{"paper_id":"bb9c2aad-de67-4215-a451-78a1aea43dad","body_text":"Chronic pelvic pain and menorrhagia: \nAssessing treatment effectiveness \nJane P . Daniels\n\n\n\nChronic pelvic pain and menorrhagia: \nAssessing treatment effectiveness \nJane P . Daniels\n\nThis work was supported by:\nWellbeing of Women, UK\nUK National Institute of Health Research Health Technology Assessment Programme (05/45/02)\nUniversity of Birmingham\nThe views and opinions expressed therein are those of the authors and do not necessarily reflect \nthose of the HTA programme, NIHR, NHS or the Department of Health.\nChronic Pelvic pain and menorrhagia: Assessing treatment effectiveness / Jane P . Daniels\nAmsterdam: Thela Thesis\nThesis Universiteit van Amsterdam – with references – with summary in Dutch\n© J.P . Daniels, 2013\nCover and layout by: Tickle Me Creative\nPrinted by: Clarkeprint\n\nChronic pelvic pain and menorrhagia: \nAssessing treatment effectiveness\nACADEMISCH PROEFSCHRIFT\nTer verkrijging van der grad van doctor \naan de Universiteit van Amsterdam\n op gezag van de Rector Magnificus\n prof. dr. D.C. van den Boom\n ten overstaan van een door het college voor promoties ingestelde\n commissie, in het openbaar te verdedigen in de Agnietenkapel \nop dinsdag 16 april 2013, te 10.00 uur\ndoor\nJane Patricia Daniels\ngeboren te Derby, Groot Brittannië \n\nPromotiecommissie\nPromotor:  prof. dr. B.W.J. Mol\nCo-promotor:  prof. dr. K.S. Khan\nOverige leden:  prof. dr. P .M.M. Bossuyt\n   dr. J.P .W.R. Roovers\n   dr. M.Y. Bongers\n   prof. dr. R.J. Lilford\n   dr. G ter Riet\n   prof. dr. S. Bhattacharya \nFaculteit der Genneskunde\n\nTable of Contents\nChapter 1 ........................................................................................................................9 \nIntroduction\nPart One: Chronic Pelvic Pain\nChapter 2 ......................................................................................................................19 \nChronic pelvic pain in women\nChapter 3 ......................................................................................................................35 \nLaparoscopic uterine nerve ablation:  \nA survey of gynaecological practice in the UK\nChapter 4 ......................................................................................................................43 \nStudy protocol - A randomised controlled trial to assess the efficacy of  \nLaparoscopic Uterosacral Nerve Ablation (LUNA) in the treatment of chronic  \npelvic pain: the trial protocol ISRCTN41196151\nChapter 5 ......................................................................................................................61 \nLaparoscopic Uterosacral Nerve Ablation (LUNA)  \nfor alleviating Chronic Pelvic Pain: a Randomized Trial\nChapter 6 ......................................................................................................................77 \nMeta-analysis using individual patient data from randomised trials to assess  \nthe effectiveness of laparoscopic uterosacral nerve ablation in the treatment  \nof chronic pelvic pain: a proposed protocol\nChapter 7 ......................................................................................................................87 \nIndividual patient data meta-analysis of randomised evidence to assess the  \neffectiveness of laparoscopic uterosacral nerve ablation in chronic pelvic pain\nChapter 8 ....................................................................................................................103 \nPsychological Therapies for Chronic Pelvic Pain:  \nSystematic review of Randomized Controlled Trials\nPart Two: Heavy Menstrual Bleeding\nChapter 9 ....................................................................................................................117 \nHysterectomy, endometrial destruction and Mirena® for heavy menstrual bleeding:  \nA systematic review and individual patient data meta-analysis\nChapter 10 ..................................................................................................................141 \nSecond generation endometrial ablation techniques  \nfor heavy menstrual bleeding: network meta-analysis\nChapter 11 ..................................................................................................................163 \nThe measurement properties of the menorrhagia multi-attribute  \nquality-of-life scale: a psychometric analysis\nChapter 12 ..................................................................................................................173 \nSummary and Discussion\nSamenvatting .............................................................................................................182\nBibliography ...............................................................................................................189\nAuthors’ affiliations ...................................................................................................193\nDankwoord (Acknowledgements) ............................................................................195\nCurriculum vitae ........................................................................................................ 197\n\n\n\n9\nChapter 1\nIntroduction\n\n10\nChapter 1\nChronic pelvic pain\nChronic pelvic pain is described in many ways. 1 The most frequently cited definition is cyclical \nor non-cyclical pain in the lower abdomen or pelvis, of at least six months’ duration, occurring \ncontinuously or intermittently, that causes functional disability or limits activities of daily living. \nChronic pelvic pain (CPP) has a major impact on health-related quality of life, work attendance \nand productivity,2 and health care utilization, accounting for 40% of referrals for diagnostic lapa-\nroscopy,3 and is an important contributor to health care expenditure.4\nChronic pelvic pain can be a symptom and also a syndrome in its own right when underlying pa-\nthology has been excluded, which may be termed idiopathic CPP . It may encompass dyspareu-\nnia, dyschesia, dysuria, or dysmenorrhoea, but it can also be independent of these symptoms. \nThere are several underlying aetiological factors which may overlap, including endometriosis, \nchronic pelvic inflammatory infection, adhesions, irritable bowel syndrome, interstitial cystitis, and \npelvic congestion syndrome.5\nAnalgesics are often the first line of management, and women often try to manage their pain \nwith over the counter painkillers before consulting with their general practitioner. Pharmaceuti-\ncal approaches considered first include the combined oral contraceptive pill, progestogens or \ngonadotropin-releasing hormone agnosists, yet there are very few randomised data to support \ntheir use in idiopathic chronic pelvic pain. Methoxyprogesterone acetate provides more improve-\nment in pain scores than placebo at 3 months (OR: 2.64, 95%CI 1.33 to 5.25; p=0.005, n=146) \nwhich is sustained to 6 months whilst on treatment. No improvement in pain scores was seen \nin a further small trial of the selective serotonin uptake inhibitor sertraline compared to placebo.\n7 \nThe combined oral contraceptive pill is frequently prescribed. Only 20-25% of patients respond \nto conservative treatment.8  Neuropathic pain treatments are the focus of ongoing investigations \nfollowing initial promising results,9 although there is no evidence to suggest CPP is neuropathic \nin origin.\nGiven that the symptoms experienced are chronic, varied and non-specific, a differential diag-\nnosis in chronic pelvic pain can be hard to establish and women often present to health care \nproviders repeatedly. This wastes both the patients’ time and NHS resources. The diagnosis of \nendometriosis may be delayed by over 8 years after first presentation with CPP symptoms, 10,11 \npotentially demoralising the patient and missing the opportunity to improve their life quality \nthrough early effective treatment, and may provoke women to seek having a hysterectomy with \nChapter 1\n\n11\nits inherent risks and consequences. The Royal College of Obstetricians and Gynaecologists \nguidelines provide a number of suggested initial investigations, including history, microbiological \nscreening and vaginal examination, all with weak evidence for utility. 12 If no cause of the pain is \nfound, gynaecologist frequently perform a diagnostic laparoscopy.\nAs diagnoses emerge through careful history and examination and directed investigation, so do \ntreatment strategies. These should be tailored to the needs of individual patients as, whatever \nthe cause, chronic symptoms need long term management. Surgical treatments tend to focus \non the organic causes of pelvic pain. Laparoscopic vaporisation or excision of endometriotic \nlesions is effective in reducing pain.11 Division of intraperitoneal adhesions is of unclear benefit. 7 \nInterruption of the Lee-Frankenhauser sensory nerve plexuses by laparoscopic uterosacral nerve \nablation was thought to alleviate pain, based on mainly observational data.14\nA multidisciplinary approach has been advocated 15 and fits with the biopsychosocial model of \npain.  Counselling, psychotherapy and attempts to provide reassurance using laparoscopy are \nemployed. Whilst less invasive, psychological approaches are time consuming and may not be \nacceptable to all women. Trials of cognitive behavioural therapy have been favourable in other \nchronic pain conditions, but evidence for psychological interventions within chronic pelvic pain \nis lacking.\nHeavy menstrual bleeding\nHeavy menstrual bleeding is another common problem that can significantly impact on women’s \nlives, and burden individuals and healthcare systems. 16 With an annual community prevalence \nof 25%,16 HMB accounts for 20% of gynecologist consultations in the UK,18 over 5% of women \naged 30-49 consult general practitioners each year in the UK with this problem.18 Rates of surgi-\ncal procedures for HMB are 14.3 per 10,000 in the UK.19\nThere is discordance between women’s subjective experience of menstruation, actual menstrual \nblood loss,\n20 and clinicians’ assessment.21 Only about half of women presenting to health provid-\ners with heavy menstrual bleeding have blood loss greater than the traditional clinical threshold \nof 80 ml per cycle considered as heavy menstruation.20 A shift in emphasis away from menstrual \nblood loss to a more patient-centered definition, to define heavy menstrual bleeding as that sig-\nnificant as to interfere with women’s physical, emotional and social life, is now advocated, as are \nthe use of patient reported outcomes.\n22\nIntroduction\n\n12\nA range of non-hormonal and hormonal medical treatments are available as first line therapy \nfor women presenting with heavy menstrual bleeding in primary care. For over a decade, the \nlevonorgestrel-releasing intrauterine system (LNG-IUS; [Mirena ®]) has been available for heavy \nmenstrual bleeding.  Although developed as a contraceptive, it also reduces menstrual blood \nloss.\n21 In 2007, UK guidance 22 introduced the option of LNG-IUS for heavy menstrual bleeding \nbased on limited evidence.24\nHysterectomy offers a definitive treatment for heavy menstrual bleeding and guarantees amenor-\nrhoea, but it is particularly invasive and carries significant morbidity.25 It is also expensive – due to \nthe need for general anaesthesia, prolonged hospital stay and delayed recovery.  \nEndometrial ablative techniques aimed at destruction of the functionally active endometrium, \nalong with some of the underlying myometrium, 26 offer a conservative surgical alternative. The \nfirst generation ablative techniques, including endometrial laser ablation,27 transcervical resection \nof the endometrium 28 and rollerball endometrial ablation, were all endoscopic procedures. Al-\nthough they do not guarantee amenorrhoea, their effectiveness in comparison with hysterectomy \nhas been demonstrated in a number of randomised  controlled trials.\n29-34 However, first genera-\ntion ablative techniques carry some risks including uterine perforation, cervical laceration, false \npassage creation, haemorrhage, sepsis and bowel injury.\n35,36\nSecond generation ablative techniques represent simpler, quicker and potentially more efficient \noptions of treating heavy menstrual bleeding, which require less skill on the part of the operator. \nSecond generation techniques employ fluid filled thermal balloons, high temperature saline irriga-\ntion, 3D bipolar radiofrequency or microwave energy, freezing or photodynamic therapy to cause \nendometrial destruction. Complications associated with second generation techniques include \nequipment failure, uterine infection, perforation, visceral burn, bleeding and cyclical pain.37\nThese procedures appear to be as effective as first generation ablative techniques in a limited \nnumber of randomised trials,\n37 and some have the added benefit of being performed under lo-\ncal anaesthetic.38 However, some women treated by endometrial ablation will eventually require \nrepeat ablation or hysterectomy. A proportion of women will also develop subsequent problems \nsuch as post surgical adhesions and pelvic floor dysfunction which may lead to further hospital \ntreatment and surgery.\nChapter 1\n\n13\nAims of the thesis\nTo provide an overview of the treatments for chronic pelvic pain and of the interventions for heavy \nmenstrual bleeding.\nSpecific questions:\n•\t What is the effectiveness of LUNA for chronic pelvic pain?\n•\t Are there any subgroups of women who may benefit from LUNA?\n•\t What is the current evidence for psychological treatments for chronic pelvic pain?\n•\t What is the relative effectiveness of hysterectomy, endometrial destruction techniques and \nthe levonorgestrel releasing intra-uterine system for heavy menstrual bleeding?\n•\t What is the relative effectiveness of different second generation endometrial ablation \ntechniques?\nOutline of the thesis\nPart One: Chronic Pelvic Pain\nChapter 2 reviews the current thinking and evidence base for diagnostic tests and medical, \nsurgical and psychological therapies.\nChapter 3 illustrates the situation regarding use of laparoscopic uterosacral nerve ablation \n(LUNA) in the UK, presenting the results of a survey of clinicians.\nChapter 4 describes the protocol for the LUNA randomised controlled trial, detailing the out-\ncomes, analytical methods and practicalities of the trial.\nChapter 5 evaluates the clinical effectiveness of LUNA in alleviating chronic pelvic pain in a\nlarge randomised controlled trial.\nChapter 6 describes the protocol for an individual patient data meta-analysis of five randomised \nLUNA trials.\nChapter 7 presents results from the individual patient data meta-analysis, whereupon data from \nare subject to further subgroup analyses. \nIntroduction\n\n14\nChapter 8 contains a systematic review of the few psychological treatments for chronic pelvic \npain.\nPart Two: Heavy Menstrual Bleeding\nChapter 9 describes the results of an individual patient data meta-analysis of the clinical effec-\ntiveness of hysterectomy, endometrial destruction and the levonorgestrel releasing intra-uterine \nsystem.\nChapter 10 extends the previous review to by undertaking an indirect comparison of the relative \neffectiveness of the different types of second generation endometrial destruction techniques.\nChapter 11 examines the measurement properties of a patient reported questionnaire assessing \nthe impact of heavy menstrual bleeding on women’s lives.\nChapter 12 summarises the previous chapters and describes the results from the studies in this \nthesis in the context of the current literature. Unanswered questions, implications for clinical prac-\ntice and directions for future research are discussed and the main conclusions are described. \nChapter 1\n\n15\nReferences\n1  Williams RE, Hartmann KE, Steege JF. Documenting the current definitions of chronic \npelvic pain: implications for research. Obstet Gynecol 2004 Apr;103(4):686-91.\n2  Sundell G, Milsom I, Andersch B. Factors influencing the prevalence and severity of dys-\nmenorrhoea in young women. BJOG 1990 Jul;97(7):588-94.\n3  Howard FM. The role of laparoscopy in chronic pelvic pain: promise and pitfalls. Obstet \nGynecol Surv 1993;48(6):357-87.\n4  Winkel CA. Modeling of medical and surgical treatment costs of chronic pelvic pain: New \nparadigms for making clinical decisions. Managed Care 1999;5(5):S276-S290.\n5  Latthe P , Mignini L, Gray R, Hills RK, Khan KS. Factors predisposing women to chronic \npelvic pain: systematic review. BMJ 2006;332(7544):749-55.\n6  Henzl MR. Dysmenorrhoea: achievements and challenges. Sexual Medicine Today \n1985;9:8-12.\n7  Stones W, Cheong YC, Howard FM. Interventions for treating chronic pelvic pain in wom-\nen. Cochrane Database of Systematic Reviews 2005;2.\n8 Engel CC, Jr., Walker EA, Engel AL, Bullis J, Armstrong A. A randomized, double-\nblind crossover trial of sertraline in women with chronic pelvic pain. J Psychosom Res \n1998;44(2):203-7.\n9  Sator-Katzenschlager SM, Scharbert G, Kress HG, Frickey N, Ellend A, Gleiss A, et al. \nChronic pelvic pain treated with gabapentin and amitriptyline: a randomized controlled \npilot study. Wien Klin Wochenschr 2005;117(21-22):761-8.\n10  Ballard K, Lowton K, Wright J. What’s the delay? A qualitative study of women’s experi-\nences of reaching a diagnosis of endometriosis. 2006;86(5):1296-301.\n11  Hadfield R, Mardon H, Barlow D, Kennedy S. Delay in the diagnosis of endometriosis: a \nsurvey of women from the USA and the UK. Hum Reprod 1996;11(4):878-80.\n12  RCOG Guidelines and Audit Committee. The initial management of chronic pelvic pain.  \nRoyal College of Obstetricians and Gynaecologists; 2005. Guideline No.41.\n13  Jacobson TZ, Duffy J MN, Barlow D, Koninckx PR, Garry R. Laparoscopic surgery for \npelvic pain associated with endometriosis. Cochrane Database of Systematic Reviews \n2009;4.\n14  Khan KS, Khan SF, Nwosu CR, Dwarakanath LS, Chien PFW. Laparoscopic uterosacral \nnerve ablation in chronic pelvic pain: An overview. Gynaecol Endosc 1999;8(5):257-65.\n15  Stones RW, Price C. Health services for women with chronic pelvic pain. J R Soc Med \n2002;95(11):531-5.\n16  Frick KD, Clark MA, Steinwachs DM, Langenberg P , Stovall D, Munro MG, et al. Financial \nand quality-of-life burden of dysfunctional uterine bleeding among women agreeing to \nobtain surgical treatment. Womens Health Issues 2009;19(1):70-8.\nIntroduction\n\n16\n17  Shapley M, Jordan K, Croft PR. An epidemiological survey of symptoms of menstrual loss \nin the community. Practice 2004;54(502):359-63.\n18  Coulter A, Bradlow J, Agass M. Outcomes of referrals to gynaecology outpatient clinics \nfor menstrual problems: An audit of general practice records. BJOG 1991;98:789-96.\n19  Cromwell DA, Mahmood TA, Templeton A, van der Meulen JH. Surgery for menorrhagia \nwithin English regions: variation in rates of endometrial ablation and hysterectomy. rit Ob-\nstet Gynaecol 2009;116(10):1373-9.\n20  Higham JM, Shaw RW. Clinical associations with objective menstrual blood volume. Eur \nJ Obstet Gynecol Reprod Biol 1999;82(1):73-6.\n21  Warner P , Critchley HO, Lumsden MA, Campbell-Brown M, Douglas A, Murray G. Refer-\nral for menstrual problems: cross sectional survey of symptoms, reasons for referral, and \nmanagement. BMJ 2001;323(7303):24-8.\n22 National Collaborating Centre for Women’s and Children’s Health. Heavy Menstrual \nBleeding. London: Royal College of Obstetricians and Gynaecologists; 2007. Report No.: \nCG44.\n23  Endrikat J, Vilos G, Muysers C, Fortier M, Solomayer E, Lukkari-Lax E. The levonorg-\nestrel-releasing intrauterine system provides a reliable, long-term treatment option for \nwomen with idiopathic menorrhagia. Arch Gynecol Obstet 2012;285(1):117-21.\n24  Stewart A, Cummins C, Gold L, Jordan R, Phillips W. The effectiveness of the lev-\nonorgestrel-releasing intrauterine system in menorrhagia: a systematic review. BJOG \n2001;108(1):74-86.\n25  Lethaby A, Sheppard S, Cooke I, Farquhar C. Endometrial resection and abation versus \nhysterectomy for heavy menstrual bleeding. Cochrane Database of Systematic Reviews \n1999;(2).\n26  Duffy S, Reid P C, Sharp F. In-vivo studies of uterine electrosurgery. BJOG 1992;99(7):579-\n82.\n27  Goldrath M H, Fuller TASS. Laser photovaporisation of endometrium for the treatment of \nmenorrhagia. Am J Obstet Gynaecol 1981;140:14-9.\n28  Magos A, Baumann R, Turnbull A C. Transcervical resection of the endometrium in wom-\nen with menorrhagia. BMJ 1989;298:1209-12.\n29  Gannon M, Holt E M, Fairbank J, Fitzgerald M, Milne M A, Crystal A M, et al. A ran-\ndomised control trial comparing endometrial resection and abdominal hysterectomy for \nthe treatment of menorrhagia. BMJ 1991;303:1362-4.\n30  Dwyer N, Hutton J, Stirrat G M. Randomised controlled trial comparing endometrial re-\nsection with abdominal hysterectomy for the surgical treatment of menorrhagia. BJOG \n1993;100:237-43.\n31  Pinion S B, Parkin D E, Abramovich D R, Naji A, Alexander D A, Russell I T Randomised \ntrial of hysterectomy, endometrial laser ablation, and transcervical endometrial resection \nfor dysfunctional uterine bleeding. BMJ 1994;309(6960):979-83.\nChapter 1\n\n17\n32  O’Connor H, Broadbent J A, Magos A L, McPherson K. Medical Research Council ran-\ndomised trial of endometrial resection versus hysterectomy in the management of menor-\nrhagia. Lancet 1997;349(9056):891-901.\n33  Crosignani P G, Vercellini P , Apolone G, De Giorgi O, Cortesi I, Meschia M. Endometrial \nresection versus vaginal hysterectomy for menorrhagia: long-term clinical and quality-of-\nlife outcomes. Am J Obstet Gynaecol 1997;177(1):95-101.\n34  Cooper K G, Jack S A, Parkin D E, Grant A M. Five-year follow up of women randomised to \nmedical management or transcervical resection of the endometrium as treatment for heavy \nmenses. BJOG 2001;108:1222-8.\n35  Overton C, Hargreaves J, Maresh M. A national survey of the complications of endome-\ntrial destruction for menstrual disorders: the MISTLETOE study. Minimally Invasive Surgi-\ncal Techniques- Laser, EndoThermal or Endoresection. BJOG 1997;104(12):1351-9.\n36  Scottish Hysteroscopy Audit Group. A Scottish audit of hysteroscopic surgery for menor-\nrhagia: complications and follow up. BJOG 1995;102(3):249-54.\n37  Lethaby A, Hickey M, Garry R, Penninx J. Endometrial resection / ablation techniques for \nheavy menstrual bleeding. Cochrane Database of Systematic Reviews 2009.\n38  Clark TJ, Samuel N, Malick S, Middleton LJ, Daniels J, Gupta JK. Bipolar radiofrequency \ncompared with thermal balloon endometrial ablation in the office: a randomized controlled \ntrial. Obstet Gynecol 2011 Jan;117(1):109-18.\nIntroduction\n\n18\n\n\n19\nChapter 2\nChronic pelvic pain in women\n\n20\nChapter 2\nSummary points\n•\t Chronic pelvic pain is a common and debilitating condition\n•\t Women with chronic pelvic pain want to be taken seriously and attach a high value to \nidentifying a cause for their pain\n•\t No organic cause is found on laparoscopy in at least a third of women with chronic pelvic \npain, and patients should be made aware of this before they consent to investigations\n•\t Psychological symptoms, such as anxiety and depression, are frequently present\n•\t Gynaecological treatment for chronic pelvic pain is often unsatisfactory even when directed \nby an identified cause\n•\t A multidisciplinary approach is thought to be essential for effective management\nChronic pelvic pain in women is a debilitating condition that impairs quality of life. Studies using \nvarious definitions estimated its prevalence ranges from 2.1% to 24% of the female population \nworldwide.\n1 It is a common presentation in UK primary care, with 38 per 1000 women affected \nannually—a rate comparable to those of asthma (37 per 1000) and back pain (41 per 1000). 2 \nBecause pelvic pain is associated with conditions such as endometriosis and interstitial cystitis, \na diagnosis is often difficult to establish, leading to a delay in appropriate treatment. Social and \npsychological factors are strongly associated with chronic pelvic pain, so tailored, effective treat-\nment can be challenging to provide.\nSources and selection criteria\nWe searched EMBASE and Medline for articles with titles that included the keywords “chronic \npelvic pain” with the limits “meta-analysis, review and randomised controlled trial”. We restricted \nthe search to articles published in English in the past five years. We identified all systematic \nreviews and trial citations in the Cochrane Library under the MeSH terms pelvic pain and dys-\nmenorrhoea. Articles of interest cited in these papers were identified. Finally we reviewed pub-\nlished clinical guidelines from international pain associations and gynaecological societies and \nsearched the National Guidelines Clearing House.\nChapter 2\n\n21\nWhat is chronic pelvic pain and who gets it?\nChronic pelvic pain is described in many ways.3 The most frequently cited definition is cyclical or \nnon-cyclical pain in the lower abdomen or pelvis, of at least six months’ duration, occurring con-\ntinuously or intermittently, that causes functional disability or limits activities of daily living. 4 The \ncondition is perhaps best seen from a biopsychosocial perspective; organic pathology, beliefs, \ncoping skills, and social interactions all contribute to the woman’s experience of pain.\n5\nChronic pelvic pain can be a symptom and also a syndrome in its own right when underlying \npathology has been excluded. It may encompass dyspareunia (pain during sexual intercourse), \ndyschesia (painful bowel motions), dysuria (painful micturition), or exacerbation of dysmenor -\nrhoea (painful periods), but it can also be independent of these symptoms. Possible pathological \ncauses include endometriosis, chronic pelvic inflammatory infection, adhesions,\n6 irritable bowel \nsyndrome, interstitial cystitis, and pelvic congestion syndrome. 7 Pain may also arise from mus-\nculoskeletal conditions, from pelvic organ prolapse, or from adaptive posture as a result of lower \nback pain. \nAt diagnostic laparoscopy, a substantial proportion of women with chronic pelvic pain (55% in a \nrecent study) have no obvious pathological cause for their pain;\n8 this syndrome is also called idio-\npathic chronic pelvic pain. Several neural paths within the pelvis transmit pain from the reproductive \norgans,\n9 but these nerve fibres may also refer pain to somatic receptors and manifest as chronic \npelvic pain.10 Changes to inflammatory cytokines observed in patients with interstitial cystitis, irrita-\nble bowel syndrome, and vulvar vestibulitis might indicate neurogenic inflammation.11\nA review of factors predisposing women to chronic pelvic pain identified that drug or alcohol \nmisuse, miscarriage, heavy periods, previous caesarean section, pelvic pathology, abuse, and \npsychological comorbidities were associated with a higher risk of non-cyclical pain. 6 Figure 1 \nsummarises these findings using composite categories. Psychological factors affect how people \nreact to and report pain, and they result in coping strategies that can have either a positive or \nnegative effect on wellbeing. Twin studies indicate an aspect of heritability, although this can be \naccounted for by genetic variance underlying the related conditions of endometriosis, fibroids, \ndysmenorrhoea, and also somatic distress: no independent genetic factors were identified as \nunique to idiopathic chronic pelvic pain.\n12\nChronic pelvic pain in women\n\n22\nFigure 1  Composite forest plot of data from studies of individual risk factors associated with  \n dyspareunia, dysmenorrhea, and non-cyclical pelvic pain6\nWhat are the personal and societal costs of chronic pelvic pain?\nThe negative effect of chronic pelvic pain on women’s general quality of life is substantial. Women \nwith chronic pelvic pain tend to report lower general physical health scores than controls without \npain.13,14 Women with chronic pelvic pain describe loss, social isolation, and effects on relation-\nships. They have a high incidence of comorbidity, sleep disturbance, and fatigue. A community \nbased study found that 41% of women with chronic pelvic pain had not seen a healthcare provid-\ner in the previous year,\n13 suggesting that most women are coping outside of the system. Pain af-\nfects daily activities; around 18% of employed women in the UK take at least a day off work each \nyear because of such pain.13 The economic burden to healthcare systems is difficult to establish \nand no recent data are available. Hospital episode data estimated the direct cost of health care \nfor chronic pelvic pain at £158m (€188m, $237m) with a further £24m in indirect costs in 1992.\n15\nHow is chronic pelvic pain diagnosed?\nBecause the symptoms are varied and non-specific, a differential diagnosis can be hard to es-\ntablish. The condition is chronic and women present repeatedly over several years. It is important \nto consider the whole range of possible causes at every new encounter in the general practice \nor gynaecology clinic, whether first or repeat, bearing in mind the biopsychosocial perspective \n(Figure 2). A previously diagnosed condition  may have recurred (for example, endometriosis) \nor a new condition may have developed (such as interstitial cystitis or depression in a woman \npreviously diagnosed with endometriosis or idiopathic chronic pelvic pain). A multidisciplinary \napproach is ideal.\nChapter 2\n\n23\nFigure 2  Biopsychosocial model of chronic pelvic pain\nWomen with chronic pelvic pain want to feel they are being taken seriously, desire reassurance, \nand attach importance to having an explanation for their pain. 16 Failure to address these con-\ncerns can mean a loss of opportunity to improve the patient’s quality of life. 17 The International \nPelvic Pain Society recommends the use of body charts, simple visual analogue scales, and \nthe short-form McGill questionnaire. 18 Although these measures may help women to articulate \ntheir experience of pain, qualitative research shows that the McGill pain questionnaire does not \ncapture the full range of descriptions women attribute to pain.\n19,20 The clinician should explore \ndescriptions and locations of pain with the patient. 21 Enabling the woman to tell her story and \nconsider the causes of pain helps to improve the therapeutic relationship and serves as a positive \nexperience for the woman.\n22\nBy taking a history of the patterns and intensity of pain and its association with gynaecological, \nurogenital, and bowel symptoms, a potential cause may be revealed. 23,24 Women with endome-\ntriosis report throbbing pain and pain during defecation more frequently that those with idiopathic \nchronic pelvic pain.21 Symptoms suspicious of pelvic inflammatory disease should prompt a cer-\nvical swab, although a negative culture does not always exclude the diagnosis. 25 Psychological \nhistory is equally important in view of the association between chronic pelvic pain and anxiety and \nChronic pelvic pain in women\n\n24\ndepressive symptoms.6 Routine use of a simple screening questionnaire, such as the two ques-\ntion Patient Health Questionnaire-2, may help to identify depression. 26 A more sensitive subject \nto raise is that of physical and sexual abuse. 6 Few valid tools exist to guide questionning 27 and \nsuch investigations should always be supported with appropriate counselling services. Physical \nexamination can also inform further testing. For example, tenderness or nodularity of the cul-de-\nsac and uterosacral ligaments are highly predictive of deep endometriosis,28,29 and tenderness of \nthe uterus to palpation is a non-specific marker for adenomyosis. Pelvic organ prolapse can be \nassessed and described with the POP-Q system.30\nThe accuracy of tests used for chronic pelvic pain is affected by the range of target conditions that \nhavedifferent diagnostic standards and criteria. Additionally, a woman may have more than one \nunderlying cause for her symptoms. Women should be made aware that a substantial proportion \nof women with chronic pelvic pain do not have an identifiable organic cause before consenting to \nunpleasant and potentially risky investigations. Radiological tests including transvaginal ultrasound \nand magnetic resonance (MRI) scanning can delineate normal pelvic organs. They can also identify \nmany pathological conditions such as deep infiltrating endometriosis29 and ovarian endometrio-\nma,31 adenomyosis,32 adhesions,33 dilated pelvic veins,34 and other soft markers.35\nAlthough its accuracy is not perfect, the use of MRI for the differential diagnosis of chronic pelvic \npain has the potential to replace diagnostic laparoscopy in a proportion of women.36,37 Laparos-\ncopy is regarded as the “gold standard”38,39 although many target conditions cannot be diagnosed \nin this way (for example, ademomyosis). Laparoscopically diagnosable diseases are more common \n(65%) among women with chronic pelvic pain than in those without such pain (28%).39 Although \ndiagnostic criteria such as perihepatic adhesions in pelvic inflammatory disease40 or typical lesions \nseen in endometriosis41 are easy to identify, subtle lesions and deep infiltrating endometriosis may \nbe missed, possibly because of a lack of training in laparoscopic diagnosis. The relevant question \nis not so much whether MRI or laparoscopy is more accurate, but which of the two has a greater \neffect on treatment decisions. This issue will be the subject of research through our recently funded \nNIHR health technology assessment project (www.medal.bham.ac.uk).