tailor A. Levonorgestrel-releasing intrauterine system \n(LNG-IUS) as a therapy for endometrial carcinoma. Gynecologic Oncology 2004; 95(3):762- 764. \n (92)  Montz FJ, Bristow RE, Bovicelli A, Tomacruz R, Kurman RJ. Intrauterine progesterone \ntreatment of early endometrial cancer. Am J Obstet Gynecol 2002; 186(4):651- 657. \n (93)  Jones K, Georgiou M, Hyatt D, Spencer T, Thomas H. Endometrial adenocarcinoma \nfollowing the insertion of a Mirena IUCD. Gynecologic Oncology 2002; 87(2):216- 218. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n433 \n \n (94)  Sivin I, Alvarez F, Diaz J, Diaz S, el Mahgoub S, Coutinho E et al. Intrauterine contraception \nwith copper and with levonorgestrel: a randomized study of the TCu 380Ag and levonorgestrel 20 mcg/day \ndevices. Contraception 1984; 30(5):443-456. \n (95)  Luukkainen T, Allonen H, Haukkamaa M, Lahteenmaki P, Nilsson CG, Toivonen J. Five \nyears' experience with levonorgestrel-releasing IUDs. Contraception 1986; 33(2):139- 148. \n (96)  Nilsson CG, Luukkainen T, Diaz J, Allonen H. Clinical performance of a new levonorgestrel-\nreleasing intrauterine device. A randomized comparison with a nova-T-copper device. Contraception 1982; \n25(4):345-356. \n (97)  Ronnerdag M, Odlind V. Health effects of long-term use of the intrauterine levonorgestrel-\nreleasing system. A follow-up study over 12 years of continuous use. Acta Obstetricia et Gynecologica \nScandinavica 1999; 78(8):716-721. \n (98)  Baldaszti E, Wimmer-Puchinger B, Loschke K. Acceptability of the lo ng-term contraceptive \nlevonorgestrel-releasing intrauterine system (Mirena): a 3-year follow-up study. Contraception 2003; 67(2):87-\n91. \n (99)  Dolan LM, Mulholland M, Price J. The levonorgestrel intra-uterine system: therapeutic \napplication in family planning. J Fam Plann Reprod Health Care 2001; 27(1):19- 21. \n (100)  Andersson K, Odlind V, Rybo G. Levonorgestrel-releasing and copper-releasing (Nova T) \nIUDs during five years of use: a randomized comparative trial. Contraception 1994; 49(1):56- 72. \n (101)  Chi IC. An evaluation of the levonorgestrel-releasing IUD: its advantages and disadvantages \nwhen compared to the copper-releasing IUDs. Contraception 1991; 44(6):573- 588. \n (102)  Odlind V. Long-term experience of a levonorgestrel-releasing intrauterine system. European \nJournal of Contraception & Reproductive Health Care 1996; 1(4):319-323. \n (103)  Baveja R, Bichille LK, Coyaji KJ, Engineer AD, Gogoi MP, Hazra MN et al. Randomized \nclinical trial with intrauterine devices (levonorgestrel intrauterine device (LNG), CuT 380Ag, CuT 220C and \nCuT 200B). A 36-month study. Indian Council of Medical Research Task Force on IUD. Contraception 1989; \n39(1):37-52. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n434 \n \n (104)  Cox M, Tripp J, Blacksell S. Clinical performance of the levonorgestrel intrauterine system in \nroutine use by the UK Family Planning and Reproductive Health Research Network: 5-year report. J Fam Plann \nReprod Health Care 2002; 28(2):73-77. \n (105)  French RS, Cowan FM, Mansour D, Higgins JP, Robinson A, Procter T et al. Levonorgestrel-\nreleasing (20 microgram/day) intrauterine systems (Mirena) compared with other methods of reversible \ncontraceptives. BJOG: an International Journal of Obstetrics & Gynaecology 2000; 107(10):1218-1225. \n (106)  Luukkainen T, Pakarinen P, Toivonen J. Progestin-releasing intrauterine systems. Semin \nReprod Med 2001; 19(4):355-363. \n (107)  Backman T. Benefit-risk assessment of the levonorgestrel intrauterine system in \ncontraception. Drug Saf 2004; 27(15):1185-1204. \n (108)  Hidalgo M, Bahamondes L, Perrotti M, Diaz J, Dantas-Monteiro C, Petta C. Bleeding patterns \nand clinical performance of the levonorgestrel-releasing intrauterine system (Mirena) up to two years. \nContraception 2002; 65(2):129-132. \n (109)  Backman T, Huhtala S, Tuominen J, Luoto R, Erkkola R, Blom T et al. Sixty thousand \nwoman-years of experience on the levonorgestrel intrauterine system: an epidemiological survey in Finland. Eur \nJ Contracept Reprod Health Care 2001; 6 Suppl 1:23-26. \n (110)  Sivin I, Stern J, Diaz J, Diaz MM, Faundes A, el Mahgoub S et al. Two years of intrauterine \ncontraception with levonorgestrel and with copper: a randomized comparison of the TCu 380Ag and \nlevonorgestrel 20 mcg/day devices. Contraception 1987; 35(3):245-255. \n (111)  Sivin I, Stern J, Coutinho E, Mattos CE, el Mahgoub S, Diaz S et al. Prolonged intrauterine \ncontraception: a seven-year randomized study of the levonorgestrel 20 mcg/day (LNg 20) and the Copper T380 \nAg IUDS. Contraception 1991; 44(5):473-480. \n (112)  Cox M, Blacksell S. Clinical performance of the levonorgestrel intra-uterine system in routine \nuse by the UK Family Planning and Reproductive Health Research Network: 12-month report. Br J Fam Plann \n2000; 26(3):143-147. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n435 \n \n (113)  Abou-Setta A, Al-Inany H, Farquhar C. Levonorgestrel-releasing intrauterine device (LNG-\nIUD) for symptomatic endometriosis following surgery. Cochrane Database of Systematic Reviews 2005; 1, \n2005. \n (114)  Yap C. Pre and post operative medical therapy for endometriosis surgery. Cochrane Database \nof Systematic Reviews 2005; 1, 2005. \n (115)  Riphagen FE. Intrauterine application of progestins in hormone replacement therapy: a \nreview. Climacteric 2000; 3(3):199-211. \n (116)  Luukkainen T. Issues to debate on the Women's Health Initiative: failure of estrogen plus \nprogestin therapy for prevention of breast cancer risk. Hum Reprod 2003; 18(8):1559-1561. \n (117)  Stahlberg C, Pederson AT, Lynge E, Ottesen B. Hormone replacement therapy and risk of \nbreast cancer: the role of progestins. Acta Obstetricia et Gynecologica Scandinavica 2003; 82(7):335-344. \n (118)  Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H et al. Effects of \nconjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative \nrandomized controlled trial. JAMA 2004; 291(14):1701-1712. \n (119)  Beral V. Breast cancer and hormone-replacement therapy in the Million Women Study. \nLancet 2003; 362(9382):419-427. \n (120)  Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML et al. \nRisks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the \nWomen's Health Initiative randomized controlled trial. JAMA 2002; 288(3):321- 333. \n (121)  Hurskainen R, Teperi J, Aalto AM, Grenman S, Kivela A, Kujansuu E et al. Levonorgestrel-\nreleasing intrauterine system or hysterectomy in the treatment of essential menorrhagia: predictors of outcome. \nActa Obstetricia et Gynecologica Scandinavica 2004; 83(4):401-403. \n (122)  Bourdrez P, Bongers MY, Mol BW. Treatment of dysfunctional uterine bleeding: patient \npreferences for endometrial ablation, a levonorgestrel-releasing intrauterine device, or hysterectomy. Fertil \nSteril 2004; 82(1):160-6, quiz. \n\nREFERENCES  Chapter 3.2. Systematic review of LNG-IUS (Mirena) \n436 \n \n (123)  Chabbert-Buffet N, Meduri G, Bouchard P, Spitz IM. Selective progesterone receptor \nmodulators and progesterone antagonists: mechanisms of action and clinical applications. Hum Reprod Update \n2005; 11(3):293-307. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n437 \n \nChapter 4 \n \n(1)  RCOG. Birth after previous caesarean birth. RCOG Green top guideline No.45.  2007.  RCOG \nPress.  \n (2)  Varma R, Smith GC. Management of women with previous caesarean section. In Press. In: \nWarren R, Arulkumaran S, editors. Best Practice in Labour and Delivery. Cambridge University Press, \nCambridge, UK.; 2008. \n (3)  Varma R, Gupta JK. Ectopic Pregnancy. \nhttp://clinicalevidence.bmj.com/ceweb/conditions/pac/1406/1406_background.jsp. BMJ Clinical Evidence . \n2006.  \n (4)  Varma R, Gupta JK. Laparoscopic entry techniques: clinical guideline, national survey, and \nmedicolegal ramifications. Surg Endosc 2008; 22(12):2686-2697. \n (5)  Varma R, Gupta JK. Failed sterilisation: evidence-based review and medico-legal \nramifications. BJOG 2004; 111(12):1322-1332. \n (6)  Varma R, Gupta JK. Minimising the risk of sterilisation failure: an evidence based approach. \nIn: O'Donovan P, editor. Complications in Gynaecological Surgery. Springer-Verlag, London; 2008. 106-126. \n (7)  RCOG. Development of RCOG Green-top Guidelines: Producing a Clinical Practice \nGuideline. Clinical Governance Advice No. 1c.  2006.  Royal College of Obstetricians and Gynaecologists, \nLondon, UK.  \n (8)  Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S et al. Grading quality of \nevidence and strength of recommendations. BMJ 2004; 328(7454):1490. \n (9)  SIGN. Scottish Intercollegiate Guidelines Network (SIGN). Forming guideline \nrecommendations. In: A guideline developers' handbook. Edinburgh: SIGN, 2001. (Publication No 50.). www \nsign ac uk/guidelines/fulltext/50/section6 html [ 2001  \n (10)  The Parliamentary Office of Science and Technology. Caesarean Sections. PostNote. Report \nNumber 184. http://www.parliament.uk/post/pn184.pdf.  2002.  \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n438 \n \nRef Type: Report \n (11)  Menacker F. Trends in cesarean rates for first births and repeat cesarean rates for low-risk \nwomen: United States, 1990-2003. Natl Vital Stat Rep 2005; 54(4):1-8. \n (12)  Liu S, Rusen ID, Joseph KS, Liston R, Kramer MS, Wen SW et al. Recent trends in caesarean \ndelivery rates and indications for caesarean delivery in Canada. Journal of Obstetrics & Gynaecology Canada: \nJOGC 2004; 26(8):735-742. \n (13)  Black C, Kaye JA, Jick H. Cesarean delivery in the United Kingdom: time trends in the \ngeneral practice research database. Obstet Gynecol 2005; 106(1):151-155. \n (14)  Yeh J, Wactawski-Wende J, Shelton JA, Reschke J. Temporal trends in the rates of trial of \nlabor in low-risk pregnancies and their impact on the rates and success of vaginal birth after cesarean delivery. \nAm J Obstet Gynecol 2006; 194(1):144. \n (15)  Landon MB, Hauth JC, Leveno KJ, Spong CY, Leindecker S, Varner MW et al. Maternal and \nperinatal outcomes associated with a trial of labor after prior cesarean delivery. New England Journal of \nMedicine 2004; 351(25):2581-2589. \n (16)  Smith GC, Pell JP, Cameron AD, Dobbie R. Risk of perinatal death associated with labor after \nprevious cesarean delivery in uncomplicated term pregnancies. JAMA 2002; 287(20):2684- 2690. \n (17)  NICE. National Institute of Clinical Excellence, RCOG Caesarean Section.Guideline Number \n13.(www.nice.org.uk/CG013NICEguideline).  2004.  \n (18)  ACOG Committee on Obstetric Practice. ACOG Committee Opinion No. 342: induction of \nlabor for vaginal birth after cesarean delivery. Obstet Gynecol 2006; 108(2):465-468. \n (19)  ACOG Committee on Obstetric Practice. ACOG Practice Bulletin No. 54: Vaginal Birth After \nPrevious Cesarean. Obstetrics & Gynecology 2004; 104(1):203-212. \n (20)  SOGC. The Society of Obstetricians and Gynaecologists of Canada. SOGC Clinical Practice \nGuidelines. Guidelines for vaginal birth after previous caesarean birth. Number 155 (Replaces guideline \nNumber 147), February 2005. Int J Gynaecol Obstet 2005; 89(3):319-331. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n439 \n \n (21)  Guise JM, McDonagh MS, Hashima J, Kraemer DF, Eden KB, Berlin M et al. Vaginal birth \nafter cesarean (VBAC). Evidence Report: Technology Assessment (Summary) 2003;(71):1- 8. \n (22)  New Zealand Guidelines Group. Care of women with breech presentation or previous \ncaesarean birth. Evidence based practice guideline. New Zealand Guidelines Group \n(NZGG).http://www.nzgg.org.nz. November 2004.  2004.  \n (23)  Dodd JM, Crowther CA, Huertas E, Guise JM, Horey D. Planned elective repeat caesarean \nsection versus planned vaginal birth for women with a previous caesarean birth. Cochrane Database Syst Rev \n2006;(4):CD004224. \n (24)  Dodd JM, Crowther CA, Hiller JE, Haslam RR, Robinson JS. Birth after caesarean study --\nplanned vaginal birth or planned elective repeat caesarean for women at term with a single previous caesarean \nbirth: protocol for a patient preference study and randomised trial. BMC Pregnancy & Childbirth 2007; 7:17. \n (25)  Lieberman E, Ernst EK, Rooks JP, Stapleton S, Flamm B. Results of the national study of \nvaginal birth after cesarean in birth centers. Obstetrics & Gynecology 2004; 104(5 Pt 1):933- 942. \n (26)  Macones GA, Cahill A, Pare E, Stamilio DM, Ratcliffe S, Stevens E et al. Obstetric outcomes \nin women with two prior cesarean deliveries: is vaginal birth after cesarean delivery a viable option? Am J \nObstet Gynecol 2005; 192(4):1223-1228. \n (27)  Guise JM, Hashima J, Osterweil P. Evidence-based vaginal birth after Caesarean section. Best \nPract Res Clin Obstet Gynaecol 2005; 19(1):117-130. \n (28)  Spaans WA, van der Vliet LM, Roell-Schorer EA, Bleker OP, van Roosmalen J. Trial of \nlabour after two or three previous caesarean sections. European Journal of Obstetrics, Gynecology, & \nReproductive Biology 2003; 110(1):16-19. \n (29)  Landon MB, Leindecker S, Spong CY, Hauth JC, Bloom S, Varner MW et al. The MFMU \nCesarean Registry: factors affecting the success of trial of labor after previous cesarean delivery. American \nJournal of Obstetrics & Gynecology 2005; 193(3 Pt 2):1016-1023. \n (30)  Turner MJ. Uterine rupture. Best Practice & Research in Clinical Obstetrics & Gynaecology \n2002; 16(1):69-79. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n440 \n \n (31)  Seracchioli R, Manuzzi L, Vianello F, Gualerzi B, Savelli L, Paradisi R et al. Obstetric and \ndelivery outcome of pregnancies achieved after laparoscopic myomectomy. Fertil Steril 2006; 86(1):159- 165. \n (32)  Dubuisson JB, Fauconnier A, Babaki-Fard K, Chapron C. Laparoscopic myomectomy: a \ncurrent view. Hum Reprod Update 2000; 6(6):588-594. \n (33)  Seracchioli R, Rossi S, Govoni F, Rossi E, Venturoli S, Bulletti C et al. Fertility and obstetric \noutcome after laparoscopic myomectomy of large myomata: a randomized comparison with abdominal \nmyomectomy. Hum Reprod 2000; 15(12):2663-2668. \n (34)  Landon MB, Spong CY, Thom E, Hauth JC, Bloom SL, Varner MW et al. Risk of uterine \nrupture with a trial of labor in women with multiple and single prior cesarean delivery. Obstet Gynecol 2006; \n108(1):12-20. \n (35)  Spaans WA, van der Vliet LM, Roell-Schorer EA, Bleker OP, van Roosmalen J. Trial of \nlabour after two or three previous caesarean sections. Eur J Obstet Gynecol Reprod Biol 2003; 110(1):16- 19. \n (36)  Caughey AB, Shipp TD, Repke JT, Zelop CM, Cohen A, Lieberman E. Rate of uterine rupture \nduring a trial of labor in women with one or two prior cesarean deliveries. American Journal of Obstetrics & \nGynecology 1999; 181(4):872-876. \n (37)  Miller DA, Diaz FG, Paul RH. Vaginal birth after cesarean: a 10-year experience. Obstet \nGynecol 1994; 84(2):255-258. \n (38)  Lyerly AD, Mitchell LM, Armstrong EM, Harris LH, Kukla R, Kuppermann M et al. Risks, \nvalues, and decision making surrounding pregnancy. Obstetrics & Gynecology 2007; 109(4):979-984. \n (39)  Montgomery AA, Emmett CL, Fahey T, Jones C, Ricketts I, Patel RR et al. Two decision aids \nfor mode of delivery among women with previous caesarean section: randomised controlled trial. BMJ 2007; \n334(7607):1305. \n (40)  Wen SW, Rusen ID, Walker M, Liston R, Kramer MS, Baskett T et al. Comparison of \nmaternal mortality and morbidity between trial of labor and elective cesarean section among women wit h \nprevious cesarean delivery. American Journal of Obstetrics & Gynecology 2004; 191(4):1263-1269. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n441 \n \n (41)  Chauhan SP, Martin JN, Jr., Henrichs CE, Morrison JC, Magann EF. Maternal and perinatal \ncomplications with uterine rupture in 142,075 patients who attempted vaginal birth after cesarean delivery: A \nreview of the literature. American Journal of Obstetrics & Gynecology 2003; 189(2):408-417. \n (42)  Guise JM, Berlin M, McDonagh M, Osterweil P, Chan B, Helfand M. Safety of vaginal birth \nafter cesarean: a systematic review. Obstetrics & Gynecology 2004; 103(3):420-429. \n (43)  Mozurkewich EL, Hutton EK. Elective repeat cesarean delivery versus trial of labor: a meta-\nanalysis of the literature from 1989 to 1999. American Journal of Obstetrics & Gynecology 2000; 183(5):1187-\n1197. \n (44)  Smith GC, White IR, Pell JP, Dobbie R. Predicting cesarean section and uterine rupture \namong women attempting vaginal birth after prior cesarean section. PLoS Med 2005; 2(9):871- 878. \n (45)  Gyamfi C, Juhasz G, Gyamfi P, Stone JL. Increased success of trial of labor after previous \nvaginal birth after cesarean. Obstetrics & Gynecology 2004; 104(4):715-719. \n (46)  Hibbard JU, Gilbert S, Landon MB, Hauth JC, Leveno KJ, Spong CY et al. Trial of labor or \nrepeat cesarean delivery in women with morbid obesity and previous cesarean delivery. Obstet Gynecol 2006; \n108(1):125-133. \n (47)  Goodall PT, Ahn JT, Chapa JB, Hibbard JU. Obesity as a risk factor for failed trial of labor in \npatients with previous cesarean delivery. Am J Obstet Gynecol 2005; 192(5):1423- 1426. \n (48)  Juhasz G, Gyamfi C, Gyamfi P, Tocce K, Stone JL. Effect of body mass index and excessive \nweight gain on success of vaginal birth after cesarean delivery. Obstet Gynecol 2005; 106(4):741-746. \n (49)  Bujold E, Hammoud AO, Hendler I, Berman S, Blackwell SC, Duperron L et al. Trial of labor \nin patients with a previous cesarean section: does maternal age influence the outcome? American Journal of \nObstetrics & Gynecology 2004; 190(4):1113-1118. \n (50)  Coassolo KM, Stamilio DM, Pare E, Peipert JF, Stevens E, Nelson DB et al. Safety and \nefficacy of vaginal birth after cesarean attempts at or beyond 40 weeks of gestation. Obstet Gynecol 2005; \n106(4):700-706. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n442 \n \n (51)  Hollard AL, Wing DA, Chung JH, Rumney PJ, Saul L, Nageotte MP et al. Ethnic disparity in \nthe success of vaginal birth after cesarean delivery. J Matern Fetal Neonatal Med 2006; 19(8):483-487. \n (52)  Rochelson B, Pagano M, Conetta L, Goldman B, Vohra N, Frey M et al. Previous preterm \ncesarean delivery: identification of a new risk factor for uterine rupture in VBAC candidates. J Matern Fetal \nNeonatal Med 2005; 18(5):339-342. \n (53)  Durnwald C, Mercer B. Vaginal birth after Cesarean delivery: predicting success, risks of \nfailure. Journal of Maternal-Fetal & Neonatal Medicine 2004; 15(6):388-393. \n (54)  Hoskins IA, Gomez JL. Correlation between maximum cervical dilatation at cesarean delivery \nand subsequent vaginal birth after cesarean delivery. Obstet Gynecol 1997; 89(4):591-593. \n (55)  Flamm BL, Geiger AM. Vaginal birth after cesarean delivery: an admission scoring system. \nObstetrics & Gynecology 1997; 90(6):907-910. \n (56)  Macones GA, Hausman N, Edelstein R, Stamilio DM, Marder SJ. Predicting outcomes of \ntrials of labor in women attempting vaginal birth after cesarean delivery: a comparison of multivariate methods \nwith neural networks. American Journal of Obstetrics & Gynecology 2001; 184(3):409-413. \n (57)  Hashima JN, Eden KB, Osterweil P, Nygren P, Guise JM. Predicting vaginal birth after \ncesarean delivery: a review of prognostic factors and screening tools. American Journal of Obstetrics & \nGynecology 2004; 190(2):547-555. \n (58)  Dinsmoor MJ, Brock EL. Predicting failed trial of labor after primary cesarean delivery. \nObstetrics & Gynecology 2004; 103(2):282-286. \n (59)  Macones GA, Cahill AG, Stamilio DM, Odibo A, Peipert J, Stevens EJ. Can uterine rupture in \npatients attempting vaginal birth after cesarean delivery be predicted? Am J Obstet Gynecol 2006; 195(4):1148-\n1152. \n (60)  Ofir K, Sheiner E, Levy A, Katz M, Mazor M. Uterine rupture: risk factors and pregnancy \noutcome. American Journal of Obstetrics & Gynecology 2003; 189(4):1042-1046. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n443 \n \n (61)  Guise JM, McDonagh MS, Osterweil P, Nygren P, Chan BK, Helfand M. Systematic review \nof the incidence and consequences of uterine rupture in women with previous caesarean section. BMJ 2004; \n329(7456):19-25. \n (62)  Turner MJ, Agnew G, Langan H. Uterine rupture and labour after a previous low transverse \ncaesarean section. BJOG 2006; 113(6):729-732. \n (63)  Gardeil F, Daly S, Turner MJ. Uterine rupture in pregnancy reviewed. Eur J Obstet Gynecol \nReprod Biol 1994; 56(2):107-110. \n (64)  Smith GC, Pell JP, Pasupathy D, Dobbie R. Factors predisposing to perinatal death related to \nuterine rupture during attempted vaginal birth after caesarean section: retrospective cohort study. BMJ 2004; \n329(7462):375. \n (65)  Shipp TD, Zelop C, Cohen A, Repke JT, Lieberman E. Post-cesarean delivery fever and \nuterine rupture in a subsequent trial of labor. Obstetrics & Gynecology 2003; 101(1):136-139. \n (66)  Durnwald C, Mercer B. Uterine rupture, perioperative and perinatal morbidity after single-\nlayer and double-layer closure at cesarean delivery. American Journal of Obstetrics and Gynecology 2003; \n189(4):925-929. \n (67)  Farmer RM, Kirschbaum T, Potter D, Strong TH, Medearis AL. Uterine rupture during trial of \nlabor after previous cesarean section. Am J Obstet Gynecol 1991; 165(4 Pt 1):996-1001. \n (68)  Smith GC, Pell JP, Dobbie R. Caesarean section and risk of unexplained stillbirth in \nsubsequent pregnancy. Lancet 2003; 362(9398):1779-1784. \n (69)  Smith GC. Life-table analysis of the risk of perinatal death at term and post term in singleton \npregnancies. Am J Obstet Gynecol 2001; 184(3):489-496. \n (70)  Badawi N, Felix JF, Kurinczuk JJ, Dixon G, Watson L, Keogh JM et al. Cerebral palsy \nfollowing term newborn encephalopathy: a population-based study. Dev Med Child Neurol 2005; 47(5):293-\n298. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n444 \n \n (71)  Levine EM, Ghai V, Barton JJ, Strom CM. Mode of delivery and risk of respiratory diseases \nin newborns. Obstet Gynecol 2001; 97(3):439-442. \n (72)  Morrison JJ, Rennie JM, Milton PJ. Neonatal respiratory morbidity and mode of delivery at \nterm: influence of timing of elective caesarean section. Br J Obstet Gynaecol 1995; 102(2):101- 106. \n (73)  Richardson BS, Czikk MJ, daSilva O, Natale R. The impact of labor at term on measures of \nneonatal outcome. Am J Obstet Gynecol 2005; 192(1):219-226. \n (74)  Hook B, Kiwi R, Amini SB, Fanaroff A, Hack M. Neonatal morbidity after elective repeat \ncesarean section and trial of labor. Pediatrics 1997; 100(3 Pt 1):348-353. \n (75)  Stutchfield P, Whitaker R, Russell I. Antenatal betamethasone and incidence of neonatal \nrespiratory distress after elective caesarean section: pragmatic randomised trial. ASTECS study. BMJ 2005; \n331(7518):662. \n (76)  Dalziel SR, Walker NK, Parag V, Mantell C, Rea HH, Rodgers A et al. Cardiovascular risk \nfactors after antenatal exposure to betamethasone: 30-year follow-up of a randomised controlled trial. Lancet \n2005; 365(9474):1856-1862. \n (77)  CEMACH. Confidential Enquiries into Maternal and Child Health. Why Mothers Die 2000-\n2002.The Sixth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. \nhttp://www.cemach.org.uk/publications/WMD2000_2002/content.htm.  2004.  RCOG Press.  \n (78)  Bloom SL, Spong CY, Weiner SJ, Landon MB, Rouse DJ, Varner MW et al. Complications of \nanesthesia for cesarean delivery. Obstet Gynecol 2005; 106(2):281-287. \n (79)  Getahun D, Oyelese Y, Salihu HM, Ananth CV. Previous cesarean delivery and risks of \nplacenta previa and placental abruption. Obstet Gynecol 2006; 107(4):771-778. \n (80)  Faiz AS, Ananth CV. Etiology and risk factors for placenta previa: an overview and meta-\nanalysis of observational studies. J Matern Fetal Neonatal Med 2003; 13(3):175-190. \n (81)  Silver RM, Landon MB, Rouse DJ, Leveno KJ, Spong CY, Thom EA et al. Maternal \nmorbidity associated with multiple repeat cesarean deliveries. Obstet Gynecol 2006; 107(6):1226-1232. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n445 \n \n (82)  Makoha FW, Felimban HM, Fathuddien MA, Roomi F, Ghabra T. Multiple cesarean section \nmorbidity. Int J Gynaecol Obstet 2004; 87(3):227-232. \n (83)  RCOG. Placenta Praevia and Placenta Praevia Accreta: Diagnosis and Management (27).  \n2005.  \n (84)  CEMACH. The Confidential Enquiry into Maternal and Child Health (CEMACH). Saving \nMothers' Lives: reviewing maternal deaths to make motherhood safer - 2003-2005. The Seventh Report on \nConfidential Enquiries into Maternal Deaths in the United Kingdom. London: CEMACH.  2007.  \n (85)  Quinones JN, Stamilio DM, Pare E, Peipert JF, Stevens E, Macones GA. The effect of \nprematurity on vaginal birth after cesarean delivery: success and maternal morbidity. Obstet Gynecol 2005; \n105(3):519-524. \n (86)  Durnwald CP, Rouse DJ, Leveno KJ, Spong CY, MacPherson C, Varner MW et al. The \nMaternal-Fetal Medicine Units Cesarean Registry: safety and efficacy of a trial of labor in preterm pregnancy \nafter a prior cesarean delivery. Am J Obstet Gynecol 2006; 195(4):1119- 1126. \n (87)  Varner MW, Leindecker S, Spong CY, Moawad AH, Hauth JC, Landon MB et al. The \nMaternal-Fetal Medicine Unit cesarean registry: trial of labor with a twin gestation. American Journal of \nObstetrics & Gynecology 2005; 193(1):135-140. \n (88)  Cahill A, Stamilio DM, Pare E, Peipert JP, Stevens EJ, Nelson DB et al. Vaginal birth after \ncesarean (VBAC) attempt in twin pregnancies: is it safe? Am J Obstet Gynecol 2005; 193(3 Pt 2):1050-1055. \n (89)  Ford AA, Bateman BT, Simpson LL. Vaginal birth after cesarean delivery in twin gestations: \na large, nationwide sample of deliveries. Am J Obstet Gynecol 2006; 195(4):1138- 1142. \n (90)  Elkousy MA, Sammel M, Stevens E, Peipert JF, Macones G. The effect of birth weight on \nvaginal birth after cesarean delivery success rates. American Journal of Obstetrics & Gynecology 2003; \n188(3):824-830. \n (91)  Peaceman AM, Gersnoviez R, Landon MB, Spong CY, Leveno KJ, Varner MW et al. The \nMFMU Cesarean Registry: impact of fetal size on trial of labor success for patients with previous cesarean for \ndystocia. Am J Obstet Gynecol 2006; 195(4):1127-1131. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n446 \n \n (92)  Bujold E, Mehta SH, Bujold C, Gauthier RJ. Interdelivery interval and uterine rupture. \nAmerican Journal of Obstetrics & Gynecology 2002; 187(5):1199-1202. \n (93)  Esposito MA, Menihan CA, Malee MP. Association of interpregnancy interval with uterine \nscar failure in labor: a case-control study. American Journal of Obstetrics & Gynecology 2000; 183(5):1180-\n1183. \n (94)  Shipp TD, Zelop CM, Repke JT, Cohen A, Lieberman E. Interdelivery interval and risk of \nsymptomatic uterine rupture. Obstetrics & Gynecology 2001; 97(2):175- 177. \n (95)  Sakala EP, Kaye S, Murray RD, Munson LJ. Epidural analgesia. Effect on the likelihood of a \nsuccessful trial of labor after cesarean section. J Reprod Med 1990; 35(9):886- 890. \n (96)  Rowbottom SJ, Critchley LA, Gin T. Uterine rupture and epidural analgesia during trial of \nlabour. Anaesthesia 1997; 52(5):486-488. \n (97)  Fisler RE, Cohen A, Ringer SA, Lieberman E. Neonatal outcome after trial of labor compared \nwith elective repeat cesarean section. Birth 2003; 30(2):83-88. \n (98)  Arulkumaran S, Chua S, Ratnam SS. Symptoms and signs with scar rupture --value of uterine \nactivity measurements. Aust N Z J Obstet Gynaecol 1992; 32(3):208-212. \n (99)  Beckley S, Gee H, Newton JR. Scar rupture in labour after previous lower uterine segment \ncaesarean section: the role of uterine activity measurement. Br J Obstet Gynaecol 1991; 98(3):265- 269. \n (100)  Rodriguez MH, Masaki DI, Phelan JP, Diaz FG. Uterine rupture: are intrauterine pressure \ncatheters useful in the diagnosis? Am J Obstet Gynecol 1989; 161(3):666-669. \n (101)  Madanes AE, David D, Cetrulo C. Major complications associated with intrauterine pressure \nmonitoring. Obstet Gynecol 1982; 59(3):389-391. \n (102)  NICE. National Institute of Clinical Excellence. RCOG Induction of labour. Evidence-based \nClinical guideline Number 9.  2001.  National Institute of Clinical Excellence.  \nRef Type: Report \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n447 \n \n (103)  McDonagh MS, Osterweil P, Guise JM. The benefits and risks of inducing labour in patients \nwith prior caesarean delivery: a systematic review. BJOG 2005; 112(8):1007- 1015. \n (104)  Dodd J, Crowther C. Induction of labour for women with a previous Caesarean birth: a \nsystematic review of the literature. Aust N Z J Obstet Gynaecol 2004; 44(5):392- 395. \n (105)  Dodd JM, Crowther CA. Elective repeat caesarean section versus induction of labour for \nwomen with a previous caesarean birth. Cochrane Database Syst Rev 2006;(4):CD004906. \n (106)  Grobman WA, Gilbert S, Landon MB, Spong CY, Leveno KJ, Rouse DJ et al. Outcomes of \ninduction of labor after one prior cesarean. Obstetrics & Gynecology 2007; 109(2:Pt 1):t- 9. \n (107)  Kelly AJ, Kavanagh J, Thomas J. Vaginal prostaglandin (PGE2 and PGF2a) for induction of \nlabour at term. Cochrane Database Syst Rev 2006;(2):CD003101. \n (108)  Gray R, Quigley MA, Hockley C, Kurinczuk JJ, Goldacre M, Brocklehurst P. Caesarean \ndelivery and risk of stillbirth in subsequent pregnancy: a retrospective cohort study in an English population. \nBJOG: an International Journal of Obstetrics & Gynaecology 2007; 114(3):264-270. \n (109)  Arulkumaran S, Gibb DM, Ingemarsson I, Kitchener HC, Ratnam SS. Uterine activity during \nspontaneous labour after previous lower-segment caesarean section. Br J Obstet Gynaecol 1989; 96(8):933- 938. \n (110)  Arulkumaran S, Ingemarsson I, Ratnam SS. Oxytocin augmentation in dysfunctional labour \nafter previous caesarean section. Br J Obstet Gynaecol 1989; 96(8):939-941. \n (111)  Goetzl L, Shipp TD, Cohen A, Zelop CM, Repke JT, Lieberman E. Oxytocin dose and the risk \nof uterine rupture in trial of labor after cesarean. Obstetrics & Gynecology 2001; 97(3):381-3 84. \n (112)  Zelop CM, Shipp TD, Repke JT, Cohen A, Caughey AB, Lieberman E. Uterine rupture during \ninduced or augmented labor in gravid women with one prior cesarean delivery. American Journal of Obstetrics \n& Gynecology 1999; 181(4):882-886. \n (113)  Hamilton EF, Bujold E, McNamara H, Gauthier R, Platt RW. Dystocia among women with \nsymptomatic uterine rupture. American Journal of Obstetrics & Gynecology 2001; 184(4):620- 624. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n448 \n \n (114)  National Insititute of Clinical Excellence. Intrapartum care: care of healthy women and their \nbabies during childbirth. NICE Clinical Guideline 55. www.nice.org.uk.  2007.  \nRef Type: Report \n (115)  Cahill AG, Stamilio DM, Odibo AO, Peipert JF, Stevens EJ, Macones GA. Does a maximum \ndose of oxytocin affect risk for uterine rupture in candidates for vaginal birth after cesarean delivery? American \nJournal of Obstetrics & Gynecology 2007; 197(5):495. \n (116)  Smith GC, Shah I, White IR, Pell JP, Dobbie R. Previous Preeclampsia, Preterm Delivery, and \nDelivery of a Small for Gestational Age Infant and the Risk of Unexplained Stillbirth in the Second Pregnancy: \nA Retrospective Cohort Study, Scotland, 1992-2001. Am J Epidemiol 2006. \n (117)  Eden KB, Hashima JN, Osterweil P, Nygren P, Guise JM. Childbirth preferences after \ncesarean birth: a review of the evidence. Birth 2004; 31(1):49-60. \n (118)  Horey D, Weaver J, Russell H. Information for pregnant women about caesarean birth. \nCochrane Database Syst Rev 2004;(1):CD003858. \n (119)  Gamble JA, Creedy DK. Women's preference for a cesarean section: incidence and associated \nfactors. Birth 2001; 28(2):101-110. \n (120)  Shorten A, Shorten B, Keogh J, West S, Morris J. Making choices for childbirth: a \nrandomized controlled trial of a decision-aid for informed birth after cesarean. Birth 2005; 32(4):252-261. \n (121)  Mankuta DD, Leshno MM, Menasche MM, Brezis MM. Vaginal birth after cesarean section: \ntrial of labor or repeat cesarean section? A decision analysis. American Journal of Obstetrics & Gynecology \n2003; 189(3):714-719. \n (122)  Pare E, Quinones JN, Macones GA. Vaginal birth after caesarean section versus elective \nrepeat caesarean section: assessment of maternal downstream health outcomes. BJOG 2006; 113(1):75- 85. \n (123)  Dodd J, Pearce E, Crowther C. Women's experiences and preferences following Caesarean \nbirth. Aust N Z J Obstet Gynaecol 2004; 44(6):521-524. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n449 \n \n (124)  David S, Mamelle N, Riviere O. Estimation of an expected caesarean section rate taking into \naccount the case mix of a maternity hospital. Analysis from the AUDIPOG Sentinelle Network (France). \nObstetricians of AUDIPOG. Association of Users of Computerised Files in Perinatalogy, Obstetrics and \nGynaecology. BJOG 2001; 108(9):919-926. \n (125)  Sur S, Mackenzie IZ. Does discussion of possible scar rupture influence preferred mode of \ndelivery after a caesarean section? J Obstet Gynaecol 2005; 25(4):338-341. \n (126)  Paglia M, Murtha A, Smith T. Factors influencing a woman's desire to choose vaginal birth \nafter cesarean section. American Journal of Obstetrics and Gynecology 2003; 189(6, Supplement 1):S142. \n (127)  Garrison F, Smith L, Givens L, Strassner H, Studee L, Judith H et al. Provider and hospital \nfactors associated with vaginal birth after cesarean: A survey of obstetric providers at two inner city teaching \nhospitals. American Journal of Obstetrics and Gynecology 2005; 193(6, Supplement 1):S124. \n (128)  Dunn EA, O'Herlihy C. Comparison of maternal satisfaction following vaginal delivery after \ncaesarean section and caesarean section after previous vaginal delivery. Eur J Obstet Gynecol Reprod Biol \n2005; 121(1):56-60. \n (129)  Cleary-Goldman J, Cornelisse K, Simpson LL, Robinson JN. Previous cesarean delivery: \nunderstanding and satisfaction with mode of delivery in a subsequent pregnancy in patients participating in a \nformal vaginal birth after cesarean counseling program. Am J Perinatol 2005; 22(4):217- 221. \n (130)  Gerber S, Sharp L, O'Toole C. Comparison of postpartum quality of life between patients with \nrepeat cesarean delivery and vaginal birth after cesarean. American Journal of Obstetrics and Gynecology 2003; \n189(6, Supplement 1):S157. \n (131)  Bouyer J, Coste J, Fernandez H, Pouly JL, Job-Spira N. Sites of ectopic pregnancy: a 10 year \npopulation-based study of 1800 cases. Human Reproduction 17(12):3224-30, 2002. \n (132)  Bakken IJ, Skjeldestad FE. Incidence and treatment of extrauterine pregnancies in Norway \n1990-2001. Tidsskr Nor Laegeforen 2003; 123(21):3016-3020. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n450 \n \n (133)  Boufous S, Quartararo M, Mohsin M, Parker J. Trends in the incidence of ectopic pregnancy \nin New South Wales between 1990-1998. Australian & New Zealand Journal of Obstetrics & Gynaecology \n41(4):436-8, 2001. \n (134)  Lewis G, Drife J, editors. Why Mothers Die 2000-2002. The sixth report of the Confidential \nEnquiry into Maternal Deaths in the United Kingdom.  2004. Royal College of Obstetricians and \nGynaecologists, RCOG Press, London.  \nRef Type: Report \n (135)  Bouyer J. Epidemiology of ectopic pregnancy: incidence, risk factors and outcomes. Journal \nde Gynecologie, Obstetrique et Biologie de la Reproduction 32(7 Suppl):S8- 17, 2003. \n (136)  Coste J, Bouyer J, Ughetto S, Gerbaud L, Fernandez H, Pouly JL et al. Ectopic pregnancy is \nagain on the increase. Recent trends in the incidence of ectopic pregnancies in France (1992-2002). Hum \nReprod 2004; 19(9):2014-2018. \n (137)  Condous G, Okaro E, Khalid A, Lu C, Van Huffel S, Timmerman D et al. The accuracy of \ntransvaginal ultrasonography for the diagnosis of ectopic pregnancy prior to surgery. Hum Reprod 2005. \n (138)  Mol BW, Van D, V, Bossuyt PM. Implementation of probabilistic decision rules improves the \npredictive values of algorithms in the diagnostic management of ectopic pregnancy. Hum Reprod 1999; \n14(11):2855-2862. \n (139)  Bouyer J, Fernandez H, Coste J, Pouly JL, Job-Spira N. Fertility after ectopic pregnancy: 10-\nyear results in the Auvergne Registry. J Gynecol Obstet Biol Reprod (Paris) 2003; 32(5):431- 438. \n (140)  Ankum WM, Mol BW, Van D, V, Bossuyt PM. Risk factors for ectopic pregnancy: a meta-\nanalysis. Fertil Steril 1996; 65(6):1093-1099. \n (141)  Dart RG, Kaplan B, Varaklis K. Predictive value of history and physical examination in \npatients with suspected ectopic pregnancy. Ann Emerg Med 1999; 33(3):283- 290. \n (142)  Latchaw G, Takacs P, Gaitan L, Geren S, Burzawa J. Risk factors associated with the rupture \nof tubal ectopic pregnancy. Gynecol Obstet Invest 2005; 60(3):177-180. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n451 \n \n (143)  Job-Spira N, Fernandez H, Bouyer J, Pouly JL, Germain E, Coste J. Ruptured tubal ectopic \npregnancy: risk factors and reproductive outcome: results of a population-based study in France. Am J Obste t \nGynecol 1999; 180(4):938-944. \n (144)  Condous G, Lu C, Van Huffel SV, Timmerman D, Bourne T. Human chorionic gonadotrophin \nand progesterone levels in pregnancies of unknown location. Int J Gynaecol Obstet 2004; 86(3):351- 357. \n (145)  Bangsgaard N, Lund CO, Ottesen B, Nilas L. Improved fertility following conservative \nsurgical treatment of ectopic pregnancy. BJOG: an International Journal of Obstetrics & Gynaecology 2003; \n110(8):765-770. \n (146)  Bouyer J, Job-Spira N, Pouly JL, Coste J, Germain E, Fernandez H. Fertility following \nradical, conservative-surgical or medical treatment for tubal pregnancy: a population-based study.[comment]. \nBJOG: an International Journal of Obstetrics & Gynaecology 2000; 107(6):714-721. \n (147)  Mol BW, Matthijsse HC, Tinga DJ, Huynh T, Hajenius PJ, Ankum WM et al. Fertility after \nconservative and radical surgery for tubal pregnancy. Hum Reprod 1998; 13(7):1804-1809. \n (148)  Silva PD, Schaper AM, Rooney B. Reproductive outcome after 143 laparoscopic procedures \nfor ectopic pregnancy. Obstet Gynecol 1993; 81(5 ( Pt 1)):710-715. \n (149)  Yao M, Tulandi T. Current status of surgical and nonsurgical management of ectopic \npregnancy. Fertil Steril 1997; 67(3):421-433. \n (150)  Hajenius PJ, Mol BW, Bossuyt PM, Ankum WM, Van D, V. Interventions for tubal ectopic \npregnancy.Cochrane Menstrual Disorders and Subfertility Group. Cochrane Database of Systematic Reviews \n2006;(2):CD000324. \n (151)  Morlock RJ, Lafata JE, Eisenstein D. Cost-effectiveness of single-dose methotrexate \ncompared with laparoscopic treatment of ectopic pregnancy. Obstet Gynecol 2000; 95(3):407- 412. \n (152)  Lund CO, Nilas L, Bangsgaard N, Ottesen B. Persistent ectopic pregnancy after linear \nsalpingotomy: a non-predictable complication to conservative surgery for tubal gestation. Acta Obstetricia et \nGynecologica Scandinavica 81(11):1053-9, 2002. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n452 \n \n (153)  Graczykowski JW, Mishell DR, Jr. Methotrexate prophylaxis for persistent ectopic pregnancy \nafter conservative treatment by salpingostomy. Obstetrics & Gynecology 1997; 89(1):118-1 22. \n (154)  Ego A, Subtil D, Cosson M, Legoueff F, Houfflin-Debarge V, Querleu D. Fertility after \nectopic pregnancy: the population-based register of the urban area around Lille, Northern France. Gynecologie, \nObstetrique & Fertilite 30(3):195-203, 2002. \n (155)  Korell M, Albrich W, Hepp H. Fertility after organ-preserving surgery of ectopic pregnancy: \nresults of a multicenter study. Fertility & Sterility 1997; 68(2):220-223. \n (156)  Dubuisson JB, Morice P, Chapron C, De Gayffier A, Mouelhi T. Salpingectomy - the \nlaparoscopic surgical choice for ectopic pregnancy. Hum Reprod 1996; 11(6):1199-1203. \n (157)  Barnhart KT, Gosman G, Ashby R, Sammel M. The medical management of ectopic \npregnancy: a meta-analysis comparing \"single dose\" and \"multidose\" regimens. Obstetrics & Gynecology 2003; \n101(4):778-784. \n (158)  Alleyassin A, Khademi A, Aghahosseini M, Safdarian L, Badenoosh B, Hamed EA. \nComparison of success rates in the medical management of ectopic pregnancy with single-dose and multiple-\ndose administration of methotrexate: a prospective, randomized clinical trial. Fertility & Sterility 85(6):1661-6, \n2006. \n (159)  Lipscomb GH, Givens VM, Meyer NL, Bran D. Comparison of multidose and single-dose \nmethotrexate protocols for the treatment of ectopic pregnancy. Obstet Gynecol Surv 2005; 60(10):646- 647. \n (160)  Sowter MC, Farquhar CM, Petrie KJ, Gudex G. A randomised trial comparing single dose \nsystemic methotrexate and laparoscopic surgery for the treatment of unruptured tubal pregnancy. BJOG: an \nInternational Journal of Obstetrics & Gynaecology 2001; 108(2):192-203. \n (161)  Nieuwkerk PT, Hajenius PJ, Ankum WM, Van D, V, Wijker W, Bossuyt PM. Systemic \nmethotrexate therapy versus laparoscopic salpingostomy in patients with tubal pregnancy. Part I. Impact on \npatients' health-related quality of life. Fertility & Sterility 1998; 70(3):511- 517. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n453 \n \n (162)  Parker J, Bisits A, Proietto AM. A systematic review of single-dose intramuscular \nmethotrexate for the treatment of ectopic pregnancy. Australian & New Zealand Journal of Obstetrics & \nGynaecology 1998; 38(2):145-150. \n (163)  Gervaise A, Masson L, de Tayrac R, Frydman R, Fernandez H. Reproductive outcome after \nmethotrexate treatment of tubal pregnancies. Fertil Steril 2004; 82(2):304- 308. \n (164)  Chapron C, Fernandez H, Dubuisson JB. Treatment of ectopic pregnancy in 2000. Journal de \nGynecologie, Obstetrique et Biologie de la Reproduction 29(4):351-61, 2000. \n (165)  Floridon C, Thomsen SG. Methotrexate treatment of ectopic pregnancy. Acta Obstet Gynecol \nScand 1994; 73(10):746-752. \n (166)  Elson J, Tailor A, Banerjee S, Salim R, Hillaby K, Jurkovic D. Expectant management of \ntubal ectopic pregnancy: prediction of successful outcome using decision tree analysis. Ultrasound Obstet \nGynecol 2004; 23(6):552-556. \n (167)  RCOG. The management of tubal pregnancy. Guideline no 21.  2004. Royal College of \nObstetricians and Gynaecologists, London, UK, RCOG Press, London.  \nRef Type: Report \n (168)  Gamzu R, Almog B, Levin Y, Avni A, Jaffa A, Lessing JB et al. Efficacy of methotrexate \ntreatment in extrauterine pregnancies defined by stable or increasing human chorionic gonadotropin \nconcentrations. Fertil Steril 2002; 77(4):761-765. \n (169)  Lecuru F, Robin F, Bernard JP, Maizan dM, Mac-Cordick C, Boucaya V et al. Single-dose \nmethotrexate for unruptured ectopic pregnancy. International Journal of Gynaecology & Obstetrics 1998; \n61(3):253-259. \n (170)  Lipscomb GH, McCord ML, Stovall TG, Huff G, Portera SG, Ling FW. Predictors of success \nof methotrexate treatment in women with tubal ectopic pregnancies. New England Journal of Medicine 1999; \n341(26):1974-1978. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n454 \n \n (171)  Nazac A, Gervaise A, Bouyer J, de Tayrac R, Capella-Allouc S, Fernandez H. Predictors of \nsuccess in methotrexate treatment of women with unruptured tubal pregnancies. Ultrasound in Obstetrics & \nGynecology 2003; 21(2):181-185. \n (172)  Tawfiq A, Agameya AF, Claman P. Predictors of treatment failure for ectopic pregnancy \ntreated with single-dose methotrexate. Fertility & Sterility 74(5):877-80, 2000. \n (173)  Erdem M, Erdem A, Arslan M, Oc A, Biberoglu K, Gursoy R. Single-dose methotrexate for \nthe treatment of unruptured ectopic pregnancy. Arch Gynecol Obstet 2003. \n (174)  Cho GJ, Lee SH, Shin JW, Lee NW, Kim T, Kim HJ et al. Predictors of success of repeated \ninjections of single-dose methotrexate regimen for tubal ectopic pregnancy. Journal of Korean Medical Science \n21(1):86-9, 2006. \n (175)  Lipscomb GH, Givens VA, Meyer NL, Bran D. Previous ectopic pregnancy as a predictor of \nfailure of systemic methotrexate therapy. Fertil Steril 2004; 81(5):1221-1224. \n (176)  Bixby S, Tello R, Kuligowska E. Presence of a yolk sac on transvaginal sonography is the \nmost reliable predictor of single-dose methotrexate treatment failure in ectopic pregnancy. Journal of Ultrasound \nin Medicine 24(5):591-8, 2005. \n (177)  Potter MB, Lepine LA, Jamieson DJ. Predictors of success with methotrexate treatment of \ntubal ectopic pregnancy at Grady Memorial Hospital. American Journal of Obstetrics & Gynecology 2003; \n188(5):1192-1194. \n (178)  Isaacs JD, Jr., McGehee RP, Cowan BD. Life-threatening neutropenia following methotrexate \ntreatment of ectopic pregnancy: a report of two cases. Obstet Gynecol 1996; 88(4 Pt 2):694- 696. \n (179)  Straka M, Zeringue E, Goldman M. A rare drug reaction to methotrexate after treatment for \nectopic pregnancy. Obstetrics & Gynecology 103(5 Pt 2):1047-8, 2004. \n (180)  Zullo F, Pellicano M, Di Carlo C, De Stefano R, Mastrantonio P, Nappi C. Late complications \nafter systemic methotrexate treatment of unruptured ectopic pregnancies: a report of three cases. Eur J Obstet \nGynecol Reprod Biol 1996; 70(2):213-214. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n455 \n \n (181)  Kelly H, Harvey D, Moll S. A cautionary tale: fatal outcome of methotrexate therapy given \nfor management of ectopic pregnancy. Obstetrics & Gynecology 107(2 Pt 2):439-41, 2006. \n (182)  Strobelt N, Mariani E, Ferrari L, Trio D, Tiezzi A, Ghidini A. Fertility after ectopic \npregnancy. Effects of surgery and expectant management. Journal of Reproductive Medicine 45(10):803-7, \n2000. \n (183)  Olofsson JI, Poromaa IS, Ottander U, Kjellberg L, Damber MG. Clinical and pregnancy \noutcome following ectopic pregnancy; a prospective study comparing expectancy, surgery and systemic \nmethotrexate treatment. Acta Obstetricia et Gynecologica Scandinavica 2001; 80(8):744- 749. \n (184)  Cohen MA, Sauer MV. Expectant management of ectopic pregnancy. Clinical Obstetrics & \nGynecology 1999; 42(1):48-54. \n (185)  Rantala M, Makinen J. Tubal patency and fertility outcome after expectant management of \nectopic pregnancy. Fertil Steril 1997; 68(6):1043-1046. \n (186)  Banerjee S, Aslam N, Woelfer B, Lawrence A, Elson J, Jurkovic D. Expectant management of \nearly pregnancies of unknown location: a prospective evaluation of methods to predict spontaneous resolution of \npregnancy. BJOG: an International Journal of Obstetrics & Gynaecology 2001; 108(2):158- 163. \n (187)  Condous G, Okaro E, Khalid A, Timmerman D, Lu C, Zhou Y et al. The use of a new logistic \nregression model for predicting the outcome of pregnancies of unknown location. Hum Reprod 2004; \n19(8):1900-1910. \n (188)  Dart RG, Mitterando J, Dart LM. Rate of change of serial beta-human chorionic gonadotropin \nvalues as a predictor of ectopic pregnancy in patients with indeterminate transvaginal ultrasound findings. Ann \nEmerg Med 1999; 34(6):703-710. \n (189)  Hahlin M, Thorburn J, Bryman I. The expectant management of early pregnancies of \nuncertain site. Hum Reprod 1995; 10(5):1223-1227. \n (190)  Kooi S, Kock HC, van Etten FH. Tubal rupture despite low and declining serum hCG levels. \nEur J Obstet Gynecol Reprod Biol 1992; 46(1):56-59. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n456 \n \n (191)  Tulandi T, Hemmings R, Khalifa F. Rupture of ectopic pregnancy in women with low and \ndeclining serum beta-human chorionic gonadotropin concentrations. Fertil Steril 1991; 56(4):786-787. \n (192)  Chapron C, Fauconnier A, Goffinet F, Breart G, Dubuisson JB. Laparoscopic surgery is not \ninherently dangerous for patients presenting with benign gynaecologic pathology. Results of a meta-analysis. \nHuman Reproduction 17(5):1334-42, 2002. \n (193)  Lundorff P, Hahlin M, Kallfelt B, Thorburn J, Lindblom B. Adhesion formation after \nlaparoscopic surgery in tubal pregnancy: a randomized trial versus laparotomy. Fertil Steril 1991; 55(5):911-\n915. \n (194)  Lundorff P, Thorburn J, Lindblom B. Fertility outcome after conservative surgical treatment \nof ectopic pregnancy evaluated in a randomized trial. Fertil Steril 1992; 57(5):998-1002. \n (195)  Jansen FW, Kapiteyn K, Trimbos-Kemper T, Hermans J, Trimbos JB. Complications of \nlaparoscopy: a prospective multicentre observational study. Br J Obstet Gynaecol 1997; 104(5):595-600. \n (196)  Garry R. Towards evidence-based laparoscopic entry techniques: clinical problems and \ndilemmas. Gynecological Endoscopy 1999; 8:315-326. \n (197)  Gazvani MR, Baruah DN, Alfirevic Z, Emery SJ. Mifepristone in combination with \nmethotrexate for the medical treatment of tubal pregnancy: a randomized, controlled trial. Human Reproduction \n1998; 13(7):1987-1990. \n (198)  Rozenberg P, Chevret S, Camus E, de Tayrac R, Garbin O, de Poncheville L et al. Medical \ntreatment of ectopic pregnancies: a randomized clinical trial comparing methotrexate-mifepristone and \nmethotrexate-placebo. Human Reproduction 2003; 18(9):1802-1808. \n (199)  Perdu M, Camus E, Rozenberg P, Goffinet F, Chastang C, Philippe HJ et al. Treating ectopic \npregnancy with the combination of mifepristone and methotrexate: a phase II nonrandomized study. American \nJournal of Obstetrics & Gynecology 1998; 179(3:Pt 1):t-3. \n (200)  Lundorff P, Thorburn J, Hahlin M, Kallfelt B, Lindblom B. Laparoscopic surgery in ectopic \npregnancy. A randomized trial versus laparotomy. Acta Obstet Gynecol Scand 1991; 70(4-5):343-348. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n457 \n \n (201)  Murphy AA, Nager CW, Wujek JJ, Kettel LM, Torp VA, Chin HG. Operative laparoscopy \nversus laparotomy for the management of ectopic pregnancy: a prospective trial. Fertil Steril 1992; 57(6):1180-\n1185. \n (202)  Vermesh M, Silva PD, Rosen GF, Stein AL, Fossum GT, Sauer MV. Management of \nunruptured ectopic gestation by linear salpingostomy: a prospective, randomized clinical trial of laparoscopy \nversus laparotomy. Obstet Gynecol 1989; 73(3 Pt 1):400-404. \n (203)  Dias PG, Hajenius PJ, Mol BW, Ankum WM, Hemrika DJ, Bossuyt PM et al. Fertility \noutcome after systemic methotrexate and laparoscopic salpingostomy for tubal pregnancy. Lancet 1999; \n353(9154):724-725. \n (204)  Hajenius PJ, Engelsbel S, Mol BW, Van D, V, Ankum WM, Bossuyt PM et al. Randomised \ntrial of systemic methotrexate versus laparoscopic salpingostomy in tubal pregnancy. Lancet 1997; \n350(9080):774-779. \n (205)  Fernandez H, Yves Vincent SC, Pauthier S, Audibert F, Frydman R. Randomized trial of \nconservative laparoscopic treatment and methotrexate administration in ectopic pregnancy and subsequent \nfertility. Human Reproduction 1998; 13(11):3239-3243. \n (206)  Saraj AJ, Wilcox JG, Najmabadi S, Stein SM, Johnson MB, Paulson RJ. Resolution of \nhormonal markers of ectopic gestation: a randomized trial comparing single-dose intramuscular methotrexate \nwith salpingostomy. Obstet Gynecol 1998; 92(6):989-994. \n (207)  Sowter MC, Farquhar CM, Gudex G. An economic evaluation of single dose systemic \nmethotrexate and laparoscopic surgery for the treatment of unruptured ectopic pregnancy. BJOG: an \nInternational Journal of Obstetrics & Gynaecology 2001; 108(2):204-212. \n (208)  Van Der Voort M, Heijnsdijk EA, Gouma DJ. Bowel injury as a complication of laparoscopy. \nBr J Surg 2004; 91(10):1253-1258. \n (209)  Tarik A, Fehmi C. Complications of gynaecological laparoscopy --a retrospective analysis of \n3572 cases from a single institute. J Obstet Gynaecol 2004; 24(7):813-816. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n458 \n \n (210)  Orlando R, Palatini P, Lirussi F. Needle and trocar injuries in diagnostic laparoscopy under \nlocal anesthesia: what is the true incidence of these complications? J Laparoendosc Adv Surg Tech A 2003; \n13(3):181-184. \n (211)  Roviaro GC, Varoli F, Saguatti L, Vergani C, Maciocco M, Scarduelli A. Major vascular \ninjuries in laparoscopic surgery. Surg Endosc 2002; 16(8):1192-1196. \n (212)  Bhoyrul S, Vierra MA, Nezhat CR, Krummel TM, Way LW. Trocar injuries in laparoscopic \nsurgery. J Am Coll Surg 2001; 192(6):677-683. \n (213)  Munro MG. Laparoscopic access: complications, technologies, and techniques. Curr Opin \nObstet Gynecol 2002; 14(4):365-374. \n (214)  Philips PA, Amaral JF. Abdominal access complications in laparoscopic surgery. J Am Coll \nSurg 2001; 192(4):525-536. \n (215)  Chapron C, Pierre F, Querleu D, Dubuisson JB. Major vascular complications from \ngynecologic laparoscopy. Gynecol Obstet Fertil 2000; 28(12):880-887. \n (216)  Chapron CM, Pierre F, Lacroix S, Querleu D, Lansac J, Dubuisson JB. Major vascular injuries \nduring gynecologic laparoscopy. J Am Coll Surg 1997; 185(5):461-465. \n (217)  Chapron C, Querleu D, Mage G, Madelenat P, Dubuisson JB, Audebert A et al. Complications \nof gynecologic laparoscopy. Multicentric study of 7,604 laparoscopies. J Gynecol Obstet Biol Reprod (Paris) \n1992; 21(2):207-213. \n (218)  Leonard F, Lecuru F, Rizk E, Chasset S, Robin F, Taurelle R. Perioperative morbidity of \ngynecological laparoscopy. A prospective monocenter observational study. Acta Obstet Gynecol Scand 2000; \n79(2):129-134. \n (219)  Leng J, Lang J, Huang R, Liu Z, Sun D. Complications in laparoscopic gynecologic surgery. \nChin Med Sci J 2000; 15(4):222-226. \n (220)  Mac CC, Lecuru F, Rizk E, Robin F, Boucaya V, Taurelle R. Morbidity in laparoscopic \ngynecological surgery: results of a prospective single-center study. Surg Endosc 1999; 13(1):57- 61. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n459 \n \n (221)  Marret H, Harchaoui Y, Chapron C, Lansac J, Pierre F. Trocar injuries during laparoscopic \ngynaecological surgery. Report from the French Society of Gynaecological Laparoscopy. Gynacological \nEndoscopy 1998; 7(5):235-241. \n (222)  Rosen DM, Lam AM, Chapman M, Carlton M, Cario GM. Methods of creating \npneumoperitoneum: a review of techniques and complications. Obstet Gynecol Surv 1998; 53(3):167-174. \n (223)  Nezhat C, Childers J, Nezhat F, Nezhat CH, Seidman DS. Major retroperitoneal vascular \ninjury during laparoscopic surgery. Hum Reprod 1997; 12(3):480-483. \n (224)  Soderstrom RM. Injuries to major blood vessels during endoscopy. J Am Assoc Gynecol \nLaparosc 1997; 4(3):395-398. \n (225)  Bateman BG, Kolp LA, Hoeger K. Complications of laparoscopy --operative and diagnostic. \nFertil Steril 1996; 66(1):30-35. \n (226)  Champault G, Cazacu F. Laparoscopic surgery: injuries caused by trocars. (French Survey \n1994) in reference to 103,852 interventions. J Chir (Paris) 1995; 132(3):1 09-113. \n (227)  Saville LE, Woods MS. Laparoscopy and major retroperitoneal vascular injuries (MRVI). \nSurg Endosc 1995; 9(10):1096-1100. \n (228)  Nordestgaard AG, Bodily KC, Osborne RW, Jr., Buttorff JD. Major vascular injuries during \nlaparoscopic procedures. Am J Surg 1995; 169(5):543-545. \n (229)  Hanney RM, Alle KM, Cregan PC. Major vascular injury and laparoscopy. Aust N Z J Surg \n1995; 65(7):533-535. \n (230)  Geers J, Holden C. Major vascular injury as a complication of laparoscopic surgery: a report \nof three cases and review of the literature. Am Surg 1996; 62(5):377-379. \n (231)  Sokol AI, Chuang K, Milad MP. Risk factors for conversion to laparotomy during \ngynecologic laparoscopy. Journal of the American Association of Gynecologic Laparoscopists 2003; 10(4):469-\n473. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n460 \n \n (232)  Baggish MS. Analysis of 31 Cases of Major-Vessel Injury Associated with Gynecologic \nLaparoscopy Operations. Journal of Gynecologic Surgery 2003; 19(2):63- 73. \n (233)  Fuller J, Ashar BS, Carey-Corrado J. Trocar-associated injuries and fatalities: an analysis of \n1399 reports to the FDA. J Minim Invasive Gynecol 2005; 12(4):302-307. \n (234)  Jansen FW, Kolkman W, Bakkum EA, de Kroon CD, Trimbos-Kemper TC, Trimbos JB. \nComplications of laparoscopy: an inquiry about closed- versus open-entry technique. Am J Obstet Gynecol \n2004; 190(3):634-638. \n (235)  Chapron C, Cravello L, Chopin N, Kreiker G, Blanc B, Dubuisson JB. Complications during \nset-up procedures for laparoscopy in gynecology: open laparoscopy does not reduce the risk of major \ncomplications. Acta Obstetricia et Gynecologica Scandinavica 2003; 82(12):1125-1129. \n (236)  Chapron C, Pierre F, Querleu D, Dubuisson JB. Complications of gynaecological laparoscopy. \nGynecologie Obstetrique Fertilite 2001; 29(9):605-612. \n (237)  Chapron C, Fauconnier A, Goffinet F, Breart G, Dubuisson JB. Laparoscopic surgery is not \ninherently dangerous for patients presenting with benign gynaecologic pathology. Results of a meta-analysis. \nHum Reprod 2002; 17(5):1334-1342. \n (238)  Chapron C, Querleu D, Bruhat MA, Madelenat P, Fernandez H, Pierre F et al. Surgical \ncomplications of diagnostic and operative gynaecological laparoscopy: a series of 29,966 cases. Hum Reprod \n1998; 13(4):867-872. \n (239)  Merlin TL, Hiller JE, Maddern GJ, Jamieson GG, Brown AR, Kolbe A. Systematic review of \nthe safety and effectiveness of methods used to establish pneumoperitoneum in laparoscopic surgery. Br J Surg \n2003; 90(6):668-679. \n (240)  Catarci M, Carlini M, Gentileschi P, Santoro E. Major and minor injuries during the creation \nof pneumoperitoneum. A multicenter study on 12,919 cases. Surg Endosc 2001; 15(6):566- 569. \n (241)  Schafer M, Lauper M, Krahenbuhl L. Trocar and Veress needle injuries during laparoscopy. \nSurg Endosc 2001; 15(3):275-280. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n461 \n \n (242)  Wang PH, Lee WL, Yuan CC, Chao HT, Liu WM, Yu KJ et al. Major complications of \noperative and diagnostic laparoscopy for gynecologic disease. Journal of the American Association of \nGynecologic Laparoscopists 2001; 8(1):68-73. \n (243)  Harkki-Siren P, Sjoberg J, Kurki T. Major complications of laparoscopy: a follow-up Finnish \nstudy. Obstetrics & Gynecology 1999; 94(1):94-98. \n (244)  Harkki-Siren P, Kurki T. A nationwide analysis of laparoscopic complications. Obstetrics & \nGynecology 1997; 89(1):108-112. \n (245)  Jansen FW, Kapiteyn K, Trimbos-Kemper T, Hermans J, Trimbos JB. Complications of \nlaparoscopy: a prospective multicentre observational study. Br J Obstet Gynaecol 1997; 104(5):595-600. \n (246)  Brosens I, Gordon A, Campo R, Gordts S. Bowel injury in gynecologic laparoscopy. J Am \nAssoc Gynecol Laparosc 2003; 10(1):9-13. \n (247)  Chandler JG, Corson SL, Way LW. Three spectra of laparoscopic entry access injuries. J Am \nColl Surg 2001; 192(4):478-490. \n (248)  Corson SL, Chandler JG, Way LW. Survey of laparoscopic entry injuries provoking litigation. \nJ Am Assoc Gynecol Laparosc 2001; 8(3):341-347. \n (249)  El Banna M, Abdel-Atty M, El Meteini M, Aly S. Management of laparoscopic-related bowel \ninjuries. Surg Endosc 2000; 14(9):779-782. \n (250)  Chapron C, Pierre F, Harchaoui Y, Lacroix S, Beguin S, Querleu D et al. Gastrointestinal \ninjuries during gynaecological laparoscopy. Hum Reprod 1999; 14(2):333- 337. \n (251)  Schrenk P, Woisetschlager R, Rieger R, Wayand W. Mechanism, management, and \nprevention of laparoscopic bowel injuries. Gastrointest Endosc 1996; 43(6):572- 574. \n (252)  Bishoff JT, Allaf ME, Kirkels W, Moore RG, Kavoussi LR, Schroder F. Laparoscopic bowel \ninjury: incidence and clinical presentation. J Urol 1999; 161(3):887-890. \n (253)  Ferriman A. Laparoscopic surgery: two thirds of injuries initially missed. West J Med 2000; \n173(6):372. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n462 \n \n (254)  Gordts S, Watrelot A, Campo R, Brosens I. Risk and outcome of bowel injury during \ntransvaginal pelvic endoscopy. Fertil Steril 2001; 76(6):1238-1241. \n (255)  Gett RM, Joseph MG. A safe technique for the insertion of the Hasson cannula. ANZ J Surg \n2004; 74(9):797-798. \n (256)  Hasson HM, Rotman C, Rana N, Kumari NA. Open laparoscopy: 29-year experience. Obstet \nGynecol 2000; 96(5 Pt 1):763-766. \n (257)  Hasson HM. Open laparoscopy as a method of access in laparoscopic surgery. Gynacological \nEndoscopy 1999; 8(6):353-362. \n (258)  Gunenc MZ, Yesildaglar N, Bingol B, Onalan G, Tabak S, Gokmen B. The safety and \nefficacy of direct trocar insertion with elevation of the rectus sheath instead of the skin for pneumoperitoneum. \nSurg Laparosc Endosc Percutan Tech 2005; 15(2):80-81. \n \n (259)  Agresta F, De Simone P, Ciardo LF, Bedin N. Direct trocar insertion vs Veress needle in \nnonobese patients undergoing laparoscopic procedures: a randomized prospective single-center study. Surg \nEndosc 2004; 18(12):1778-1781. \n (260)  Rahman MM, Mamun AA. Direct trocar insertion: alternative abdominal entry technique for \nlaparoscopic surgery. Mymensingh Med J 2003; 12(1):45-47. \n (261)  Kaloo P, Cooper M, Reid G. A prospective multicentre study of laparoscopic complications \nrelated to the direct-entry technique. Gynaecological Endoscopy 2002; 11(2):67-70. \n (262)  Nezhat FR, Silfen SL, Evans D, Nezhat C. Comparison of direct insertion of disposable and \nstandard reusable laparoscopic trocars and previous pneumoperitoneum with Veress needle. Obstet Gynecol \n1991; 78(1):148-150. \n (263)  Kaali SG, Barad DH. Incidence of bowel injury due to dense adhesions at the sight of direct \ntrocar insertion. J Reprod Med 1992; 37(7):617-618. \n (264)  Byron JW, Markenson G, Miyazawa K. A randomized comparison of Verres needle and direct \ntrocar insertion for laparoscopy. Surg Gynecol Obstet 1993; 177(3):259- 262. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n463 \n \n (265)  Hill DJ, Maher PJ. Direct cannula entry for laparoscopy. J Am Assoc Gynecol Laparosc 1996; \n4(1):77-79. \n (266)  Jacobson MT, Osias J, Bizhang R, Tsang M, Lata S, Helmy M et al. The direct trocar \ntechnique: an alternative approach to abdominal entry for laparoscopy. Journal of the Society of \nLaparoendoscopic Surgeons 6(2):169-74, 2002;-Jun. \n (267)  Schoonderwoerd L, Swank DJ. The role of optical access trocars in laparoscopic surgery. Surg \nTechnol Int 2005; 14:61-67. \n (268)  Jirecek S, Drager M, Leitich H, Nagele F, Wenzl R. Direct visual or blind insertion of the \nprimary trocar. Surg Endosc 2002; 16(4):626-629. \n (269)  Kaali SG. Complications associated with optical-access laparoscopic trocars. Obstet Gynecol \n2002; 100(3):614. \n (270)  Lombezzi R, Galleano R, Lucarini L, Falchero F. New technique for optical control of the first \ntrocar insertion. Minerva Chirurgica 57(4):527-9, 2002. \n (271)  Sharp HT, Dodson MK, Draper ML, Watts DA, Doucette RC, Hurd WW. Complications \nassociated with optical-access laparoscopic trocars. Obstet Gynecol 2002; 99(4):553- 555. \n (272)  String A, Berber E, Foroutani A, Macho JR, Pearl JM, Siperstein AE. Use of the optical \naccess trocar for safe and rapid entry in various laparoscopic procedures. Surg Endosc 2001; 15(6):570- 573. \n (273)  Marcovich R, Del Terzo MA, Wolf JS, Jr. Comparison of transperitoneal laparoscopic access \ntechniques: Optiview visualizing trocar and Veress needle. J Endourol 2000; 14(2):175-179. \n (274)  Hallfeldt KK, Trupka A, Kalteis T, Stuetzle H. Safe creation of pneumoperitoneum using an \noptical trocar. Surg Endosc 1999; 13(3):306-307. \n (275)  Mettler L, Schmidt EH, Frank V, Semm K. Optical trocar systems: Laparoscopic entry and its \ncomplications (a study of cases in Germany). Gynaecological Endoscopy 1999; 8(6):383-389. \n (276)  Kaali SG, Merkatz IR. Clinical experience with an optical access trocar in gynecological \nlaparoscopy-pelviscopy. JSLS 1998; 2(3):315. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n464 \n \n (277)  Kaali SG. Introduction of the Opti-trocar. J Am Assoc Gynecol Laparosc 1993; 1(1):50-53. \n (278)  Rubenstein JN, Blunt LW, Jr., Lin WW, User HM, Nadler RB, Gonzalez CM. Safety and \nefficacy of 12-mm radial dilating ports for laparoscopic access. BJU International 92(3):327-9, 2003. \n (279)  Bhoyrul S, Payne J, Steffes B, Swanstrom L, Way LW. A randomized prospective study of \nradially expanding trocars in laparoscopic surgery. J Gastrointest Surg 2000; 4(4):392-397. \n (280)  Galen DI, Jacobson A, Weckstein LN, Kaplan RA, DeNevi KL. Reduction of cannula-related \nlaparoscopic complications using a radially expanding access device. J Am Assoc Gynecol Laparosc 1999; \n6(1):79-84. \n (281)  Ternamian AM. Laparoscopy without trocars. Surg Endosc 1997; 11(8):815- 818. \n (282)  Ternamian AM, Deitel M. Endoscopic threaded imaging port (EndoTIP) for laparoscopy: \nexperience with different body weights. Obes Surg 1999; 9(1):44-47. \n (283)  Yim SF, Yuen PM. Randomized double-masked comparison of radially expanding access \ndevice and conventional cutting tip trocar in laparoscopy. Obstet Gynecol 2001; 97(3):435- 438. \n (284)  Molloy D, Kaloo PD, Cooper M, Nguyen TV. Laparoscopic entry: a literature review and \nanalysis of techniques and complications of primary port entry. Aust N Z J Obstet Gynaecol 2002; 42(3):246-\n254. \n (285)  Vilos GA, Ternamian A, Dempster J, Laberge PY. Laparoscopic entry: a review of \ntechniques, technologies, and complications. Society of Obstetricians and Gynaecologists of Canada Clinical \nPractice Guideline. J Obstet Gynaecol Can 2007; 29(5):433-447. \n (286)  Larobina M, Nottle P. Complete evidence regarding major vascular injuries during \nlaparoscopic access. Surg Laparosc Endosc Percutan Tech 2005; 15(3):119-123. \n (287)  Moberg AC, Montgomery A. Primary access-related complications with laparoscopy: \ncomparison of blind and open techniques. Surg Endosc 2005; 19(9):1196-1199. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n465 \n \n (288)   A systematic review of the methods used to establish laparoscopic pneumoperitoneum. \nASERNIP-S Report No. 13. Adelaide, South Australia: ASERNIP-S. October \n2001.http://www.surgeons.org/asernip-s . 2001. \n (289)  Pasic RP, Kantardzic M, Templeman C, Levine RL. Insufflation techniques in gynecologic \nlaparoscopy. Surg Laparosc Endosc Percutan Tech 2006; 16(1):18-23. \n (290)  Hender K. What is the safety of open (Hasson) technique versus closed (blind Veress needle) \ntechnique for laparoscopy? Centre for Clinical Effectiveness - Evidence Report.  2001.  Centre for Clinical \nEffectiveness (CCE),Clayton, Victoria.  \n (291)  Woolcot R. The efficacy and safety of different techniques for trocar insertion in laparoscopic \nsurgery. Minimally Invasive Therapy and Allied Technologies 2001; 10(1):11-14. \n (292)  Bemelman WA, Dunker MS, Busch OR, Den Boer KT, de Wit LT, Gouma DJ. Efficacy of \nestablishment of pneumoperitoneum with the Veress needle, Hasson trocar, and modified blunt trocar \n(TrocDoc): a randomized study. J Laparoendosc Adv Surg Tech A 2000; 10(6):325-330. \n (293)  Bonjer HJ, Hazebroek EJ, Kazemier G, Giuffrida MC, Meijer WS, Lange JF. Open versus \nclosed establishment of pneumoperitoneum in laparoscopic surgery. Br J Surg 1997; 84(5):599- 602. \n (294)  McKernan JB, Champion JK. Access techniques: Veress needle --initial blind trocar insertion \nversus open laparoscopy with the Hasson trocar. Endosc Surg Allied Technol 1995; 3(1):35- 38. \n (295)  Mayol J, Garcia-Aguilar J, Ortiz-Oshiro E, Diego Carmona JA, Fernandez-Represa JA. Risks \nof the minimal access approach for laparoscopic surgery: multivariate analysis of morbidity related to umbilical \ntrocar insertion. World J Surg 1997; 21(5):529-533. \n (296)  Cravello L, Banet J, Agostini A, Bretelle F, Roger V, Blanc B. Open laparoscopy: analysis of \ncomplications due to first trocar insertion. French. Gynecologie, Obstetrique & Fertilite 30(4):286-90, 2002. \n (297)  Ahmad G, Duffy JMN, Watson AJS. Laparoscopic entry techniques and complications. \nInternational Journal of Gynecology & Obstetrics 2007; 99(1):52-55. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n466 \n \n (298)  Chin K, Newton J. Survey of training in minimal access surgery in the West Midlands region \nof the UK. Gynacological Endoscopy 1996; 5(6):329-333. \n (299)  Lalchandani S, Philips K. Laparoscopic entry technique-a survey of practices of consultant \ngynaecologists. Gynecological Surgery 2005; 2(4):245-249. \n (300)  Lingam K, Cole RA. Laparoscopic entry port visited: a survey of practices of consultant \ngynaecologists in Scotland. Gynaecological Endoscopy 2001; 10(5):335-342. \n (301)  Kaloo P, Cooper M, Molloy D. A survey of entry techniques and complications of members \nof the Australian Gynaecological Endoscopy Society. Aust N Z J Obstet Gynaecol 2002; 42(3):264-266. \n (302)  Marret H, Golfier F, Cassignol A, Raudrant D. Methods for laparoscopy: open laparoscopy or \nclosed laparoscopy? Attitude of the French Central University Hospital. Gynecol Obstet Fertil 2001; \n29(10):673-679. \n (303)  Garry R. A consensus document concerning laparoscopic entry techniques: Middlesbrough, \nMarch 19–20 1999. Gynacological Endoscopy 1999;(8):403-406. \n (304)   Preventing Gynaecological Laparoscopic Injury. Guideline No.48. 2007. \n (305)  RANZCOG. Use of the Veress needle to obtain pneumoperitoneum prior to laparoscopy. \nStatement C-Gyn 7. Consensus statement of the Royal Australian & New Zealand College of Obstetricians & \nGynaecologists (RANZCOG) and the Australian Gynaecological Endoscopy Society (AGES).   2006.  Royal \nAustralian and New Zealand College of Obstetricians and Gynaecologists.  \n (306)  Neudecker J, Sauerland S, Neugebauer E, Bergamaschi R, Bonjer HJ, Cuschieri A et al. \n(EAES) The European Association for Endoscopic Surgery clinical practice guideline on the pneumoperitoneum \nfor laparoscopic surgery. Surg Endosc 2002; 16(7):1121-1143. \n (307)  SAGES. Society of American Gastrointestinal Endoscopic Surgeons (SAGES). SAGES \nguidelines for diagnostic laparoscopy. Los Angeles (CA): Society of American Gastrointestinal Endoscopic \nSurgeons (SAGES); 2002 Mar. 5 p. [13 references] www.sages.org.  2002.  Society of American \nGastrointestinal Endoscopic Surgeons (SAGES).  \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n467 \n \n (308)  Pierre F, Chapron C, Deshayes M, Madelenat P, Magnin G, Querleu D. Initial access for \nlaparoscopic gynecologic surgery. French Society of Endoscopic Gynecology, International Society of Pelvic \nSurgery and the National College of French Gynecologists-Obstetricians. J Gynecol Obstet Biol Reprod (Paris) \n2000; 29(1):8-12. \n (309)  Bakkum EA, Timmermans A, Admiraal JF, Brolmann HAM, Jansen FW. Laparoscopic entry \ntechniques: a protocol for daily gynaecological practice in The Netherlands. Gynecological Surgery 2006; \n3(2):84-87. \n (310)  Garry R. Laparoscopic surgery. Best Pract Res Clin Obstet Gynaecol 2006; 20(1):89-104. \n (311)  Vilos GA. The ABCs of a safer laparoscopic entry. J Minim Invasive Gynecol 2006; \n13(3):249-251. \n (312)  Semm K, Semm I. Safe insertion of trocars and the Veress needle using standard equipment \nand the 11 security steps . Gynaecological Endoscopy 1999; 8(6):339-347. \n (313)  Jones KD, Fan A, Sutton C. Safe entry during laparoscopy: a prospective audit in a district \ngeneral hospital. Gynaecological Endoscopy 2002; 11(2):85-89. \n (314)  Richardson RE, Sutton CJG. Complications of first entry: a pospective laparoscopy audit. \nGynacological Endoscopy 1999; 8(6):327-334. \n (315)  Vilos GA, Vilos AG. Safe laparoscopic entry guided by Veress needle CO2 insufflation \npressure. J Am Assoc Gynecol Laparosc 2003; 10(3):415-420. \n (316)  SIGN. Scottish Intercollegiate Guidelines Network . SIGN 50: A guideline developers' \nhandbook. Section 6: Forming guideline recommendations. \nhttp://www.sign.ac.uk/guidelines/fulltext/50/section6.html.  2007.  \n (317)  Hurd WH, Bude RO, DeLancey JO, Gauvin JM, Aisen AM. Abdominal wall characterization \nwith magnetic resonance imaging and computed tomography. The effect of obesity on the laparoscopic \napproach. J Reprod Med 1991; 36(7):473-476. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n468 \n \n (318)  Hurd WW, Bude RO, DeLancey JO, Pearl ML. The relationship of the umbilicus to the aortic \nbifurcation: implications for laparoscopic technique. Obstet Gynecol 1992; 80(1):48- 51. \n (319)  Nezhat F, Brill AI, Nezhat CH, Nezhat A, Seidman DS, Nezhat C. Laparoscopic appraisal of \nthe anatomic relationship of the umbilicus to the aortic bifurcation. J Am Assoc Gynecol Laparosc 1998; \n5(2):135-140. \n (320)  Parker J, Rahimpanah F. The advantages of microlaparoscopic left upper quadrant entry in \nselected patients. Aust N Z J Obstet Gynaecol 2001; 41(3):314-316. \n (321)  Pasic R, Levine RL, Wolf WM, Jr. Laparoscopy in morbidly obese patients. J Am Assoc \nGynecol Laparosc 1999; 6(3):307-312. \n (322)  Pelosi MA, III, Pelosi MA. Alignment of the umbilical axis: an effective maneuver for \nlaparoscopic entry in the obese patient. Obstet Gynecol 1998; 92(5):869-872. \n (323)  Schwartz ML, Drew RL, Andersen JN. Induction of pneumoperitoneum in morbidly obese \npatients. Obes Surg 2003; 13(4):601-604. \n (324)  Agarwala N, Liu CY. Safe entry techniques during laparoscopy: left upper quadrant entry \nusing the ninth intercostal space--a review of 918 procedures. J Minim Invasive Gynecol 2005; 12(1):55-61. \n (325)  Audebert AJ, Gomel V. Role of microlaparoscopy in the diagnosis of peritoneal and visceral \nadhesions and in the prevention of bowel injury associated with blind trocar insertion. Fertil Steril 2000; \n73(3):631-635. \n (326)  Brill AI, Nezhat F, Nezhat CH, Nezhat C. The incidence of adhesions after prior laparotomy: \na laparoscopic appraisal. Obstet Gynecol 1995; 85(2):269-272. \n (327)  Chang FH, Lee CL, Soong YK. Use of Palmer's Point for Insertion of the Operative \nLaparoscope in Patients with Severe Pelvic Adhesions: Experience of Seventeen Cases. J Am Assoc Gynecol \nLaparosc 1994; 1(4, Part 2):S7. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n469 \n \n (328)  Chang FH, Chou HH, Lee CL, Cheng PJ, Wang CW, Soong YK. Extraumbilical insertion of \nthe operative laparoscope in patients with extensive intraabdominal adhesions. J Am Assoc Gynecol Laparosc \n1995; 2(3):335-337. \n (329)  Chi DS, Abu-Rustum NR, Sonoda Y, Awtrey C, Hummer A, Venkatraman ES et al. Ten-year \nexperience with laparoscopy on a gynecologic oncology service: Analysis of risk factors for complications and \nconversion to laparotomy. Am J Obstet Gynecol 2004; 191(4):1138-1145. \n (330)  Childers JM, Brzechffa PR, Surwit EA. Laparoscopy using the left upper quadrant as the \nprimary trocar site. Gynecol Oncol 1993; 50(2):221-225. \n (331)  Durand-Reville M, Guichard-Checchi C, Ejnes L, Boulanger JP, Gilly V, Bongain A et al. \nGynecologic laparoscopy and abdominal scars: what approach for the peritoneal cavity? J Gynecol Obstet Biol \nReprod (Paris) 2003; 32(7):625-633. \n (332)  Gersin KS, Heniford BT, Arca MJ, Ponsky JL. Alternative site entry for laparoscopy in \npatients with previous abdominal surgery. J Laparoendosc Adv Surg Tech A 1998; 8(3):125- 130. \n (333)  Golan A, Sagiv R, Debby A, Glezerman M. The minilaparoscope as a tool for localization and \npreparation for cannula insertion in patients with multiple previous abdominal incisions or umbilical hernia. \nJournal of the American Association of Gynecologic Laparoscopists 2003; 10(1):14- 16. \n (334)  Howard FM, El Minawi AM, DeLoach VE. Direct laparoscopic cannula insertion at the left \nupper quadrant. J Am Assoc Gynecol Laparosc 1997; 4(5):595-600. \n (335)  Jacobson TZ, Davis C. A prospective studiy of Palmer's point entry and the rate of sub-\numbilical adhesions in women undergoing laparoscopy with previous abdominal surgery. Abstract P04. \nReviews in Gynaecological Practice 3[1]. 2003.  \nRef Type: Abstract \n (336)  Kumakiri J, Takeuchi H, Sato Y, Kitade M, Kikuchi I, Shimanuki H et al. A novel method of \nninth-intercostal microlaparoscopic approach for patients with previous laparotomy. Acta Obstetricia et \nGynecologica Scandinavica 85(8):977-81, 2006. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n470 \n \n (337)  Lam KW, Pun TC. Left upper quadrant approach in gynecologic laparoscopic surgery using \nreusable instruments. Journal of the American Association of Gynecologic Laparoscopists 9(2):199-203, 2002. \n (338)  Lecuru F, Leonard F, Philippe JJ, Rizk E, Robin F, Taurelle R. Laparoscopy in patients with \nprior surgery: results of the blind approach. JSLS 2001; 5(1):13-16. \n (339)  Levrant SG, Bieber EJ, Barnes RB. Anterior abdominal wall adhesions after laparotomy or \nlaparoscopy. J Am Assoc Gynecol Laparosc 1997; 4(3):353-356. \n (340)  Parker J, Reid G, Wong F. Microlaparoscopic left upper quadrant entry in patients at high risk \nof periumbilical adhesions. Aust N Z J Obstet Gynaecol 1999; 39(1):88-92. \n (341)  Parker MC, Ellis H, Moran BJ, Thompson JN, Wilson MS, Menzies D et al. Postoperative \nadhesions: ten-year follow-up of 12,584 patients undergoing lower abdominal surgery. Dis Colon Rectum 2001; \n44(6):822-829. \n (342)  Rafii A, Camatte S, Lelievre L, Darai E, Lecuru F. Previous abdominal surgery and closed \nentry for gynecologic laparoscopy: a prospective study. Obstet Gynecol Surv 2005; 60(4):229- 230. \n (343)  Sepilian V, Ku L, Wong H, Liu CY, Phelps JY. Prevalence of infraumbilical adhesions in \nwomen with previous laparoscopy. JSLS 2007; 11(1):41-44. \n (344)  Szigetvari I, Feinman M, Barad D, Bartfai G, Kaali SG. Association of previous abdominal \nsurgery and significant adhesions in laparoscopic sterilization patients. J Reprod Med 1989; 34(7):465-466. \n (345)  Kolecki RV, Golub RM, Sigel B, Machi J, Kitamura H, Hosokawa T et al. Accuracy of \nviscera slide detection of abdominal wall adhesions by ultrasound. Surg Endosc 1994; 8(8):871- 874. \n (346)  Tu FF, Lamvu GM, Hartmann KE, Steege JF. Preoperative ultrasound to predict \ninfraumbilical adhesions: a study of diagnostic accuracy. Am J Obstet Gynecol 2005; 192(1):74- 79. \n (347)  Hsu WC, Chang WC, Huang SC, Torng PL, Chang DY, Sheu BC. Visceral sliding technique \nis useful for detecting abdominal adhesion and preventing laparoscopic surgical complications. Gynecol Obstet \nInvest 2006; 62(2):75-78. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n471 \n \n (348)  Kothari SN, Fundell LJ, Lambert PJ, Mathiason MA. Use of transabdominal ultrasound to \nidentify intraabdominal adhesions prior to laparoscopy: a prospective blinded study. American Journal of \nSurgery 2006; 192(6):843-847. \n (349)  Nezhat CH, Adib T. A Diagnostic Study to Predict Subumbilical Adhesions using \nPreoperative Ultrasound for Visceral Slide and a Novel Peri-umbilical Ultrasound-guided Saline Infusion \n(PUGSI) Technique. Abstract 212. Journal of Minimally Invasive Gynecology 2007; 14(6, Supplement 1):S78. \n (350)  Larciprete G, Cirese E, Flora R, Fanning J. 171: Safe Peritoneal Access for Laparoscopy in \nWomen With Previous Abdominal Open Surgery. Ultrasound Preoperative Evaluation of the Subumbilical \nField. Journal of Minimally Invasive Gynecology 2007; 14(6, Supplement 1):S63-S64. \n (351)  Sriprasad S, Yu DF, Muir GH, Poulsen J, Sidhu PS. Positional anatomy of vessels that may be \ndamaged at laparoscopy: new access criteria based on CT and ultrasonography to avoid vascular injury. J \nEndourol 2006; 20(7):498-503. \n (352)  Narendran M, Baggish MS. Mean Distance Between Primary Trocar Insertion Site and Major \nRetroperitoneal Vessels During Routine Laparoscopy. Journal of Gynecologic Surgery 2002; 18(4):121- 127. \n (353)  Roy GM, Bazzurini L, Solima E, Luciano AA. Safe technique for laparoscopic entry into the \nabdominal cavity. J Am Assoc Gynecol Laparosc 2001; 8(4):519-528. \n (354)  Angelini L, Lirici MM, Papaspyropoulos V, Sossi FL. Combination of subcutaneous \nabdominal wall retraction and optical trocar to minimize pneumoperitoneum-related effects and needle and \ntrocar injuries in laparoscopic surgery. Surg Endosc 1997; 11(10):1006-1009. \n (355)  Briel JW, Plaisier PW, Meijer WS, Lange JF. Is it necessary to lift the abdominal wall when \npreparing a pneumoperitoneum? A randomized study. Surg Endosc 2000; 14(9):862-864. \n (356)  Teoh B, Sen R, Abbott J. An evaluation of four tests used to ascertain Veres needle placement \nat closed laparoscopy. J Minim Invasive Gynecol 2005; 12(2):153-158. \n (357)  Vilos AG, Vilos GA, Abu-Rafea B, Hollett-Caines J, Al Omran M. Effect of body habitus and \nparity on the initial Veres intraperitoneal CO2 insufflation pressure during laparoscopic access in women. J \nMinim Invasive Gynecol 2006; 13(2):108-113. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n472 \n \n (358)  Azevedo OC, Azevedo JL, Sorbello AA, Miguel GP, Wilson Junior JL, Godoy AC. \nEvaluation of tests performed to confirm the position of the Veress needle for creation of pneumoperitoneum in \nselected patients: a prospective clinical trial. Acta Cir Bras 2006; 21(6):385-391. \n (359)  Reich H, Ribeiro SC, Rasmussen C, Rosenberg J, Vidali A. High-pressure trocar insertion \ntechnique. JSLS 1999; 3(1):45-48. \n (360)  Reich H, Rasmussen C, Vidali A. Peritoneal hyperdistention for trocar insertion. \nGynacological Endoscopy 1999; 8(6):375-377. \n (361)  Tsaltas J, Pearce S, Lawrence A, Meads A, Mezzatesta J, Nicolson S. Safer laparoscopic \ntrocar entry: It's all about pressure. Aust N Z J Obstet Gynaecol 2004; 44(4):349- 350. \n (362)  Abu-Rafea B, Vilos GA, Vilos AG, Ahmad R, Hollett-Caines J, Al Omran M. High-pressure \nlaparoscopic entry does not adversely affect cardiopulmonary function in healthy women. J Minim Invasive \nGynecol 2005; 12(6):475-479. \n (363)  Abu-Rafea B, Vilos GA, Vilos AG, Hollett-Caines J, Al Omran M. Effect of body habitus and \nparity on insufflated CO2 volume at various intraabdominal pressures during laparoscopic access in women. J \nMinim Invasive Gynecol 2006; 13(3):205-210. \n (364)  Epstein J, Arora A, Ellis H. Surface anatomy of the inferior epigastric artery in relation to \nlaparoscopic injury. Clin Anat 2004; 17(5):400-408. \n (365)  Saber AA, Meslemani AM, Davis R, Pimentel R. Safety zones for anterior abdominal wall \nentry during laparoscopy: a CT scan mapping of epigastric vessels. Ann Surg 2004; 239(2):182-185. \n (366)  Nezhat CH, Nezhat F, Brill AI, Nezhat C. Normal variations of abdominal and pelvic anatomy \nevaluated at laparoscopy. Obstet Gynecol 1999; 94(2):238-242. \n (367)  Hurd WW, Bude RO, DeLancey JO, Newman JS. The location of abdominal wall bl ood \nvessels in relationship to abdominal landmarks apparent at laparoscopy. Am J Obstet Gynecol 1994; \n171(3):642-646. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n473 \n \n (368)  Hurd WW, Amesse LS, Gruber JS, Horowitz GM, Cha GM, Hurteau JA. Visualization of the \nepigastric vessels and bladder before laparoscopic trocar placement. Fertility & Sterility 80(1):209-12, 2003. \n (369)  Wind J, Cremers JE, Berge Henegouwen MI, Gouma DJ, Jansen FW, Bemelman WA. \nMedical liability insurance claims on entry-related complications in laparoscopy. Surg Endosc 2007. \n (370)  Vilos GA. Laparoscopic bowel injuries: forty litigated gynaecological cases in Canada. \nJournal of Obstetrics & Gynaecology Canada: JOGC 2002; 24(3):224-230. \n (371)  Vilos GA. Litigation of laparoscopic major vessel injuries in Canada. J Am Assoc Gynecol \nLaparosc 2000; 7(4):503-509. \n (372)  Soderstrom RM. Bowel injury litigation after laparoscopy. J Am Assoc Gynecol Laparosc \n1993; 1(1):74-77. \n (373)  Sutton CJ. Medico-legal implications of keyhole surgery. Medico-Legal Journal 64 ( Pt \n3):101-13, 1996. \n (374 )  Rein H. Complications and litigation in gynecologic endoscopy. Curr Opin Obstet Gynecol \n2001; 13(4):425-429. \n (375)  Ellis H. Medicolegal consequences of postoperative intra-abdominal adhesions. J R Soc Med \n2001; 94(7):331-332. \n (376)  Jansen FW, Wind J, Cremeres JEL, Bemelman WA. 146: Entry Related Complications in \nLaparoscopy and Their Medical Liability Insurance. Journal of Minimally Invasive Gynecology 2007; 14(6, \nSupplement 1):S54-S55. \n (377)  Driscoll V. Bowel injury during laparoscopic sterilization - Vanessa Palmer v Cardiff & Vale \nNHS Trust. The AvMA Medical & Legal Journal 2004; 10(3):109-111. \n (378)  Hart R, Doherty DA, Karthigasu K, Garry R. The value of virtual reality-simulator training in \nthe development of laparoscopic surgical skills. J Minim Invasive Gynecol 2006; 13(2):126- 133. \n (379)  Kolkman W, Wolterbeek R, Jansen FW. Gynecological laparoscopy in residency training \nprogram: Dutch perspectives. Surg Endosc 2005; 19(11):1498-1502. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n474 \n \n (380)  RCOG. Royal College of Obstetricians and Gynaecologists. Male and female sterilisation. \nNational Evidence-Based Clinical Guideline Number 4.  2004.  RCOG Press, London.  \n (381)  NICE. Hysteroscopic sterilisation by tubal cannulation and placement of intrafallopian \nimplants. Interventional Procedure Guidance 44.http://www.nice.org.uk/page.aspx?o=104525.  2004.  \nRef Type: Report \n (382)  Karthigasu KA, Garry R, Hart R. Case report of failed tubal occlusion using Essure pbc \n(permanent birth control) hysteroscopic sterilisation procedure. Aust N Z J Obstet Gynaecol 2006; 46(4):365-\n367. \n (383)  Thoma V, Chua I, Garbin O, Hummel M, Wattiez A. Tubal perforation by ESSURE \nmicroinsert. J Minim Invasive Gynecol 2006; 13(2):161-163. \n (384)  Ubeda A, Labastida R, Dexeus S. Essure: a new device for hysteroscopic tubal sterilization in \nan outpatient setting. Fertil Steril 2004; 82(1):196-199. \n (385)  Kerin JF, Carignan CS, Cher D. The safety and effectiveness of a new hysteroscopic method \nfor permanent birth control: results of the first Essure pbc clinical study. Aust N Z J Obstet Gynaecol 2001; \n41(4):364-370. \n (386)  Cooper JM, Carignan CS, Cher D, Kerin JF. Microinsert nonincisional hysteroscopic \nsterilization. Obstet Gynecol 2003; 102(1):59-67. \n (387)  Duffy S, Marsh F, Rogerson L, Hudson H, Cooper K, Jack S et al. Female sterilisation: a \ncohort controlled comparative study of ESSURE versus laparoscopic sterilisation. BJOG 2005; 112(11):1522-\n1528. \n (388)  Peterson HB, Xia Z, Hughes JM, Wilcox LS, Tylor LR, Trussell J. The risk of pregnancy after \ntubal sterilization: findings from the U.S. Collaborative Review of Sterilization. Am J Obstet Gynecol 1996; \n174(4):1161-1168. \n (389)  Trussell J, Guilbert E, Hedley A. Sterilization failure, sterilization reversal, and pregnancy \nafter sterilization reversal in Quebec. Obstet Gynecol 2003; 101(4):677-684. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n475 \n \n (390)  Kovacs GT, Krins AJ. Female sterilisations with Filshie clips: what is the risk failure? A \nretrospective survey of 30,000 applications. J Fam Plann Reprod Health Care 2002; 28(1):34- 35. \n (391)  Peterson HB, Xia Z, Hughes JM, Wilcox LS, Tylor LR, Trussell J. The risk of ectopic \npregnancy after tubal sterilization. U.S. Collaborative Review of Sterilization Working Group. N Engl J Med \n1997; 336(11):762-767. \n (392)  Penfield AJ. The Filshie clip for female sterilization: a review of world experience. Am J \nObstet Gynecol 2000; 182(3):485-489. \n (393)  Stovall TG, Ling FW, O'Kelley KR, Coleman SA. Gross and histologic examination of tubal \nligation failures in a residency training program. Obstet Gynecol 1990; 76(3 Pt 1):461-465. \n (394)  Stovall TG, Ling FW, Henry GM, Ryan GM, Jr. Method failures of laparoscopic tubal \nsterilization in a residency training program. A comparison of the tubal ring and spring-loaded clip. J Reprod \nMed 1991; 36(4):283-286. \n (395)  Varma R, Gupta JK. Predicting negligence in female sterilization failure using time interval to \nsterilization failure: analysis of 131 cases. Hum Reprod 2007; 22(9):2437-2443. \n (396)  Rock JA, Parmley TH, King TM, Laufe LE, Su BS. Endometriosis and the development of \ntuboperitoneal fistulas after tubal ligation. Fertil Steril 1981; 35(1):16-20. \n (397)  McCausland A. Endosalpingosis (\"endosalpingoblastosis\") following laparoscopic tubal \ncoagulation as an etiologic factor of ectopic pregnancy. Am J Obstet Gynecol 1982; 143(1):12- 24. \n (398)  Stock RJ, Nelson KJ. Ectopic pregnancy subsequent to sterilization: histologic evaluation and \nclinical implications. Fertil Steril 1984; 42(2):211-215. \n (399)  Badawy S, Gilman T, Mroziewicz E. The role of recanalization in tubal pregnancy afte r \nsterilization. Int Surg 1979; 64(5):49-51. \n (400)  Grunert GM. Late tubal patency following tubal ligation. Fertil Steril 1981; 35(4):406-408. \n (401)  Hernandez FJ. Tubal ligation and pregnancy: mechanism of recanalization after tubal ligation. \nFertil Steril 1975; 26(5):392-396. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n476 \n \n (402)  Makar AP, Vanderheyden JS, Schatteman EA, Albertyn GP, Verkinderen JJ, Van Marck EA. \nFemale sterilization failure after bipolar electrocoagulation: a 6 year retrospective study. Eur J Obstet Gynecol \nReprod Biol 1990; 37(3):237-246. \n (403)  McCausland AM. Recanalization and fistulization of the fallopian tubes are thought to be the \ncauses of pregnancies following female sterilization. Am J Obstet Gynecol 1981; 139(1):114- 115. \n (404)  Argent V. Failed sterilization and the law. Br J Obstet Gynaecol 1988; 95(2):113- 115. \n (405)  Cattanach v Melchior [2003] HCA 38. The costs of raising a child: Cattanach v Melchior and \nthe Justice and other legislation amendement bill 2003 (Qld). Cattanach v Melchior [2003] HCA 38 (16 July \n2003).  2003.  \n (406)  Nardin JM, Kulier R, Boulvain M. Techniques for the interruption of tubal patency for female \nsterilisation. Cochrane Database Syst Rev 2006;(1):CD003034. \n (407)  RCOG. Clinical Governance Advice No 1.Guidance for the Development of RCOG Green-top \nGuidelines.http://www.rcog.org.uk/index.asp?PageID=480.  2000.  Royal College of Obstetricians and \nGynaecologists, London, UK.  \n (408)  Chi I, Mumford SD, Laufe LE. Technical failures in tubal ring sterilization: Incidence, \nperceived reasons, outcome, and risk factors. Am J Obstet Gynecol 1980; 138(3):307- 312. \n (409)  Chi IC, Potts M, Wilkens L. Rare events associated with tubal sterilizations: an international \nexperience. Obstet Gynecol Surv 1986; 41(1):7-19. \n (410)  Feldblum PJ, Champion CB, Chi IC, Lamptey P. Technical failures in female sterilization \nusing the tubal ring: a case-control analysis. Contraception 1986; 34(5):505-512. \n (411)  Maker AP, Keersmaekers GH, Vanderheyden JS, Hansch C. Development of \nendosalpingoblastosis and tuboperitoneal fistulas following tubal sterilization: relation with uterine \nadenomyosis. Eur J Obstet Gynecol Reprod Biol 1993; 52(3):187-191. \n (412)  Pati S, Cullins V. Female sterilization. Obstet Gynecol Clin North Am 2000; 27(4):859- 899. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n477 \n \n (413)  Levgur M, Duvivier R. Pelvic inflammatory disease after tubal sterilization: a review. Obstet \nGynecol Surv 2000; 55(1):41-50. \n (414)  Lipscomb GH, Spellman JR, Ling FW. The effect of same-day pregnancy testing on the \nincidence of luteal phase pregnancy. Obstet Gynecol 1993; 82(3):411-413. \n (415)  Birdsall MA, Pattison NS, Wilson P. Female sterilisation: National Women's Hospital 1988-9. \nN Z Med J 1994; 107(990):473-475. \n (416)  Chi IC, Siemens AJ, Champion CB, Gates D, Cilenti D. Pregnancy following minilaparotomy \ntubal sterilization--an update of an international data set. Contraception 1987; 35(2):171-178. \n (417)  Hillis SD, Marchbanks PA, Tylor LR, Peterson HB. Poststerilization regret: findings from the \nUnited States Collaborative Review of Sterilization. Obstet Gynecol 1999; 93(6):889-895. \n (418)  Kulier R, Boulvain M, Walker D, de Candolle G, Campana A. Minilaparotomy and \nendoscopic techniques for tubal sterilisation. Cochrane Database Syst Rev 2003;(1). \n (419)  Huber AW, Mueller MD, Ghezzi F, Cromi A, Dreher E, Raio L. Tubal sterilization: \nComplications of laparoscopy and minilaparotomy. Eur J Obstet Gynecol Reprod Biol 2006. \n (420)  Yan JS, Hsu J, Yin CS. Comparative study of Filshie clip and Pomeroy method for \npostpartum sterilization. International Journal of Gynaecology & Obstetrics 1990; 33(3):263- 267. \n (421)  Kohaut BA, Musselman BL, Sanchez-Ramos L, Kaunitz AM. Randomized trial to compare \nperioperative outcomes of Filshie clip vs. Pomeroy technique for postpartum and intraoperative cesarean tubal \nsterilization: a pilot study. Contraception 2004; 69(4):267-270. \n (422)  Dominik R, Gates D, Sokal D, Cordero M, Lasso dl, V, Remes RA et al. Two randomized \ncontrolled trials comparing the Hulka and Filshie Clips for tubal sterilization. Contraception 2000; 62(4):169-\n175. \n (423)  Sokal D, Gates D, Amatya R, Dominik R. Two randomized controlled trials comparing the \ntubal ring and filshie clip for tubal sterilization. Fertil Steril 2000; 74(3):525-533. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n478 \n \n (424)  Filshie GM, Helson K, Teper S. Day case sterilization with the Filshie Clip in Nottingham. \n10-year follow up study: the first 200 cases. In: Kruger T, Gome V, Van der Wat J, editors. 7th Annual Meeting \nof the International Society for Gynecologic Endoscopy. Bologna: Monduzzi Editore International Proceedings \nDivision; 1998. p.145-58.  1998.  \n (425)  Filshie GM. Laparoscopic sterilization. Semin Laparosc Surg 1999; 6(2):112- 117. \n (426)  Lammes FB. Spontaneous opening of the Filshie clip as a cause of sterilisation failure. BJOG \n2001; 108(6):657-658. \n (427)  Jones KP. Failed laparoscopic sterilisation due to interlaced Hulka-Clemens clips. \nContraception 1987; 36(3):317-320. \n (428)  Soderstrom RM. Sterilization failures and their causes. Am J Obstet Gynecol 1985; \n152(4):395-403. \n (429)  Chi IC. Use of multiple clips for tubal occlusion in interval laparoscopic sterilization: \ncircumstances and consequences. Contraception 1994; 50(5):409-416. \n (430)  Hammerstein J. Legal liability in failed sterilization from the physician's viewpoint. Z Arztl \nFortbild (Jena) 1995; 89(6):678-681. \n (431)  Newton J. Failure of female sterilisation. Poster presentation no. 356. BJOG: an International \nJournal of Obstetrics & Gynaecology 1998; 105(Suppl 17):113-114. \n (432)  Sharma D, Singhal SR, Singhal SK. Uterus didelphys, a rare cause for tubal sterilization \nfailure. Aust N Z J Obstet Gynaecol 1998; 38(3):327-328. \n (433)  Roy KK, Banerjee N, Takkar D. Pregnancy following tubal sterilization: an 11-year survey. \nInt J Gynaecol Obstet 2000; 68:53-54. \n (434)  Hughes GJ. Sterilisation failure. BMJ 1977; 2(6098):1337-1339. \n (435)  Heisterberg L, Jessen P, Schroeder E, Wohlk P, Pedersen LM. Comparison of interval and \npostabortal/puerperal laparoscopic sterilization with the tubal ring procedure. Acta Obstet Gynecol Scand 1985; \n64(3):223-225. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n479 \n \n (436)  Cook CL. Evaluation of Falope Ring sterilization by hysterosalpingogram. J Reprod Med \n1982; 27(5):243-248. \n (437)  Hertz JB. Laparoscopic sterilization with the Falope-ring technique. Acta Obstet Gynecol \nScand 1982; 61(1):13-15. \n (438)  Sheikh HH. Hysterosalpingographic follow-up of laparoscopic sterilization. Am J Obstet \nGynecol 1976; 126(2):181-185. \n (439)  Pellicer A, Serra V. Female sterilization using tubal coagulation. Adv Contraceptive Delivery \nSyst 1988; 4(4):349-367. \n (440)  Klumper F, Peters AA. Migrating clips; a complication following sterilization. Nederlands \nTijdschrift voor Geneeskunde 1991; 135(6):233-235. \n (441)  Amu O, Husemeyer RP. Migration of sterilisation clips: case report and review. Br J Fam Plan \n1999; 25:27-28. \n (442)  Connolly D, McGookin RR, Wali J, Kernohan RM. Migration of Filshie clips --report of two \ncases and review of the literature. Ulster Med J 2005; 74(2):126-128. \n (443)  Hasan A, Evgenikos N, Daniel T, Gatongi D. Filshie clip migration with recurrent perianal \nsepsis and low fistula in ano formation. BJOG 2005; 112(11):1581. \n (444)  Pandit M. Early extrusion of bilateral Filshie clips after laparoscopic sterilisation. BJOG 2005; \n112(5):680. \n (445)  Tan BL, Chong C, Tay EH. Migrating Filshie clip. Aust N Z J Obstet Gynaecol 2004; \n44(6):583-584. \n (446)  Miliauskas JR. Migration of a Filshie clip into the urinary bladder with abscess formation. \nPathology 2003; 35(4):356-357. \n (447)  Lok IH, Lo KW, Ng JS, Tsui MH, Yip SK. Spontaneous expulsion of a Filshie clip through \nthe anterior abdominal wall. Gynecol Obstet Invest 2003; 55(3):183-185. \n\nREFERENCES   Chapter 4. Clinical Guideline Development \n480 \n \n (448)  Buckett W, Carlin A, Kingsland C. Prolapse of Filshie clips following vaginal hysterectomy. \nActa Obstet Gynecol Scand 1998; 77(4):471-472. \n (449)  Kesby GJ, Korda AR. Migration of a Filshie clip into the urinary bladder seven years after \nlaparoscopic sterilisation. Br J Obstet Gynaecol 1997; 104(3):379-382. \n (450)  Anderson J, Gunn E, Hunter M, Owen P. Documentation of preoperative counselling for \nfemale sterilisation: a complete audit cycle. J Fam Plann Reprod Health Care 2005; 31(1):24- 25 \n.\n\nREFERENCES   Chapter 5. Thesis Conclusion \n481 \n \nChapter 5  \n \n(1)  Hill AB. The environment and disease:association or causation. Proc R Soc Med 1965; \n58:295-300. \n (2)   Genome wide SNP 100K analysis of endometriosis: demonstration of genomic imbalance. \n6th July 2007, EXCEL Conference centre, London. Oral Presentation. British Congress of Obstetrics and \nGynaecology (BCOG); Royal College of Obstetricians and Gynaecologists; 2007. \n (3)  Prowse AH, Manek S, Varma R, Liu J, Godwin AK, Maher ER et al. Molecular genetic \nevidence that endometriosis is a precursor of ovarian cancer. Int J Cancer 2006; 119(3):556- 562. \n (4)  Varma R, Rollason T, Gupta JK, Maher ER. Endometriosis and the neoplastic process. \nReproduction 2004; 127(3):293-304. \n (5)   Genetic evidence for malignant transformation of endometriosis. Oral Presentation. 23rd \nAnnual Meeting of European Society of Human Reproduction and Embryology (ESHRE), Tête d'Or , Lyon \nCongress Centre, France; 1st July 2007. 2007. \n (6)  Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100(1):57- 70. \n (7)  Gogusev J, Bouquet de JJ, Telvi L, Doussau M, du MS, Stojkoski A et al. Genetic \nabnormalities detected by comparative genomic hybridization in a human endometriosis-derived cell line. Mol \nHum Reprod 2000; 6(9):821-827. \n (8)  Guo SW, Wu Y, Strawn E, Basir Z, Wang Y, Halverson G et al. Genomic alterations in the \nendometrium may be a proximate cause for endometriosis. Eur J Obstet Gynecol Reprod Biol 2004; 116(1):89-\n99. \n (9)  Wu Y, Strawn E, Basir Z, Wang Y, Halverson G, Jailwala P et al. Genomic alterations in \nectopic and eutopic endometria of women with endometriosis. Gynecol Obstet Invest 2006; 62(3):148-159. \n (10)  Gaetje R, Holtrich U, Engels K, Kourtis K, Cikrit E, Kissler S et al. Expression of membrane-\ntype 5 matrix metalloproteinase in human endometrium and endometriosis. Gynecol Endocrinol 2007; \n23(10):567-573. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n482 \n \n (11)  Burney RO, Talbi S, Hamilton AE, Vo KC, Nyegaard M, Nezhat CR et al. Gene expression \nanalysis of endometrium reveals progesterone resistance and candidate susceptibility genes in women with \nendometriosis. Endocrinology 2007; 148(8):3814-3826. \n (12)  Chand AL, Murray AS, Jones RL, Hannan NJ, Salamonsen LA, Rombauts L. Laser capture \nmicrodissection and cDNA array analysis of endometrium identify CCL16 and CCL21 as epithelial-derived \ninflammatory mediators associated with endometriosis. Reproductive Biology & Endocrinology 2007; 5:18. \n (13)  Eyster KM, Klinkova O, Kennedy V, Hansen KA. Whole genome deoxyribonucleic acid \nmicroarray analysis of gene expression in ectopic versus eutopic endometrium. Fertil Steril 2007; 88(6):1505-\n1533. \n (14)  Huber A, Hudelist G, Knofler M, Saleh L, Huber JC, Singer CF. Effect of highly purified \nhuman chorionic gonadotropin preparations on the gene expression signature of stromal cells derived from \nendometriotic lesions: potential mechanisms for the therapeutic effect of human chorionic gonadotropin in vivo. \nFertil Steril 2007; 88(4 Suppl):1232-1239. \n (15)  Mettler L, Salmassi A, Schollmeyer T, Schmutzler AG, Pungel F, Jonat W. Comparison of c-\nDNA microarray analysis of gene expression between eutopic endometrium and ectopic endometrium \n(endometriosis). Journal of Assisted Reproduction & Genetics 2007; 24(6):249-258. \n (16)  Wren JD, Wu Y, Guo SW. A system-wide analysis of differentially expressed genes in ectopic \nand eutopic endometrium. Human Reproduction 2007; 22(8):2093-2102. \n (17)  Yunus D, Sarkar PK. Audit of documentation of female sterilisation. J Fam Plann Reprod \nHealth Care 2006; 32(1):53-54. \n (18)  Matsuzaki S, Canis M, Pouly JL, Botchorishvili R, Dechelotte PJ, Mage G. Differential \nexpression of genes in eutopic and ectopic endometrium from patients with ovarian endometriosis. Fertil Steril \n2006; 86(3):548-553. \n (19)  Matsuzaki S, Canis M, Vaurs-Barriere C, Boespflug-Tanguy O, Dastugue B, Mage G. DNA \nmicroarray analysis of gene expression in eutopic endometrium from patients with deep endometriosis using \nlaser capture microdissection. Fertility & Sterility 2005; 84:Suppl-90. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n483 \n \n (20)  Matsuzaki S, Canis M, Vaurs-Barriere C, Pouly JL, Boespflug-Tanguy O, Penault-Llorca F et \nal. DNA microarray analysis of gene expression profiles in deep endometriosis using laser capture \nmicrodissection. Molecular Human Reproduction 2004; 10(10):719-728. \n (21)  Smith SK. Study of the profile of gene expression in the endometrium. Journal de \nGynecologie, Obstetrique et Biologie de la Reproduction 2004; 33(6:Pt 2):t- 8. \n (22)  Arimoto T, Katagiri T, Oda K, Tsunoda T, Yasugi T, Osuga Y et al. Genome-wide cDNA \nmicroarray analysis of gene-expression profiles involved in ovarian endometriosis. International Journal of \nOncology 2003; 22(3):551-560. \n (23)  Kao LC, Germeyer A, Tulac S, Lobo S, Yang JP, Taylor RN et al. Expression profiling of \nendometrium from women with endometriosis reveals candidate genes for disease-based implantation failure \nand infertility. Endocrinology 2003; 144(7):2870-2881. \n (24)  Eyster KM, Boles AL, Brannian JD, Hansen KA. DNA microarray analysis of gene \nexpression markers of endometriosis. Fertility & Sterility 2002; 77(1):38-42. \n (25)  Flores I, Rivera E, Mousses S, Chen Y, Rozenblum E. Identification of molecular markers for \nendometriosis in blood lymphocytes by using deoxyribonucleic acid microarrays. Fertil Steril 2006; 85(6):1676-\n1683. \n (26)  Flores I, Rivera E, Ruiz LA, Santiago OI, Vernon MW, Appleyard CB. Molecular profiling of \nexperimental endometriosis identified gene expression patterns in common with human disease. Fertility & \nSterility 2007; 87(5):1180-1199. \n (27)  Konno R, Fujiwara H, Netsu S, Odagiri K, Shimane M, Nomura H et al. Gene expression \nprofiling of the rat endometriosis model. American Journal of Reproductive Immunology 2007; 58(4):330-343. \n (28)  Kamat AA, Younes PS, Sayeeduddin M, Wheeler TM, Simpson JL, Agoulnik AI. Protein \nexpression profiling of endometriosis: validation of 2-mm tissue microarrays. Fertility & Sterility 2004; \n82(6):1681-1683. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n484 \n \n (29)  Kyama CM, T'Jampens D, Mihalyi A, Simsa P, Debrock S, Waelkens E et al. ProteinChip \ntechnology is a useful method in the pathogenesis and diagnosis of endometriosis: a preliminary study. Fertil \nSteril 2006; 86(1):203-209. \n (30)  Wang L, Zheng W, Yu JK, Jiang WZ, Mu L, Zhang SZ. Artificial neural networks combined \nwith surface-enhanced laser desorption/ionization mass spectra distinguish endometriosis from healthy \npopulation. Fertil Steril 2007; 88(6):1700-1702. \n (31)  Hever A, Roth RB, Hevezi PA, Lee J, Willhite D, White EC et al. Molecular characterization \nof human adenomyosis. Molecular Human Reproduction 2006; 12(12):737-748. \n (32)  Hoque MO, Lee CC, Cairns P, Schoenberg M, Sidransky D. Genome-wide genetic \ncharacterization of bladder cancer: a comparison of high-density single-nucleotide polymorphism arrays and \nPCR-based microsatellite analysis. Cancer Research 2003; 63(9):2216-2222. \n (33)  Liu W, Chang B, Sauvageot J, Dimitrov L, Gielzak M, Li T et al. Comprehensive assessment \nof DNA copy number alterations in human prostate cancers using Affymetrix 100K SNP mapping array. Genes, \nChromosomes & Cancer 2006; 45(11):1018-1032. \n (34)  Borthwick JM, Charnock- Jones DS, Tom BD, Hull ML, Teirney R, Phillips SC et al. \nDetermination of the transcript profile of human endometrium. Molecular Human Reproduction 2003; 9(1):19-\n33. \n (35)  Levy M, Mittal K, Chiriboga L, Zhang X, Yee H, Wei JJ. Differential expression of selected \ngene products in uterine leiomyomata and adenomyosis. Fertil Steril 2007; 88(1):220-223. \n (36)  Varma R, Soneja H, Clark TJ, Gupta JK. Hysteroscopic myomectomy for menorrhagia using \nVersascopetrade mark bipolar system: Efficacy and prognostic factors at a minimum of one year follow up. Eur \nJ Obstet Gynecol Reprod Biol 2008. \n (37)  Varma R, Soneja H, Bhatia K, Ganesan R, Rollason T, Clark TJ et al. The effectiveness of a \nlevonorgestrel-releasing intrauterine system (LNG-IUS) in the treatment of endometrial hyperplasia --a long-\nterm follow-up study. Eur J Obstet Gynecol Reprod Biol 2008; 139(2):169-175. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n485 \n \n (38)  Varma R, Soneja H, Samuel N, Sangha E, Clark TJ, Gupta JK. Hospital recovery following \nThermachoice ablation is not dependent on setting (outpatient or daycase) or rescue analgesia: unexpected \nresult. Eur J Obstet Gynecol Reprod Biol 2008; 140(1):76-81. \n (39)  Varma R, Soneja H, Clark TJ, Gupta JK. Outpatient Thermachoice endometrial balloon \nablation: long term, prognostic and quality of life measures. In Submission 2008. Eur J Obstet Gynecol Reprod \nBiol 2008. \n (40)  Varma R, Gupta JK. Predicting negligence in female sterilization failure using time interval to \nsterilization failure: analysis of 131 cases. Hum Reprod 2007; 22(9):2437-2443. \n (41)  RCOG. Development of RCOG Green-top Guidelines: Producing a Clinical Practice \nGuideline. Clinical Governance Advice No. 1c.  2006.  Royal College of Obstetricians and Gynaecologists, \nLondon, UK.  \n (42)  Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S et al. Grading quality of \nevidence and strength of recommendations. BMJ 2004; 328(7454):1490. \n (43)  Varma R, Gupta JK. Antibiotic treatment of bacterial vaginosis in pregnancy: Multiple meta-\nanalyses and dilemmas in interpretation. Eur J Obstet Gynecol Reprod Biol 2006; 124(1):10- 14. \n (44)  Varma R, Gupta JK, James DK, Kilby MD. Do screening-preventative interventions in \nasymptomatic pregnancies reduce the risk of preterm delivery--a critical appraisal of the literature. Eur J Obstet \nGynecol Reprod Biol 2006; 127(2):145-159. \n (45)  Varma R, Sinha D, Gupta JK. Non-contraceptive uses of levonorgestrel-releasing hormone \nsystem (LNG-IUS)--a systematic enquiry and overview. Eur J Obstet Gynecol Reprod Biol 2006; 125(1):9-28. \n (46)  Noorani HZ, Husereau DR, Boudreau R, Skidmore B. Priority setting for health technology \nassessments: a systematic review of current practical approaches. Int J Technol Assess Health Care 2007; \n23(3):310-315. \n (47)  Varma R, Gupta JK, Smith GC. Birth after previous caesarean section. Royal College of \nObstetricians and Gynaecologists Clinical Green top guideline No.45. . \nhttp://www.rcog.org.uk/index.asp?PageID=1913.  2007.  RCOG Press, London.  \n\nREFERENCES   Chapter 5. Thesis Conclusion \n486 \n \n (48)  Varma R, Gupta JK. Ectopic Pregnancy. \nhttp://clinicalevidence.bmj.com/ceweb/conditions/pac/1406/1406_background.jsp. BMJ Clinical Evidence . \n2006.  \n (49)  Varma R, Gupta JK. Laparoscopic entry techniques: clinical guideline, national survey and \nmedicolegal ramifications.  In Press. Surgical Endoscopy . 2008.  \n (50)  Nicholls S, Cullen R, O'Neill S, Halligan A. Clinical governance: its origins and its \nfoundations. Clin Perform Qual Health Care 2000; 8(3):172-178. \n (51)  Varma R, Gupta JK. Failed sterilisation: evidence-based review and medico-legal \nramifications. BJOG 2004; 111(12):1322-1332. \n (52)  SIGN. Scottish Intercollegiate Guidelines Network (SIGN). Forming guideline \nrecommendations. In: A guideline developers' handbook. Edinburgh: SIGN, 2001. (Publication No 50.). www \nsign ac uk/guidelines/fulltext/50/section6 html [ 2001  \n (53)  US Preventive Services Task Force. Guide to clinical preventive services. 2nd ed. Baltimore: \nWilliams and Wilkins, 1996: xxxix-lv.  http://www ahrq gov/clinic/uspstfix htm [ 1996  \n (54)  Guyatt GH, Sackett DL, Sinclair JC, Hayward R, Cook DJ, Cook RJ. Users' guides to the \nmedical literature. IX. A method for grading health care recommendations. Evidence-Based Medicine Working \nGroup. JAMA 1995; 274(22):1800-1804. \n (55)  Grilli R, Magrini N, Penna A, Mura G, Liberati A. Practice guidelines developed by specialty \nsocieties: the need for a critical appraisal. Lancet 2000; 355(9198):103-106. \n (56)  Grol R, Cluzeau FA, Burgers JS. Clinical practice guidelines: towards better quality \nguidelines and increased international collaboration. Br J Cancer 2003; 89 Suppl 1:S4-S8. \n (57)  Graham ID, Calder LA, Hebert PC, Carter AO, Tetroe JM. A comparison of clinical practice \nguideline appraisal instruments. Int J Technol Assess Health Care 2000; 16(4):1024-1038. \n (58)  Cluzeau FA, Littlejohns P, Grimshaw JM, Feder G, Moran SE. Development and application \nof a generic methodology to assess the quality of clinical guidelines. Int J Qual Health Care 1999; 11(1):21-28. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n487 \n \n (59)  Cluzeau FA, Littlejohns P, Grimshaw JM, Feder G, Moran SE. Development and application \nof a generic methodology to assess the quality of clinical guidelines. Int J Qual Health Care 1999; 11(1):21-28. \n (60)  AGREE Collaboration. Development and validation of an international appraisal instrument \nfor assessing the quality of clinical practice guidelines: the AGREE project. Qual Saf Health Care 2003; \n12(1):18-23. \n (61)  Marnellos G. High-throughput SNP analysis for genetic association studies. [Review] Current \nOpinion in Drug Discovery & Development 2003; 6(3):317-321. \n (62)  Meaburn E, Butcher LM, Schalkwyk LC, Plomin R. Genotyping pooled DNA using 100K \nSNP microarrays: a step towards genomewide association scans. Nucleic Acids Research 2006; 34(4):e27. \n (63)  Dobrin SE, Stephan DA. Integrating microarrays into disease-gene identification strategies. \nExpert Rev Mol Diagn 2003; 3(3):375-385. \n (64)  Strausberg RL, Simpson AJ, Wooster R. Sequence-based cancer genomics: progress, lessons \nand opportunities. [Review] [74 refs]. Nature Reviews Genetics 2003; 4(6):409- 418. \n (65)  Bayani J, Brenton JD, Macgregor PF, Beheshti B, Albert M, Nallainathan D et al. Parallel \nanalysis of sporadic primary ovarian carcinomas by spectral karyotyping, comparative genomic hybridization, \nand expression microarrays. Cancer Research 2002; 62(12):3466-3476. \n (66)  Falconer H, D'Hooghe T, Fried G. Endometriosis and genetic polymorphisms. Obstetrical & \nGynecological Survey 2007; 62(9):616-628. \n (67)  Medical Research Council. Patient Research Cohorts Initiative. http://www mrc ac \nuk/ApplyingforaGrant/CallsForProposals/CohortsInitiative/index htm [ 2008  \n (68)  Ollier W, Sprosen T, Peakman T. UK Biobank: from concept to reality. Pharmacogenomics \n2005; 6(6):639-646. \n (69)  Bianconi L, Hummelshoj L, Coccia ME, Vigano P, Vittori G, Veit J et al. Recognizing \nendometriosis as a social disease: the European Union-encouraged Italian Senate approach. Fertility & Sterility \n2007; 88(5):1285-1287. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n488 \n \n (70)  Perel P, Roberts I, Sena E, Wheble P, Briscoe C, Sandercock P et al. Comparison of treatment \neffects between animal experiments and clinical trials: systematic review. BMJ 2007; 334(7586):197. \n (71)  Becker CM, Sampson DA, Short SM, Javaherian K, Folkman J, D'Amato RJ. Short synthetic \nendostatin peptides inhibit endothelial migration in vitro and endometriosis in a mouse model. Fertil Steril 2006; \n85(1):71-77. \n (72)  Matsuzaki S, Canis M, Darcha C, Dechelotte PJ, Pouly JL, Mage G. Effects of a protein \nkinase C inhibitor on the initial development of ectopic implants in a syngeneic mouse model of endometriosis. \nFertil Steril 2008; 89(1):206-211. \n (73)  Lebovic DI, Kir M, Casey CL. Peroxisome proliferator-activated receptor-gamma induces \nregression of endometrial explants in a rat model of endometriosis. Fertil Steril 2004; 82 Suppl 3:1008-1013. \n (74)  D'Hooghe TM, Bambra CS, Raeymaekers BM, De J, I, Lauweryns JM, Koninckx PR. \nIntrapelvic injection of menstrual endometrium causes endometriosis in baboons (Papio cynocephalus and Papio \nanubis). Am J Obstet Gynecol 1995; 173(1):125-134. \n (75)  Dinulescu DM, Ince TA, Quade BJ, Shafer SA, Crowley D, Jacks T. Role of K-ras and Pten in \nthe development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005; 11(1):63-\n70. \n (76)  Lebovic DI, Mwenda JM, Chai DC, Mueller MD, Santi A, Fisseha S et al. PPAR-gamma \nreceptor ligand induces regression of endometrial explants in baboons: a prospective, randomized, placebo- and \ndrug-controlled study. Fertil Steril 2007; 88(4 Suppl):1108-1119. \n (77)  Benbrook DM. Promise and problems of translational research. Gynecol Oncol 2006; 103(2 \nSuppl 1):S14-S17. \n (78)  Dodd JM, Crowther CA, Hiller JE, Haslam RR, Robinson JS. Birth after caesarean study --\nplanned vaginal birth or planned elective repeat caesarean for women at term with a single previous caesarean \nbirth: protocol for a patient preference study and randomised trial. BMC Pregnancy & Childbirth 2007; 7:17. \n (79)  Smith GC, Pell JP, Dobbie R. Caesarean section and risk of unexplained stillbirth in \nsubsequent pregnancy. Lancet 2003; 362(9398):1779-1784. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n489 \n \n (80)  Smith GC, Pell JP, Pasupathy D, Dobbie R. Factors predisposing to perinatal death related to \nuterine rupture during attempted vaginal birth after caesarean section: retrospective cohort study. BMJ 2004; \n329(7462):375. \n (81)  Black N, Barker M, Payne M. Cross sectional survey of multicentre clinical databases in the \nUnited Kingdom. BMJ 2004; 328(7454):1478. \n (82)  National Insititute of Clinical Excellence. Intrapartum care: care of healthy women and their \nbabies during childbirth. NICE Clinical Guideline 55. www.nice.org.uk.  2007.  \n (83)  Department of Health. NHS Care Records Service. Department of Health, UK. http://www \nnhscarerecords nhs uk/ [ 2008  \n (84)  Richesson RL, Krischer J. Data standards in clinical research: gaps, overlaps, challenges and \nfuture directions. J Am Med Inform Assoc 2007; 14(6):687-696. \n (85)  Black N. Maximising research opportunities of new NHS information systems. BMJ 2008; \n336(7636):106-107. \n (86)  Moher D, Altman DG, Schulz KF, Elbourne DR. Opportunities and challenges for improving \nthe quality of reporting clinical research: CONSORT and beyond. CMAJ 2004; 171(4):349- 350. \n (87)  Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N, Hamel C et al. Development of \nAMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med Res \nMethodol 2007; 7:10. \n (88)  Plint AC, Moher D, Morrison A, Schulz K, Altman DG, Hill C et al. Does the CONSORT \nchecklist improve the quality of reports of randomised controlled trials? A systematic review. Med J Aust 2006; \n185(5):263-267. \n (89)  Shea B, Moher D, Graham I, Pham B, Tugwell P. A comparison of the quality of Cochrane \nreviews and systematic reviews published in paper-based journals. Eval Health Prof 2002; 25(1):116-129. \n\nREFERENCES   Chapter 5. Thesis Conclusion \n490 \n \n (90)  Moja LP, Telaro E, D'Amico R, Moschetti I, Coe L, Liberati A. Assessment of \nmethodological quality of primary studies by systematic reviews: results of the metaquality cross sectional \nstudy. BMJ 2005; 330(7499):1053. \n (91)  Grimshaw JM, Eccles MP. Is evidence-based implementation of evidence-based care \npossible? Med J Aust 2004; 180(6 Suppl):S50-S51. \n (92)  Brook RH. Implementing medical guidelines. Lancet 1995; 346(8968):132. \n (93)  Grimshaw JM, Russell IT. Effect of clinical guidelines on medical practice: a systematic \nreview of rigorous evaluations. Lancet 1993; 342(8883):1317-1322. \n (94)  Bahtsevani C, Uden G, Willman A. Outcomes of evidence-based clinical practice guidelines: \na systematic review. Int J Technol Assess Health Care 2004; 20(4):427-433. \n (95)  Foy R, Eccles MP, Jamtvedt G, Young J, Grimshaw JM, Baker R. What do we know about \nhow to do audit and feedback? Pitfalls in applying evidence from a systematic review. BMC Health Serv Res \n2005; 5:50. \n (96)  Shekelle PG, Ortiz E, Rhodes S, Morton SC, Eccles MP, Grimshaw JM et al. Validity of the \nAgency for Healthcare Research and Quality clinical practice guidelines: how quickly do guidelines become \noutdated? JAMA 2001; 286(12):1461-1467.","source_license":"CC0","license_restricted":false}