\nConsumers, and some professionals, may see diagnostic laparoscopy as an invasive, expensive, and \noverused, and risky procedure.\n4 Rates of its use vary greatly across regions42 and its use as the initial \ntest in secondary care is by no means universal.43 The procedure is associated with about a 3% risk \nof minor complications (for example, nausea and vomiting, shoulder tip pain), a 0.24% risk of unan-\nticipated injury causing major complications (such as bowel perforation)—of which two-thirds require \nlaparotomy—and an estimated risk of death of 3.3-8 per 100,000. 44,45 Consent for laparoscopic \nprocedures is often poor, with women being confused as to whether the procedure is diagnostic or \ntherapeutic, and complications are a frequent cause for medical negligence claims.\n46\nChapter 2\n\n25\nWhat is the initial therapeutic approach for chronic pelvic pain?\nAnalgesics are often the first line of management, and women often try to manage their pain \nwith over the counter painkillers before consulting with their general practitioner. As diagnoses \nemerge, so do treatment strategies. These strategies should be tailored to the needs of individual \npatients, since chronic symptoms, whatever the cause, need long term management and a \nmultimodal approach. The combined oral contraceptive pill is frequently used with good results, \ndespite no direct evidence for efficacy in chronic pelvic pain and limited evidence in dysmenor -\nrhoea.47 The contraceptive pill can be used in the usual way with monthly cycles or continuously \nto avoid monthly periods and associated pain.48 Two small randomised controlled trials have as-\nsessed use of medroxyprogesterone for women with a diagnosis of pelvic congestion syndrome; \none factorial trial comparing this treatment with psychotherapy and placebo, and one trial against \nplacebo alone. Medroxyprogesterone alone improved pain scores (odds ratio 2.64, 95% confi-\ndence interval 1.33 to 5.25, n=146) and self reported improvement (6.81, 95% CI 1.83 to 25.3, \nn=44) at the end of the four month treatment period, but the benefit was not sustained to nine \nmonths later.49,50 For many women, progestogens have side effects, such as weight gain and \nacne, that may lead to discontinuation. The orally vasoactive lofexidine hydrochloride, which was \npostulated to affect pain by reducing vasospasms associated with pelvic venous congestion, did \nnot show a benefit in a small placebo controlled trial.\n51\nTricyclic antidepressants and venlafaxine provide benefit for neuropathic pain 52 but not lower \nback pain. 53 Only one small randomised controlled trial of sertraline compared with placebo \nhas been reported for chronic pelvic pain, with no evidence of improvement in pain scores. 54 \nOne randomised controlled trial of gabapentin, which is frequently used for neuropathic pain, \nassessed this drug either alone or in combination with amitriptyline. Fifty-six women with pain \nrefractory to the nonopioid drug metamizol together with weak opioid tramadol were followed \nup for 24 months of treatment. All women reported improvements in pain, but gabapentin alone \nor with amitriptyline was better than amitriptyline alone in the reduction of pelvic pain reported \non a visual analogue scale.\n55 The method of randomisation was unstated and the trial was not \nblinded, to allow careful titration of drugs to minimise adverse events, so ideally this result should \nbe substantiated in a larger, higher quality trial.\nWomen with chronic pelvic pain tend to be referred to gynaecology clinics. A multidisciplinary \napproach has been advocated56 and fits with the biopsychosocial model of pain, but such clinics \nare few and far between, leading to lack of a holistic approach. One Dutch randomised con-\ntrolled trial lends support to this approach.\n57 In the group randomised to standard management, \ndiagnostic laparoscopy was routinely done to exclude organic pelvic pain, which was treated ap-\npropriately. If no somatic cause could be found psychological therapies were considered. In the \nChronic pelvic pain in women\n\n26\ngroup randomised to integrated management, somatic, psychological, dietary, environmental, \nand physiotherapeutic factors were addressed equally and laparoscopy was not routinely done. \nThe latter group had greater self reported improvement (odds ratio 4.15, 95% CI 1.91 to 8.99), \nbetter activities of daily living scores (3.53, 95% CI 1.65 to 7.58), and improved non-pain symp-\ntoms (6.99, 95%CI 3.26 to 14.96), but no difference with the standard group in improvement in \nthe McGill pain score (1.52, 95%CI 0.71 to 3.27). It is difficult to identify which components of the \nintegrated approach are responsible, but trials of cognitive behavioural therapy—which aims to \ninfluence dysfunctional emotions, behaviours, and cognitions through a goal-oriented, system-\natic procedure—have been favourable in other chronic pain conditions. The role of counselling \nand psychological therapy is not widely appreciated by gynaecologists, general practitioners, \nand patients. Meta-analysis of two small randomised controlled trials including 142 women with \nchronic pelvic pain gave a mean difference in pain scores at least six months after intervention of \n−1.57 (95% CI −2.53 to −0.61, P<0.00001) in favour of the psychological intervention group.\n58,59\nSurgical treatments tend to focus on the organic causes of pelvic pain. Laparoscopic vaporisa-\ntion or excision of endometriotic lesions is effective in reducing pain.\n60 Division of intraperitoneal \nadhesions is of unclear benefit.61,62 Interruption of the Lee-Frankenhauser sensory nerve plexuses \nby laparoscopic uterosacral nerve ablation was thought to alleviate pain. However, considerable \nevidence,8 including an individual patient data meta-analysis, 63 now shows that the procedure \nis not effective in reducing dysmenorrhoea, dyspareunia, or non-cyclical pain. Presacral neurec-\ntomy has been assessed in primary and secondary dysmenorrhoea\n64 and seems to be more \neffective than uterosacral nerve ablation, but it is associated with a high incidence of later consti-\npation and is not widely practiced.\nWhat are the recommendations from clinical guidelines?\nThe UK Royal College of Obstetricians and Gynaecologists has produced an algorithm for as-\nsessment and initial treatment of chronic pain, which was the basis of the information in this \nreview.43 Further guidelines recommend that, if no pathology is seen at laparoscopy, a letter of re-\nassurance should be offered and the woman contacted for review a year later.65 Guidelines from \nthe European Association of Urology also recommend assessment and treatment of nerve en-\ntrapment and pelvic floor dysfunction through nerve blocks and transcutaneous electrical nerve \nstimulation for interstitial cystitis.\n66 Both UK and American guidelines encourage an empirical trial \nof gonadotropin releasing hormone agonists for women suspected of endometriosis who do not \ndesire a definitive diagnosis or wish to defer surgical investigation.\n 67-69\nChapter 2\n\n27\nQuestions for future research\n•\t Analysis of distribution of the underlying causes of chronic pelvic pain\n•\t Role of MRI in differential diagnosis of chronic pelvic pain\n•\t Estimation of accuracy of individual and combinations of diagnostic tests\n•\t High quality assessments of treatment packages, including psychological therapies\n•\t Development of a new pain assessment tool to capture the multidimensional experience \nof pelvic pain\n•\t Economic analysis to determine most cost effective diagnosis and management strategies\n \nTips for non-specialists\n•\t Let the woman tell her story—taking a good history is important.\n•\t Consider gynaecological, urological, bowel, and psychological symptoms\n•\t Refer to a gynaecological pain specialist or gynaecologist with access to a multidisciplinary \npain team\nAdditional educational resources\nResources for healthcare professionals\nInternational Association for the Study of Pain (IASP) (www.iasp-pain.org) - IASP brings together \nscientists, clinicians, health care providers, and policy makers to stimulate and support the study \nof pain and to translate that knowledge into improved pain relief worldwide\nInternational Pelvic Pain Society (www.pelvicpain.org) - United States educational society for \nhealthcare providers. A pelvic pain assessment proforma can be downloaded from the website\nMultidisciplinary Approach to the Study of Chronic Pelvic Pain (MAPP) Research Network  \n(www.mappnetwork.org) - collaborative research on urological chronic pelvic pain disorders \nResources for patients\nPelvic Pain Support Network (www.pelvicpain.org.uk)—charity that provides support, informa-\ntion, and advocacy for people with chronic pelvic pain, and promotes education and research in \nchronic pelvic pain amongst health professionals.\nEndometriosis UK (www.endometriosis-uk.org)—provides information and support and works to \nincrease understanding and awareness.\nThe Gut Trust (www.theguttrust.org)—has a telephone helpline and a web based self manage-\nment programme for people with irritable bowel syndrome\nChronic pelvic pain in women\n\n28\nA patient’s perspective\nI have suffered with period pain since I was fifteen. I had taken various oral contraceptive pills \nover the years. Sixteen years later, by chance, a retired doctor noticed that I was using a hot \nwater bottle to ease my period pain, which was unbearable. He told me this was not normal and \nthat it could be endometriosis. Two months later, I was diagnosed with severe endometriosis via \nlaparoscopy. Last September, I had laser surgery by laparoscopy and a levonorgestrel intrauter-\nine device was inserted. This procedure did not alleviate the pain but the hormonal treatment has \nreduced the bleeding.\nLiving with endometriosis is extremely frustrating, especially because the path of diagnosis has \nbeen so long. I have always been determined to take charge of my life, but the lack of good \nquality information has hindered me in being able to do this. Recently I was lucky to be able to \nattend an international congress on endometriosis, a source of reliable information that has given \nme some directions in which to exert my efforts: finding the best treatments to manage pain and \nthe best options to boost my emotional wellbeing. Although I am scared, now I am in a better \nposition to manage pain and my whole life, which is being affected at all levels by the severity of \nmy condition. \nFrom a member of the Pelvic Pain Support Network.\nChapter 2\n\n29\nReferences\n1 Latthe P , Latthe M, Say L, Gulmezoglu M, Khan KS. WHO systematic review of prev-\nalence of chronic pelvic pain: a neglected reproductive health morbidity. BMC Public \nHealth 2006;6:177.\n2 Zondervan KT, Yudkin PL, Vessey MP , Dawes MG, Barlow DH, Kennedy SH. Prevalence \nand incidence of chronic pelvic pain in primary care: evidence from a national general \npractice database. BJOG 1999;106:1149-55.\n3 Williams RE, Hartmann KE, Steege JF. Documenting the current definitions of chronic \npelvic pain: implications for research. Obstet Gynecol 2004;103:686-91.\n4 Howard FM. The Role of Laparoscopy in the Chronic Pelvic Pain Patient. Clinical Obstet-\nrics & Gynecology 2003;46(4):749-66.\n5 Engel GL. The need for a new medical model: a challenge for biomedicine. Science \n1977;196(4286):129-36.\n6 Latthe P , Mignini L, Gray R, Hills RK, Khan KS. Factors predisposing women to chronic \npelvic pain: systematic review. BMJ 2006;332:749-55.\n7 Beard RW. Pelvic congestion: A functional disorder of the menstrual cycle? In: Maclean \nAB, Stones RW, Thornton S, editors. Pain in obstetrics and gynaecology London: RCOG \nPress; 2001.\n8 Daniels J, Gray R, Hills RK, Latthe P , Buckley L, Gupta J, et al. Laparoscopic uterosa-\ncral nerve ablation for alleviating chronic pelvic pain: a randomized controlled trial. JAMA \n2009;302:955-61.\n9 Wesselmann U, Czakanski PP . Pelvic pain: a chronic visceral pain syndrome. Curr Pain \nHeadache Rep 2001;5(1):13-9.\n10 Alsom N, Harrison G, Khan KS, Patwardhan S. Visceral hyperalgesia in chronic pelvic \npain. BJOG 2009;116:1551-5.\n11 Wesselmann U. Neurogenic inflammation and chronic pelvic pain. World J Urol \n2001;19:180-5.\n12 Zondervan KT, Cardon LR, Kennedy SH, Martin NG, Treloar SA. Multivariate genetic anal-\nysis of chronic pelvic pain and associated phenotypes. Behav Genet 2005;35(2):177-88.\n13 Zondervan KT, Yudkin PL, Vessey MP , Jenkinson CP , Dawes MG, Barlow DH, et al. The \ncommunity prevalence of chronic pelvic pain in women and associated illness behaviour. \nBr J Gen Pract 2001;51:541-7.\n14 Grace V, Zondervan K. Chronic pelvic pain in women in New Zealand: comparative well-\nbeing, comorbidity, and impact on work and other activities. Health Care Women Int \n2006;27(7):585-99.\n15 Davies L, Ganger K, Drummond M, Saunders D, Beard R. The economic burden of in-\ntractable gynaecological pain. J Obstet Gynaecol 1992;12(S2):54-6.\nChronic pelvic pain in women\n\n30\n16 Price J, Farmer G, Harris J, Hope T, Kennedy S, Mayou R. Attitudes of women with \nchronic pelvic pain to the gynaecological consultation: a qualitative study. Brit J Obstet \nGynaecol 2006;113(4):446-52.\n17 Ballard K, Lowton K, Wright J What’s the delay? A qualitative study of women’s experi-\nences of reaching a diagnosis of endometriosis. Fertil Steril 2006;86(5):1296-301.\n18 Melzack R. The McGill Pain Questionnaire: Major properties and scoring methods. Pain \n1975;1(277):299.\n19 Grace VM, MacBride-Stewart S. “How to say it”: women’s descriptions of pelvic pain. \nWomen Health 2007;46(4):81-98.\n20 Grace VM, MacBride-Stewart S. ‘Women get this’: gendered meanings of chronic pelvic \npain. Health (London) 2007;11(1):47-67.\n21 Ballard K, Lane H, Hudelist G, Banerjee S, Wright J. Can specific pain symptoms help in \nthe diagnosis of endometriosis? A cohort study of women with chronic pelvic pain. Fertil \nSteril 2010;94:20-7.\n22 Stones RW, Lawrence WT, Selfe SA. Lasting impressions: influence of the initial hospital \nconsultation for chronic pelvic pain on dimensions of patient satisfaction at follow-up. J \nPsychosom Res 2006;60(2):163-7.\n23 Parsons C.L. Diagnosing chronic pelvic pain of bladder origin. J Reprod Med 2004;49(Sup-\npl. 3):235-42.\n24 Rome Foundation Inc. Rome III Diagnostic Criteria for Functional Gastrointestinal Disor -\nders. Gastroenterology 2006;20(5):1377-90.\n25 RCOG Guidelines and Audit Committee. Management of Acute Pelvic Inflammatory \nDisease. Lonson: Royal College of Obstetricians and Gynaecologists; 2003. Greentop \nguideline 32.\n26 Kroenke K, Spitzer RL, Williams JB. The Patient Health Questionnaire-2: validity of a two-\nitem depression screener. Med Care 2003;41(11):1284-92.\n27 Leserman J, Drossman DA, Li Z. The reliability and validity of a sexual and physical \nabuse history questionnaire in female patients with gastrointestinal disorders. Behav Med \n1995;21(3):141-50.\n28 Bazot M, Lafont C, Rouzier R, Roseau G, Thomassin-Naggara I, Darai E. Diagnostic ac-\ncuracy of physical examination, transvaginal sonography, rectal endoscopic sonography, \nand magnetic resonance imaging to diagnose deep infiltrating endometriosis. Fertil Steril \n2009;92(6):1825-33.\n29  Hudelist G, Oberwinkler KH, Singer CF, Tuttlies F, Rauter G, Ritter O, et al. Combination \nof transvaginal sonography and clinical examination for preoperative diagnosis of pelvic \nendometriosis. Hum Reprod 2009;24(5):1018-24.\nChapter 2\n\n31\n30 Bump RC, Mattiasson A, Bo K, Brubaker LP , DeLancey JO, Klarskov P , et al. The stand-\nardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am \nJ Obstet Gynecol 1996;175(1):10-7.\n31 Moore J, Copley S, Morris J, Lindsell D, Golding S, Kennedy S. A systematic review of \nthe accuracy of ultrasound in the diagnosis of endometriosis. Ultrasound Obstet Gynecol \n2002;20(6):630-4. \n32 Champaneria R, Abedin P , Daniels JP , Balogun M, Khan KS. Ultrasound scan and mag-\nnetic resonance imaging for the diagnosis of adenomyosis: Systematic review comparing \ntest accuracy. Acta Obstet Gynecol Scand 2010; 89(11):1374-84.\n33 Katayama M, Masui T, Kobayashi S, Ito T, Sakahara H, Nozaki A, et al. Evaluation of pel-\nvic adhesions using multiphase and multislice MR imaging with kinematic display. Am J \nRoentgenol 2001;177(1):107-10.\n34 Asciutto G, Mumme A, Marpe B, Koster O, Asciutto KC, Geier B. MR venography in the \ndetection of pelvic venous congestion. Eur J Vasc Endovasc Surg 2008;36(4):491-6.\n35 Okaro E., Condous G, Khalid A, Timmerman D, Ameye L, Huffel SV, et al. The use of ultra-\nsound-based ‘soft markers’ for the prediction of pelvic pathology in women with chronic \npelvic pain - Can we reduce the need for laparoscopy? BJOG 2006;113(3):251-6.\n36 Cody J, Ascher SM. Diagnostic value of radiological tests in chronic pelvic pain. Baillieres \nBest Pract Res Clin Obstet Gynaecol 2000;14:433-66.\n37 Bazot M, Thomassin-Naggara I, Darai E, Marsault C. Imaging of chronic pelvic pain. J \nRadiol 2008;89(1 Pt 2):107-14.\n38 Howard FM. The role of laparoscopy in chronic pelvic pain: promise and pitfalls. Obstet \nGynecol Surv 1993;48:357-87.\n39 Howard FM. The role of laparoscopy as a diagnostic tool in chronic pelvic pain. Best Pract \nRes Clin Obstet Gynaecol 2000;14:467-94.\n40 Kim JY, Kim Y, Jeong WK, Song S-Y, Cho OK. Perihepatitis with pelvic inflammatory dis-\nease (PID) on MDCT: Characteristic findings and relevance to PID. Abdom Imaging 2009; \n34(6): 737-742\n41 Engemise S, Gordon C, Konje JC. Endometriosis. BMJ 2010;340:c2168.\n42 Moses SH, Clark TJ. Current practice for the laparoscopic diagnosis and treatment of \nendometriosis: A national questionnaire survey of consultant gynaecologists in UK. BJOG \n2004;111(11):1269-1272. \n43 RCOG Guidelines and Audit Committee. The initial management of chronic pelvic pain. \nLondon: Royal College of Obstetricians and Gynaecologists; 2005. Greentop guideline 41.\n44 Jansen FW, Kapiteyn K, Trimbos-Kemper T, Hermans J, Trimbos JB. Complications of \nlaparoscopy: a prospective multicentre observational study. BJOG 1997;104(5):595-600.\nChronic pelvic pain in women\n\n32\n45 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 \ncases. Hum Reprod 1998;13(4):867-72.\n46 Physician Insurers Association of America. Laparoscopic Injury Study. Rockville, MD. \nUSA: Physician Insurers Association of America; 2000. \n47 Wong CL, Farquhar C, Roberts H, Proctor M. Oral contraceptive pill for primary dysmen-\norrhoea. Cochrane Database Syst Rev 2009;4\n48 Kwiecien M, Edelman A, Nichols MD, Jensen JT. Bleeding patterns and patient accept-\nability of standard or continuous dosing regimens of a low-dose oral contraceptive: a \nrandomized trial. Contraception 2003;67(1):9-13.\n49 Farquhar CM, Rogers V, Franks S, Pearce S, Wadsworth J, Beard RW. A randomized \ncontrolled trial of medroxyprogesterone acetate and psychotherapy for the treatment of \npelvic congestion. BJOG 1989;96(10):1153-62.\n50 Walton SM, Batra HK. The use of medroxyprogesterone acetate 50 mg in the treatment \nof painful pelvic conditions: Preliminary results from a multicentre trial. BJOG 1992;12(6 \n(Supp 2)):S50-S53.\n51 Stones RW, Bradbury L, Anderson D. Randomized placebo controlled trial of lofexidine \nhydrochloride for chronic pelvic pain in women. Hum Reprod 2001;16(8):1719-21.\n52 Saarto T, Wiffen PJ. Antidepressants for neuropathic pain. Cochrane Database of Sys-\ntematic Reviews 2010;1.\n53 Urquhart DM, Hoving JL, Assendelft-Willem JJ, Roland M, van-Tulder MW. Antidepres-\nsants for non-specific low back pain. Cochrane Database of Systematic Reviews 2008;1.\n54 Engel CC, Jr., Walker EA, Engel AL, Bullis J, Armstrong A. A randomized, double-\nblind crossover trial of sertraline in women with chronic pelvic pain. J Psychosom Res \n1998;44(2):203-7.\n55 Sator-Katzenschlager SM, Scharbert G, Kress HG, Frickey N, Ellend A, Gleiss A, et al. \nChronic pelvic pain treated with gabapentin and amitriptyline: a randomized controlled \npilot study. Wien Klin Wochenschr 2005;117(21-22):761-8.\n56 Stones RW, Price C. Health services for women with chronic pelvic pain. J R Soc Med \n2002;95:531-5.\n57 Peters AA, van Dorst E, Jellis B, van Zurren E, Hermans J, Trimbos JB. A randomized \nclinical trial to compare two different approaches in women with chronic pelvic pain. Ob-\nstet Gynecol 1991;77:740-4.\n58 Haugstad GK, Haugstad TS, Kirste UM, Leganger S, Klemmetsen I, Malt UF, et \nal. Mensendieck somatocognitive therapy as treatment approach to chronic pel-\nvic pain: results of a randomized controlled intervention study. Am J Obstet Gynecol \n2006;194(5):1303-10.\nChapter 2\n\n33\n59 Onwude JL, Thornton JG, Morely S, Lilleyman J, Currie I, Lilford R. A randomised trial of \nphotographic reinforcement during post-operative counselling after diagnostic laparos-\ncopy for plevic pain. BJOG 2004;112:89-94.\n60 Jacobson TZ, Duffy-J MN, Barlow D, Koninckx PR, Garry R. Laparoscopic surgery for \npelvic pain associated with endometriosis. Cochrane Database of Systematic Reviews \n2009;4.\n61 Stones W, Cheong YC, Howard FM. Interventions for treating chronic pelvic pain in wom-\nen. Cochrane Database Syst Rev 2005;2\n62 Keltz MD, Gera PS, Olive DL, Keltz MD, Gera PS, Olive DL. Prospective randomized \ntrial of right-sided paracolic adhesiolysis for chronic pelvic pain. J Soc Laparo Surg \n2006;10(4):443-6.\n63 Daniels JP , Middleton L, Xiong T, Champaneria R, Johnson N, Lichten EM, et al. Individual \npatient data meta-analysis of randomised evidence to assess the effectiveness of lapa-\nroscopic uterosacral nerve ablation in chronic pelvic pain. Hum Reprod Update 2010; \n6(6):568-76\n64 Chen F.P ., Chang S.D., Chu K.K., Soong Y.K. Comparison of laparoscopic presacral \nneurectomy and laparoscopic uterine nerve ablation for primary dysmenorrhea. J Reprod \nMed 1996;41(7):463-6.\n65 Kennedy S, Bergqvist A, Chapron C, D’Hooghe T, Dunselman G, Hummerlshoj L, et \nal. ESHRE guideline for the diagnosis and treatment of endometriosis. Hum Reprod \n2005;20:2698-704.\n66 Fall M, Baranowski AP , Elneil S, Engler D, Hughes J, Messelink EJ et al Guidelines on \nchronic pelvic pain. European Urology 2010;57: 35-48 .\n67 American College of Obstetricians and Gynaecologists. Chronic pelvic pain. ACOG Prac-\ntice Bulletin 51. American College of Obstetricians and Gynaecologists, 2004. \n68 RCOG Guidelines and Audit Committee. The investigation and management of endome-\ntriosis. Royal College of Obstetricians and Gynaecologists, 2006. Greentop guideline 24.\n69 Ling FW. Randomized controlled trial of depot leuprolide in patients with chronic pelvic \npain and clinically suspected endometriosis. Pelvic Pain Study Group. Obstet Gynecol \n1999;93(1):51-8.\nChronic pelvic pain in women\n\n34\n\n\n35\nChapter 3\nLaparoscopic uterine nerve ablation:  \nA survey of gynaecological practice in the UK\n\n36\nChapter 3\nAbstract\nObjective: To determine the current use of and the indications for laparoscopic uterine nerve \nablation (LUNA) in the UK, and to establish practitioners’ views regarding the desirability of a \ncontrolled clinical trial to assess the efficacy of LUNA.\nSetting: Gynaecologists associated with the British Society of Gynecological Endoscopy.\nDesign: A self-administered structured questionnaire was posted to 457 UK gynaecologists. \nTheir surgical practice and willingness to participate in a randomized trial  were surveyed.  The \npostal questionnaire was returned by 249 (55% response rate).\nResults: The  survey indicated  that  113 (45%)  of  the  responders performed LUNA. Of these, \n86 (76%) carried out LUNA for chronic pelvic pain, 72 (64%) for endometriosis, and 16 (14%) for \ndyspaurenia/dysmenorrhoea. Diathermy was  used as  the surgical tool  by 66 (59%)  gynae- \ncologists and laser was used by 57 (51%). Of the 113 currently using LUNA, 91 (81%)  were \nwilling  to  enter  patients into  a proposed clinical  trial of LUNA.\nConclusion: Current clinical practice regarding LUNA  for  pelvic pain is hung in equipoise. This \nmeans that the technique has been introduced without reliable evidence of efficacy but opinion  \nregarding its use is not yet solidified. Hence, a rigorous randomized trial of LUNA’s efficacy is \nurgently needed and is practicable.\nIntroduction\nChronic pelvic pain, defined as constant or intermittent,  cyclic or  acyclic pelvic pain that  persists \nfor  6 months or more, 1 has a major impact on health-related quality of life, work productivity \nand health care utilisation, costing an estimated 158 million  pounds per year in  the UK.2   Dys-\nmenorrhoea, deep dyspareunia and intermenstrual pain constitute the main symptom complex \nin chronic pelvic pain, which frequently leads to surgical interventions including laparoscopy.3,4\nWith recent developments in minimal access surgery, laparoscopic uterosacral nerve ablation \n(LUNA)  is a practicable treatment option at the time of diagnostic laparoscopy in  chronic  pelvic \npain.4\nBy interrupting pain pathways in  the nerve plexuses and ganglia of the uterosacral ligaments \nLUNA is expected to alleviate pelvic pain.5,6\nChapter 3\n\n37\nHowever, the therapeutic role of this intervention has not been adequately assessed by rigorous \nclinical research.7 Despite the lack of definitive evidence, many gynaecologists  familiar  with  the  \noperative technique offer LUNA as a therapeutic option to their patients with pelvic pain. This dis-\ncrepancy between evidence and practice induced us to examine current  surgical practices re-\ngarding LUNA  and to determine  practitioners’ willingness to participate in a clinical trial of LUNA.\nMethods and results\nWe undertook  a  self-administered postal survey of gynaecologists  associated with  the  British  \nSociety of Gynaecological Endoscopy. In view of declining return rates to mailed questionnaires,  \nwe designed our survey strategy to include the factors known to be associated with  higher  \nresponse.\n8,9 Questionnaires were posted to 457 gynaecologists. Of these, 249 returned  the \nquestionnaires, which gave a 55% response rate. Responses to  individual  items were missing \nfrom some respondents.\nTable 1 shows the results of the survey. There were 113 (45%) gynaecologists who carried out \nLUNA. Of these, 86 (76%) performed LUNA for chronic pelvic pain, 72 (64%) for endometriosis, \nand 16 (14%) for dyspaurenia/dysmenorrhoea. Diathermy was  used as the surgical tool by 66 \n(59%) gynaecologists and laser was used by 57 (51%). More importantly, of the 113 currently \nusing LUNA, 91 (81%) gynaecologists were willing to enter patients into a proposed clinical trial \nof LUNA.\nDiscussion\nThe generalisability of our findings may be somewhat diminished  by the relatively low rate of response \nto our questionnaire. Although we made all efforts within our means to enhance the return rate,\n8,9 we \nreceived replies from just 55% of gynaecologists, a rate lower than that reported in other recent sur-\nveys.10-12  There are several reasons for non-response, including out-of-date addresses, but the  main  \nreason might have been that LUNA has been promoted as an effective treatment in endometriosis \nbased on one influential trial.13 This trial was rigorous, but because of its design it was not possible to \ndistinguish whether the treatment effect was caused by the transection of the uterosacral ligaments or \nby the ablation of the endometriotic lesions. Indeed, subsequent research has shown that LUNA may \nnot have been the main factor alleviating pain  in endometriosis.\n14 Lack of this information might  have \nled  some gynaecologists to believe that further research on LUNA lacks relevance, leading to  an un-\nenthusiastic attitude towards our questionnaire. However, there has been a trend towards reduction in \nresponse to questionnaire surveys8 and so our 55% response rate should be considered reasonable \nand does not invalidate our findings. \nSurvey of LUNA practice\n\n38\nWilling to  \nparticipate  \nn (%)\nUnlikely to  \nparticipate  \nn (%)\nIndications for LUNA\nChronic pelvic pain\n Only 25 (28) 3 (14)\n + Endometriosis 36 (40) 8 (36)\n + Other 7 (8) 1 (4)\n + Endometriosis + other 4 (4) 2 (9)\nEndometriosis\n Only 12 (13) 3 (14)\n + Other 4 (4) 3 (14)\nOther\n Only 1 (1)                      1 (4)\nNot recorded                                   2 (2)                      1 (4)\nTotal 91 22\nSurgical technique for LUNA\nDiathermy\n Only 37 (41) 1 (4)\n + Laser 10 (11) 6 (27)\n + Scissors 8 (9) 2 (9)\nLaser\n Only 28 (31) 10 (45)\n + Scissors 1 (1) 0 (0)\nScissors\n Only 3 (3) 0 (0)\nNot recorded                                   3 (3)                      2 (9)\nTotal   91                         22\nTable 1  Current UK gynaecological practice concerning laparoscopic uterosacral nerve  \n ablation (LUNA), according to gynaecologist’s willingness to participate in a   \n proposed trial of LUNA\nChapter 3\n\n39\nThe assessment of new surgical interventions requires an initial assessment of safety and practicability \nfollowed by a proper randomized evaluation.15 We undertook a systematic literature review concern-\ning LUNA,7 which identified only a few publications with evaluative evidence on LUNA.  Moreover, \nthis literature consisted largely of observational uncontrolled case series and poor quality controlled  \nstudies which could not be regarded as anything more than suggestive evidence. The initial  evaluative \nevidence alone is not sufficient to assess  the clinical effectiveness of LUNA for which a larger, prop-\nerly randomized trial, unconfounded by other variables, is needed. This need has been highlighted in \nseveral recent publications.16-18 From our survey, it is disturbing to note that many gynaecologists  are \nperforming LUNA as a routine therapeutic procedure when evidence is insufficient to establish efficacy. \nOn  the other hand it is encouraging that many gynaecologists wish to participate in a clinical trial. This \nis a situation where equipoise applies (i.e. the technique has been introduced without definite evidence \nbut opinion regarding its use is not yet solidified) making the need for a trial even more urgent.\nEncouraged by our survey, we have launched a multi-centre, prospective, blind, randomized con-\ntrolled trial assessing LUNA. Eligible consenting patients are randomly allocated at diagnostic laparos-\ncopy either to LUNA or to no pelvic denervation.19 The main endpoints for assessing the efficacy of \nLUNA are pain, sexual function  and quality of life. Long-term follow up will be undertaken as well as \nshort-term monitoring. This study is being coordinated from the University of Birmingham Clinical Tri-\nals Unit, and interested gynaecologists  are welcome to participate. We await with interest the results \nof this trial.\nSurvey of LUNA practice\n\n40\nReferences\n1  Vercellini P , Fedele L, Arcani L, Bianchi S, Rognoni MT, Candiani GB. Laparoscopy in the \ndiagnosis of chronic pelvic pain in adolescent women. J Reprod Med1989; 34: 827-30.\n2  Davies L, Ganger KF, Drummond M, Saunders D, Beard RW. The economic burden of \nintractable gynaecological pain. J Obstet Gynaeco 1992; 12: s46-s54.\n3  Howard FM. The role of laparoscopy in chronic pelvic pain: promise and pitfalls. Obstet \nGynecol Survey 1993; 48: 357-87.\n4  Dwarakanath LS, Parsad PS, Khan KS. Role of laparoscopy in the management of chron-\nic pelvic pain. Hospital Medicine 1998; 59 (8): 627-31.\n5  Frankenhauser G. Die Bewegungenerven der Gebarmutter. Zeitsccrift für Medizin Nat \nWiss 1864; 1: 35.\n6  Doyle JB, Des Rosiers  JJ. Paracervical uterine denervation for the relief of pelvic pain. \nClinical Obstetrics and Gynecol 1963;  6: 742-53.\n7  Khan KS, Khan SF, Nowsu CR, Dwarakanath LS, Chein PFW. Laparoscopic  uterine \nnerve ablation in chronic pelvic pain: an overview.  Gynaecol Endoscopy,\n 1998; 8: 257-67.\n8  McAvoy BR, Kaner EF. General practice postal surveys: a questionnaire too far? BMJ  \n1996; 313: 732-3.\n9  Heywood A, Mudge P , Ring I, Sanson-Fisher R. Reducing systematic bias in studies of \ngeneral practitioners: the use of a medical peer in the recruitment of general practioners \nin research.  Family Practice 1995; 12: 227-31.\n10  McColl A, Smith H, White P , Field J. General practitioners’ perceptions of the route to \nevidence based medicine: a questionnaire survey. BMJ 1998; 316: 361-5.\n11  Olatunbosun OA, Edouard L, Pierson RA. Physicians’ attitudes toward evidence based \nobstetrics practice: a questionnaire survey. BMJ 1998; 316: 765-6.\n12  Gulmezoglu AM, Duley L. Use of anticonvulsants in eclampsia and pre-eclampsia: survey \nof obstetricians in the United Kingdom and Republic of Ireland. BMJ 1998; 316: 975-6.\n13  Sutton CJG, Ewen SP , Whitelaw N, Haines B. Prospective, randomised, double blind \ncontrolled trial of laser laparoscopy in the treatment of pelvic pain associated with mini-\nmal, mild, and moderate endometriosis. Fertil St teril 1994; 62: 696-700.\n14  Dover RW, Pooley A, Haines P , Sutton CJG. A prospective randomised double-blind trial \nof laparoscopic laser uterine nerve ablation in the treatment of pelvic pain associated with \nendometriosis: a proviaional report. BJOG 1998;  105 (Suppl. 17): 48.\n15  Cushcieri A. Minimal access therapy: scope,evaluation and future direction. Healthcare \nBulletin 1996; 54: 514-21.\n16  Stones RW. Chronic pelvic pain. Personal  Assessment in Continuing Education. Review \nNo 97/01. London: Royal College of Obstetricians and Gynaecologists, 1997.\nChapter 3\n\n41\n17  Stones RW, Mountfield  J, Farquhar C. Management of chronic pelvic pain in women. \nCochcrane Database of  Systematic Reviews. 1998.\n18  Porpora MG, Victor G. The role of laparoscopy in the management of pelvic pain in wom-\nen of reproductive age. Fertil Steril 1997; 68: 765-79.\n19  Birmingham Clinical Trials Unit. The LUNA trial. \nSurvey of LUNA practice\n\n42\n\n\n43\nChapter 4\nStudy protocol  \nA randomised controlled trial to assess the efficacy of \nLaparoscopic Uterosacral Nerve Ablation (LUNA) in \nthe treatment of chronic pelvic pain: the trial protocol \nISRCTN41196151\n\n44\nChapter 4\nAbstract\nBackground: Chronic pelvic pain is a common condition with a major impact on health-related \nquality of life, work productivity and health care utilisation. The cause of the pain is not always \nobvious as no pathology is seen in 40-60% of the cases. In the absence of pathology there is \nno established treatment. The Lee-Frankenhauser sensory nerve plexuses and parasympathetic \nganglia in the uterosacral ligaments carry pain from the uterus, cervix and other pelvic struc-\ntures. Interruption of these nerve trunks by laparoscopic uterosacral nerve ablation (LUNA) may \nalleviate pain. However, the balance of benefits and risks of this intervention have not been reli-\nably assessed. LUNA has, nevertheless, been introduced into practice, although there remains \ncontroversy regarding indications for LUNA. Hence, there is an urgent need for a randomised \ncontrolled trial to confirm, or refute, any worthwhile effectiveness. The principal hypothesis is that, \nin women with chronic pelvic pain in whom diagnostic laparoscopy reveals either no pathology or \nmild endometriosis (AFS score ≤ 5), LUNA alleviates pain and improves life quality at 12 months.\nMethods/ Design: The principal objective is to test the hypothesis that, in women with chronic \npelvic pain in whom diagnostic laparoscopy reveals either no pathology or mild endometriosis \n(AFS score ≤ 5), LUNA alleviates pain and improves life quality at 12 months. A multi-centre, \nprospective, randomised-controlled-trial will be carried out with blind assessment of outcomes in \neligible consenting patients randomised at diagnostic laparoscopy to LUNA (experimental group) \nor to no pelvic denervation (control group). Postal questionnaires including visual analogue scale \nfor pain (primary outcome), an index of sexual satisfaction and the EuroQol EQ-5D instrument will \nbe administered at 3, 6 and 12 months. The primary assessment of the effectiveness of LUNA \nwill be from comparison of outcomes at the one-year follow-up, although the short, medium and \nlong-term risks and benefits of LUNA will also be evaluated.\nThe sample size for this trial has been estimated as 420 patients in total using the hypothesis that \nLUNA will alleviate pain symptoms (i.e. reduce pain scores on a VAS) more than no intervention at \none-year following diagnostic laparoscopy and taking into consideration 20% loos to follow-up. \nThe intention to treat analysis to address the principal research question will be conducted using \nthe one-year follow-up data.\nChapter 4\n\n45\nBackground\nPelvic pain remains the single most common indication for referral to a gynaecology clinic ac-\ncounting for 20% of all outpatient appointments. 1,2 Five percent of all new appointments are \nfor chronic pelvic pain. 1 It continues to be one of the most difficult and perplexing problems \nencountered in gynaecology. Pelvic pain has a major impact on health-related quality of life, \nwork productivity and health care utilisation. It is also a major cause of workplace absenteeism.2 \nAn estimated £158 million are spent annually on the management of this condition in the health \nservice.\n3 In primary care, the annual prevalence is 38/1000 in women aged 15-73, a rate com-\nparable to that of asthma (37/1000) and chronic back pain (41/1000).4\nAn initial troublesome problem in addressing research issues related to pelvic pain is the difficulty \nin making a definite diagnosis. The following definition is commonly used and will be adopted for \nour research proposal: constant or intermittent, cyclic or acyclic pain, that persists for 6 months \nor more and includes dysmenorrhoea, deep dyspareunia and intermenstrual pain.5\nAn effective treatment for pelvic pain has evaded gynaecologists for centuries. Even today only \n20-25% patients respond to conservative management.\n6 When such treatment fails, a diagnostic \nlaparoscopy is usually performed.1,7,8 Chronic pelvic pain accounts for up to 40-60% of all diag-\nnostic laparoscopies.7 The cause of the pain is not always obvious as no pathology is seen in \n40-60% of the cases. When definite pathology is found the likely causes are endometriosis (25%) \nand adhesions (25%).7 There is evidence that negative findings at laparoscopy and follow up with \nultrasound will provide reassurance and relief to some patients.9 Randomised trials demonstrate \nsome benefit from ablation of mild and moderate endometriosis10 and from lysis of vascular ad-\nhesions.11 For patients without obvious pathology there is no established treatment.\nIn recent years, operative laparoscopy has developed rapidly and now makes laparoscopic pelvic \ndenervation a practicable option in the management of chronic pelvic pain. The nerve plexuses \nand parasympathetic ganglia in the uterosacral ligaments, first described in the last century, 12 \nand subsequently confirmed,13 carry “pain” from the uterus, cervix and other pelvic structures. \nThe surgical approach to the management of pelvic pain is based on interruption of these nerve \ntrunks, which can be achieved by transection of uterosacral ligaments.\nLUNA study protocol\n\n46\nFigure 1  Pelvic sensory nerve pathways and site for LUNA\nThe transection of the uterosacral ligaments and the nerve plexuses it contains is the simplest \nof the surgical procedures for pelvic pain. The original work by Doyle described vaginal and \nabdominal approaches to divide the attachments of the uterosacral ligaments to the cervix. 14,15 \nWith the wider use of minimal access therapy there is a renewed interest in the division of the \nFrankenhauser nerve plexus in the uterosacral ligaments laparoscopically using lasers or electro-\ndiathermy. In an attempt to relieve patients’ symptoms clinicians frequently perform laparoscopic \nuterosacral nerve ablation, or LUNA as it is sometimes called. However the efficacy of this proce-\ndure has not been assessed objectively using methodologically sound research. \nThe Need for the LUNA Trial\nHealth technology assessment in surgical interventions requires an initial evaluation of the safety \nand stability of new interventions followed by randomised trials. The initial evaluative evidence \nalone is not sufficient to assess the clinical effectiveness of LUNA for which randomised research \nremains the gold standard. \nChapter 4\n\n47\nEvidence from the medical literature\nIn the past, pelvic denervation had been proposed for treatment of chronic pelvic pain, 14 but \nit has not been widely adopted due to the need for open invasive surgery and the lack of reli-\nable evidence of benefit. However, recent developments in minimal access surgery now makes \nLUNA an easily practicable treatment option. Our systematic review has shown that the currently \navailable research evidence on LUNA is inconclusive. 16 Therefore further research is required to \ngenerate effectiveness evidence in the form of a high quality randomised controlled trial. This \nneed is also highlighted by the recent Cochrane Reviews\n17,18 that recommend rigorous research \nto assess surgical interventions in chronic pelvic pain.\n \nProfessional consensus\nWe conducted a survey of UK gynaecologists associated with the British Society of Gynaecologi-\ncal Endoscopy in 1998 to determine the extent to which LUNA was being used in practice.\n19 Our \nsurvey indicated that despite the lack of definitive evidence, many gynaecologists familiar with \nthe technique were using LUNA as a therapeutic option. Of the 247 respondents, 108 (44%) of-\nfered LUNA to patients: 78% for chronic pelvic pain, 66% endometriosis and 18% for dyspaure-\nnia/ others reasons. Crucially, this survey indicated that 93 of 108 (86%) gynaecologists currently \nperforming LUNA were willing to recruit patients in a randomised trial of LUNA. In this situation \nequipoise applies (i.e. the technique has been introduced without definite evidence but opinion \nregarding its use is not yet solidified) making the need for a trial even more urgent. \nPilot study\nWe have undertaken a pilot of the LUNA trial with the objective of assessing its feasibility. It has \nshown acceptability to participants. It has also established trial management procedures, piloted \nquestionnaires, measured compliance and standardised operating procedures. A confidential \ninterim analysis was recently reviewed by an independent data monitoring committee when \nthe first 60 participants had completed 6 months follow-up. The committee’s report regarding \ncompliance and follow-up was encouraging but the recommendation was that a larger study is \nneeded for adequate statistical power in the trial to evaluate LUNA reliably. \nResearch Proposal\nObjectives\n1.  To test the hypothesis that in women with chronic pelvic pain in whom diagnostic laparos-\ncopy reveals either no pathology or mild endometriosis (AFS score ≤ 5) LUNA alleviates \npain and improves life quality at 12 months (principal objective).\n2.  To test the hypothesis that response to LUNA differs according to the site and cause of \nthe pain by two secondary analyses:  (i) women with central pain, (ii) women with no vis-\nible pathology.\nLUNA study protocol\n\n48\n3.  To explore the variation in LUNA’s effectiveness at different periods of follow-up (3 months, \n6 months, 2, 3, 5 and 10 years)\n4.  To explore any long-term complications of LUNA and to examine the incidence of further \ngynaecological procedures.\nDesign\nA multi-centre, prospective, randomised-controlled-trial will be carried out with single blind as-\nsessment of outcomes in eligible consenting patients randomised at diagnostic laparoscopy to \nLUNA (experimental group) or to no pelvic denervation (control group). Postal questionnaires \nincluding visual analogue scale for pain (primary outcome), an index of sexual satisfaction and the \nEuroQol 5D-EQ instrument (secondary outcomes) will be administered at 3, 6 and 12 months. \nThe primary assessment of the effectiveness of LUNA will be from comparison of outcomes at \nthe one-year follow-up. The medium and long-term risks and benefits of LUNA will also be evalu-\nated by postal questionnaires to the women at 2, 3, 5 and 10 years after laparoscopy.\nFigure 2  The Laparoscopic Uterosacral Nerve Ablation (LUNA) Trial Schema\nBMC Women's Health 2003, 3 http://www.biomedcentral.com/1472-6874/3/6\nPage 9 of 10\n(page number not for citation purposes)\nCross Hospital, Wolverhampton, UK); S Keay (Walsgrave\nHospital, Coventry, UK); E Shaxted (Northampton Gen-\neral Hospital, UK) and J Thornton (City Hospital, Not-\ntingham, UK) The following centres have obtained ethics\napproval and await recruitment: A Ali (Darlington Memo-\nrial Hospital, UK); C Chandler (Billinge Hospital, Wigan,\nUK); E Downes (Chasefarm Hospital, Enfield, UK); A\nHextall (St Albans City Hospital, UK);J Kabukoba (Sand-\nwell Hospital, West Bromwich, UK); M Katesmark\n(Epsom General Hospital, UK); P. Kirwan (Leicester Gen-\neral Hospital, UK); K Louden (Royal Hampshire Hospital,\nUK);J Moohan (Altnagelvin Area Hospital, UK); S Najja\n(Queen's Park Hospital, Blackburn, UK); K Olah (War-\nwickshire General Hospital, UK); M Padwick (Watford\nGeneral Hospital, UK); P Owen (Stobhill Hospital, Glas-\ngow, UK) and A Tapp (Royal Shrewsbury Hospital, UK).\nAdditional material\nAcknowledgements\nFunding: WellBeing, UK (Charity grant no. CF/371)\nData Monitoring Committee: P Brocklehurst, D Braunholtz and JA Jordan\nReferences\n1. M Morris, D O'Niell: Out patient Gynaecology. BMJ 1958, 2:1038.\n2. Howard FM: The role of laparoscopy in chronic pelvic pain:\npromise and pitfalls. Obstet Gynecol Surv 1993, 48:357-387.\n3. Stones RW: Chronic pelvic pain. Personal Assessment in Con-\ntinuing Education. 97/01th edition. London, Royal College of Obstetri-\ncians and Gynaecologists; 1997. \n4. Sundell G, Milsom I, Andersch B: Factors influencing the preva-\nlence and severity of dys menorrhoea in young women.  Br J\nObstet Gynaecol 1990, 97:588-594.\n5. Davies L, Ganger K, Drummond M, Saunders D, Beard R: The eco-\nnomic burden of intractable gynaecological pain.  J Obstet\nGynecol 1992, 12:46-54.\n6. Zondervan KT, Yudkin PL, Vessey MP, Dawes MG, Barlow DH,\nKennedy SH: Prevalence and incidence of chronic pelvic pain\nin primary care: evid ence from a nation al general practice\ndatabase. Br J Obstet Gynaecol 1999, 106:1149-1155.\n7. Vercellini P, Fedele L,  Arcaini L, Bianchi S, Rognoni MT, Candiani GB:\nLaparoscopy in the diagnosis of chronic pelvic pain in adoles-\ncent women. J Reprod Med 1989, 34:827-830.\n8. MR Henzl: Dysmenorrhoea; achievements and challenge.  Sex\nMed Today 1985, 9:8-12.\n9. Porpora MG, Gomel V: The role of laparoscopy in the manage-\nment of pelvic pain in women of reproductive age.  Fertil Steril\n1997, 68:765-779.\n10. G Frankenhauser: Die Bewegungenerven der Gebarmutter.  Z\nMed Nat Wiss 1864, 1:35.\n11. Latarjet A Roget P: Le plexus hypogastrique chez la femme.\nGynecologie et Obstetrique 1922, 6:225.\n12. RM Campbell: Anatomy and physiology of sacrouterine\nligaments. Am J Obstet Gynecol 1950, 59:1.\n13. JB Doyle: Paracervical uterine denervation by transection of\nthe cervical plexus for th e relief of dysmenorrhoea.  Am J\nObstet Gynecol 1955, 70:1.\n14. Doyle JB Des Rosiers JJ: Paracervical uterine denervation for\nthe relief of pelvic pain. Clinical Obstet Gynecol 1963, 6:742-753.\n15. A Cushcieri: Minimal Access Therapy: scope, evaluation and\nfuture direction. Health Bull 1996, 54:514-521.\n16. Khan KS, Khan SF, Nwosu CR, Dwarakanath LS, Chien PFW: Lapar-\noscopic uterosacral nerve ablation in chronic pelvic pain: An\noverview. Gynaecological Endoscopy 1999, 8:257-265.\n17. Wilson ML, Farquhar CM, Sinclair OJ, Johnson NP: Surgical inter-\nruption of pelvic nerve pathways for primary and secondary\ndysmenorrhoea. [Review] [6 refs]. Cochrane Database Syst Rev\n2000:CD001896.\n18. Stones RW, Mountfield J: Interventions for treating chronic pel-\nvic pain in women. [update of Cochrane Database Syst Rev.\n2000;(2):CD000387 ; 10796713.]. [Review] [15 refs]. Cochrane\nDatabase Syst Rev 2000:CD000387.\n19. Daniels J, Gray R, Khan K, Gupta J: Laparoscopic uterine nerve\nablation: A survey of gynaecological practice in the UK.\nGynaecol Endosc 2000, 9:157-159.\n20. Unit University of Birm ingham Clinical Trials: The LUNA Trial.\nhttp://www luna bham ac uk 2003.\nThe laparoscopic uterosacral nerve ablation (luna) trial schemaFigure 2\nThe laparoscopic uterosacral nerve ablation (luna) trial \nschema\nIdentification of eligible patient\n•Chronic pelvic pain >6mth\n•Diagnostic laparoscopy planned\n•Obtain informed consent\nFollow-up Questionnaire\nat 3, 6, 12, 24, 36, 60 and 120 \nmonths\nDuring diagnostic laparoscopy\n•No obvious pathology\n•Technically feasible\n•Randomisation\nNO LUNALUNA\nAdditional File 1\nAppendix A1: Patient information leaflet. Appendix A2: Consent form. \nAppendix B: Registration and randomisation form. Appendix C: On study \nform. Appendix D: Post surgery form.\nClick here for file\n[http://www.biomedcentral.com/content/supplementary/1472-\n6874-3-6-S1.doc]\nChapter 4\n\n49\nEligibility\nAll new patients presenting to the gynaecology outpatient clinic with pelvic pain (cyclical or non-\ncyclical) and/or dyspareunia, and requiring diagnostic laparoscopy for evaluation of these condi-\ntions, will be invited to participate.\nInclusion criteria\n•\t Pelvic\tpain\tof\tlonger\tthan\t6\tmonth\tduration.\n•\t Pain\tlocated\twithin\tthe\ttrue\tpelvis\tor\tbetween\tand\tbelow\tthe\tanterior\tiliac\tcrests.\n•\t Associated\tfunctional\tdisability.\n•\t Lack\tof\tresponse\tto\tmedical\ttreatment.\n•\t Diagnostic\tlaparoscopy\tplanned.\nExclusion criteria\n•\t Previous LUNA\n•\t Moderate and severe endometriosis (American Fertility Society score >5)\n•\t Previous surgery for endometriosis.\n•\t Previous surgery for pelvic inflammatory disease.\n•\t Previous hysterectomy.\n•\t Adnexal pathology.\nInterventions\nDiagnostic laparoscopy plus uterosacral nerve ablation (experimental group) or laparoscopy \nwithout pelvic denervation (control group). \nLaparoscopic uterosacral nerve ablation will be carried out in a uniform manner by named sur -\ngeons in each of the participating centres following a common protocol as described in the \nstandard surgical text.20 Routine preparation will be made for a diagnostic laparoscopy with the \npatient intubated for general anaesthesia. Following pneumoperitonium, a laparoscope will be \nused to visualise the pelvis. Before embarking on operative laparoscopy an anatomical pelvic \nassessment will be performed to identify pelvic structures and pathology. At this stage patients \nwith pathology outlined in the exclusion criteria will be excluded. It is expected that around 30% \nof women will be unsuitable for LUNA at operation and will be ‘registered only’ cases. Eligible \npatients will be randomised by a telephone call to the Birmingham University Clinical Trials Unit.\nClear identification of the uterosacral ligaments is a prerequisite to treatment with lasers or elec-\ntro-diathermy. The posterior leaf of the broad ligament will be carefully inspected to identify the \ncourse of the ureters, which on rare occasions could be particularly close to the uterosacral liga-\nments. Care will also be taken to note thin walled pelvic veins, which often lie just lateral to the \nuterosacral ligaments. If accidentally punctured, they may cause troublesome bleeding requiring\nLUNA study protocol\n\n50\nfurther endoscopic endocoagulation. The uterosacral ligaments will be identified by manipulation \nof the uterus in the right and left lateral planes. The ligaments will then be ablated with laser or \nmicropoint electro-diathermy or endocoagulation depending upon the surgeons’ preferences. \nThe ablation will start as close to the posterior aspect of the cervix as possible and continue for \na minimum of 1 cm posterolaterally on either side. The aim of the procedure is to destroy the \nsensory nerve fibres and the secondary ganglia as they leave the uterus and come to lie within \nthe uterosacral ligaments. \nThe safe conduct of operative laparoscopy for LUNA requires the use of two ports, one for deliv-\nery of the energy source (laser or diathermy) and another for manipulation. These are in addition \nto the umbilical port used for the laparoscope itself. In contrast, diagnostic laparoscopy in wom-\nen with no pathology requires only one port in addition to the umbilical laparoscopic port. This \ndifference in number of ports has potential for introducing bias by compromising patient-blinding \nto group allocation. A sham incision (see Blinding and Ethical Considerations) in the control group \nis used to overcome this problem.\nRandomisation\nConsenting eligible patients will be randomised to diagnostic laparoscopy plus uterosacral nerve \nablation (experimental group) or to no pelvic denervation at the time of diagnostic laparoscopy \n(control group). Important issues related to randomisation are considered below:\nAllocation Sequence\nThe subjects will be allocated to groups using a chance procedure, blocking and stratification.\n21 \nStratified block randomisation will be employed to ensure that there will be nearly equal numbers \nof patients in the two groups within the prognostic subgroups, even if the study ends prema-\nturely. Variable block size will be used to avoid any possibility of foreknowledge. Stratified alloca-\ntion is used so that chance imbalances in the stratification variable do not have an effect on the \noutcome. Since response to treatment may depend on the surgeons’ technique for laparoscopic \nuterosacral nerve ablation, analyses will be retrospectively stratified according to the surgeons \nparticipating in the trial. \nIntraoperative randomisation\nAllocation concealment is a crucial factor in avoiding bias in randomised trials.22 Although it is not \npossible to blind the surgeon, it is essential to keep the surgeon unaware of the group allocation \nuntil after the irrevocable decision to enter the woman into the trial has been made.  So, in LUNA, \ntreatment allocation will be issued at diagnostic laparoscopy, after the surgeon has inspected the \npelvis and ensured that the patient fulfils all of the inclusion criteria and she does not have any \nof the exclusion criteria. Women may be randomised or registered into the study by telephoning \nthe toll free Randomisation Line on 0800 953 0274 (+44 121 687 2319 from outside the UK) or \nChapter 4\n\n51\nby internet randomisation at https://www.trials.bham.ac.uk/luna and clicking on the randomisa-\ntion button. Passwords for internet randomisation will only be allocated to centres with ethical \napproval.\nFollowing surgery, the surgeon fills in operation details on the post-surgery form.\nBlinding\nIn the LUNA Trial, patients kept blind to their treatment allocation until the follow-up in the trial \nis completed. Patients may show a placebo response if they know they have received the ac-\ntive treatment. The magnitude of placebo effects should not be underestimated. There is clear \nevidence that inadequacies of blinding in randomisation lead to exaggeration of treatment effect \nin randomised trials.22\nHowever, there is a potential problem in the maintenance of blinding in the LUNA trial. As men-\ntioned above, women allocated to have LUNA will have the standard operative laparoscopy with \nthree ports (one 10mm umbilical port and two 5mm lateral ports), whereas women allocated to \nthe control group under normal circumstances would have standard diagnostic laparoscopy with \ntwo ports (one 10mm umbilical port and one 5mm lateral or midline port). By noting the different \nnumber of incisions some women might become aware of their group allocation and this might \nalter their response. In order to maintain participant blinding, a sham 5mm skin incision is made \nsuperficially in a lateral port site. This approach in avoiding bias due to lack of blinding has been \nused in a previous trial of laparoscopic nerve ablation,\n10 and has received ethical approval in the \nLUNA trial.\nA second purpose of blinding is to prevent differences in other aspects of patient management \nintroducing biases affecting the results. The patient’s GP will therefore be kept blind to treatment \nallocation. Double blinding is not possible in LUNA, however, as the surgeons performing the \nsurgical intervention on the participants will be aware of the group allocation. However, the likeli-\nhood that this will lead to bias in outcome assessment is low as the outcome assessments in this \nstudy will be conducted by patient self-administered questionnaires, avoiding any possible bias \nfrom surgeons’ knowledge of group allocation.\nEndpoints\nPrimary Outcome measure: Visual analogue scale for assessment of pain\nPain is difficult to measure, partly because it is accompanied by other sensations and partly \nbecause the reaction component affects the judgement of the pain regardless of the intensity \nof the stimulus. A measure of pain is nevertheless essential to the outcome of this clinical trial. \nVisual analogue scales (VAS) originally devised as measures of well-being\n23 have been success-\nLUNA study protocol\n\n52\nfully adapted to measure pain.24,25 This technique involves use of a 10 cm line on a piece of paper \nrepresenting a continuum of the participants’ opinion of the degree of pain. It is explained to the \nparticipant that the one extreme of the line represents “no pain at all” while the other represents \n“as much pain as she can possibly imagine”. The subject rates the degree of pain by placing a \nmark on the line and scale values are obtained by measuring the distance from zero to that mark. \nVisual analogue scales in the assessment of pain have been established to be reproducible and \naccurate.25 VAS has commonly been used in measurement of chronic pain. 26 All the studies of \nLUNA included in systematic reviews so far have used this measure, or a variation on it, for as-\nsessing outcome. Individual pain scores have sufficient psychometric strengths to be used in \nchronic pain research involving group comparison designs.\n27\nSecondary Outcome measures\nAssessment of sexual function\nSexual function is an important aspect of life quality in patients with pelvic pain. Pain itself is an \nanti-aphrodisiac, and together with discomfort and altered self image, it impacts upon sexual \nfunction.28 Its assessment in an objective manner is therefore an important part of the LUNA trial. \nThere are several sexual function instruments, with high levels of reliability, validity and respon-\nsiveness, which yield comparable results across occasions and individuals, making them suitable \nfor monitoring therapeutic progress in randomised research.28\nThe Sexual Activity Questionnaire (SAQ)29 will replace the Brief Index of Sexual Satisfaction (ISS)30 \nfor the assessment sexual function. The outcome measure has been changed because of poor \nacceptability and compliance with the ISS. The SAQ has excellent internal consistency and test \nretest reliability. It also has excellent concurrent and construct validity and has been shown to \nbe acceptable to women in other clinical trials.31 In the questionnaire it will be clearly stated that \nthe measure of sexual function covers material that is sensitive and personal. Participants will be \nreassured that their responses will be kept completely confidential and that if they do not wish to \nanswer any questions, they will be allowed to leave the questionnaire blank. \nHealth-related quality of life Instrument\nThe EuroQol EQ-5D questionnaire will be used to assess health-related quality of life (HRQL) \ninstruments are becoming powerful tools for outcome assessments in randomised trials. Quality \nof life instruments assess aspects of patient’s health status usually not grasped by conventional \nclinical indices; hence, they can be applied as complementary assessments together with VAS \nand SAQ. Quality of life has to be defined clearly and patient’s perception of normal performance \nserves a pivotal role in this context. HRQL instruments are administered with questionnaires as-\nsessing a number of different domains, i.e. areas of behaviour or experience that the instrument \nis attempting to measure.\n32\nChapter 4\n\n53\nResource usage and additional procedures\nLUNA is a quick, safe and inexpensive procedure for women already undergoing diagnostic \nlaparoscopy. Our hypothesis is about a clinically important effect without an excess of complica-\ntions. If the hypothesis is confirmed, then any benefits will essentially be “dominant” outweighing \nthe relatively small costs of intervention. Therefore, we do not plan a formal economic evaluation \nat this stage. However, data on health resource use will be collected partly as efficacy outcome \nmeasures (i.e. less need for medical care for pelvic pain indicates greater efficacy), and partly to \nallow an economic evaluation to be carried out, should significant complications occur. Other \nmeasures will include analgesic use, consultations at general practice and hospital, and time off \nwork. Again these are both economic outcomes and indicators of residual pain.\nFollow-up\nPostal questionnaires to assess pain and sexual function will be administered at enrolment in the \ntrial and then at 3, 6, 12, 24 and 36 months after laparoscopy. The outcomes at 12 months will \nbe used to address the primary research question. This time interval is chosen because laparos-\ncopy alone has a placebo effect for up to 3-6 months in some patients.\n33,34 The 24 and 36 month \nfollow up will be used to monitor medium term effects of the intervention. Existing participants in \nthe trial who consented to 3 years of follow-up will be asked to consent to long-term follow-up \n(10 years) once they have reached the 3 year follow up time-point. Participants randomised after \nJune 2003 will be asked to consent to 10 years of follow-up at entry.\nTable 1:  LUNA Trial data collection schedule\nClinic Theatre Follow up (months/years)\n3 6 12 2 3 5 10\nPatient information and consent x\nEligibility Checklist and Randomisation \nform x x\nEnrolment Questionnaire x\nSurgery Form x\nLetter to GP x\nFollow-up Questionnaire x x x x x x x\nLUNA study protocol\n\n54\nIneligible Patients – follow-up only\nWomen who are ineligible for the LUNA trial because of moderate to severe endometriosis, signif-\nicant adhesions, significant pelvic inflammatory disease, other significant pathology or those for \nwhom LUNA is not technically feasible should be registered with the Trial Office for follow-up only. \nThe woman should be told that she was not eligible for the trial randomisation, and the reasons \nwhy, and asked if she would agree to complete the follow-up questionnaires at 6 and 12 months.\nThis non-random cohort will provide comparative data on the natural history of patients with \nchronic pelvic pain with significant pathology.\nSide-effects quantification\nThe centres have been advised to fill an “adverse event form” in case of immediate and delayed \ncomplications if any are associated with the procedure.\nSample Size and Power Considerations\nThe sample size for this trial has been estimated using the hypothesis that LUNA will alleviate \npain symptoms (i.e. reduce pain scores on a VAS) more often than no intervention at one-year \nfollowing diagnostic laparoscopy. Cohen describes ‘effect sizes’ of 0.2 and 0.5 standard devia-\ntions (SD) as ‘small’ and ‘medium’. Interim analyses of the pilot study indicate that the SD of \nthe difference in change in VAS scales between LUNA and no pelvic denervation groups will be \nabout 4.0. This corresponds to small and medium effect sizes on VAS of 0.8 and 2.0 respectively \nand is consistent with other studies of chronic pelvic pain, where clinically important symptom \nalleviation has been defined as a reduction in pain score of 2 or more.\n35 To confirm or refute a \nsmall to medium effect of LUNA (0.3 SD difference or 1.2 VAS points), based on α=0.05 and \nß=0.2 (80% power), 175 participants in each group (i.e. 350 participants in total) will be required. \nConsidering a 20% loss to follow-up, the sample size is inflated to 220 participants in each group \n(i.e. 440 participants in total).\nStratification Variables\nRandomisation will be conducted using minimisation, stratified by the four variables:\na. Presence or absence of some minimal pathology (minimal endometriosis ± ablation; ad-\nhesions requiring adhesionolysis only; minimal pelvic inflammatory disease)\nb. Site of pain (presence of central pain or not)\nc. Parity of the woman (nulliparous or parous)\nd. Whether the woman is sexually active or not\nThe first two variables form the prespecified subgroup analyses, and the other two variables are \nincluded as having impact on dysmennorhoea and dyspareunia respectively.\nChapter 4\n\n55\nData Analysis\nType of analysis\nIntention to Treat\nThe main analysis to address the principal research questions will be undertaken using the one-\nyear follow-up data. The mean differences in VAS pain scores; sexual satisfaction and life quality \nscores in the two groups will be comparing using a two-sample t-test. The rates of women with \nclinically significant (2 VAS point) alleviation of pain symptoms will also be compared produc-\ning a Peto odds ratio estimate with 95% confidence intervals (Mantel-Haenzel test). Baseline \ncharacteristics of the participants enrolled in the two groups will be compared to ensure that \nrandomisation has produced comparable groups of participants.) The use of additional treatment \n(co-intervention) for pelvic pain following LUNA or no pelvic denervation will be assessed for any \nsystematic difference between the two groups. \nSubgroup analyses are limited by statistical power and can produce spurious results particularly \nif many are undertaken. Our literature review\n16 and consultation with gynaecologists,19 suggests \nthat the effectiveness of LUNA may be greater for central compared to non-central pain and if \nthere is no associated pathology. Therefore, we have chosen to limit secondary analyses to these \nsubgroups only. The LUNA trial is powered to detect a small to medium overall difference and if a \nlarger treatment benefit is found then other subgroup analyses will be undertaken, but interpreted \nappropriately cautiously. The LUNA trial is powered overall at 80% to detect a 0.3 SD difference \nin effect. Our pilot study shows that 60% of participants have mainly central pain and 70% have \nno pathology. Hence, in the subgroup with central pain the power will be 80% to detect a 0.4 \nSD treatment effect. In the subgroup with no pathology the power will be 80% to detect a 0.35 \nSD treatment effect. \nFrequency of analyses\nInterim analysis at completion of 6 months follow-up of the first 60 patients has already been \ncompleted as pilot study. Bi-annual analyses of recruitment, compliance and loss to follow-up are \nbeing carried out for LUNA Trial Management Committee. An annual interim analysis of effective-\nness is done for confidential review by independent Data Monitoring Committee to determine \nwhether the principal question has been answered and to monitor adverse events. Main analyses \nof effectiveness of LUNA will be done at completion of one year follow up of the total study sam-\nple. Additional analysis of long-term effects will be done at two and three-year follow-up points.\nEthical considerations\nWhen there is uncertainty about the appropriate therapy, clinical trials are the scientifically and \nethically best way to resolve this uncertainty and thereby benefit both the individual participants \nand all others concerned in their care.\n36 The need for a “sham” incision in this trial is the main \nethical issue and it is required because without it the participants cannot be kept blinded. The \nLUNA study protocol\n\n56\npurpose of blinding is to prevent various biases from affecting the results. The need for blinding \nin surgical trials has been emphasised in the medical literature,37 and there is empirical evidence \nthat inadequacies of blinding in randomisation lead to exaggeration of therapeutic efficacy in ran-\ndomised trials.\n22 Blinding of participants in surgical trials is clearly indicated when the intervention \nprimarily treats symptoms and when the outcomes are based on participants’ own assessments. \nLUNA is an intervention for treating chronic pelvic pain (a diagnosis based on symptoms) and the \noutcome assessment is based on participants’ responses on a visual analogue pain scale and \nquality of life instruments. Hence, blinding through the use of a “sham” incision is justified if bias \nis to be avoided in the LUNA trial and this approach has been used in a previous trial of LUNA.\n10 \nEthical approval for the LUNA trial has already been obtained from the West Midlands Multicentre \nResearch Ethics Committee (MREC).  \nConduct of the Study and Data Management\nThe trial will be coordinated from the University of Birmingham Clinical Trials Unit (BCTU). Each \ninvestigating centre will carry out the study in accordance with the study protocol and to the \nMRC guidelines on Good Clinical Practice in Clinical Research (1998). Patients will be invited to \nparticipate if they fulfil all the inclusion criteria and do not have any exclusion criteria. They will \nbe provided with a laparoscopy and LUNA trial information leaflet and signed consent obtained \nprior to laparoscopy. Consenting patients should be asked to complete the Enrolment Question-\nnaire. The final decision to enrol patients in the trial will depend on the findings at laparoscopy \nwhen the surgeon will perform LUNA or not after determining eligibility as shown in Baseline \nData and Randomisation Form. At the end of the procedure the surgeon will complete the Post \nSurgery Form. All three forms for each participant enrolled will be photocopied to keep a record \nat the participating centre and the originals will be sent to the BCTU. At 3, 6, 12 months after \nenrolment, the Follow-up Questionnaire will be mailed to the participants with a pre-paid self \naddressed envelope. Recruitment is expected to take 12 months and follow up for the main \nendpoints another 12 months. At completion of the main study, further Follow-up Questionnaire \nwill be mailed out at 2, 3, 5 and 10 years.\nChapter 4\n\n57\nReferences\n(1)  Stones RW. Chronic Pelvic Pain. Personal Assessment in Continuing Education. 97/01th \nedition ed. London: Royal College of Obstetricians and Gynaecologists; 1997.\n(2)  Sundell G, Milsom I, Andersch B. Factors influencing the prevalence and severity of dys-\nmenorrhoea in young women. BJOG 1990 Jul;97(7):588-94.\n(3)  Davies L, Ganger K, Drummond M, Saunders D, Beard R. The economic burden of in-\ntractable gynaecological pain. J Obstet Gynaecol 1992;12(S2):54-6.\n(4)  Zondervan KT, Yudkin PL, Vessey MP , Dawes MG, Barlow DH, Kennedy SH. Prevalence \nand incidence of chronic pelvic pain in primary care: evidence from a national general \npractice database. BJOG 1999;106(11):1149-55.\n(5)  Vercellini P , Fedele L, Arcaini L, Bianchi S, Rognoni MT, Candiani GB. Laparosco-\npy in the diagnosis of chronic pelvic pain in adolescent women. J Reprod Med 1989 \nOct;34(10):827-30.\n(6)  Henzl MR. Dysmenorrhoea; achievements and challenge. Sex Med Today 1985;9:8-12.\n(7)  Howard FM. The role of laparoscopy in chronic pelvic pain: promise and pitfalls. Obstet \nGynecol Surv 1993;48(6):357-87.\n(8)  Porpora MG, Gomel V. The role of laparoscopy in the management of pelvic pain in \nwomen of reproductive age. Fertil Steril 1997;68(5):765-79.\n(9)  Ghaly AFF. The psychological and physical benefits of pelvic ultrasonography in patients \nwith chronic pelvic pain and negative laparoscopy. A random allocation trial. J Obstet \nGynaecol 1994;14:269-71.\n(10)  Sutton CJ, Ewen SP , Whitelaw N, Haines P . Prospective, randomized, double-blind, con-\ntrolled trial of laser laparoscopy in the treatment of pelvic pain associated with minimal, \nmild, and moderate endometriosis. Fertil Steril 1994;62(4):696-700.\n(11)  Peters AA, Trimbos-Kemper GC, Admiraal C, Trimbos JB, Hermans J. A randomized \nclinical trial on the benefit of adhesiolysis in patients with intraperitoneal adhesions and \nchronic pelvic pain. BJOG 1992 Jan;99(1):59-62.\n(12)  Frankenhauser G. Die Bewegungenerven der Gebarmutter. Z Med Nat Wiss 1864;1:35.\n(13)  Latarjet A, Roget P . Le plexus hypogastrique chez la femme. Gynecol Obstet (Paris) \n1922;6:225.\n(14)  Doyle JB. Paracervical uterine denervation by transection of the cervical plexus for the \nrelief of dysmenorrhoea. Am J Obstet Gynecol 1955;70:1.\n(15)  Doyle JB, Des Rosiers JJ. Paracervical uterine denervation for the relief of pelvic pain. \nClinical Obstet Gynecol 1963;6:742-53.\n(16)  Khan KS, Khan SF, Nwosu CR, Dwarakanath LS, Chien PFW. Laparoscopic uterosacral \nnerve ablation in chronic pelvic pain: An overview. Gynaecol Endosc 1999;8(5):257-65.\nLUNA study protocol\n\n58\n(17)  Wilson ML, Farquhar CM, Sinclair OJ, Johnson NP . Surgical interruption of pelvic nerve \npathways for primary and secondary dysmenorrhoea. Cochrane Database Syst Rev \n2000.\n(18)  Stones RW, Mountfield J. Interventions for treating chronic pelvic pain in women. \nCochrane Database of Systematic Reviews 2000.\n(19)  Daniels J, Gray R, Khan K, Gupta J. Laparoscopic uterine nerve ablation: A survey of \ngynaecological practice in the UK. Gynaecol Endosc 2000;9(3):157-9.\n(20)  Sutton C, Whitelaw N. Laparoscopic uterosacral nerve ablation for intractable dysmenor-\nrhoea. In: Sutton C, editor. Endoscopic surgery for gynaecologists.London: WB Saun-\nders; 1993. p. 159-63.\n(21)  Zelen M. The randomization and stratification of patients to clinical trials. J Chronic Dis \n1974;27(7-8):365-75.\n(22)  Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of \nmethodological quality associated with estimates of treatment effects in controlled trials. \nJAMA 1995;273(5):408-12.\n(23)  Clarke PR, Spear FG. Reliability and sensitivity in the self assessment of wellbeing. Bull Br \nPsychol Soc 1964;17:55.\n(24)  Bond MR, Pilowsky I. Subjective assessment of pain and its relationship to the administra-\ntion of analgesics in patients with advanced cancer. J Psychosom Res 1966;10(2):203-8.\n(25)  Revill SI, Robinson JO, Rosen M, Hogg MI. The reliability of a linear analogue for evaluat-\ning pain. Anaesthesia 1976;31(9):1191-8.\n(26)  Carlsson AM. Assessment of chronic pain. I. Aspects of the reliability and validity of the \nvisual analogue scale. Pain 1983;16(1):87-101.\n(27)  Jensen MP , Turner JA, Romano JM, Fisher LD. Comparative reliability and validity of \nchronic pain intensity measures. Pain 1999;83(2):157-62.\n(28)  Jehu D. Sexual dysfunction- a behavioural approach to causation, assessment, and \ntreatment. Toronto: John Wiley and sons; 1979.\n(29)  Thirlaway K, Fallowfield L, Cuzick J. The Sexual Activity Questionnaire: a measure of \nwomen’s sexual functioning. Qual Life Res 1996;5(1):81-90.\n(30)  Hudson W. Index of sexual satisfaction. The clinical measurement package: a field man-\nual. Chicago: Dorsey Press; 1982. p. 190-1.\n(31)  Stead ML, Crocombe WD, Fallowfield LJ, Selby P , Perren TJ, Garry R, et al. Sexual activ-\nity questionnaires in clinical trials: acceptability to patients with gynaecological disorders. \nBJOG 1999;106(1):50-4.\n(32)  Guyatt GH, Feeny DH, Patrick DL. Measuring health-related quality of life. Ann Intern Med \n1993;118(8):622-9.\n(33)  Fedele L, Marchini M, Acaia B, Garagiola U, Tiengo M. Dynamics and significance of \nplacebo response in primary dysmenorrhea. Pain 1989;36(1):43-7.\nChapter 4\n\n59\n(34)  Baker PN, Symonds EM. The resolution of chronic pelvic pain after normal laparoscopy \nfindings. Am J Obstet Gynecol 1992;166(3):835-6.\n(35)  Zullo F, Pellicano M, DeStefano R. Efficacy of Laparoscopic Pelvic Denervation in Central-\ntype Chronic Pelvic Pain: A multicentre study. J Gynecol Surg 1996;12:35-40.\n(36)  Chalmers T. The clinical trial. Milbank Memorial Fund Quarterly 1981;59:339-42.\n(37)  Johnson AG, Dixon JM. Removing bias in surgical trials. BMJ 1997;314(7085):916-7.\nLUNA study protocol\n\n60\n\n\n61\nChapter 5\nLaparoscopic Uterosacral Nerve Ablation \n(LUNA) for alleviating Chronic Pelvic Pain:  \na Randomized Trial \n\n62\nChapter 5\nAbstract\nContext\nChronic pelvic pain (CPP) is a common condition with a major impact on health-related quality \nof life, work productivity and health care use. Operative interruption of nerve trunks in the utero-\nsacral ligaments, believed to carry pain from the uterus and cervix, by laparoscopic uterosacral \nnerve ablation (LUNA) is a treatment option for patients with chronic pelvic pain. \nObjective\nTo assess the effectiveness of LUNA in patients with chronic pelvic pain.\nDesign, Setting, and Participants\nRandomized controlled trial of 487 women with chronic pelvic pain lasting longer than 6 months \nwithout or with minimal endometriosis, adhesions, or pelvic inflammatory disease, who were \nrecruited to the study by consultant gynecological surgeons from 18 UK hospitals between Feb-\nruary 1998 and December 2005. Follow-up was by postal questionnaires at 3 and 6 months, 1, \n2, 3 and 5 years.\nIntervention\nBilateral LUNA or laparoscopy without pelvic denervation (no LUNA); participants blinded to the \ntreatment allocation.\nMain outcome measures\nThe primary outcome was pain, assessed by a visual analogue scale. Data concerning the 3 \ntypes of pain (non-cyclical, dysmenorrheal, and dyspareunia) were analyzed separately as was \nthe worst pain level experienced from any of these 3 types of pain. Secondary outcomes were \nquality of life (EuroQoL EQ-5D and EQ-VAS).\nResults\nAfter a median follow up of 69 months, there were no significant differences between the visual \nanalogue pain scales for for the worst pain reported (mean difference between LUNA group \nand No LUNA group: -0·04cm, 95% confidence interval (CI), -0·33 to 0·25cm; p=0·80), non-\ncyclical pain (-0·11cm, 95%CI -0·50 to 0·29cm; p=0·60) dysmenorrhoea (-0·09cm, 95%CI -0·49 \nto 0·30cm; p=0·60), or dyspareunia (0·18cm, 95%CI  -0·22 to 0·62cm; p=0·40). No differences \nwere observed between the LUNA group and the no LUNA group for quality of life.\nChapter 5\n\n63\nConclusions\nAmong women with chronic pelvic pain, LUNA did not result in improvements in pain, dysmenor-\nrheal, dyspareunia or quality of life, compared to laparoscopy without pelvic denervation.\nTrial Registration\nControlled-trials.com identifier ISRCTN number: 41196151 \nIntroduction\nChronic pelvic pain (CPP) in women is as common as asthma and chronic back pain in primary \ncare1.2 CPP continues to be one of the most difficult and perplexing of women’s health problems \nwith a multifactorial etiology.3 Chronic pelvic pain has a major impact on health-related quality of \nlife, work attendance and productivity4 and health care utilization, accounting for 40% of referrals \nfor diagnostic laparoscopy,5 and is an important contributor to health care expeditures.6\nTreatmentsfor CPP are often unsatisfactory.7 As part fo the evaluation and management phase, \npatients often undergo diagnostic laparoscopy, 8 but actionable pathology is found only occa-\nsionally identified.9, 10 Negative findings at laparoscopy, and follow up with ultrasound, may pro-\nvide reassurance and relief to some patients, 11 but in the absence of underlying pathology no \nestablished gynaecological treatments are available. \nNerve plexuses and parasympathetic ganglia in the uterosacral ligaments 12, 13 are thought to \ncarry pain signals from the uterus, cervix, and other pelvic structures. Conventional open vagi-\nnal and abdominal procedures have been used to interrupt these nerve trunks by dividing the \nattachments of the uterosacral ligaments to the cervix in women with dysmenorrhoea. 14, 15 In \npart because these procedures are invasive and carry risk they have not been widely adopted. \nLaparoscopic uterosacral nerve ablation (LUNA) is performed  after diagnostic laparoscopy and \ncan be completed using lasers or electro-diathermy, and has become increasingly used. 16, 17 \nSystematic reviews of the current research evidence on LUNA’s efficacy are inconclusive.18, 19 and \na National Institute of Clinical Excellence report suggested that there was not sufficient evidence \nof its value. 20 Clinicians’ beliefs about LUNA’s effectiveness vary widely21 and LUNA remains a \ncontroversial procedure. We conducted a single-blind, randomized controlled trial comparing \nLUNA with laparoscopy without pelvic denervation.\nMethods\nThe LUNA Trial was a multicenter, prospective randomized intervention trial with patient-blinded, \npatient-rated assessment of outcomes to evaluate LUNA. 22, 23 Ethics committee approval was \nobtained from the West Midlands Multi-centre Research Ethics Committee (reference number \nLUNA: a randomised trial\n\n64\n99/7/03). Patients presenting to gynaecology outpatient clinics with CPP (dysmenorrhoea, non-\ncyclical pain or dyspareunia) of longer than 6-month duration located within between and below \nthe anterior iliac crests, and who were undergoing diagnostic laparoscopy for differential diagno-\nsis of CPP , were invited to participate in our study.\nWomen were ineligible if they had previous LUNA, hysterectomy or therapeutic procedures for, \nor diagnosis of, moderate to severe endometriosis or major pelvic inflammatory disease. Written \nconsent was obtained before surgery. At laparoscopy, women were excluded if they were found \nto have more than minimal pathology (i.e. American Fertility Society endometriosis score >5 or \nsignificant adhesions or serious adnexal pathology) or if bilateral LUNA was technically unfea-\nsible. Intraoperatively, eligible patients were randomized by telephone call to the Birmingham \nUniversity Clinical Trials Unit, or through its Internet based randomization service, to LUNA or no \ndenervation.\nRandomization involved a computer minimization program to balance group allocations for site \nof pain, parity, self reported sexual activity status and presence or absence of minimal pathology. \nTo avoid eligibility classification bias, treatment allocation was issued only after the surgeon had \ninspected the pelvis and ensured that the patient fulfilled all of the inclusion criteria and did not \nhave any of the exclusion criteria. \nThose allocated to LUNA had the procedure performed immediately by the same laparoscopic \nsurgeon, who had prior experience of the technique and who followed a common protocol.\n24 In \na typical case, after inspection of the posterior leaf of the broad ligament to identify ureters and \nany pelvic venous congestion, the ablation was performed as close to the posterior aspect of the \ncervix as possible and continued for a minimum of 1 cm posterolaterally on either side with the \nintended aim of destroying the sensory nerve fibres and the secondary ganglia as they left the \nuterus and lie within the uterosacral ligaments.\nFull or partial transaction of the ligaments was achieved bilaterally with laser or electro-diathermy, \naccording to the surgeons’ preference. In centers where surgeons used 2 secondary ports to \nperform LUNA, a second 5 mm incision was made in the patients in the no LUNA group through \nthe skin in an area corresponding to where an additional port site was made. This approach of \nsham incisions has been used in a previous trial\n25 and was ethically justified to help avoid bias in \nthe patient rated assessment of a subjective outcome like pain.26 All women were asked at least \n12 months after randomization whether they believed they had LUNA or no LUNA.\nBaseline data were collected following consent and prior to laparoscopy. At 3 and 6 months \nfollowing randomization, and at 1, 2, 3 and 5 years, the same questionnaires were mailed to \nChapter 5\n\n65\npatients with a pre-paid return envelope. Non responders were followed up through postal and/\nor telephone reminders or, if this failed, via their general practitioners.\nThe primary outcome of pain was rated using a 10 cm visual analogue scale (VAS), 27 anchored \nat one end as  no pain at all  and the other as worst imaginable pain. The VAS ratings were \nobtained for each of the 3 types of pain: non-cyclical pain (pain at any other times, other than \nduring periods or during intercourse), dysmenorrhoea (pain during your periods) and dyspareunia \n(pain during intercourse. Women rated the degree of pain by placing a mark on the line and scale \nvalues were obtained by measuring the distance from zero to that mark. This measurement is \nvalidated as a sensitive measure for large group comparisons.28\nThe secondary outcomes were health-related quality of life measured using a generic instrument \n(EuroQOL EQ-5D and EQ-VAS),29 and a proforma recording the need for additional treatments, \nresource usage, days off work and complications of surgery. The EQ-5D is measured on a 0-1 \nscale based on responses to five questions about life quality and the EQ-VAS is measured on a \n0 - 100 scale.\nThe sample size was powered to detect a small to medium effect of LUNA in alleviating pain \nsymptoms compared to laparoscopy without pelvic denervation. To confirm or refute a 0·3 \nstandard deviation effect size,30 equivalent to a difference between groups of 1.2cm on the visual \nanalogue scale, at a two-sided α=0·05 and ß=0·2 (80% power), 175 women in each group (i.e. \n350 in total) were required. Allowing for a 20% loss to follow-up, the recruitment target was \ninflated to 420.  Considering that existing research in this condition has shown substantial rates \nof loss to follow-up, recruitment continued until the end of the funding period when 487 women \nwere included. An independent data monitoring committee reviewed confidential interim analy-\nses annually and, recommended at each meeting continuing recruitment because the data were \ninconclusive.\nAll participants were analyzed in the group to which they were allocated, using all available data, \nand SAS statistical software version 9 (SAS Institute, Cary, North Carolina). Baseline character -\nistics of the patients enrolled in the two groups were compared to ensure that randomization \nhas produced comparable groups. Data for the various outcome measures were presented as \nmeans and mean differences over time, with 95% confidence intervals (CIs).\nFor the primary outcome, we analyzed data separately concerning the three types of pain (non-\ncyclical pain, dysmenorrhoea, and dyspareunia) and performed an analysis of the worst pain level \nexperienced from any of the three types of pain. Comparisons between groups over time were \nundertaken using repeated measures analyses,\n31 a statistically efficient approach that includes all \nLUNA: a randomised trial\n\n66\nof the follow up data collated during the study, increasing power over analysis of data at individual \ntimepoints. Pain scores at 12 months were compared using standard two sample t-tests.\nThe principal analysis for the worst pain experienced from any of the three types of pain was an \nintention to treat analysis using multiple imputation. 32,33 To investigate the impact of missing data, \nthe analyses of the individual types of pain at 12 months were repeated using the last observa-\ntion carried forward method of imputation. All comparisons were two-sided and were considered \nstatistically significant if p<0.05.\nSub group analyses were chosen on the basis of anticipated variations in pain and potential \nbenefit from LUNA, but were considered hypothesis generating. Pre-specified sub-groups were \nthose used to stratify the randomization, namely site of pain (central, not central) and presence \nor absence of some minimal pathology for all types of pain, parity (nulliparous, multiparous) and \nwhether women were sexually active or not. \nResults\nBetween February 1998 and December 2005, 487 women were randomized into the LUNA trial \nfrom 18 centers. A further 105 were consented but were found at laparoscopy to have pathol-\nogy that made them ineligible for randomization, or anatomy that precluded LUNA from being \nperformed. No women in the control group received LUNA, whereas five women allocated LUNA \nultimately did not have LUNA performed bilaterally due to technical difficulties, but were analyzed \nin the LUNA group. Figure 1 shows the trial profile. Baseline pain data were missing on 26 (5%) \nparticipants, including 10 participants from one centre where a batch of baseline data was lost.\n \nBaseline characteristics of the women did not differ between the two groups (Table 1) and there \nwere not any differences in cointerventions between the groups. Blinding was maintained, apart \nfrom 13 women who asked to be informed of their allocation, although in a sub-sample of 211 \nwho had been blinded to their allocation, 122 correctly guessed their allocation (58%) .\nThe majority of women had multiple types of pain. A total of 168 participants reporting non-\ncyclical pain dysmenorrhoea, and dyspareunia (36%). and 113 reporting dysmenorrhoea and \ndyspareunia (24%). In 266 (54%) women, there was no obvious pathology identified, suggestive \nof primary dysmenorrhoea. A further 146 had minimal to mild endometriosis (30%), 86 had adhe-\nsions (18%), 13 had pelvic inflammatory disease (3%), and 35 had other visible pathologies (7%), \nthe majority being subserosal fibroids and small ovarian cysts.\nChapter 5\n\n67\nFigure 1: Selection of participants for the Laparoscopic Uterosacral Nerve Ablation (LUNA) Trial\na  Twenty-six patients had more than 1 reason for exclusion.\nb  Indicates worst pain level experienced from any of the 3 types of pain analyzed \n (noncyclical pain, dysmenorrhea, and dyspareunia).\nMedian time in the study was 69 months with 72% of participants having reached 5 years of \nfollow-up. The differences in pain are shown in Figure 2. The results are presented as the mean \npain scores from the VAS for each randomized group in the upper graph, and as the difference \nbetween the two groups in the lower graph, where a positive value indicates better pain relief with \nLUNA than without LUNA. There was no difference reported for the worst pain level experienced \nfrom any of the 3 types of pain (mean difference between the LUNA group vs the no LUNA group, \n−0.02 cm, 95% CI −0.61 to 0.65 cm; p=0.90 at 12 months; −0.04 cm, 95% CI, −0.33 to 0.25 \ncm; p =0.80 over all time points), noncyclical pain (0.17 cm, [95% CI −0.40 to 0.74 cm; p =.50 \nat 12 months; −0.11 cm, 95% CI, −0.50 to 0.29 cm; p =0.60 over all time points), dysmenorrhea \n(−0.10 cm, 95% CI −0.70 to 0.50 cm; p =0.70 at 12 months; −0.09 cm, 95% CI −0.49 to 0.30 \ncm; p =0.60 over all time points), or dyspareunia (0.34 cm, 95% CI −0.34 to 1.02 cm; p =0.30 at \n12 months; 0.18 cm, 95% CI −0.22 to 0.62 cm; p =0.40 over all time points).\nLUNA: a randomised trial\nassurance and relief to some patients11\nbut in the absence of underlying pa-\nthology, no established gynecological\ntreatments are available.\nNerve plexuses and parasympa-\nthetic ganglia in the uterosacral liga-\nments12,13 are thought to carry pain sig-\nnals from the uterus, cervix, and other\npelvic structures. Conventional open\nvaginaland abdominalprocedureshave\nbeen used to interrupt these nerve\ntrunks by dividing the attachments of\nthe uterosacral ligaments to the cervix\nin women with dysmenorrhea.14,15 In\npart because these procedures are in-\nvasive and carry risk and their efficacy\nhas not been established, they have not\nbeen widely adopted. Laparoscopic\nuterosacral nerve ablation (LUNA) is\nperformed after diagnostic laparos-\ncopy and can be completed using la-\nsers or electrodiathermy and has be-\ncome increasingly used.16,17 Systematic\nreviews of the current research evi-\ndence on LUNA’s efficacy are incon-\nclusive.18,19 A NationalInstituteof Clini-\ncal Excellence report suggested that\nthere was not sufficient evidence of\nits value. 20 Clinicians’ beliefs about\nLUNA’s effectiveness vary widely21 and\nLUNA remainsacontroversial proce-\ndure.We conductedasingle-blind, ran-\ndomized controlled trial comparing\nLUNA with laparoscopy without pel-\nvic denervation.\nMETHODS\nThe LUNA trial wasamulticenter, pro-\nspective, randomized intervention trial\nwith patient-blinded, patient-rated as-\nsessment of outcomes to evaluate\nLUNA. 22,23 Ethics committee approval\nwas obtained from the West Midlands\nMulticenter Research Ethics Commit-\ntee (reference No. 99/7/03). Patients\npresenting to gynecology outpatient\nclinics with chronic pelvic pain (non-\ncyclical pain, dysmenorrhea, or dyspa-\nreunia) lasting longer than 6 months,\nlocated within and below the anterior\niliac crests, and who were undergoing\ndiagnostic laparoscopy for differential\ndiagnosis of chronic pelvic pain were\ninvited to participate in our study.\nWomen were ineligible if they had\npreviousLUNA,hysterectomy,or thera-\npeutic procedures for, or diagnosis of,\nmoderate to severe endometriosis or\nmajor pelvic in/f_lammatory disease.\nWritten informed consent was ob-\ntained before surgery. At laparoscopy,\nwomen were excluded if they were\nfound to have more than minimal pa-\nthology (ie, American Fertility Society\nendometriosis score 5 or signi/f_icant\nadhesions or serious adnexal pathol-\nogy) or if bilateral LUNA was techni-\ncally unfeasible. Intraoperatively, eli-\ngible patients were randomized via a\ntelephone call to the Birmingham Uni-\nversity Clinical Trials Unit, or through\nits Internet-based randomization ser-\nvice, to the LUNA group or the no\nLUNA group.\nRandomization involvedacom-\nputer minimization program to bal-\nance group allocations for site of pain,\nparity, self-reported sexual activity sta-\ntus, and presence or absence of mini-\nmal pathology.To avoid eligibilityclas-\nsi/f_icationbias, treatmentallocationwas\nissued only after the surgeon had in-\nspected the pelvis and ensured that the\npatient ful/f_illedall of the inclusion cri-\nteria and did not have any of the ex-\nclusion criteria.\nThose allocated to the LUNA group\nhad the procedure performed immedi-\nately by the same laparoscopic sur-\ngeon who had prior experience of the\ntechnique and who followedacom-\nmon protocol.24 Inatypical case, after\ninspection of the posterior leaf of the\nbroad ligament to identify ureters and\nany pelvic venous congestion, the ab-\nlation was performed as close to the\nposterior aspect of the cervix as pos-\nsible and continued foraminimum of\n1 cm posterolaterally on eithersidewith\nthe intended aim of destroying the sen-\nsorynerve/f_ibersand the secondarygan-\nglia as they left the uterus and lie within\nthe uterosacral ligaments.\nFull or partial transactionof the liga-\nments was achieved bilaterally with la-\nser or electrodiathermy, according to\nthe surgeons’ preference. In centers in\nwhich surgeons used2additional ports\nto perform LUNA,asecond 5-mm in-\nFigure 1. Selection of Participants for the Laparoscopic Uterosacral Nerve Ablation (LUNA) Trial\nPrimary pain analyses at 12 mo\n187 Included in worst pain level analysisb\n185 Included in noncyclical pain analysis\n167 Included in dysmenorrhea analysis\n142 Included in dyspareunia analysis\nPrimary pain analyses at 12 mo\n192 Included in worst pain level analysisb\n185 Included in noncyclical pain analysis\n170 Included in dysmenorrhea analysis\n138 Included in dyspareunia analysis\n5 Withdrew consent prior to 12-mo\nfollow-up\n51 Missing pain data at 12-mo follow-up\n4 Withdrew consent prior to 12-mo\nfollow-up\n48 Missing pain data at 12-mo follow-up\n243 Randomized to receive LUNA\n229 Received LUNA as assigned\n5 Did not receive LUNA due to\ntechnical difficulties\n9 Not known whether LUNA\nwas received\n244 Randomized to receive laparoscopy\nwithout pelvic denervation (no LUNA)\n231 Did not receive LUNA\nas assigned\n13 Not known whether LUNA\nwas received\n487 Randomized\n592 Patients assessed for eligibility and provided\nconsent prior to diagnostic laparoscopy\n105 Excludeda\n63 Moderate or severe endometriosis\n51 LUNA technically not feasible\n26 Moderate or severe adhesions\n3 Other pathology considered significant\n6 Moderate or severe pelvic inflammatory\ndisease\na Twenty-six patients had more than1reason for exclusion.\nbIndicates worst pain level experienced from any of the3types of pain analyzed (noncyclical pain, dysmen-\norrhea, and dyspareunia).\nLAPAROSCOPIC UTEROSACRAL NERVE ABLATION FOR CHRONIC PELVIC PAIN\n956 JAMA, September 2, 2009—Vol 302, No. 9 (Reprinted) ©2009 American Medical Association. All rights reserved.\n at University of Birmingham on October 12, 2009 www.jama.comDownloaded from \n\n68\n \n 22 \nTable 1 Baseline characteristics of participants in the Laparoscopic \nUterosacral Nerve Ablation (LUNA) trial  \n LUNA \nn (%); total=243 \nNo LUNA \n n (%); total=244 \n     \nAge1  30·6 (7·53; \n17-64) \n30·5 (7·48; \n17-57) \nClinical presentation     \nNulliparous 87 (36) 92 (38) \nSexually active 195 (80) 193 (79) \nType of pain     \nDysmenorrhoea 184 (76) 179 (73) \nDyspareunia 2 158 (65) 149 (61) \nNon-cyclical pain 159 (65) 153 (63) \nAll three types of pain 89 (37) 79 (32) \nLocation of pain     \nCentral 168 (69) 167 (68) \nLaparoscopic findings     \nAny visible pathology 110 (45) 111 (45) \nAdhesions 49 (20) 37 (15) \nPelvic inflammatory disease 4 (2) 9 (4) \nEndometriosis                     None 177 (73) 164 (67) \nMinimal 41 (17) 52 (21) \nMinimal, ablated 25 (10) 28 (11) \nPain medication     \nPainkillers3 155 (64) 156 (64) \nAnti-depressants4 16 (7) 28 (11) \nHormonal contraceptives     \nCombined oral contraceptive 33 (14) 33 (13) \nMini-pill 4 (2) 5 (2) \nMirena coil 1 (<1) 0 (0) \nContraception (not specified) 30 (12) 24 (10) \nIndication for other medication     \nIrritable bowel syndrome 2 (1) 1 (0) \nBladder pain 0 (0) 1 (0) \nUrinary tract infection 2 (1) 1 (0) \nMigraine 5 (2) 2 (1) \nAsthma 19 (8) 15 (6) \nBlood pressure 4 (2) 4 (2) \n                                            \n1 Mean (standard deviation; range)  \n2 % of sexually active women only  \n3 13 and 8 missing data items for LUNA and no LUNA respectively  \n4 20 and 20 missing data items  \nTable 1:  Baseline characteristics of participants in the Laparoscopic Uterosacral Nerve  \n Ablation (LUNA) trial\nChapter 5\n\n69\nUsing last observation carried forward for imputation of missing observations at 12 months, \nwe found no significant difference between the LUNA group and the no LUNA group for non-\ncyclical pain (0.35cm, 95% CI -0.19 to 0.88cm; p=0.2), dysmenorrhea (-0.16cm, 95% CI -0.73 \nto 0.40cm; p=0.6), or dyspareunia (0.26cm, 95%CI -0.39 to 0.92cm; p=0.4). A similar lack of \nefficacy was observed for the outcome measures concerning health-related quality of life with the \nmean difference in the EQ-5D being 0.03 (95% CI -0.03 to 0.09; p=0.3) and that in EQ-VAS being \n-0.78cm (95% CI, -3.9 to 5.4; p=0.3) at 12 months. Between 9 and 12 months, no differences \nwere found between the LUNA group and the no LUNA group in terms of days off work (27% vs \n22%, respectively, took at least 1 day off work; p=0.20) or in the number of general practitioner \nsurgery visits (mean [SD] number of visits, 1.9 [1.1] vs 2.5 [2.0], respectively; p=0.08). The pre-\nspecified subgroup analyses using repeated- measures analysis appears in Table 2.\nThere were 8 cases of minor haemorrhaging while performing the LUNA procedure and one \ncase that converted to an open procedure. One participant not allocated LUNA suffered ureteric \ndamage.\nComment\nThe LUNA trial was designed to assess the effects of LUNA, compared to no denervation, among \nwomen undergoing diagnostic laparoscopy for chronic pelvic pain. LUNA did not alleviate pain - \nnon-cyclical pain, dysmenorrhoea, or dyspareunia - or improve life quality, irrespective of the pres-\nence or absence of mild endometriosis. A Cochrane review19 did suggest a subgroup benefit for \npatients with dysmenorrhoea that our study did not find. This finding came from two small studies34, \n35 totaling 68 randomised participants, one of which34 did not have strictly concealed randomization.\n \n 23 \nTable 2 Pre-specified sub-group analyses using repeated measures analysis \n \n Worst pain Dysmenorrhea Dyspareunia Non-cyclical pain \nSubgroup Treatment effect \n(95% CI) \np \nvalue \nTreatment effect \n(95% CI) \np \nvalue \nTreatment effect \n(95% CI) \np \nvalue \nTreatment effect \n(95% CI) \np \nvalu\ne \nNulliparous 0.34 (-0.06-0.74) 0.14 (-0.28-0.54) 0.44 (-0.04-0.92) -0.02 (-0.41-0.37) \nParous -0.08 (-0.41-0.24) \n0.5 \n-0.20 (-0.55-0.15) \n0.4 \n-0.32 (-0.68-0.05) \n0.1 \n-0.13 (-0.47-0.20) \n0.6 \nNo pathology 0.10 (-0.16-0.36) -0.11 (-0.39-0.17) 0.08 (-0.22-0.39)  -0.08 (-0.34-0.18) \nAny minimal \npathology \n0.19 (-0.81-1.18) 0.2 0.64 (-0.29-1.56) 0.3 -0.54 (-1.04-0.12) 0.02 0.64 (-0.68-1.39) 0.2 \nMild/minimal \nendometriosis \n    0\n.2 (-0.59-0.62) 0.8   \nCentral pain -0.03 (0.31-0.30) -0.15 (-0.46-0.16) -0.06 (-0.41-0.30) -0.15 (-0.46-0.16) \nNo central pain 0.10 (-0.49-0.49) \n0.2 \n-0.19 (-0.74-0.36) \n0.2 \n0.21 (-0.33-0.76) \n0.1 \n0.01 (-0.5-0.51) \n0.3 \nTable 2:  Pre-specified sub-group analyses using repeated measures analysis\nLUNA: a randomised trial\n\n70\nFor the primary outcome, we ana-\nlyzed data separately concerning the 3\ntypes of pain (noncyclical pain, dysmen-\norrhea, and dyspareunia) and per-\nformed an analysis of the worst pain level\nexperienced from any of these 3 types of\npain. Comparisons between groups over\ntime were undertaken using repeated-\nmeasures analyses,\n31 a statistically effi-\ncient approach that includes all of the fol-\nlow-up data collated during the study,\nincreasing power over analysis of data at\nindividual time points. Pain scores at 12\nmonths were compared using standard\n2-sample t tests.\nThe principal analysis for the worst\npain level experienced from any of the\n3 types of pain was an intention-to-\ntreat analysis using multiple imputa-\ntion.\n31-33 To investigate the effect of miss-\ning data, the analyses of the individual\ntypes of pain at 12 months were re-\npeated using the last observation car-\nried forward method of imputation. All\ncomparisons were 2-sided and were con-\nsidered statistically significant ifP/H11021.05.\nSubgroup analyses were chosen on the\nbasis of anticipated variations in pain and\npotential benefit from LUNA, but were\nconsidered hypothesis generating. Pre-\nspecified subgroups were those used to\nstratify the randomization, namely site\nof pain (central, not central), presence\nor absence of minimal pathology for all\ntypes of pain, parity (nulliparous, mul-\ntiparous), and whether women were\nsexually active or not.\nRESULTS\nBetween February 1998 and Decem-\nber 2005, 487 women were random-\nized into the LUNA trial from 18 UK\nhospitals. A further 105 patients pro-\nvided consent but were found at lapa-\nroscopy to have pathology that made\nthem ineligible for randomization, or\nanatomy that precluded LUNA from\nbeing performed. No women in the con-\ntrol (no LUNA) group received LUNA,\nwhereas 5 women in the LUNA group\nultimately did not have LUNA per-\nformed bilaterally due to technical dif-\nficulties, but were analyzed in the\nLUNA group. F\nIGURE 1 shows the trial\nprofile. Baseline pain data were miss-\nFigure 2. Effect of Laparoscopic Uterosacral Nerve Ablation (LUNA) at 12 Months and at\nEach Time Point\nNoncyclical painB\nNo. of observations\nLUNA\nNo LUNA\n219\n228\n181\n189 185\n185 157\n165 159\n147\n104\n96 219\n228\n181\n189 185\n185 157\n165 159\n147\n104\n96\n10\n8\n4\n6\n2\n0\nMean VAS Pain Score, cm\n0 3 6 12 24 36 48 60\nTime, mo\n0 3 6 12 24 36 48 60\nTime, mo\n4\n2\n0\n–2\n–4\nDifference in Mean\nVAS Pain Score, cm\nNo. of observations\nLUNA\nNo LUNA\n228\n234\n186\n195 192\n187 158\n168 161\n151\n106\n97\nLUNA\nNo LUNA\n10\n8\n4\n6\n2\n0\nMean VAS Pain Score, cm\nWorst pain levelaA\n0 3 6 12 24 36 48 60\nTime, mo\n0 3 6 12 24 36 48 60\nTime, mo\n4\n2\n0\n–2\n–4\nDifference in Mean\nVAS Pain Score, cm\nNo. of observations\nLUNA\nNo LUNA\n177\n183\n137\n146 138\n142 121\n119 112\n112\n68\n72 177\n183\n137\n146 138\n142 121\n119 112\n112\n68\n72\n10\n8\n4\n6\n2\n0\nMean VAS Pain Score, cm\nDyspareuniaD\n0 3 6 12 24 36 48 60\nTime, mo\n0 3 6 12 24 36 48 60\nTime, mo\n4\n2\n0\n–2\n–4\nDifference in Mean\nVAS Pain Score, cm\nNo. of observations\nLUNA\nNo LUNA\n216\n228\n166\n180 170\n167 143\n144 137\n124\n91\n74 216\n228\n166\n180 170\n167 143\n144 137\n124\n91\n74\n10\n8\n4\n6\n2\n0\nMean VAS Pain Score, cm\nDysmenorrheaC\n0 3 6 12 24 36 48 60\nTime, mo\n4\n2\n0\n–2\n–4\n0 3 6 12 24 36 48 60\nTime, mo\nDifference in Mean\nVAS Pain Score, cm\n228\n234\n186\n195 192\n187 158\n168 161\n151\n106\n97\nThe graph on the right in each lettered part of the figure shows the difference in mean visual analogue scale\n(VAS) pain scores and 95% confidence intervals; values greater than 0 indicate that LUNA is a better therapy\nthan no LUNA. The error bars indicate 95% confidence intervals.\na Indicates worst pain level experienced from any of the 3 types of pain analyzed (noncyclical pain, dysmen-\norrhea, and dyspareunia).\nLAPAROSCOPIC UTEROSACRAL NERVE ABLATION FOR CHRONIC PELVIC PAIN\n958 JAMA, September 2, 2009—Vol 302, No. 9 (Reprinted) ©2009 American Medical Association. All rights reserved.\n at University of Birmingham on October 12, 2009 www.jama.comDownloaded from \nFigure 2:  Effect of Laparoscopic Uterosacral Nerve Ablation (LUNA) at 12 Months and at  \n Each Time Point\nChapter 5\n\n71\nThe LUNA trial is 4 times larger than any previously published trial evaluating neuroablation in \nchronic pelvic pain. It may be more reliable than any previous study of LUNA and was also \ndesigned to minimize bias, with concealment of allocation before randomization and blinded \noutcome assessment. Women were not told whether they received LUNA or not. Although the \nmajority were not informed of their allocation, there is a suggestion that a small proportion were \nable to guess correctly. If anything, however, this would be likely to enhance the apparent value \nof LUNA.\nLUNA was adopted by many practitioners because afferent nerves from pelvic organs passed \nthrough the uterosacral ligament and it was believed that disruption of these would reduce the \nperceived pain. Lack of efficacy in this study, and in prior studies, provide evidence that the ana-\ntomical and physiological picture of chronic pelvic pain is more complicated. Anatomically, there \nare at least five pathways that transmit signals from noxious stimuli in the pelvis. These nerve \ntrunks vary in location and can intersect, with the potential for neuronal cross-talk. LUNA may \nobliterate some of the nerve fibers, but others are interwoven with the pelvic arteries and ureters.\nAggressive ablation more laterally risks damaging the ureter, so most procedures are compro-\nmised in their ability to achieve complete neurodestruction.\nFurthermore, it is conceivable that nerves regenerate, or there may be a retrograde rerouting of \nnerve pathways. \nWe followed participants for longer than 6 months because laparoscopy may have a placebo \neffect for up to 3-6 months.\n36, 37  We found no benefit for LUNA at any time point, but found im-\nprovement in pain at 3 months in both the LUNA group and the No LUNA group. This early pain \nreduction may be a placebo effect; attributable to the reassurance provided by the laparoscopy \nthat there was no serious pathology. A comparison of diagnostic laparoscopy against no laparo-\nscopic investigation would be required to establish benefit. Alternatively, it could be a regression \nto the mean effect, with women more likely to undergo laparoscopic investigation when their pain \nis at its worst, rather than average level. \nThis study has several limitations. We did not obtain follow-up data on all women but drop out \nrates were similar in each group and multiple imputation and last observation carried forward \nanalyses produced near identical findings to those of the observed data. Given that we observed \nno effect of LUNA, the question arises whether this might be due to type II error (i.e. inadequate \nstatistical power). A clinically significant difference in pain has been defined as 2 points on a \n10-point (cm) VAS for chronic pelvic pain \n36, 38 and also for other types of pain 39 whereas our trial \nhad the power to detect a 1.2 point difference at 12 months and even smaller differences over \nLUNA: a randomised trial\n\n72\ntime. In every comparison, the confidence intervals around the mean difference in VAS scores \nbetween the groups were less than ±1.2 points. Taking worst pain experienced at 12 months \nas an example, the pain score was 0.02cm lower in the LUNA group than no LUNA group and \nthe 95% confidence interval was 0.61cm lower to 0.65cm higher (i.e. well below the clinically \nsignificant improvement). \nIn conclusion, among women with chronic pelvic pain, LUNA did not result in improvements in \npain, dysmenorrheal, pelvic pain or quality of life, compared to laparoscopy without pelvic den-\nervation. \nChapter 5\n\n73\nReferences\n(1)  Zondervan KT, Yudkin PL, Vessey MP , Dawes MG, Barlow DH, Kennedy SH. Prevalence \nand incidence of chronic pelvic pain in primary care: evidence from a national general \npractice database. BJOG 1999;106(11):1149-55.\n(2)  Latthe P , Latthe M, Say L, Gulmezoglu M, Khan KS. WHO systematic review of prev-\nalence of chronic pelvic pain: a neglected reproductive health morbidity. BMC Public \nHealth 2006;6(1):177.\n(3)  Latthe P ., Mignini L, Gray R, Hills R, Khan K. Factors predisposing women to chronic \npelvic pain: systematic review. BMJ 2006;332(7544):749-51.\n(4)  Sundell G, Milsom I, Andersch B. Factors influencing the prevalence and severity of dys-\nmenorrhoea in young women. BJOG 1990;97(7):588-94.\n(5)  Howard FM. The role of laparoscopy in chronic pelvic pain: promise and pitfalls. Obstet \nGynecol Surv 1993;48(6):357-87.\n(6)  Winkel CA. Modeling of medical and surgical treatment costs of chronic pelvic pain: New \nparadigms for making clinical decisions. Managed Care 1999;5(5):S276-S290.\n(7)  Mathias SD, Kuppermann M, Liberman RF, Lipschutz RC, Steege JF. Chronic pelvic \npain: prevalence, health-related quality of life, and economic correlates. Obstet Gynecol \n1996;87(3):321-7.\n(8)  RCOG Guidelines and Audit Committee. The initial management of chronic pelvic pain.  \nRoyal College of Obstetricians and Gynaecologists; 2005. Report No.: Guideline 41.\n(9)  Howard FM. The role of laparoscopy as a diagnostic tool in chronic pelvic pain. Best \nPractice & Research Clinical Obstetrics & Gynaecology 2000;14(3):467-94.\n(10)  Porpora MG, Gomel V. The role of laparoscopy in the management of pelvic pain in \nwomen of reproductive age. Fertil Steril 1997;68(5):765-79.\n(11)  Ghaly AFF. The psychological and physical benefits of pelvic ultrasonography in patients \nwith chronic pelvic pain and negative laparoscopy. A random allocation trial. J Obstet \nGynaecol 1994;14:269-71.\n(12)  Campbell RM. Anatomy and physiology of sacrouterine ligaments. Am J Obstet Gynecol \n1950;59:1.\n(13)  Latarjet A, Roget P . Le plexus hypogastrique chez la femme. Gynecol Obstet (Paris) \n1922;6:225.\n(14)  Doyle JB. Paracervical uterine denervation by transection of the cervical plexus for the \nrelief of dysmenorrhoea. Am J Obstet Gynecol 1955;70:1.\n(15)  Doyle JB, Des Rosiers JJ. Paracervical uterine denervation for the relief of pelvic pain. \nClinical Obstet Gynecol 1963;6:742-53.\n(16)  Latthe P , Powell R, Daniels J et al. Variation in practice of Laparoscopic uterosacral nerve \nablation: A European survey. J Obstet Gynaecol 2004;24(5):547-51.\nLUNA: a randomised trial\n\n74\n(17)  Daniels J, Gray R, Khan K, Gupta J. Laparoscopic uterine nerve ablation: A survey of \ngynaecological practice in the UK. Gynaecol Endosc 2000;9(3):157-9.\n(18)  Khan KS, Khan SF, Nwosu CR, Dwarakanath LS, Chien PFW. Laparoscopic uterosacral \nnerve ablation in chronic pelvic pain: An overview. Gynaecol Endosc 1999;8(5):257-65.\n(19)  Proctor ML, Latthe PM, Farquahr CM, Khan KS, Johnson NP . Surgical interruption of \npelvic nerve pathways for primary and secondary dysmenorrhoea. Cochrane Database \nSyst Rev 2005;4.\n(20)  National Institute for Health and Clinical Excellence. Interventional Procedure Guidance \n234. London, England. National Institute for Health and Clinical Excellence; 2007\n(21)  Latthe PM, Braunholtz DA, Hills RK, Khan KS, Lilford R. Measurement of beliefs about \neffectiveness of laparoscopic uterosacral nerve ablation. BJOG 2005;112(2):243-6.\n(22)  Latthe P ., Khan KS, Gupta JK et al. A randomised controlled trial to assess the efficacy of \nLaparoscopic Uterosacral Nerve Ablation (LUNA) in the treatment of chronic pelvic pain: \nThe trial protocol. Biomed Central Women’s Health 2003 8;3.\n(23)  University of Birmingham Clinical Trials Unit. The LUNA Trial. http://www luna bham ac uk \n2003.\n(24)  Sutton C, Whitelaw N. Laparoscopic uterosacral nerve ablation for intractable dysmenor-\nrhoea. In: Sutton C, editor. Endoscopic surgery for gynaecologists.London: WB Saun-\nders; 1993. p. 159-63.\n(25)  Sutton CJ, Ewen SP , Whitelaw N, Haines P . Prospective, randomized, double-blind, con-\ntrolled trial of laser laparoscopy in the treatment of pelvic pain associated with minimal, \nmild, and moderate endometriosis. Fertil Steril 1994;62(4):696-700.\n(26)  Johnson AG, Dixon JM. Removing bias in surgical trials. BMJ 1997;314(7085):916-7.\n(27)  Revill SI, Robinson JO, Rosen M, Hogg MI. The reliability of a linear analogue for evaluat-\ning pain. Anaesthesia 1976;31(9):1191-8.\n(28)  Jensen MP , Turner JA, Romano JM, Fisher LD. Comparative reliability and validity of \nchronic pain intensity measures. Pain 1999;83(2):157-62.\n(29)  Kind P . The EuroQol instrument: An index of health-related quality of life. In: Spliker B, \neditor. Quality of life and PharmacoEconomics in clinical trials. 2nd. ed. Philadelphia: Lip-\npincott-Raven; 1996.\n(30)  Cohen J. Statistical power analysis for the behavioural sciences. Revised ed. New York: \nAcademic Press; 1977.\n(31)  Goldstein H. Multilevel statistical models. 3rd ed. London: Arnold Publishers; 2003.\n(32)  Rubin.D.B. Multiple Imputation for Nonresponse in Surveys. New York: John Wiley and \nSons, Inc; 1987.\n(33)  Schafer JL. Analysis of Incomplete Multivariate Data. New York: Chapman and Hall; 1997.\n(34)  Lichten EM, Bombard J. Surgical treatment of primary dysmenorrhea with laparoscopic \nuterine nerve ablation. J Reprod Med 1987;32(1):37-41.\nChapter 5\n\n75\n(35)  Johnson NP , Farquhar CM, Crossley S et al. A double-blind randomised controlled trial \nof laparoscopic uterine nerve ablation (LUNA) for women with chronic pelvic pain. BJOG \n2004;111(9):950-9.\n(36)  Baker PN, Symonds EM. The resolution of chronic pelvic pain after normal laparoscopy \nfindings. Am J Obstet Gynecol 1992;166(3):835-6.\n(37)  Swank DJ, Swank-Bordewijk SC, Hop WC et al. Laparoscopic adhesiolysis in patients \nwith chronic abdominal pain: a blinded randomised controlled multi-centre trial. Lancet \n2003;361(9365):1247-51.\n(38)  Zullo F, Pellicano M, DeStefano R. Efficacy of Laparoscopic Pelvic Denervation in Central-\ntype Chronic Pelvic Pain: A multicentre study. J Gynecol Surg 1996;12:35-40.\n(39)  Biljsma JW. Patient centred outcomes in arthritis. Ann Rheum Dis 2007;64:1-2.\n \nLUNA: a randomised trial\n\n76\n\n\n77\nChapter 6\nMeta-analysis using individual patient data from \nrandomised trials to assess the effectiveness of \nlaparoscopic uterosacral nerve ablation in the \ntreatment of chronic pelvic pain: a proposed \nprotocol \n\n78\nChapter 6\nAbstract\nBackground:  Currently, there are a number of clinical trials, but no international collaboration \nfor collating research on effectiveness of laparoscopic uterosacral nerve ablation (LUNA) for al-\nleviating chronic pelvic pain.\nObjective: Meta-analysis was used by collecting individual patient data (IPD) from the existing \ntrials, to provide a comprehensive assessment of the effectiveness of LUNA that will be generalis-\nable in various clinical contexts.\nMethods: IPD will be sought and collected from all relevant (both already finished and continu-\ning) randomised trials identified through previous systematic reviews. After obtaining raw data \nand cleaning the database, analysis will be of all patients ever randomised based on the inten-\ntion-to-treat principle using endpoints measured at 12 months following randomisation.\nProposal:  We will update searches, contact all authors, obtain data in whatever form it can be \nprovided, build a single database, produce results for individual studies, have them verified by \noriginal authors, explore of any heterogeneity and reasons behind it and finally pool all raw data in \nto a meta-analysis using appropriate statistical methods. The project will test the effectiveness of \nLUNA for women with chronic pelvic pain. It will also motivate collaborating primary investigators \nto undertake new primary studies to corroborate or improve upon the conclusions derived from \nthe retrospective analysis.\nIntroduction\nChronic pelvic pain is a common condition with a major impact on health-related quality of life, \nwork productivity and health care utilisation,\n1-3 costing an estimated £158 million per year to the \nNHS.4 An effective treatment for this condition has evaded the medical profession for centuries. \nEven today only 20-25% patients respond to conservative management.5 When such treatment \nfails, a diagnostic laparoscopy is performed.1;3;6 The cause of the pain is not always obvious as no \npathology is seen in 40-60% of the cases. 1 In the absence of pathology there is no established \ntreatment. The Lee-Frankenhauser sensory nerve plexuses and parasympathetic ganglia in the \nuterosacral ligaments carry pain from the uterus, cervix and other pelvic structures. 7-10 Interrup-\ntion of these nerve trunks by laparoscopic uterosacral nerve ablation (LUNA) may alleviate pain.11 \nHowever, the balance of benefits and risks of this intervention have not been reliably assessed. \nChapter 6\n\n79\nLUNA has, nevertheless, been introduced into practice, although there remains controversy re-\ngarding indications for LUNA. \nWe carried out a comprehensive systematic review of observational and randomised evidence \non the efficacy of LUNA in the treatment of chronic pelvic pain. 11 For the treatment of primary \ndysmenorrhoea there was some evidence of the effectiveness of LUNA when compared to a \ncontrol or no treatment. The comparison between LUNA and laparoscopic presacral neurec-\ntomy (LPSN) for primary dysmenorrhoea showed no significant difference in pain relief in the \nshort term; however, long-term LPSN was shown to be significantly more effective than LUNA. \nFor the treatment of secondary dysmenorrhoea six identified randomised controlled trials (RCT)\ns addressed endometriosis and one included women with uterine myomas. The treatment of \nLUNA combined with surgical treatment of endometrial implants versus surgical treatment of \nendometriosis alone showed that the addition of LUNA did not aid pain relief. For presacral neu-\nrectomy (PSN) combined with endometriosis treatment versus endometriosis treatment alone \nthere was an overall difference in pain relief, although the data suggests this may be specific to \nlaparoscopy and for midline abdominal pain only. Adverse events were significantly more com-\nmon for presacral neurectomy; however, the majority were complications such as constipation, \nwhich may spontaneously improve. This systematic review concluded that there is insufficient \nevidence to recommend the use of nerve interruption in the management of dysmenorrhoea, \nregardless of cause.\n11 \nHence, there is an urgent need for a methodologically sound and sufficiently powered analytic re-\nsearch to be undertaken, to confirm, or refute, any worthwhile effectiveness. A more realistic and \nunbiased assessment for the effectiveness of LUNA might be obtained via multivariable analyses \nthat incorporate the clinical context in which investigations are carried out. Systematic reviews \nand conventional meta-analyses of such data increase the number of outcome events but do not \npermit the clinical context to be taken into account.\nMeta-analysis using individual patient data (IPD) from primary studies has the potential to esti-\nmate informativeness of the investigations in context, which has been introduced as the gold \nstandard analytic approach in reviews of randomised controlled trials,\n12-14 however no such anal-\nyses appear to exist in the current literature. This project will undertake IPD meta-analysis includ-\ning all relevant studies identified in previously published systematic reviews.\n15-21 In this way it will \nprovide the most comprehensive and reliable means of assessing the effectiveness of LUNA in a \nclinical setting considering various clinical context.\nWe aim to answer the following purposes through this IPD meta-analysis. Firstly, to test the hy-\npothesis that in women with chronic pelvic pain in whom diagnostic laparoscopy reveals either \nno pathology or stage I-II endometriosis LUNA alleviates pain and improves life quality at 12 \nLUNA IPD Meta-analysis protocol\n\n80\nmonths, using raw data collated from studies included in systematic review. Secondly, to test \nthe hypothesis that response to LUNA differs according to the site and cause of the pain by two \nsecondary analyses – women with central pain and women with no visible pathology. Thirdly, to \nmotivate collaborating primary investigators to undertake new, mutually planned, primary studies \nto corroborate or improve upon the conclusions derived from the retrospective analysis above, \ne.g. long term follow-up of randomised trials. Lastly, to initiate and advocate the execution of IPD \nmeta-analysis in clinical research and to motivate others to experiment with it in other fields of \nmedicine – if it were proven to be of value.\nMethods\nIdentification Of Randomised Trials And Investigators\nIndividual patient (raw) data will be sought from all currently existent and relevant studies of \nrandomised trials identified in the previously published systematic review, including both already \nfinished and continuing trials.\nAll authors of included studies of randomised trials were invited formally or personally by the \ninitiator of this study for collaboration in the spring of 2006 (a related website has been devel-\noped). An e-mail system based team building exercise will be undertaken by bringing authors’ \nviewpoints together to an electronic forum. This forum will: examine the protocol for refinements; \ndiscuss the variables on which the data are to be collected, the data checking procedures and \nthe main analyses to be performed; and agree on a timetable and a publication policy (a collabo-\nrative/group authorship will be proposed).\nData Collection And Administration\nThe Birmingham Clinical Trials Unit (BCTU) will act as the group’s secretariat and will hold the \nmain database (International LUNA IPD Meta-analysis Collaborative Group, Birmingham, United \nKingdom. Email bctu@bham.ac.uk, website: birmingham.ac.uk/bctu). All data, updates and cor-\nrections should be sent to BCTU in whatever format is most convenient for each group and will \nbe stored in an Access database. All data will be held securely and treated in the strictest con-\nfidence. The data will not be used in any publication without the permission of the responsible \ntrialists.\nAdditional items may be collected if they are of interest to the Group and are available.\nIn addition, we invite the principal investigators of included studies to send us their study protocol \nand details about study execution. We expect that many of these protocols will give us details \nthat cannot be taken from the publications. Furthermore, the current project will ensure checks \non consistency and ranges of the data. The original investigators will be contacted to supply \nmissing data or to check values of doubtful validity.\nChapter 6\n\n81\nStatistical Analysis Plan\nAnalysis will be of all patients ever randomised and will be based on the intention to treat princi-\nple. Visual analogue scale scores [on the scale 0 (no pain) to 10 (worst pain)] within the first 12 \nmonths of treatment will be combined and considered in a repeated measures analysis, giving us \nthe advantage of being able to estimate overall effects over time utilising all available information. \nMultilevel modelling techniques in SAS PROC MIXED will be used, adjusting for baseline score \nand allowing for treatment type and study of origin using fixed or random effects as deemed \nappropriate. Confidence intervals for overall treatment effects will then be estimated from the \nmodel.\nPre-specified subgroup analyses will be performed by:\n1. Presence or absence of some coexisting pathology (endometriosis ± ablation; adhesions \nrequiring adhesiolysis only; minimal pelvic inflammatory disease)\n2. Site of pain (presence of central pain or not)\n3. Age group\n4. Parity (presence/ absence of children)\nWe will also perform stratified analysis at study level depending on the control arm – diagnostic \nlaparoscopy versus other intervention.\nBy collecting IPD, investigation of heterogeneity between these subgroups and estimation of \npossible difference between will be possible, which is a clear advantage over using summary \ndata from published reports. Furthermore, the collection of updated data, with longer follow up \nthan in many of the individual trial reports, will enable reliable evaluation of treatment effects over \ntime. \nFuture Publications\nThe results of the IPD meta-analysis will be presented and discussed at a meeting of the col-\nlaborating trialists.  Any subsequent report of the meta-analysis results will be published in the \nname of the collaborative group – the International LUNA IPD Meta-analysis Collaborative Group \n– and will be circulated to the collaborators for comment, amendment and approval before sub-\nmission.  In case of any disagreement, the fundamental principle that will be applied is that the \nreport should provide the meta-analysis results, presenting the totality of the available evidence, \nbut will not include any interpretations of the data, except for those that are unanimously agreed \nby all collaborators.  Any group will be free to withdraw its data from the collaboration at any time.\nLUNA IPD Meta-analysis protocol\n\n82\nConclusion\nThis proposed international collaboration will provide tghe opportunity to achieve a new meth-\nodology of IPD meta-analysis in a field requiring urgent high quality research. It will also facilitate \ninternational collaboration in medical research while allowing individual investigators to maintain \nautonomy. The proposed analyses also provide the potential to provide high-level evidence on \nmethods of pain management in a timely manner.\nChapter 6\n\n83\nReferences\n(1) Howard FM. The role of laparoscopy in chronic pelvic pain: promise and pitfalls. Obstet \nGynecol Surv. 1993;48:357-87.\n(2)  Morris M,.O’Niell D. Out patient Gynaecology. BMJ 1958;2:1038.\n(3) Stones RW. Chronic pelvic pain. Personal Assessment in Continuing Education. London: \nRoyal College of Obstetricians and Gynaecologists, 1997.\n(4)  Davies L, Ganger K, Drummond M, Saunders D, Beard R. The economic burden of in-\ntractable gynaecological pain. J Obstet Gynecol 1992;12:46-54.\n(5)  Henzl MR. Dysmenorrhoea; achievements and challenge. Sex.Med.Today 1985;9:8-12.\n(6)  Porpora MG,.Gomel V. The role of laparoscopy in the management of pelvic pain in wom-\nen of reproductive age. Fertil.Steril. 1997;68:765-79.\n(7)  Campbell RM. Anatomy and physiology of sacrouterine ligaments. Am J Obstet Gynecol \n1950;59:1.\n(8)  Davis A. Intrinsic dysmenorrhoea. Proc R Soc Med 1936;29:931.\n(9)  Frankenhauser G. Die Bewegungenerven der Gebarmutter. Z Med.Nat Wiss 1864;1:35.\n(10)  Latarjet A,.Roget P . Le plexus hypogastrique chez la femme. Gynecologie et Obstetrique \n1922;6:225.\n(11)  Proctor ML, Latthe PM, Farquhar CM, Khan KS, Johnson NP . Surgical interruption of pel-\nvic nerve pathways for primary and secondary dysmenorrhoea. The Cochrane Database \nof Systematic Reviews 2005, 4. \n(12)  Clarke MJ,.Stewart LA. Meta-analyses using individual patient data. J Eval.Clin Pract. \n1997;3:207-12.\n(13)  Stewart LA,.Parmar MK. Meta-analysis of the literature or of individual patient data: is \nthere a difference? Lancet. 1993;341:418-22.\n(14)  Stewart LA,.Clarke MJ. Practical methodology of meta-analyses (overviews) using up-\ndated individual patient data. Cochrane Working Group. Stat Med. 1995;14:2057-79.\n(15)  Chen FP , Chang SD, Chu KK, Soong YK. Comparison of laparoscopic presacral neurec-\ntomy and laparoscopic uterine nerve ablation for primary dysmenorrhea. J Reprod Med \n1996;41:463-6.\n(16)  Johnson NP , Farquhar CM, Crossley S, Yu Y, van Peperstraten AM, Sprecher M et al. A \ndouble-blind randomised controlled trial of laparoscopic uterine nerve ablation (LUNA) for \nwomen with chronic pelvic pain. BJOG 2004;111:950-9.\n(17)  Lichten EM,.Bombard J. Surgical treatment of primary dysmenorrhea with laparoscopic \nuterine nerve ablation. J Reprod.Med. 1987;32:37-41.\n(18)  Sutton C, Pooley AS, Jones KD, Dover RW, Haines P . A prospective, randomized, dou-\nble-blind controlled trial of laparoscopic uterine nerve ablation in the treatment of pelvic \npain associated with endometriosis. Gynaecol Endosc 2001;10:217-22.\nLUNA IPD Meta-analysis protocol\n\n84\n(19)  The LUNA Trial Collaboration. A randomised controlled trial to assess the efficacy of Lap-\naroscopic Uterosacral Nerve Ablation (LUNA) in the treatment of chronic pelvic pain: The \ntrial protocol [ISRCTN41196151]. BMC.Womens Health 2003;3:6.\n(20)  Vercellini P , Aimi G, Busacca M, Apolone G, Uglietti A, Crosignani PG. Laparoscopic \nuterosacral ligament resection for dysmenorrhea associated with endometriosis: results \nof a randomized, controlled trial. Fertil.Steril. 2003;80:310-9.\n(21)  Yen YK, Liu WM, Yuan CC, Ng HT. Addition of laparoscopic uterine nerve ablation to \nlaparoscopic bipolar coagulation of uterine vessels for women with uterine myomas and \ndysmenorrhea. J Am Assoc.Gynecol Laparosc. 2001;8:573-8.\nChapter 6\n\n85\nLUNA IPD Meta-analysis protocol\n\n86\n\n\n87\nChapter 7\nIndividual patient data meta-analysis of  \nrandomised evidence to assess the  \neffectiveness of laparoscopic uterosacral nerve \nablation in chronic pelvic pain \n\n88\nChapter 7\nAbstract\nBackground: There have been conflicting results in randomised trials of the effects of laparo-\nscopic uterosacral nerve ablation (LUNA) in chronic pelvic pain. Our objective was to perform a \nmeta-analysis using individual patient data (IPD) to provide the most comprehensive and reliable \nassessment of the effectiveness of LUNA. \nMethods:  Electronic searches were conducted in the Medline, Embase, PsycInfo and Cochrane \nLibrary databases from database inception to August 2009. The reference lists of known relevant \npapers were searched for any further articles. Randomised trials comparing LUNA with no ad-\nditional intervention were selected and authors contacted for IPD. Raw data were available from \n862 women randomised into five trials. Pain scores were calibrated to a 10 point scale and were \nanalysed using a multilevel model allowing for repeated measures.\nResults: There was no significant difference between LUNA and No LUNA for the worst pain \nrecorded over a 12 month time period (mean difference 0.25 points in favour of No LUNA on a \n0-10 point scale, 95% CI -0.08 to 0.58; p=0.1).\nConclusions: LUNA does not result in improved chronic pelvic pain.\nIntroduction\nChronic pelvic pain, defined as lower abdominal pain or pelvic pain of longer than 6 months’ \nduration, is a common condition. 1 It has multiple causes, 2 and prevalence varies widely glob-\nally.3 It remains one of the most challenging of women’s health problems with a major impact on \nhealth-related quality of life, productivity,4 and health care utilisation, costing an estimated £250 \nmillion (at 2005/06 prices) on its management in the UK.5 Only 20-25% patients respond to initial \nconservative management.6 When such treatment fails, a diagnostic laparoscopy is frequently \nperformed.7 The cause of the pain is not always obvious as no pathology is seen in 35% (range \n3-92%) of the cases.8,9 In the absence of pathology there is no established gynaecological treat-\nment. \nInterruption of the Lee-Frankenhauser sensory nerve plexuses by laparoscopic uterosacral nerve \nablation (LUNA) may alleviate pain. 10,11 LUNA can be carried out quickly and simply alongside \ndiagnostic laparoscopy using lasers or electro-diathermy and has become widely practised, 12,13 \nChapter 7\n\n89\nalthough clinicians’ beliefs about its effectiveness vary widely. 14 The effects of this intervention \nhave been assessed in randomised controlled trials 15-19 and summarised in published data sys-\ntematic reviews.20-22 In a Cochrane review of surgical interruption for primary and secondary \ndysmenorrhoea,21 there was some evidence of the effectiveness of LUNA when compared to a \ncontrol or no treatment in the absence of endometriosis (odds ratio 6.1, 95% confidence interval \n1.8 to 21.0), but not for secondary dysmenorrhoea (0.8, 95% CI -0.4 to 1.4). These reviews \nconcluded that there is insufficient evidence to recommend the use of nerve interruption in the \nmanagement of dysmenorrhoea, regardless of cause, which was reiterated by the UK National \nInstitute of Health and Clinical Excellence.23\nMeta-analysis using individual patient data (IPD) from primary studies has the potential to pro-\nduce a more reliable estimate of treatment effect than from meta-analysis of aggregate data and \nalso to allow exploration of sub-groups.\n24,25 The LUNA IPD Meta-analysis Collaborative Group, \na consortium of authors of all primary studies involving LUNA, was established to use these ad-\nvanced analytic techniques to evaluate the effectiveness of LUNA in reducing chronic pelvic pain. \nMethods\nWe undertook an IPD meta-analysis to examine the effectiveness of LUNA in reducing chronic \npelvic pain according to a pre-specified protocol,26 developed using recommended methods.27\nSearch Strategy\nThe search methods previously described\n21 were updated in August 2009 (from 2003 to current) \nto identify any trials published since the last search and to identify ongoing trials to seek unpub-\nlished data. A comprehensive database was constructed using Reference Manager 12.0 to store \nall identified references. No language restrictions were applied. \nStudy inclusion and data collection\nStudies were considered for inclusion if LUNA, as an intervention for women of reproductive age \nwith CPP , primary or secondary dysmenorrhoea, was compared against diagnostic laparoscopy \nin a randomised controlled trial. Trials with concomitant ablation of endometrial lesions were in-\ncluded provided ablation was an option in both study groups. Two reviewers (JPD, RC) indepen-\ndently assessed trials for inclusion and quality, with queries resolved by a third reviewer (KSK). \nCriteria for quality assessment were based initially on the reported characteristics and included \nadequacy of sequence generation and allocation concealment; blinding of participants, and as-\nsessors where used, to treatment allocation; intention to treat analysis performed and potential \nimpact of losses of follow-up data. \nLUNA IPD Meta-analysis\n\n90\nPrimary authors of included randomised trials were invited to join the collaborative group and \nto provide anonymised raw data on visual analogue score (VAS) pain ratings, the presence or \nabsence of visual pathology, age at randomisation and parity. Further assessment of data quality \non receipt of the dataset included the ability to reproduce the published results and discrepant \npatterns of data. Where discrepancies existed authors were contacted for clarification.\nStatistical Analysis\nThe primary outcome was a derived measure of worst pain level experienced if data was re-\nceived on different pain symptoms (e.g. dysmenorrhoea, non-cyclical pain and dyspareunia). \nPain scores using a visual analogue scale (VAS) were rescaled where necessary, to anchor the \nscale between zero (no pain) and ten (worst pain imaginable). Rather than dichotomising the data \nat an arbitrary threshold, as was the case in two of the included studies,\n16,17 scores were evalu-\nated as a continuous measurement, thus improving the precision of the analysis. 28 A clinically \nsignificant difference in pain has been defined as 2 points on a 10-point VAS for chronic pelvic \npain \n29 and elsewhere, moderately important difference equates to at least a 30% reduction of \nthe 10-point VAS for pain.30 \nWhere study information was available on allocated and received treatments, the intention to \ntreat principle was used to analyse the data. Comparisons between allocation groups, using \ndata up to and including 12 months, were performed using multilevel modelling techniques in \nSAS PROC MIXED, using all available data to estimate treatment effects over time.\n31 The pri-\nmary analysis model included covariate parameters for trial (to take account of differing sized \ntrials), allocation group, baseline score and time, the latter nested within patient and trial.\n32,33 \nHeterogeneity of treatment effects across trials was investigated by including a trial by treatment \ninteraction term in the model.\n34 Where this test was statistically significant (a value of p<0.05 used \nhere), sensitivity analysis were undertaken to investigate any existing inconsistency in the data.  \nSecondary analyses incorporating the allocation group as a random effect were also undertaken \nfor the primary analysis. Restricted maximum likelihood estimates of the overall treatment effect \nwere calculated along with 95% confidence intervals.  \nSubgroup analysis, including any differential treatment effect over time, was explored in a similar \nfashion by adding the relevant parameter by treatment interaction term to the basic model. Pre-\ndefined subgroup analyses were the presence or absence of visual pathology (endometriosis, \nadhesions, pelvic inflammatory disease) and site of pain (central versus peripheral), as some tri-\nals reported differential effects.\n17,35 Additional subgroups that could feasibly impact on treatment \neffects and for which data were available were age group (<30 years, ≥30 years at randomisa-\ntion) and parity (nulliparous, parous). Treatment effect estimates within subgroups were further \ninvestigated, alone, and where necessary in combination, if the relevant interaction terms were \nstatistically important at p<0.05.\nChapter 7\n\n91\nResults\nTrials and patients\nA total of 56 citations were identified by electronic searches (Figure 1). After detailed evaluation \nof the papers and removal of four duplicates, five primary articles met the selection criteria and \nwere included.15-19  Trials were excluded due to a lack of an adequate method for generating \nallocation sequence for randomisation,(36) if they used an intervention not within the scope of \nthis meta-analysis,35,37,38 did not have a non-interventional control, 39-42 or could not assess the \nindividual contribution of LUNA.43 \nFigure 1:  Study selection process for the individual patient data meta-analysis to assess the  \n effectiveness of laparoscopic uterosacral nerve ablation (LUNA) in chronic pelvic pain.\nThe characteristics of all included studies in the analysis are summarized in the Table 1. The \ninterventions in experimental and control arms were comparable across all trials. Trial popula-\ntions comprised women with CPP with or without visible pathology, apart from one,19 which only \nassessed LUNA for primary dysmenorrhea. IPD was received for all five (100%) of the included \ntrials, which involved 862 randomised women.\n1 \n \n \nFIGURES \nFigure 1: Study selection process for the individual patient data meta-analysis \nto assess the effectiveness of laparoscopic uterosacral nerve ablation (LUNA) \nin chronic pelvic pain \n \nArticles excluded  n = 13 \n- Duplicates  n = 4 \n- Comparison not of interest \n                   n = 7 \n- Unable to distinguish effect of \nLUNA   n = 1 \n- Not randomised  n = 1 \nCitations excluded after \nscreening titles/ abstracts \nn = 38 \nArticles included in \nsystematic review n = 5 \n(862 women randomised) \n \nTotal citations identified \nfrom electronic searches \nn = 45 \nArticles retrieved for \nevaluation \nn = 7 \nArticles identified through \nthe Cochrane Library  \nn = 11 \nTotal articles retrieved for \ndetailed evaluation n = 18 \nLUNA IPD Meta-analysis\n\nTable 1  Table of characteristics of included studies in the individual patient data meta-analysis \nto assess the effectiveness of laparoscopic uterosacral nerve ablation in chronic pelvic pain.\nStudies Lichten 1987 Sutton 2001 Vercellini 2003 Johnson 2004 Daniels 2009\nMethods\nAllocation\nconcealment\nInadequate. Adequate, sealed \nenvelopes\nAdequate, sealed \nenvelopes\nAdequate, sealed \nenvelopes\nAdequate, third \nparty randomisation\nRandomisation Randomised \nby last digit of \nmedical case \nnumber on day \nof surgery.\nComputer-\ngenerated \nrandomisation \nsequence\nComputer-\ngenerated \nrandomisation \nsequence\nComputer-generated \nrandomisation \nsequence\nComputer-\ngenerated \nrandomisation \nsequences using \nminimisation\nBlinding Double blinding: \nparticipant \nand clinical \npsychologist.\nDouble blinding; \nparticipant and \nresearch nurse.\nOpen Double blinding; \nparticipant and \ninvestigator.\nSingle blinding; only \npatient blinded, no \nevaluator in this trial \nFollow-up 3 and 12 \nmonths\n6 and 12 months 24 hours, 3 and 12 \nmonths\n3,6 and 12 months \nand 2, 5 and 10 yrs\nNumber \nof women \nrandomised\n21 51 180 123 (56  with no \nendometriosis)\n487 (266 with no \npathology)\nNumber \nof women \nin primary \nanalysis\n21 at 12 months 46 at 6 months 116 at 12 months 106 at 12 months 378 at 12 months\nDrop-outs/ \nwithdrawals\nNone. 5 (1 became \npregnant and 4 \nwere lost to follow \nup). However \ndata points for to \n14 women were \nmissing for some \nanalyses.\n29 pregnant, \n14 used oral \ncontraception, 15 \nlost to follow-up, \n6 miscellaneous \nreasons.\n14 were excluded \nbased on \nlaparoscopic findings.\nLoss to follow up:\n24 hours: 1\n3 months: 3 (2 LUNA \nand 0 no LUNA in \nthe population with \nno endometriosis; 0 \nLUNA and 1 no LUNA \nin the endometriosis \npopulation).\n12 months: 17 (4 \nLUNA and 2 no LUNA \nin the population with \nno endometriosis; 6 \nLUNA and 5 no LUNA \nin the endometriosis \npopulation).\nWithdrawals: 6 (3 \nLUNA 3 No LUNA) \nbefore 12 months\nLost to follow-up: \n18 (11 LUNA 8 no \nLUNA) provided \nno data up and \nincluding 12 \nmonths.\nQuality Compliance \nwith quality \ncriteria (out \nof 5)\n2 4 3 4 4\nInclusion \nCriteria\nWomen \nwith severe \ndysmenorrhoea. \nNo improvement \nduring run-\nin phase on \nmedication \nor oral \ncontraceptive.\nWomen with history \nand physical or \nlaparoscopic \nexamination \nsuggestive of \nendometriosis who \nhad not received \nmedical treatment \nfor endometriosis \nwithin the last 6 \nmonths, and had \nnot previously \nundergone surgical \ntreatment of their \ndisease\nWomen aged \n18 to 40, who \nreported pelvic \npain of more than \n6 months duration; \nundergoing  \noperative \nlaparoscopy for \nsymptomatic \nminimal to severe \nendometriosis \nWomen aged 18 to 45 \nyears, who reported \npelvic pain of more \nthan 6 months \nduration; no change \nin medication in \nprevious 3 months.\nWomen who \nreported pelvic \npain of 6 months or \nlonger in duration, \nlocated within \nthe true pelvis or \nbetween and below \nthe anterior iliac \ncrests; associated \nfunctional disability; \nlack of response to \nmedical treatment; \nundergoing \ndiagnostic \nlaparoscopy.\nChapter 7\n\n93\nStudies Lichten 1987 Sutton 2001 Vercellini 2003 Johnson 2004 Daniels 2009\nExclusion \nCriteria\nHistory of \npsychotherapy, \nmajor abdominal \nprocedures, \ndrug abuse; \ndemonstrable \npelvic pathology \nat diagnostic \nlaparoscopy.\nSevere \nendometriosis, any \nother pathology \nthat may have \nbeen responsible \nfor their pain \nsymptoms; \ncurrently pregnant \nor intending to \nget pregnant; \nmedical treatment \nfor endometriosis \nwithin last 6 \nmonths; previous \nsurgery for \nendometriosis.\nPrevious diagnosis \nof endometriosis; \nany other \npathology that \nmay have been \nresponsible \nfor their pain \nsymptoms, \npresence \nof vaginal \nendometriosis; \ntreatment for \nendometriosis \nother than \nnonsteroid anti-\ninflammatory drugs \nup to 6 months \nbefore entry in the \nstudy; previous \ndiagnosis of \ngastrointestinal, \nurologic and \northopedic \ndiseases in which \npain may radiate \nto the pelvic area; \nknown psychiatric \ndisturbances.\nPrevious \nhysterectomy or \npelvic malignancy; \nprevious LUNA; \nknown ovarian cysts; \npelvic adhesions \nwhich did not \nappear to be due to \nendometriosis; plan \nfor a pregnancy within \n12 months; intention \nto change other \nmedical treatment \nwhich could influence \npelvic pain scores \nwithin 12 months; \nLUNA technically not \npossible.\nPrevious \nhysterectomy; \nprevious LUNA; \nmoderate \nand severe \nendometriosis; \nprevious surgery for \nendometriosis or \npelvic inflammatory \ndisease; adnexal \npathology; LUNA \ntechnically not \npossible.\nPrimary \ndysmenorrhoea\nCPP with minimal \nto moderate pelvic \nendometriosis\nSecondary \ndysmenorrhoea \n(minimal to severe \nendometriosis)\nCPP with no visible \npathology or with \nminimal to severe \nendometriosis\nCPP with no \nvisible pathology \nor minimal to mild \nendometriosis, mild \npelvic inflammatory \ndisease, minimal \nadhesions\nAge Range 18-34. \nMean for each \ngroup not given.\nMean 28 (range \n20 to 41). Mean \nfor each group not \ngiven, but stated no \ndifference.\nRange 18-40. \nMean for each \ngroup not given.\nMean (SD) No \nendometriosis group \nLUNA 29 (5.83)  No \nLUNA 29 (6.49) With \nendometriosis group \nLUNA 30 (6.71) No \nLUNA 29 (5.31)\nMean (SD) LUNA \n30·6 (7·53) No \nLUNA 30·5 (7·48)\nLocation Detroit, USA Surrey, UK Milan, Italy Auckland, New \nZealand\nMulticentre, UK\nTreatment LUNA LUNA with \nlaparoscopic \nexcision or ablation \nof all visible \nendometriosis.\nLUNA with \nlaparoscopic \nexcision or ablation \nof all visible \nendometriosis \nGroup with \nendometriosis: LUNA \nwith laparoscopic \nexcision or ablation \nof all visible \nendometriosis\nGroup without \nendometriosis: LUNA \nGroup with \npathology: LUNA \nwith laparoscopic \nexcision or ablation \nof all visible \nendometriosis and/\nor adhesionolysis.\nGroup without \npathology: LUNA \nControl Diagnostic \nlaparoscopic \nsurgery only.\nLaparoscopic \nexcision or \nablation of visible \nendometriosis.\nLaparoscopic \nexcision or \nablation of visible \nendometriosis.\nGroup with \nendometriosis: \nlaparoscopic excision \nor ablation of visible \nendometriosis\nGroup without \nendometriosis: \ndiagnostic \nlaparoscopic surgery \nonly.\nGroup with \npathology: \nlaparoscopic \nexcision or \nablation of visible \nendometriosis and/\nor adhesionolysis.\nGroup without \npathology: \ndiagnostic \nlaparoscopic \nsurgery only.\nLUNA IPD Meta-analysis\n\n94\nThe methodological quality of the included studies was generally good, with adequate randomi-\nsation and blinding of the participants and assessors, where employed (Figure 2). In one trial, 19 \nrandomisation was inadequately concealed so was excluded from the primary analysis, although \na sensitivity analysis including these data was conducted. On receipt of raw data, it was not \npossible to guarantee an intention-to-treat analysis for two studies,16,18 as provided data did not \ngive details on allocated and received procedures. In one of these studies, a problem with the \ndatabase precluded use of all but the baseline and three months data.\nFigure 2  Mean pain scores over time for trials included in the analysis.   \n Lower score indicates less pain.\nEffectiveness of LUNA on relieving pain\nIn all trials, a marked improved in pain scores was seen in both LUNA and No LUNA groups \nfollowing laparoscopy. Dysmenorrhoea was invariably the worst pain reported. In the primary \nanalysis, the No LUNA group demonstrated a non-significant greater improvement in their worst \npain score than the LUNA group when a constant treatment effect over time was assumed (0.25 \npoints on a 10-point scale, 95%CI -0.08 to 0.58; p=0.1).  This result was consistent over the four \n2 \n \n \n \n \n \n \nFigure 2: Mean pain scores over time for trials included in the analysis.  Lower score indicates less pain. \nChapter 7\n\n95\nhigh quality trials (test for trial by treatment interaction: p=0.2) (Figure 3).  A similar difference was \nseen when the treatment effect was allowed to vary randomly over the trials (0.32 points, 95%CI \n-0.40 to 1.05; p=0.3). Likewise, including the inadequately concealed trial as a sensitivity analysis \ndid not change the overall result (0.1 points, 95%CI -0.23 to 0.44; p=0.5), but significant het-\nerogeneity between trials was introduced (p<0.0001) and so no further examination of the data \nincluding this trial was undertaken. The observed treatment effect was demonstrated to change \nover time (test for time by treatment interaction p=0.004); a difference in pain improvement in \nfavour of No LUNA was seen at 3 months (0.49 points, 95%CI 0.12 to 0.86; p=0.009), but not \nat 6 months (0.25 points, 95%CI -0.09 to 0.58; p=0.1) or 1 year (-0.25 points, 95%CI -0.72 to \n0.23; p=0.3) (Figure 3). \nFigure 3  The bottom graph shows the maximum likelihood estimates of visual analogue pain \nscores from a multilevel model with timepoint nested within patient, within trial and with baseline \nscore as a covariate and treatment effect allowed to vary over time. Mean pain score are meas-\nured on a visual analogue scale of 0 (no pain) to 10 (worst imaginable pain).The top graph shows \nthe difference in mean visual analogue scale (VAS) pain scores, with error bars indicating 95% \nconfidence intervals.\nThere was some evidence that presence of visible pathology altered the effectiveness of LUNA \n(p=0.01) (Table 2).Where pathology was present, the No LUNA group had greater improvement \nin pain at 3 months (0.81 points, 95%CI 0.36 to 1.26; p=0.0004) and 6 months (0.61 points, \n95%CI 0.19 to 1.04; p=0.005), whereas those without visible pathology benefited more from \n3 \n \n \n \nFigure 3: The bottom graph shows the maximum likelihood estimates of visual analogue pain scores from a multilevel model with \ntimepoint nested within patient, within trial and with baseline score as a covariate and treatment effect allowed to vary over time. \nMean pain score are measured on a visual analogue scale of 0 (no pain) to 10 (worst imaginable pain).The top graph shows the \ndifference in mean visual analogue scale (VAS) pain scores, with error bars indicating 95% confidence intervals. \nNo LUNA \nsuperior \nLUNA IPD Meta-analysis\n\n96\nLUNA at 1 year (-0.68 points, 95%CI -1.29 to 0.07; p=0.03), but not at 3 or 6 months. There was \nno evidence that age (p=0.1) or parity (p=0.7) had any differential effect on the treatment effect \nwhen these parameters were assessed in the model, so combinations of subgroups were not \ninvestigated further. Only one trial provided data on site of pain, so this subgroup analysis was \nnot attempted. There were no reports of serious peri-operative or subsequent adverse events.\nTable 2  Subgroup analysis of presence or absence of pathology on the effect of LUNA in al \n leviating pain\nDiscussion\nMain findings\nOur review provides the most comprehensive and reliable means of assessing the effectiveness \nof LUNA in women with CPP in whom diagnostic laparoscopy reveals either no or minimal pa-\nthology.  LUNA did not alleviate pain compared to the No LUNA group, indeed there was some \nevidence that women who have the LUNA procedure may suffer from more pain in the short term \nthan those who do not. Contradictory treatment effects were seen in the presence and absence \nof pathology, but were not consistent at every timepoint, whereas age and parity did not influence \nthe treatment effect.\nStrengths and limitations of the review\nThe availability of individual patient data from all included studies enabled a more thorough inves-\ntigation of overall and subgroup treatment effects than previous meta-analyses have achieved. \nProblems with database relational integrity prompted us to exclude some data from the meta-\nanalysis. One lower quality study, which was clinically and statistically heterogeneous to the other \nstudies was excluded from the primary analyses, although sensitivity analysis indicated this had \nno effect on the overall treatment effect. The multilevel model accounting for repeated measures \nincreased the accuracy of estimates by using continuous pain scores, rather than dichotomised \nTimepoint Pathology present\nMean VAS difference (95% confidence \ninterval), p-value for difference\nPathology absent\nMean VAS difference (95% \nconfidence interval), p-value for \ndifference\n3 months 0.81 (0.36,1.26) p=0.0004 -0.08 (-0.66,0.50) p=0.8\n6 months 0.61 (0.19,1.04) p=0.005 -0.28 (-0.82,0.26) p=0.3\n12 months 0.21 (-0.35,0.78) p=0.5 -0.68 (-1.29,-0.07) p=0.03\nTest for heterogenetity between subgroups p=0.01 Mean VAS difference of worst pain score between LUNA \nand No LUNA groups. Positive number indicates greater improvement in No LUNA group\nChapter 7\n\n97\nmeasures of success used in some trials. 16,17 With IPD, we were able to explore the variation \nbetween subgroups and treatment effects within subgroups, using similar statistical models. Un-\nlike previous meta-analyses,21 we chose to exclude trials with interventional controls in order to \nassess LUNA’s efficacy in pain relief, therefore cannot comment on its effectiveness compared to \npresacral neurectomy. There were no other outcome measures common to included studies, so \nanalyses were restricted to pain reduction.\nInterpretation\nThis IPD meta-analysis reinforces the conclusions drawn from the largest trial of LUNA \n15 that the \nprocedure is not effective in alleviating pain within one year of treatment. Although a long term \nassessment of LUNA was not possible here due to lack of data beyond 12 months from most \nprimary studies, the UK LUNA trial does not suggest any delayed effect of LUNA. Subgroup \nanalysis did suggest a benefit of LUNA in those without visible pathology, consistent with the \nCochrane review, although this effect was only seen at 12 months and was of borderline sta-\ntistical significance.\n21 Conversely, the IPD meta-analysis of those with visible pathology showed \ngreater decrease in pain in the No LUNA group, contradicting the Cochrane review. These sub-\ngroup effects lack biological plausibility and are likely to have arisen by chance. Moreover, given \nthat a clinically significant difference in pain has been defined as 2 points on a 10-point visual \nanalogue scale for chronic pelvic pain,\n29,30 and also for other types of pain,44 it seems implausible \nthat LUNA could produce a clinically relevant effect in women with minimal gynaecological pa-\nthology. In every comparison, the confidence intervals around the mean difference in pain scores \nbetween the groups were less than ±0.7 points: in our statistically significant subgroup analysis, \nat 12 months in the absence of pathology, the pain score was 0.68 points lower with LUNA \nand the upper 95% confidence interval was 1.29 points lower, well below a clinically significant \nimprovement. \nThe meta-analysis conclusively dismisses the use of LUNA for all women with pelvic pain in the \npresence of visible pathology,, and strongly suggests that LUNA does not provide any benefit of \nclinical importance for women without recognisable pathology, due to lack of efficacy. National \nguidelines should be updated and brought in line with that of the UK.\n23 There is no need for any \nfurther randomized controlled trials of LUNA and future research should focus on further investi-\ngation of the causes of chronic pelvic with a view to identifying novel treatments or refining exist-\ning therapies and assessing their effectiveness in large, well designed randomized trials. \nLack of efficacy in this study, and others (Personal communication E Lichten), provide strong evi-\ndence that the pathophysiology of chronic pelvic pain is complicated and a greater understand-\ning of the interaction of neural pathways and uterine physiology is needed to direct research of \nfurther interventions. \nLUNA IPD Meta-analysis\n\n98\nAn additional aim was to motivate the collaborating primary investigators to undertake new, \nmutually planned, primary studies in this field. Few IPD meta-analysis have been published in \ngynaecological research to date and it is hoped this paper, alongside other international initia-\ntives (www.ipd-meta-analysis.com) will motivate others to adopt this technique when uncertainty \nremains despite published data meta-analysis.\nAcknowledgements\nWe naturally want to thank the International LUNA IPD Meta-analysis Collaborative Group for \nkindly providing the data, responding to the various queries raised and providing valuable feed-\nback on the draft paper. The authors would like to thank the Wellbeing of Women, a registered \nUK charity, for providing the financial support for the original UK LUNA clinical trial and continued \nsupport to produce this IPD meta-analysis. They would also like to thank Tai-In Um for translation \nof the paper by Moon et al from Korean.\nChapter 7\n\n99\nReferences\n1 Zondervan KT, Yudkin PL, Vessey MP , Dawes MG, Barlow DH, Kennedy SH Prevalence \nand incidence of chronic pelvic pain in primary care: evidence from a national general \npractice database. BJOG 1999;106(11):1149-55.\n2  Latthe P , Mignini L, Gray R, Hills R, Khan K. Factors predisposing women to chronic pel-\nvic pain: systematic review. BMJ 2006;332(7544):749-51.\n3  Latthe P , Latthe M, Say L, Gulmezoglu M, Khan KS. WHO systematic review of prev-\nalence of chronic pelvic pain: a neglected reproductive health morbidity. BMC Public \nHealth 2006;6(1):177.\n4  Sundell G, Milsom I, Andersch B. Factors influencing the prevalence and severity of dys-\nmenorrhoea in young women. BJOG 1990;97(7):588-94.\n5  Davies L, Ganger K, Drummond M, Saunders D, Beard R. The economic burden of in-\ntractable gynaecological pain. J Obstet Gynaecol 1992;12(S2):54-6.\n6  Henzl M.R. Dysmenorrhoea; achievements and challenge. Sex Med Today 1985;9:8-12.\n7  Royal College of Obstetricians and Gynaecologists: Guidelines and Audit Committee. The \ninitial management of chronic pelvic pain.  2005. Guideline 41.\n8  Porpora M, Gomel V. The role of laparoscopy in the management of pelvic pain in women \nof reproductive age. Fertil Steril 1997;68(5):765-79.\n9  Howard FM. The role of laparoscopy as a diagnostic tool in chronic pelvic pain. Best Pract \nRes Clin Obstet Gynaecol 2000;14(3):467-94.\n10  Doyle JB. Paracervical uterine denervation by transection of the cervical plexus for the \nrelief of dysmenorrhoea. Am J Obstet Gynecol 1955;70:1.\n11  Doyle JB, Des Rosiers JJ Paracervical uterine denervation for the relief of pelvic pain. \nClinical Obstet Gynecol 1963;6:742-53.\n12  Daniels J, Gray R, Khan K, Gupta J. Laparoscopic uterine nerve ablation: A survey of \ngynaecological practice in the UK. Gynaecol Endosc 2000;9(3):157-9.\n13  Latthe P , Powell R, Daniels J, Hills RK, Gray R, Gupta JK, et al. Variation in practice \nof Laparoscopic uterosacral nerve ablation: A European survey. J Obstet Gynaecol \n2004;24(5):547-51.\n14 Latthe PM, Braunholtz DA, Hills RK, Khan KS, Lilford R. Measurement of beliefs about ef-\nfectiveness of laparoscopic uterosacral nerve ablation. BJOG 2005;112(2):243-6.\n15  Daniels J, Gray R, Hills RK, Buckley L, Gupta J, Latthe P , et al. Laparoscopic uterosa-\ncral nerve ablation for alleviating chronic pelvic pain: a randomized controlled trial. JAMA \n2009;302(9):955-61.\n16  Vercellini P , Aimi G, Busacca M, Apolone G, Uglietti A, Crosignani PG. Laparoscopic \nuterosacral ligament resection for dysmenorrhea associated with endometriosis: results \nof a randomized, controlled trial. Fertil Steril 2003 Sep;80(2):310-9.\nLUNA IPD Meta-analysis\n\n100\n17  Johnson NP , Farquhar CM, Crossley S, Yu Y, van Peperstraten AM, Sprecher M, et al. A \ndouble-blind randomised controlled trial of laparoscopic uterine nerve ablation (LUNA) for \nwomen with chronic pelvic pain. BJOG 2004;111(9):950-9.\n18  Sutton C, Pooley AS, Jones KD, Dover RW, Haines P . A prospective, randomized, dou-\nble-blind controlled trial of laparoscopic uterine nerve ablation in the treatment of pelvic \npain associated with endometriosis. Gynaecol Endosc 2001;10(4):217-22.\n19  Lichten EM, Bombard J. Surgical treatment of primary dysmenorrhea with laparoscopic \nuterine nerve ablation. J Reprod Med 1987;32(1):37-41.\n20  Khan KS, Khan SF, Nwosu CR, Dwarakanath LS, Chien PFW. Laparoscopic uterosacral \nnerve ablation in chronic pelvic pain: An overview. Gynaecol Endosc 1999;8(5):257-65.\n21  Proctor ML, Latthe PM, Farquhar CM, Khan KS, Johnson NP . Surgical interruption of \npelvic nerve pathways for primary and secondary dysmenorrhoea. Cochrane Database \nSyst Rev 2005;4.\n22  Stones RW, Mountfield J. Interventions for treating chronic pelvic pain in women. \nCochrane Database Syst Rev 2000;4.\n23  National Institute for Health and Clinical Excellence. Laparoscopic uterine nerve ablation \n(LUNA) for chronic pelvic pain. London; 2007. Report No. IPG234.\n24  Stewart LA, Clarke MJ. Practical methodology of meta-analyses (overviews) using indi-\nvidual patient data. Cochrane Working Group. Stat Med 1995;14(19):2057-79.\n25  Stewart LA, Tierney JF. To IPD or not to IPD? Advantages and disadvantages of system-\natic reviews using individual patient data. Eval Health Prof 2002;25(1):76-97.\n26  Xiong T, Daniels J, Middleton L, Champaneria R, Khan KS, Gray R, et al. Meta-analysis \nusing individual patient data from randomised trials to assess the effectiveness of lapa-\nroscopic uterosacral nerve ablation in the treatment of chronic pelvic pain: a proposed \nprotocol. BJOG 2007;114(12):1580, e1- e7.\n27  Stewart LA, Tierney JF, larke M. Reviews of individual patient data. In: Higgins JPT, Green \nS, editors. Cochrane Handbook for Systematic Reviews of Interventions. 5.01 ed.  The \nCochrane Collaboration; 2008.\n28  Altman DG, Royston P . The cost of dichotomising continuous variables. BMJ \n2006;332(7549):1080.\n29  Baker PN, Symonds EM. The resolution of chronic pelvic pain after normal laparoscopy \nfindings. Am J Obstet Gynecol 1992;166(3):835-6.\n30  Dworkin RH, Turk DC, Wyrwich KW, Beaton D, Cleeland CS, Farrar JT, et al. Interpreting \nthe clinical importance of treatment outcomes in chronic pain clinical trials: IMMPACT \nrecommendations. J Pain 2008;9(2):105-21.\n31  Higgins JP , Whitehead A, Turner RM, Omar RZ, Thompson SG. Meta-analysis of continu-\nous outcome data from individual patients. Stat Med 2001;20(15):2219-41.\n32  Jones AP , Riley RD, Williamson PR, Whitehead A. Meta-analysis of individual patient data \nversus aggregate data from longitudinal clinical trials. Clin Trials 2009;6(1):16-27.\nChapter 7\n\n101\n33  Brown H, Prescott R. Applied Mixed Models in Medicine. Wiley, Chichester; 2006.\n34  Whitehead A. Meta-Analysis of Controlled Clinical Trials. Wiley, Chichester; 2002.\n35  Tjaden B, Schlaff WD, Kimball A, Rock JA. The efficacy of presacral neurectomy for the \nrelief of midline dysmenorrhoea. Obstet Gynecol 1990;76:89-91.\n36  Moon M.J. The clinical significance of laparoscopic uterosacral nerve ablation (LUNA). \nKorean J Obstet Gynecol 2001;44(12):2224-8.\n37  Candiani GB, Fedele L, Vercellini P , Bianchi S, Di Nola G. Presacral neurectomy for the \ntreatment of pelvic pain associated with endometriosis: a controlled study. Am J Obstet \nGynecol 1992;167(1):100-3.\n38  Zullo F, Pellicano M, DeStefano R. Efficacy of Laparoscopic Pelvic Denervation in Central-\ntype Chronic Pelvic Pain: A multicentre study. J Gynecol Surg 1996;12:35-40.\n39  Chen FP , Chang SD, Chu KK, Soong YK. Comparison of laparoscopic presacral neurec-\ntomy and laparoscopic uterine nerve ablation for primary dysmenorrhea. J Reprod Med \n1996;41(7):463-6.\n40  Garcia LF, Oviedo OG, Reyes CH, Ibarrola BE, Von der MW. Presacral neurectomy and \nuterine nerve ablation in chronic pelvic pain. Laparoscopic management. A comparative \nstudy. Ginecol Obstet Mex 2003;71:137-42.\n41  Juang CM, Chou P , Yen MS, Horng HC, Twu NF, Chen CY. Laparoscopic uterosacral \nnerve ablation with and without presacral neurectomy in the treatment of primary dys-\nmenorrhea: a prospective efficacy analysis. J Repro Med 2007;52(7):591-6.\n42  Yen YK, Liu WM, Yuan CC, Ng HT. Addition of laparoscopic uterine nerve ablation to \nlaparoscopic bipolar coagulation of uterine vessels for women with uterine myomas and \ndysmenorrhea. J Am Assoc Gynecol Laparosc 2001;8(4):573-8.\n43  Sutton CJ, Ewen SP , Whitelaw N, Haines P . Prospective, randomized, double-blind, con-\ntrolled trial of laser laparoscopy in the treatment of pelvic pain associated with minimal, \nmild, and moderate endometriosis. Fertil Steril 1994;62(4):696-700.\n44  Biljsma JW. Patient centred outcomes in arthritis. Ann Rheum Dis 2007;64:1-2.\nLUNA IPD Meta-analysis\n\n102\n\n\n103\nChapter 8\nPsychological Therapies for Chronic Pelvic Pain: \nSystematic review of Randomized Controlled \nTrials  \n\n104\nChapter 8\nAbstract \nChronic Pelvic Pain (CPP), a common cause of disability in women, is a condition best viewed in \nthe biopsychosocial framework. Psychological interventions are frequently considered alongside \nmedical and surgical treatments. Our objective was to evaluate the effectiveness of psychological \ntherapies for the treatment of CPP . \nElectronic literature searches were conducted in Medline, Embase, PsycInfo and DARE data-\nbases from database inception to April 2010. Reference lists of selected articles were searched \nfor further articles.\nThe studies selected were randomized controlled trials of psychological therapies in patients \nwith CPP compared to no treatment, standard gynecological treatment or another form of psy-\nchological therapy. Two reviewers independently selected articles without language restrictions, \nand extracted data covering study characteristics, study quality and results. Reduction in pain, \nmeasured using visual analog scales or other measurements, was the main outcome measure. \nOf the 107 citations identified, four studies satisfied the inclusion criteria. Compared to no psy-\nchological intervention, therapy produced a standardized mean pain score of -3.27 (95% CI \n-4.52 to -2.02) and 1.11 (95% CI -0.05 to 2.27) at 3 months and -3.95 (95% CI -5.35 to -2.55) \nand 0.54 (95% CI -0.78 to 1.86) at 6 months and greater, based on a visual analog scale score \nof 0-10.\nThe current evidence does not allow us to conclude whether psychological interventions have an \neffect on self reported pain scores in women with CPP . \nIntroduction \nChronic Pelvic Pain (CPP), defined as constant or intermittent, cyclic or acyclic pain lasting 6 \nmonths or more, is a common condition.\n1  Symptoms include dysmenorrhea, deep dyspareunia \nand intermenstrual pain2. Up to 15% of women consulting primary care physicians and 40% of \nwomen consulting gynecologists in secondary care complain of CPP . 3-6  A definite diagnosis and \neffective treatment of CPP is not straightforward, as it can be a symptom of another condition \nsuch as endometriosis, or a syndrome in its own right, without an identifiable cause. \n7,8 \nChapter 8\n\n105\nChronic Pelvic Pain is perhaps best seen from a biopsychosocial perspective whereby patho-\nphysiology, health beliefs, psychological well-being and social situations all contribute to the \nwomen’s experience of pain and lead to the adoption of coping strategies which may have an \nimpact on the women’s wellbeing.8 Women with CPP are more likely to have experienced abuse, \neither physical or psychological, miscarriage, Cesarean section or drug or alcohol dependency \nthan controls without pain. 9 Psychological interventions are frequently considered alongside \nmedical and surgical treatments.  Although psychological therapies are widely used to treat other \nforms of pain, and there is good evidence to support this practice, there is a surprising lack of \nevidence for its use to treat CPP .  A Cochrane review,\n10 looked at randomized controlled trials \n(RCTs) of behavioral interventions for primary and secondary dysmenorrhea but not CPP . In this \nreview, we examine whether psychological interventions are effective in treating idiopathic CPP , \nby looking at all relevant published RCTs. \nMaterials and Methods \nThe systematic review was conducted based on a prospective protocol designed using widely \nrecommended methods. \n11,12,13 No ethical approval was needed for this review.\nIdentification of studies \nA sensitive search strategy was developed. The following databases were searched: Embase, \nMedline, PsycInfo, DARE and the Cochrane Library from database inception to April 2010. Our \nsearch term combinations consisted of MeSH subheading, text words, and word variations for \nthe concepts ‘Psychological Therapy’ and ‘chronic pelvic pain’. These were combined with filters \nfor studies of effectiveness.  For the purposes of this review, the term ‘psychological therapy’ \nwas used to encompass all therapies which address mental health problems or psychological \nissues associated with CPP through structured therapeutic conversations with trained thera-\npists, with an emphasis on a reflective, experiential process. Such therapies aim to help patients \nchange behavior, cognitions or mood which exacerbates or maintains problems.\n14 The umbrella \nterm ‘Psychological Therapy’, encompassed therapies including Cognitive Behavior Therapy, \ncounseling, art ,music and writing therapies, psychoanalysis, psychotherapy, psychiatry, Gestalt \ntherapy, group therapy, integrative psychotherapy, psychoanalytic therapy, interpersonal therapy, \npsychodynamic therapy, person/ client centered therapy, systemic therapy, nondirective therapy  \nand Neurolinguistic Programming. These terms characterizing the different forms of psychologi-\ncal therapy were found by conducting a scoping search looking at dictionaries and textbooks \nfor definitions of ‘Psychological therapies’. Results from this were then used to perform a simple \nliterature search in a handful of bibliographic databases to establish what phrases/ keywords \nwere being yielded when a broad term such as ‘psychological therapies’ was searched for. \nSystematic Review of Psychological Therapies for CPP\n\n106\nAll identified search terms then went on to form the final search strategy. This search strategy \nwas adapted to suit each database and was restricted to humans and females. We also hand \nsearched the bibliographies of all relevant primary articles and reviews to identify articles missed \nby the electronic searches. A comprehensive database was constructed using Reference Man-\nager 12.0 (Thomson Reuters) to store all identified references. No language restrictions were \napplied. \nStudy selection and data extraction procedures\nStudies were selected in two steps. First we scrutinized the citations identified by the electronic \nsearches and obtained full manuscripts of all the citations that met or were thought to have met \nthe predetermined inclusion criteria. Two reviewers (AR, RC), then independently inspected all of \nthe manuscripts to determine if they met the following criteria:\n1. Population: Women with CPP (of unknown etiology)\n2. Interventions: Psychological or behavioral therapies \n3. Comparator: No treatment, standard gynecological treatment or another form of psycho-\nlogical therapy.\n4. Outcomes: Pain relief or symptom improvement\n5. Study design: Randomized controlled trial (RCT)\nIn case of duplicate publications the most recent and complete versions were used. We extract-\ned data on study characteristics and methods onto a pre-designed pro-forma. Any disagree-\nments surrounding the selection of a manuscript or data extraction were resolved by consensus \nor arbitration by a third reviewer (JD, KSK or HMP). \nMethodological quality assessment\nAll manuscripts meeting the selection criteria were assessed for their methodological quality, \ndefined as the confidence that the study design, conduct and analysis minimized bias in the \nestimation of effectiveness. Quality items were assessed using existing texts and checklists.\n15 A \nstudy was considered as of good quality if it used a prospective randomized study design, pro-\nvided evidence of adequate allocation concealment at randomization, if groups were comparable \nat baseline, follow-up was greater than 80%, sample size estimation was reported, the target \npopulation was clearly described and if intention to treat analysis was performed. A score out of \n7 was given to each study, with a higher score indicating higher quality.\nData synthesis\nData, including number of patients, mean and standard deviation were extracted from all includ-\ned papers. Values were analyzed by student’s two- sample t-tests, to get mean difference, 95% \nconfidence intervals and p-values. Pain scores measured by visual analog scales (VAS) were \nChapter 8\n\n107\nstandardized to a 0-10 continuous scale. All statistical analyses were performed using Review \nManager 5 (version 5, Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration \n2008).  \nResults\nA total of 107 citations were identified by electronic searches (Figure 1). After detailed evalua-\ntion of the papers, 13 articles appeared to meet the selection criteria. Upon closer inspection, \nnine papers had to be excluded for the following reasons: duplicate publication, study not ran-\ndomized,16 intervention not of interest,17 type of intervention used was unclear 18 and population \nnot of interest (these papers focused on dysmenorrhea rather than idiopathic CPP).19-21 This left \nthree trials, (which were described in four papers) meeting the selection criteria for the review.22-25 \nFigure 1  Study selection process for systematic review of Psychological therapies for Chronic \nPelvic Pain *Haugstad et al described results at baseline and 90 days post treatment in their \n2006 paper (22), while their 2008 paper disseminated findings 12 months post treatment (23). \nBoth articles are describing the same trial. \nTotal citations identified from\n \nelectronic searches n\n \n= 107 \nCitations excluded after \nscreening titles/ abstracts \nn = 94 \nArticles retrieved for \ndetailed evaluation n = 13 \nArticles excluded:  n = 9 \n•\tDuplicate\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t n = 3 \n•\tStudy\tdesign\tnot\tof\tinterest \nn = 1 \n•\tIntervention\tnot\tof\tinterest \nn = 1 \n•\tIntervention\tunclear\t\t n = 1 \n•\tDisease\tgroup\tnot\tof\tinterest\nn = 3 \nArticles included in\n \nsystematic review n = 4 \n(reporting 3 trials) * \nSystematic Review of Psychological Therapies for CPP\n\n108\nThere was considerable variation in the type of psychological intervention. One trial assessed \nthe value of additional Mensendieck somatocognitive therapy, which helps train specific muscle \ngroups in defined movement patterns, compared with standard gynecological treatment and \nreported outcomes at 3 and 12 months.22,23 This intervention fulfils our definition of psychological \ntherapy, as it also encourages body awareness and cognitions about pain through reflection. The \nsecond study assessed the effect of seeing a photograph of the internal pelvic organs taken dur-\ning diagnostic laparoscopy in the post-operative consultation.24 The final trial compared behavio-\nral counseling, non-directive counseling or relaxation techniques with verbal reassurance after a \nlaparoscopic diagnosis of idiopathic CPP .\n25 This study specifically targeted the pain experienced \nby patients, who were asked to report pain free days rather than rate their pain as scores on the \nVAS. The three active interventions in this study were delivered by a psychologist. The compara-\ntive effectiveness of the interventions from the three different studies could not be quantitated \ndue to clinical heterogeneity and inconsistent reporting of pain severity.\nTable 1 provides a summary of the characteristics of trials included in the systematic review and a \nbreakdown of their study quality. Quality of the included studies was generally adequate. All stud-\nies were prospective, although not all reported concealment of allocation. Sample sizes in two tri-\nals\n23,25 were small, as a valid sample size was not calculated a priori. The small sample size of the \nPearce et al study25 also meant that their power to detect any modest but clinically meaningful \ndifference between the treatments was almost nil. Blinding of participants is not usually possible \nin participatory psychological interventions, although, the study of Mensendieck somatocognitive \ntherapy22 measured outcomes such as motor function using a blinded assessor.\nPatient reported pain \nTwo studies provided VAS pain scores.\n23,24 Figure 2 shows a forest plot of these studies, where \nrandomized groups were classified as psychological intervention versus none. VAS pain scores \nwere reported at 3 months in both trials but then at either 6 or 12 months post-randomization, so \nwere grouped as 6 months or greater. At 3 months the mean difference in pain score was -3.27 \n(95% CI -4.52 to -2.02) and 1.11 (95% CI -0.05 to 2.27), whereas at 6 months or greater the \nvalues were -3.95 (95% CI -5.35 to -2.55) and 0.54 (95% CI -0.78 to 1.86). \nChapter 8\n\n109\nFigure 2  Forest plot showing pain scores at the end of follow-up for studies included in the \nsystematic review of Psychological therapies for Chronic Pelvic Pain * IV – Inverse variance\nThe trial of Mensendieck therapy 22 used a blinded, standardized performance score of five do-\nmains (posture, gait, movement, sitting posture and respiration) on a scale of 0 - 7. At 3 months \nof treatment, a statistically significant difference was observed only in the sitting posture (mean \ndifference -0.93, 95%CI -1.77 to -0.09; p=0.03) and respiration (mean difference -1.37, 95%CI \n-2.24 to -0.50, p=0.003) in favor of the additional Mensendieck therapy.\nThe final study25 the only one using a ‘true’ psychological intervention, used the poor pain meas-\nure of the number of pain free days in the month prior to treatment and during months 3 and \ntwelve. At three months, an increase of 8, 12 and 10 pain free days from baseline was observed \nwith the relaxation, non-directive counseling and behavioral counseling groups, respectively. The \ncontrol group was not assessed at 3 months or baseline and the number of pain free days was \nunbalanced, so the comparative effectiveness of the three approaches cannot be judged. \nDiscussion \nOur review, demonstrates the scant nature of evidence on the use of psychological therapies for \ntreating idiopathic CPP .  The current evidence does not allow us to conclude whether psycholog-\nical interventions have an effect on self reported pain scores in women with CPP . Mensendieck \nsomatocognitive therapy assessed through a good quality trial\n22 had a favorable effect when \nused in addition to standard gynecological treatment. This therapeutic approach can be seen \nas a hybrid between physiotherapy and psychotherapy26, and involves teaching patients how to \nchange their posture, breathing patterns and the way they move in order to reduce their pain. It \nuses a cognitive approach to lead women to an improved understanding and experience of their \nown body.  The second study hypothesized that photograph reinforcement in the post-operative \nconsultation might reduce reported pain, but whether this approach can be formally defined as \n \nStud\ny  1.1.1 Psychological intervention vs No psycholgical intervention - 3 \nmonths Haugstad \n2006 Onwude \n2004 \n1.1.2 Psychological intervention vs No psychological intervention - 6 months and \ngreater Haugstad \n2008 Onwude \n2004 \nMea\nn \n2.8\n9 4.3\n4 \n2 \n3.7\n6 \nS\nD \n1.7\n3 3.4\n3 \n1.5\n7 3.\n5 \nTota\nl \n1\n9 5\n8 \n1\n8 4\n5 \nMea\nn \n6.1\n6 3.2\n3 \n5.9\n5 3.2\n2 \nS\nD \n2.1\n7 2.9\n7 \n2.6\n6 3.2\n1 \nTota\nl \n1\n9 6\n0 \n1\n9 5\n7 \nWeigh\nt \nInverse Variance, Random Effects, 95% \nCI \n-3.27 [-4.52, -\n2.02] 1.11 [-0.05, \n2.27] \n-3.95 [-5.35, -\n2.55] 0.54 [-0.78, \n1.86] \nInterventio\nn \nContro\nl \nMean \nDifference \nMean \nDifference Inverse Variance, Random effects, 95% \nCI \n-\n10 \n-\n5 \n0 5 1\n0 Favours \nintervention \nFavours \ncontrol \nSystematic Review of Psychological Therapies for CPP\n\n110\n“psychological therapy” is debatable. It is possible that there could be an element of reassur -\nance for patients in being shown a picture of their abdomen, but further exploration of how the \npatient interprets presence and absence of visual pathology, in the context of the diagnosis, is \nwarranted27. \nAlthough several reviews discussing the management of CPP using psychotherapy exist, to \nour knowledge this is the first systematic review of randomized controlled trials of psychological \ntherapies in women with idiopathic CPP . Quality of the included studies varied, with the oldest \nstudy failing to report an adequate randomization method and incurring >20% loss of data at \nfollow-up28. The small size of the studies increased the chance of a type 2 error. However one \ntrial23  showed a positive result beyond the play of chance. Our diverse results can be attributed \nto the heterogeneous studies identified and included in our review, as there were no common \ninterventions or outcomes, limiting the clinical impact of our findings. \nThe need for more robust studies, involving women with CPP receiving an established form of \npsychological therapy in which pain is measured as a continuous outcome, is obvious. Future \ntrials should also consider the recommendations of the IMMPACT initiative and employ outcome \nmeasures that not only capture the intensity of pain, but the effect on physical and emotional \nfunctioning29. \nFuture trials should have larger numbers 30,31, aiming to recruit 350 or more patients to detect \na more realistic, but worthwhile, effect of any psychological intervention. To explore when the \nbenefit emerges and whether it is maintained, longer follow-up of participants is recommended. \nIn chronic conditions where outcomes may vary over time, repeated measures analyses will \nmaximize statistical power from the trial sample. An ideal trial would involve cognitive behavioral \ntherapy (CBT), since this is the mainstay of psychological treatment for CPP . Furthermore, there \nis evidence that this form of therapy has a small effect on pain, pain behavior and mood when \nused to treat other types of pain\n32,33.\nAcknowledgements\nWe would like to thank Derick Yates, clinical librarian from Birmingham Womens Hospital NHS \nTrust, for carrying out the literature searches on our behalf. \nChapter 8\n\n111\nTable 1  Table of characteristics of studies included in the systematic review of Psychological \ntherapies for Chronic Pelvic Pain\nPaper Population Sample \nsize \nIntervention Comparator Outcome \nmeasures \nStudy quality \nHaugstad \net al 2006 \n(22)\n(1 year \nfollow-up \ngiven in \nHaugstad \n2008) (23)\nWomen \nwith chronic \npelvic pain \nunexplained \nby pelvic \npathology\n40 Standard \ngynecological \ntreatment \n+ Somato-\ncognitive therapy \n(Mensendieck \nsomatocognitive \ntherapy) \nStandard \ngynecological \ntreatment\nReduction in \nphysical pain, by \nchanging posture, \nmovement and \nrespiration patterns, \nmeasured using \nmensendieck \ntest of motor \nfunction (posture, \nmovement, gait, \nsitting posture and \nrespiration) and \na visual analogue \nscore of pain \nobtained before and \nafter the 90 day \ntreatment period\n- Prospective randomized \nstudy design: Yes\n- Adequate allocation \nconcealment: Yes\n- Populations comparable at \nbaseline: Yes\n- Follow-up > 80%: Yes\n- Appropriate sample size \ncalculation done: No\n- Target population clearly \ndescribed: Yes\n- Intention to treat analysis \ndone: Yes\nTOTAL: 6/7\nOnwude \net al 2004 \n(24)\nWomen \nundergoing \ndiagnostic \nlaparoscopy \nfor the \ninvestigation \nof chronic \npelvic pain\n233 Being shown \na Polaroid \nphotograph of \ntheir pelvis\nNot being \nshown a \nPolaroid \nphotograph \nof their pelvis\nPain severity and \npain belief scores at \n3 and 6 months. \n- Prospective randomized \nstudy design: Yes\n- Adequate allocation \nconcealment: Yes\n- Populations comparable at \nbaseline: Yes\n- Follow-up >80%: Yes\n- Appropriate sample size \ncalculation done: Yes\n- Target population clearly \ndescribed: Yes\n- Intention to treat analysis \ndone: Yes\nTOTAL: 7/7\nPearce et \nal 1982 \n(25)\nWomen who \nattended a \ngynecological \noutpatient \nclinic, \ncomplaining \nof chronic \npelvic pain \nof at least \n6 months’ \nduration. \n32 Women were \nrandomized to \neither: \n1. Behavioral \ncounseling\n2. Non-directive \ncounseling\n3. Relaxation\nVerbal \nreassurance \nEffectiveness of \ntreatments for \ntreating pelvic pain \nof non-organic \norigin\n- Prospective randomized \nstudy design: Yes   \n- Adequate allocation \nconcealment: Not stated\n- Populations comparable at \nbaseline: Yes\n- Follow-up >80%: No\n- Appropriate sample size \ncalculation done: No \n- Target population clearly \ndescribed: Yes\n- Intention to treat analysis \ndone: Yes\nTOTAL: 4/7\nSystematic Review of Psychological Therapies for CPP\n\n112\nReferences\n1  Latthe P , Latthe M, Say L, Gulmezoglu M, Khan KS. WHO systematic review of prev-\nalence of chronic pelvic pain: a neglected reproductive health morbidity. BMC Public \nHealth 2006;6:177-93.\n2  Vercellini P , Fedele L, Arcani L, Bianchi S, Rognoni MT, Candiani GB. Laparoscopy in the \ndiagnosis of chronic pelvic pain in adolescent women. J Reprod Med 1989;34(10):827-\n30.\n3  Jamieson DJ, Steege JF. The prevalence of dysmenorrhoea, dyspareunia, pelvic pain and \nirritable bowel syndrome in primary care practices. Obstet Gynecol 1996;87:332-7.\n4 Moen M, Schei B. Epidemiology of endometriosis in a Norwegian county. Acta Obstet \nGynecol Scand 1997;76:559-62.\n5 Stones RW, Mountfield J. Interventions for treating chronic pelvic pain in women. \nCochrane Database of Systematic Reviews 2003;4.\n6 Zondervan KT, Yudkin PL, Vessey MP , Jenkinson CP , Dawes MG, Barlow DH, et al. \nChronic pelvic pain in the community - symptoms, investigations, and diagnoses. Am \nObstet Gynecol 2001;184:1149-55.\n7 Aslam N, Harrison G, Khan K, Patwardhan S. Visceral hyperalgesia in chronic pelvic pain. \nBJOG 2009;116(12):1551-5.\n8 Daniels JP , Khan KS. Chronic Pelvic Pain in women. BMJ 2010;341:c4834.\n9 Latthe P , Mignini L, Gray R, Hills R, Khan K. Factors predisposing women to chronic pel-\nvic pain: systematic review. BMJ 2006;332(7544):749-55.\n10 Proctor ML, Murphy PA, Pattison HM, Suckling J, Farquhar CM. Behavioural interventions \nfor primary and secondary dysmenorrhea. Cochrane Database of Systematic Reviews \n2007;3.\n11 Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of \nreports of meta-analyses of randomised controlled trials: the QUOROM statement. Qual-\nity of Reporting of Meta-analyses. Lancet 1999;354(9193):1896-900.\n12 Khan KS, Kunz R, Kleijnen J, Antes G. Systematic reviews to support evidence-based \nmedicine. How to review and apply findings of healthcare research.  2nd ed. 2003. Royal \nSociety of Medicine Press Limited, London, UK\n13 Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items \nfor systematic reviews and meta-analyses: the PRISMA statement. PLoS Med \n2009;6(7):e1000097.\n14 British Psychological Society Centre for Outcomes Research and Effectiveness., Depart-\nment of Health. Treatment choice in Psychological therapies and counselling - Evidence \nbased clinical practice guideline.  2001. Report No.: 23044. www.doh.gov.uk/en/publica-\ntionsandstatistics/publications/Pubilcationspolicyandguidance/DH_4007323\nChapter 8\n\n113\n15 Moher D, Schulz KF, Altman DG. The CONSORT statement: revised recommenda-\ntions for improving the quality of reports of parallel-group randomised trials. Lancet \n2001;357(9263):1191-4.\n16 Albert H. Psychosomatic group treatment helps women with chronic pelvic pain. J Psy-\nchosomatic Obstet Gynaecol 1999;20:216-25.\n17 Heyman J, Ohrvik J, Leppert J. Distension of painful structures in the treatment for chron-\nic pelvic pain in women. Acta Obstet Gynecol Scand 2006;85(5):599-603.\n18 Peters AAW, van Dorst E, Jellis B, van Zuuren E, Hermans J, Trimbos JB. A randomised \nclinical trial to compare two different approaches in women with chronic pelvic pain. Ob-\nstet Gynecol 1991;77:740-4.\n19 Amodei N, Nelson RO, Jarrett RB, Sigmon S. Psychological treatments of dysmenorrhea: \nDifferential effectiveness for spasmodics and congestives. J Behavior Ther Experim Psy-\nchiatry 1987;18(2):95-103.\n20 Chesney MA, Tasto DL. The effectiveness of behavior modification with spasmodic and \ncongestive dysmenorrhea. Behaviour Res Ther 1975;13:245-53.\n21 Quillen MA. Self-control of dysmenorrheic symptoms through pain management training. \nJ Behavior Ther Experim Psychiatry 1982;13(2):123-30.\n22 Haugstad GK, Haugstad TS, Kirste UM, Leganger S, Klemmetsen I, Malt UF. Mensendieck \nsomatocognitive therapy as treatment approach to chronic pelvic pain: Results of a ran-\ndomised controlled intervention study. Am Obstet Gynecol 2006;194:1303-10.\n23 Haugstad GK, Haugstad TS, Kirste UM, Leganger S, Wojniusz S, Klemmetsen I, et al. \nContinuing improvement of chronic pelvic pain in women after short-term Mensendieck \nsomatocognitive therapy: results of a 1-year follow-up study. Am Obstet Gynecol \n2008;199(6):615.e1-615.e8.\n24 Onwude JL, Thornton JG, Morley S, Lilleyman J, Currie I, Lilford RJ. A randomised trial \nof photographic reinforcement during postoperative counselling after diagnostic laparos-\ncopy for pelvic pain. Eur Obstet Gynecol Reprod Biol 2004;112(1):89-94.\n25 Pearce S, Knight C, Beard RW. Pelvic Pain - A common gynaecological problem. J Psy-\nchosomatic Obstet Gynaecol 1982;1:12-7.\n26 Haugstad GK, Kirste U, Leganger S, Haakonsen E, Haugstad TS. Somatocognitive therapy \nin the management of chronic gynaecological pain. A review of the historical background \nand results of a current approach. Scandinavian Journal of Pain 2011;2:124-9.\n27 Linton SJ, McCracken LM, Vlaeyen JWS. Reassurance: Help or hinder in the treatment of \npain. Pain 2008;134:5-8.\n28 Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assess-\ning the quality of reports of randomized clinical trials: is blinding necessary? Controlled \nClinical Trials 1996;17(1):1-12.\nSystematic Review of Psychological Therapies for CPP\n\n114\n29 Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, et al. Core outcomes measures for \nchronic pain clinical trials: IMMPACT guidelines. Pain 2005;113:9-19.\n30 Hills RK, Daniels JP . Assessing new interventions in women’s health. Best Pract Res Clin \nObstet Gynaecol 2006;20(5):713-28.\n31 Khan KS, Selman T, Daniels J. Gynecology Randomized Control Trials. Handbook of \nClinical Trials. Wiley and Sons, Inc. Hoboken, New Jersey, USA; 2009. p. 519-30. Editor: \nShayne Cox Gad\n32 Bernardy K, Fuber N, Kollner V, Hauser W. Efficacy of cognitive-behavioral therapies in \nfibromyalgia syndrome - a systematic review and metaanalysis of randomized controlled \ntrials. J Rheumatol 2010;37(10):1991-2005.\n33 Johnson RE, Jones GT, Wiles NJ, Chaddock C, Potter RG, Roberts C, et al. Active exer-\ncise, education, and cognitive behavioral therapy for persistent disabling low back pain: a \nrandomized controlled trial. Spine 2007;32(15):1578-85.\nChapter 8\n\n115\nSystematic Review of Psychological Therapies for CPP\n\n116\n\n\n117\nChapter 9\nHysterectomy, endometrial destruction and \nMirena® for heavy menstrual bleeding:  \nA systematic review and individual patient data \nmeta-analysis  \n\n118\nChapter 9\nAbstract\nObjective: Heavy menstrual bleeding is a common condition that does not always respond to \noral medication. We undertook an individual patient data (IPD) meta-analysis to evaluate the rela-\ntive effectiveness of hysterectomy, endometrial destruction (both ‘first generation’ hysteroscopic \nand ‘second generation’ non-hysteroscopic techniques) and Mirena® (levonorgestrel releasing \nintrauterine system). \nMethods: Individual patient data on 2814 women were available from 17 of the 30 randomised \ncontrolled trials identified (14 trials including 2448 women for first versus second generation \nendometrial destruction; 7 trials including 1127 women for hysterectomy versus first generation \nendometrial destruction; 5 trials including 304 women for second generation endometrial de-\nstruction versus Mirena@; 3 trials including 190 women for first generation endometrial destruc-\ntion versus Mirena@; 1 trial including 236 women for hysterectomy versus Mirena @).  Direct and \nindirect comparisons were made where appropriate to assess the effect of interventions on the \nprimary outcome measure of patient dissatisfaction.\nResults: At around 12 months, 7.3% (12.6% v 5.3%) more women were dissatisfied with out-\ncome of first generation hysteroscopic techniques than hysterectomy (OR [Odds Ratio]:2.46, \n95%CI 1.54 to 3.93, p=0.0002), but hospital stay (WMD [Weighted Mean Difference]:3.0 days, \n95%CI 2.9 to 3.1, p<0.00001) and time to resumption of normal activities (WMD:5.2 days, \n95%CI 4.7 to 5.7, p<0.00001) were longer for hysterectomy.  Unsatisfactory outcomes were \ncomparable with first and second generation techniques (OR:1.20, 95%CI 0.88 to 1.62, p=0.2), \nalthough second generation techniques were quicker (WMD:14.5 minutes, 95%CI 13.7 to 15.3, \np<0.00001) and women recovered sooner (WMD:0.48 days, 95%CI 0.20 to 0.75, p=0.0008) \nwith fewer procedural complications.  Indirect comparison suggested more unsatisfactory out-\ncomes with second generation techniques than hysterectomy (10.6% v 5.3%; OR:2.32, 95%CI \n1.27 to 4.24, p=0.006). \nConclusion: More women are dissatisfied following endometrial destruction than hysterectomy. \nHowever, dissatisfaction rates are low after all treatments and hysterectomy is associated with \nincreased hospital stay and recovery period.  Definitive evidence on effectiveness of Mirena\n® \ncompared to more invasive procedures is lacking.   \nKeywords: Endometrial destruction, hysterectomy, heavy menstrual bleeding, menorrhagia, \nMirena®\nChapter 9\n\n119\nIntroduction\nHeavy menstrual bleeding (HMB) is a common problem among women of reproductive age 1. It \nis often incapacitating, expensive to treat and significantly impacts on women’s quality of life2,3.  \nMany women are not happy with medical treatment and end up undergoing surgery 4.  Hyster-\nectomy was once the only surgical option for HMB, and indeed, almost half of the hysterecto-\nmies currently performed worldwide are for the treatment of HMB5. Endometrial destruction (ED) \ntechniques, which aim to destroy or remove the endometrial tissue 6 have become increasingly \npopular alternatives and, as a result, the number of hysterectomies in the UK has declined by \n64% between 1995 and 2002\n7. ED techniques were introduced in the 1980’s, with rollerball \nablation and transcervical resection emerging as the predominant approaches under direct hyst-\neroscopic vision\n8. Subsequently, there have been a development of a second generation of non-\nhysteroscopic techniques which are easier to perform.  Here, endometrial destruction is achieved \nthrough a variety of modalities, including high temperature fluids, bipolar electrical, or microwave \nenergy. Intrauterine coil devices were initially introduced as contraceptives, but the addition of \nprogestogen resulted in reduced menstrual bleeding. Mirena\n®, a levonorgesterol releasing intrau-\nterine system provides a non-surgical alternative, which is reversible and fertility sparing9. \nWomen and clinicians now have greater choice of treatment, although evidence to support deci-\nsion making is inadequate. In the UK, guidelines from the National Institute of Healthcare and \nClinical Excellence\n10 recommend the use of Mirena ® in the first instance for women with be-\nnign HMB, followed by ED if pharmaceutical treatments fail to resolve symptoms. Syntheses \nof evidence from randomised controlled trials (RCTs) comparing these treatments have been \nlimited11-13, partly due to scarcity of head-to-head comparisons and variation in outcome meas-\nurements used to evaluate effectiveness. We undertook a meta-analysis of individual patient data \n(IPD) from all relevant trials, to address previous deficiencies in evidence synthesis. IPD meta-\nanalysis has a number of advantages over traditional published data reviews14,  including the abil-\nity to carry out data checks, standardise analytical methods and undertake subgroup analyses. \nMethods\nWe sought IPD from RCTs of hysterectomy, ED techniques and Mirena ® to examine their rela-\ntive efficacy as second line treatment of HMB. The systematic review was conducted based on \na protocol designed using widely recommended methods 15,16 that complied with meta-analysis \nreporting guidelines17 (http://www.bctu.bham.ac.uk/systematicreview/hmb/protocol.shtml).\nLiterature search and study selection \nThe Cochrane Library, Medline (1966-2008), Embase (1980 to July 2008) and CINAHL data-\nbases (1982 to July 2008) were searched using relevant terms and word variants for population \nHMB IPD Meta-analysis\n\n120\nand interventions.  We also hand-searched the bibliographies of all relevant primary articles and \nreviews, to identify any articles missed by the electronic searches.  Experts were contacted to \nidentify further studies. To identify any ongoing RCTs, the Meta-Register of Controlled Trials and \nthe ISRCTN register were searched.  No language restriction was applied. \nStudies were selected in a two step process. Firstly we scrutinised the citations identified by the \nelectronic searches and obtained full manuscripts of all the citations that met, or were thought \nlikely to meet, the predetermined inclusion criteria based on patient entry criteria (women with \nHMB or abnormal/excessive/prolonged uterine bleeding that was unresponsive to other medical \ntreatment) and study design, the latter limited to RCTs only. We then considered four catego-\nries of intervention: Hysterectomy (performed abdominally, vaginally or laparoscopically); ‘first \ngeneration’ ED techniques (using operative hysteroscopy, including endometrial laser ablation, \ntranscervical resection of the endometrium (TCRE) and rollerball endometrial ablation); ‘second \ngeneration’ ED techniques (including thermal balloon (Cavaterm\n®, Thermachoice® and Vesta®), \nmicrowave (Microsulis®), laser (ELITT ®), bipolar radio frequency (NovaSure ®), cryoablation and \nhydrothermal ablation); and levonorgesterol releasing intrauterine system (Mirena®).  Our intention \nwas to compare these categories of treatment against each other; studies were not included in \nthe meta-analysis if a comparison between relevant categories did not exist. Data from studies \nmaking a comparison within these categories were also requested to allow further exploration of \npossible predictors of the primary outcome measure.\nData collection and study quality assessment \nRepeated attempts were made to contact corresponding authors via post, email or telephone to \naccess data.  Where initial attempts failed, we attempted personal contact via our links through \nthe British and European Societies for Gynaecological Endoscopy.  Authors were asked to supply \nanonymised data for each of the pre-specified outcome measures and were invited to become \npart of the collaborative group with joint ownership of the final publication.  Where the investiga-\ntors declined to take part in the study or could not be contacted, published data were extracted \nfrom manuscripts using pre-designed pro-formas by two independent reviewers (RC and LJM). \nAny disagreements were resolved by consensus or arbitration by a third reviewer (JPD). Received \ndata were merged into a master database, specifically constructed for the IPD meta-analysis. \nThe data were cleaned and results cross-checked against published reports of the trials.  Where \ndiscrepancies existed authors were contacted for clarification. \nPatient dissatisfaction was used as the primary outcome measure. Level of satisfaction was the \nmost frequently measured outcome, but dissatisfaction was used here to simplify interpreta-\ntion of statistical output. Responses were dichotomised, with an “unsure” or “dissatisfied” re-\nsponse interpreted as indicative of an unsatisfactory rating of treatment.  Perception of treatment \nChapter 9\n\n121\n(“moderate”/”poor”), recommendation to a friend (“not sure”/”not recommended”), or resolution \nof the major presenting problem (“no”) were used as substitutes if dissatisfaction was not report-\ned explicitly, however not all trials could contribute to this end-point as a suitable measure was \nnot always available. Ideally, a more disease specific quality of life tool would have been used 18, \nbut the majority of studies did not employ such a measure. \nOther outcome measures were bleeding scores (ranging from a minimum of 0 with no upper \nlimit)19, amenorrhoea rate (converted from a bleeding score of 0 where data existed, otherwise \nas reported), HMB rate (converted from bleeding scores of >10019 where data existed, otherwise \nas reported), EQ-5D utility score 20, SF-36 scores 21, duration of surgery/hospital stay, general \nanaesthesia rates, post-operative pain score (standardised onto a 0-10 scale), time to return to \nwork/normal activities/sexual activity, dysmenorrhoea/dyspareunia rate and proportion undergo-\ning subsequent ablation/hysterectomy or discontinuing use of Mirena\n®.  Pre-defined subgroups \nwere age at randomisation (≤40 v >40 years), parity (nulliparous v parous), uterine cavity length \n(≤8 v >8 cm), presence or absence of fibroids/polyps and, where available, severity of bleeding \nat baseline (bleeding score≤350 or >350).  Time points up to 2 years were considered; results \nfrom the limited number of studies with longer term follow-up are given in the Data Supplement \nbut are not referred to in detail.  Data at less than 12 months were combined, and are described \nas results at 6 months.\nAll selected trials were assessed for their methodological quality, using received datasets where \navailable in addition to the reported information.  Quality was scrutinised by checking adequacy \nof randomisation, group comparability at baseline (examining baseline characteristics for any \nsubstantive differences), blinding (where appropriate), use of intention-to-treat (ITT) analysis, \ncompleteness of follow-up, compliance, reliability using a priori sample size estimation and gen-\neralisability using description of the sample recruited.  Adequacy of randomisation was assessed \nwith sub-questions examining information on sequence generation, the process of allocation, \nand allocation concealment.\nStatistical analysis \nTo minimise the possibility of bias IPD and aggregate data (AD) were combined in a two-stage \napproach\n22. IPD were reduced to AD to allow studies with AD only to be combined with those \nwhere IPD were obtained. Point estimates and 95% confidence intervals (CI) were calculated for \nindividual studies at each time point. Differences in effect estimates between trials and the pre-\ndefined subgroups of patients are displayed using odds ratio plots, with heterogeneity investigat-\ned using Cochran’s Q Statistic23. Sensitivity analyses to explore the causes of heterogeneity were \nundertaken if the p-values of these tests were <0.1. Differences between studies contributing IPD \nand those with AD only were examined in the same fashion to check that the latter results were \nHMB IPD Meta-analysis\n\n122\nconsistent with those we received IPD for. Further details are given in the results section if any \ninconsistency exists. Likewise, further details are given on any obvious publication bias if noted \nfrom the assessment of funnel plots.  Only a limited amount of data were available for studies \ncomparing Mirena® with ED, so Mirena@ was compared with first and second generation studies \ncombined as well as separately.  Assumption free ‘fixed effect’ methods were used to combine \ndichotomous outcome measures and estimate pooled odds ratios (OR) using the method of \nPeto24, and, for continuous variables, weighted mean differences (WMD) were calculated 25 at \neach time point.  \nThe primary outcome measure of dissatisfaction was investigated comprehensively using re-\nceived data. Results at 12 months, where the majority of studies had collected data, were used \nas the focus for analysis. Where responses were not available at this time point data were sub-\nstituted, in the first instance from 2 years and failing that 6 months. If it was not possible to make \na direct comparison between treatments (e.g. hysterectomy vs second generation ED), indirect \nestimates were made26 using a logistic regression model27 allowing for trial and treatment28. Es-\ntimates using dissatisfaction at any time were also examined, along with an analysis allowing \nfor the correlation of the repeated measurements using generalised estimating equations (IPD \nonly)\n29.  \nAccess to IPD also allowed the inclusion of patient-level covariates to examine possible predic-\ntors of dissatisfaction. First, covariates were considered individually, whilst allowing for differ -\nences between trial estimates. If considered statistically important (p<0.1), covariate parameters \nwere included together in a multivariate analysis to examine adjusted estimates. In addition to \nthe analysis of the primary outcome measure described above, as a sensitivity analysis, IPD \nwere also used to explore the effect observed in compliance rates for comparisons between first \nand second generation ED (unfortunately there were insufficient data to extend this analysis to \nMirena\n@ comparisons). For example, for those women ‘satisfied’ with treatment but subsequently \nundergoing a hysterectomy, positive responses were substituted with negative ones. Revman \nv5.0 (Cochrane Collaboration, Denmark) and SAS v9.2 (SAS Institute, Cary, USA) software were \nused for analysis.   \nResults\nTrials and patients\nA total of 556 potentially relevant citations were identified by electronic searches. After detailed \nevaluation, 30 trials were eligible for inclusion in the review (Figure 1). Of these trials, seven \ncompared hysterectomy with ED techniques. Six of these studies involved first generation tech-\nniques\n30-35. The seventh study used a combination of first and second generation in equal pro-\nChapter 9\n\n123\nportions36 and was included here as a first generation comparison, with a sensitivity analysis \nperformed without the trial. One study compared hysterectomy37 with Mirena®. Fourteen studies \ncompared first generation ED techniques with those of second generation38-51 and eight studies \ncompared Mirena® with ED, three of which were first generation 52-54 and five second genera-\ntion55-59. Characteristics of these studies are shown in the Data Supplement. Data from a further \nfive studies60-64, which involved comparisons within first and second generation ED, were also \nreceived.  \nTrials comparing hysterectomy with ED and those comparing first and second generation ED \ninvolved women of a similar age, with averages of 40.6 years (95%CI 40.3 to 40.9) and 41.0 \nyears (95%CI 40.8 to 41.2), respectively. Women involved in trials comparing Mirena ® with ED \nwere slightly older, with an average age of 43.6 years (95%CI 43.3 to 44.0). Eligibility criteria for \nwomen with uterine pathology varied between trials; inclusion of women with fibroids was gen-\nerally limited by size or number. Where included, they amounted to a maximum of 30% of the \nwomen in each individual study.\nA high proportion of data was received from trials involving hysterectomy (7/8 studies, 1278 of \n1363 women), less for trials of ED techniques (7/14 studies, 1359 of 2448 women) and those \ninvolving Mirena\n® (3/8 studies, 177 of 494 women). Overall, we received some IPD from 65% \n(2814/4305) of women involved in the trials, although only 8 studies were able to provide all \nrequested variables31,32,38,39,42,45,51,59.  The remaining studies had some missing information, with \nlimited details on patient follow-up covering subsequent operations (e.g. hysterectomy following \nMirena\n®).  See section on statistical analysis for details on how data from studies providing IPD \nwas utilised. \nStudy quality\nThe methodological quality of the published data from the studies was variable (Figure 2, ). Over \nhalf the studies failed to give adequate information about their randomisation procedure and de-\ntails on allocation concealment.  There was a general lack of true ITT analysis, with some studies \nstating that an ITT had been performed yet only analysing those women that had received treat-\nment.  For 4 studies that reported non-ITT analyses39,42,45,57 ITT analyses were undertaken using \nthe available IPD, although it was not always clear if protocol deviant patients were followed-up \ncorrectly in these cases. Small sample sizes often lacked a sensible justification, especially in \nstudies involving Mirena\n®. In the nine trials involving Mirena @, only four had greater than 80% of \nwomen with Mirena® in situ at 12 months post-randomisation.\nHMB IPD Meta-analysis\n\n124\nFigure 1  Study selection process for systematic review and individual patient data (IPD) meta-\nanalysis of randomised trials comparing hysterectomy, ED techniques and Mirena® for HMB\nSatisfaction\nas an outcome\nwithin 2 years\n(with IPD)\nArticles retrieved for detailed evaluation (n=111)\nTotal citations identified (n=557)\nArticles included (n=39)\nArticles included in systematic review and meta-analysis (n=40) \nUnpublished studies (n=1*)\nArticles excluded (n=72):\n  Abstract/letter (n=6)\n  Data not extractable (n=9) \n  Duplicates (n=57)\nTrials unable to contribute to meta-analysis (n=10): \n  First generation v first generation (n=4)\n  Second generation v second generation (n=4) \n  Hysterectomy v hysterectomy (n=2)\nMirena v first generation\nendometrial destruction \n(n=3; 190 women)\nMirena v second\ngeneration\nendometrial destruction\n(n=5; 304 women)\nFirst v second generation\nendometrial destruction\n(n=14; 2448 women)\nHysterectomy\nv Mirena\n(n=1; 236 women)\nHysterectomy v\nfirst generation\nendometrial destruction\n(n=7; 1127 women)\nNo IPD obtained\n(0 women)\nIPD obtained from\n3 trials (177 women)\nIPD obtained from\n7 trials (1359 women)\nIPD obtained from\n1 trial (236 women)\nIPD obtained from\n6 trials (1042 women)\n0 1 trial (83 women)6 trials (1083 women)03 trials (570 women)\nChapter 9\n\n125\nFigure 2  Quality of studies included in systematic review and individual patient data (IPD) meta-\nanalysis of randomised trials comparing hysterectomy, ED techniques and Mirena ® for HMB. \nNumbers inside bars are numbers of studies.\nEffectiveness in reducing dissatisfaction with treatment\nHysterectomy versus first generation endometrial destruction \nMore women were dissatisfied at 12 months following first generation ED than hysterectomy \n(12.6% vs 5.3%; OR:2.46, 95%CI 1.54 to 3.93, p=0.0002) (Figure 3), with no significant het-\nerogeneity between study estimates (p=0.9). This estimate of effect size was consistent with, \nalthough slightly less, than the estimate from the repeated measures analysis (IPD only) over all \ntime points (OR:3.75, 95%CI 2.18-6.46, p<0.0001) and an analysis using dissatisfaction at any \ntime point (OR:3.37, 95%CI 2.14-5.31, p<0.0001). There was no evidence of any differences be-\n Ablation, Mirena® and Hysterectomy for Heavy Menstrual Bleeding  \n   31 \n \n \nFigure 2: Quality of studies included in systematic review and individual patient data (IPD) meta-analysis of randomised trials comparing hysterectomy, \nED techniques and Mirena® for HMB. Numbers inside bars are numbers of studies (details given in appendix 3). \nHMB IPD Meta-analysis\n\n126\ntween subgroups (see data collection and study quality assessment section), including between \nstudies providing IPD or AD (test for heterogeneity: p=0.9).   \nFigure 3 Dissatisfaction at 12 months - Hysterectomy vs First Generation ED\nFirst versus second generation endometrial destruction techniques \nSimilar dissatisfaction rates were seen with second and first generation ED (10.6% vs 12.2%; \nOR:1.20, 95%CI 0.88 to 1.62, p=0.2; test for heterogeneity: p=0.7) (Figure 4). Similar estimates \nwere obtained from the repeated measures analysis of IPD (OR:1.21, 95%CI 0.84 to 1.74, \np=0.3), the analysis using dissatisfaction at any time (OR:1.22, 95%CI 0.91 to 1.62, p=0.2) and \nalso an analysis adjusting for patients who went onto receive hysterectomy (OR:1.25, 95%CI \n0.93 to 1.67, p=0.1). Results were consistent over all subgroups, including those studies provid-\ning IPD or AD only (test for heterogeneity p=0.8).\nFigure 4  Dissatisfaction at 12 months - First vs Second Generation ED\n Ablation, Mirena® and Hysterectomy for Heavy Menstrual Bleeding  \n   32 \n \n \n \n \n \nFigure 3: Dissatisfaction at 12 months - Hysterectomy vs First Generation ED \n \n \n Ablation, Mirena® and Hysterectomy for Heavy Menstrual Bleeding  \n   33 \n \n \n \n \n \nFigure 4: Dissatisfaction at 12 months - First vs Second Generation ED  \n \nChapter 9\n\n127\nMirena® versus endometrial destruction techniques\nRates of dissatisfaction with Mirena® and second generation ED were similar (18.1% vs 22.5%; \nOR:0.76, 95%CI 0.38 to 1.53, p=0.4) (Figure 5). The combined estimate of this and the one \nstudy which compared Mirena @ with first generation ED 54 (test for differences between sub-\ngroups: p=0.2), also showed no evidence of a difference (OR:0.94, 95%CI 0.50 to 1.77, p=0.9; \ntest for heterogeneity over all studies: p=0.1). Overall rates of dissatisfaction were 17.2% for \nMirena\n@ and 18.2% for both first and second generation ED.  Lack of IPD prohibited any further \ninvestigation of subgroups or repeated measures.\nIndirect comparisons of hysterectomy with second generation endometrial destruction tech-\nniques and  Mirena@  \nIndirect estimates (Figure 6) suggest hysterectomy is also preferable to second generation ED \n(OR:2.32, 95%CI 1.27 to 4.24, p=0.006) in terms of patient dissatisfaction. This is confirmed by \nthe repeated measures analysis over all 3 time points, which perforce only include IPD (OR:3.06, \n95%CI 1.59 to 5.90, p=0.0008). The evidence to suggest hysterectomy is preferable to Mirena\n@ \nwas weaker (OR:2.22, 95%CI:0.94-5.29, p=0.07), but given the lack of precision from Mirena @ \ncomparisons this was not a surprising result. \nFigure 5  Dissatisfaction at 12 months – First and Second generation ED v Mirena@\n Ablation, Mirena® and Hysterectomy for Heavy Menstrual Bleeding  \n   34 \n \n \n \n \nFigure 5: Dissatisfaction at 12 months – First and Second generation ED v Mirena \n \nHMB IPD Meta-analysis\n\n128\nFigure 6  Dissatisfaction at 12 months summary.  Estimates >1 indicate increased dissatisfac-\ntion for the second treatment listed.  Indirect estimates are represented by a dotted line.\nPredictors of dissatisfaction\nFor second generation ED, uterine cavity length was the strongest predictor of dissatisfaction \n(p=0.02), with shorter uterine cavity length (<=8cm v >8cm) associated with reduced rates (OR: \n0.59, 95%CI 0.38-0.93; p=0.02) (Table 1).  Absence of fibroids/polyps also showed a trend to-\nwards reduced dissatisfaction (p=0.07), although no further adjusted estimates including both \nparameters were attempted as only 3 studies had data on fibroids/polyps.  There were no con-\nvincing associations with any of the variables for hysterectomy or first generation ED. \nEffectiveness in improving other outcomes\nHysterectomy versus endometrial destruction and Mirena\n®\nThese comparisons focused on recovery times and quality of life, as estimates of post operative \nmenstrual blood loss are redundant after hysterectomy.  ED offered quicker surgery (WMD:32 \nminutes, 95%CI 30 to 34, p<0.0001), shorter hospital stay (WMD:3.0 days, 95%CI 2.9 to 3.1, \np<0.00001), faster recovery periods (time to return to normal activities: WMD:5.2 days, 95%CI \n4.7 to 5.7, p<0.00001) and less pain post-operatively (WMD:2.5 points, 95%CI 2.2 to 2.9, \np<0.0001), although estimates of differences for some of these parameters should be used with \ncaution given high variability between studies. One study 36 suggested no obvious difference in \nEQ-5D utility score, whilst another35 suggested differences in favour of hysterectomy in the gen-\neral health (WMD:9.6 points, 95%CI 5.7 to 13.5, p<0.0001), social functioning (WMD:24 points, \n95%CI 21 to 27, p<0.0001) and vitality (WMD:13 points, 95%CI 9.3 to 16, p<0.0001) domains \n Ablation, Mirena® and Hysterectomy for Heavy Menstrual Bleeding  \n   35 \n \n \nFigure 6:  Dissatisfaction at 12 months summary.  Estimates >1 indicate increased dissatisfaction for \nthe second treatment listed.  Indirect estimates are represented by a dotted line. \n \nChapter 9\n\n129\nof the SF-36 questionnaire. Peri-operative adverse events associated with hysterectomy were \nrelatively few (each 0.5% to 2.0%) but urinary tract infections were more common with hysterec-\ntomy (43/530, 8.1%) than ED (9/585, 1.5%).  Of the women who were initially treated with ED, \n15% had undergone a hysterectomy by two years.\nTable 1  Results from logistic regression analysis with dissatisfaction at 12 months as the outcome\nHysterectomy 1st generation ED 2nd generation ED\nOR (95% CI)# P-value OR (95% CI)# P-value OR (95% CI)# P-value\nIndividual estimates*:\nUterine cavity length \n(<=8 v >8) - - 0.97 (0.38-2.44) 0.9 0.59 (0.38-0.93) 0.02\nAge (<=40 v >40) 2.28 (0.66-\n7.89) 0.2 1.21 (0.81-1.81) 0.4 1.30 (0.87-1.93) 0.2 \nFibroids/polyps \n(absence v presence)\n0.51 (0.14-\n1.93) 0.3 1.15 (0.55-2.38) 0.7 0.36 (0.12-1.07) 0.07\nParity (nullparous v \nparous) - - 1.27 (0.36-4.43) 0.7 0.84 (0.33-2.16) 0.7\nBaseline bleeding \nscore (<=350 v 350) - - 0.73 (0.27-1.97) 0.5 0.96 (0.48-1.91) 0.9 \n#  Estimates <1 favour the first subgroup listed, i.e. have reduced dissatisfaction \n* after allowing for study\nNo differences in EQ-5D scores were seen at 6 or 12 months in the single study comparing \nhysterectomy with Mirena®, whilst the only statistically significant effect observed in the SF-36 \nquestionnaire was in the pain domain, favouring hysterectomy (WMD:9.6 points, 95%CI 2.7 to \n16.6, p=0.007). All results were consistent over subgroups.\nFirst versus second generation endometrial destruction techniques \nThe proportion of women with amenorrhoea or still experiencing HMB was similar in both groups \nat all time points apart from at 2 years, where there was a borderline significant difference in \nfavour of 2nd generation techniques ([amenorrhea] OR:0.64, 95%CI 0.41 to 0.99, p=0.04; \n[HMB] OR:1.85, 95%CI 1.04 to 3.32, p=0.04). Change from baseline analysis of bleeding scores \nshowed no evidence of a difference at any of the time points. Two studies 45,47 using the SF-36 \nquestionnaire and one small study38 using the EQ-5D questionnaire showed no consistent differ-\nence between first and second generation techniques. \nSecond generation ED was quicker (WMD:15 minutes; 95%CI 14 to 15, p<0.0001), and less \nlikely to need general anaesthesia (OR:0.16, 95%CI 0.12 to 0.20, p<0.0001), although highly \nsignificant heterogeneity makes estimates difficult to interpret.  Lower use of general anaesthesia \nHMB IPD Meta-analysis\n\n130\nwith second generation ED translated to a slightly quicker time to normal activities (WMD:0.48 \ndays, 95%CI 0.20 to 0.75, p=0.0008) and time to return to work (WMD:1.36 days, 95%CI 0.69 \nto 2.03, p<0.0001).  Post-operative pain was similar following either method of ED. Adverse \nevents were relatively low in both groups (each <2%), but peri-operative complications such as \nuterine perforation, excessive bleeding, fluid overload and cervical laceration were higher with \nfirst generation ED. The number of women requiring a subsequent hysterectomy was lower for \nsecond generation ED, but these differences were not large enough to be statistically significant \nwithin the first two years ([1 year] OR:0.77, 95%CI 0.47 to 1.24, p=0.3; [2 years] OR:0.68, 95%CI \n0.41 to 1.13, p=0.1). Overall rates were 3% (74/2265) and 8% (71/939) at these time points.  Any \ndifferences amongst subgroups were confined to single time points only.  Results from studies \nproviding IPD were consistent with those with AD only.\nMirena\n® versus endometrial destruction techniques\nFewer women experienced HMB after Mirena ® at 6 months (OR:0.23, 95%CI 0.09 to 0.57, \np=0.001) and at 2 years (OR:0.08, 95%CI 0.01 to 0.50, p=0.007), although not at 12 months \n(OR:0.74, 95%CI 0.34 to 1.61, p=0.5), where most trials had made this measurement. Amenor-\nrhoea rates were similar at all time points. Change in bleeding scores favoured ED at 12 months \nonly (WMD:38 points; 95%CI 15 to 60, p=0.0009). Other outcome measures could not separate \nthe two treatments. Two studies 57,59 provided SF-36 change from baseline scores and no dif-\nferences were found in any of the domains. The number of women subsequently undergoing a \nhysterectomy were similar at each time point; rates at 12 months were 2% (2/86) for ED and 7% \n(6/89) for Mirena\n@. A high proportion of women originally prescribed Mirena® discontinued use of \nthis treatment: 30/191(16%) at 12 months, rising to 29/105 (28%) by 2 years. Reported adverse \nevents were low with Mirena®; only around 3% reported an expelled/migrated coil within the first \nmonth. These results were from studies of first and second generation studies combined, where \nfirst generation data existed, and were consistent over both types of ED.\nDiscussion \nMain findings\nIn this review, access to IPD enabled a more rigorous analysis than is possible from published \ndata from trials comparing second line treatments for HMB.  Based on direct and indirect com-\nparisons using all available data, the review found that both first and second generation ED \ntechniques were associated with greater dissatisfaction than hysterectomy, although rates were \nlow for all treatments and absolute differences were small.  Recovery times and length of hospi-\ntal stay were longer for hysterectomy. Dissatisfaction levels with second generation techniques \nwere slightly lower than those associated with first generation techniques.  In addition, second \ngeneration methods of ED were quicker, had faster recovery times, were associated with fewer \nChapter 9\n\n131\nadverse procedural events and could be offered under local anaesthetic. Less women subse-\nquently converted to hysterectomy after second generation ED compared with first generation \nED, but this difference was not statistically significant. Shorter uterine cavity length showed was \nassociated with lower levels of dissatisfaction for second generation ED. Comparisons of ED with \nMirena\n® suggest comparable efficacy, although trials were generally small and suffered from poor \ncompliance in the Mirena@ arm. \nStrengths and limitations of the review\nWe used optimal methodology, complying with guidelines on reporting of systematic reviews and \nmeta-analyses\n65. An extensive literature search was conducted, with no language restrictions, \nminimising the risk of missing information. The collection of IPD allowed us to use previously \nunreported data, to improve the assessment of study quality, to standardise outcome measures, \nundertake ITT analysis and use optimal analytical methods. Subgroup, repeated measures and \nmultivariable analyses would not have been possible without the collection of IPD, and along with \nthe indirect measures analysis, have not previously been reported.  \nThe review was hampered by the unavailability of IPD from at least 35% of randomised women, \nwhich could not be accessed as a number of trialists did not agree to collaborate or could not be \ncontacted. Received data were sometimes incomplete and on occasions failed quality checks \nand so were unusable. The review’s inferences are also limited by the inconsistent outcome \nmeasure used across trials; studies involving ED and Mirena\n® focused on comparing reduction \nin bleeding, whilst hysterectomy trials focused on patient satisfaction, quality of life and resource \nusage. \nInterpretation\nIn this review we found that more women were dissatisfied following ED when compared to \nhysterectomy, although operating time, total hospital stay and recovery period are much less for \nED. Nevertheless, rates of dissatisfaction are relatively low for ED and it is an effective alterna-\ntive to hysterectomy for women with abnormal uterine bleeding who do not seek amenorrhoea.  \nWhilst this review has shown that hysterectomy is a relatively safe operation, other studies with \na more comprehensive follow-up of large populations have shown higher levels of morbidity \nfollowing hysterectomy\n5. In contrast, ED has low rates of complication 66. All these factors need \nto be taken into consideration when considering any potential benefit of hysterectomy.  Indeed, \nnearly fourteen women need to be treated with hysterectomy to prevent a single woman being \ndissatisfied with ED (number needed to treat from 1st generation comparison:13.7, 95%CI 9.1 to  \n28.4). Given the extra resources and recovery period needed for hysterectomy a benefit of this \nsize may not be large enough to warrant its use.  The cost-effectiveness of these treatments will \nbe an important factor and this is to be examined separately in a concurrent study.\nHMB IPD Meta-analysis\n\n132\nWe found that second generation techniques, such as thermal balloon ablation (Thermachoice® \nand Cavaterm®) 42,67,68, the Novasure® device44, or microwave (Microsulis®)47,69, were at least as \neffective as first generation techniques. Moreover, they are simpler and quicker, require less skill \non the part of the operator and can be attempted under local anaesthetic. Importantly, fewer op-\nerative complications have also been recorded. Thus they are clearly preferable to first generation \ntechniques. The association of shorter uterine cavity length and lower dissatisfaction with second \ngeneration ED could be because endoscopic treatment is technically more difficult, although \ngiven the borderline statistical significance it could also have arisen by chance.\nThe comparisons involving Mirena® were encouraging and given that it is a relatively cheap and \nminimally invasive procedure, it could be considered first if drug treatment for heavy bleeding \nfails70. It may even be an alternative to drug treatment as a first line agent but we did not address \nthis question in our review. However, research on Mirena® presents some difficulties in interpreta-\ntion, as seen in the trial by Hurskainen et al37, which compared Mirena® with hysterectomy. Whilst \nthe study was well conducted and reported, the lack of further investigation into the analysis of \nthe primary outcome measure (EQ-5D quality of life of life measure) made the interpretation of \nno evidence of a difference questionable.  Of the women allocated Mirena\n®, 20% had received \nhysterectomy before the main analysis time point at 12 months, with a further 12% no longer \nusing the Mirena®. Unfortunately, missing IPD from this trial meant we could not examine further.\nImplications for practice\nOur review provides evidence that hysterectomy reduces dissatisfaction compared with ED and \nthis information should be used as part of consultation with women making a choice about treat-\nment options when initial drug treatment fails to control heavy menstrual bleeding. ED is satisfac-\ntory for a very high proportion of women, but, if complete cessation of bleeding is sought, then \nhysterectomy may be offered. Although the evidence is not strong, our findings concur with a \nrecent NICE recommendation that women should be offered Mirena® before more invasive pro-\ncedures, particularly as this can be fitted in primary care10.  \nImplications for research\nThis review has shown that further investment in a randomised controlled trial comparing hyster-\nectomy with second generation ED would be of limited value, given the similar efficacy of first and \nsecond generation techniques. Questions remain about the long term clinical effectiveness of all \nthe treatments; evidence from trials with longer term follow-up (4/5 years or more) is limited to a \nhandful of studies involving differing comparisons\n36,37,71,72. Mirena®, in particular, versus alternative \nforms of surgical treatment requires further research. Whilst the small studies included in this review \nhave indicated promising results for this treatment, the substantial levels of non-compliance makes \ninterpretation of outcomes difficult and casts some doubt on the equivalent efficacy conclusions. \nChapter 9\n\n133\nIndividual patient data meta-analysis is an extremely powerful tool if used correctly73 and provides \nthe most definitive possible synthesis of the available evidence. Such collaborative meta-analy-\nses are well established in cancer and have greatly influenced clinical practice resulting in striking \nimprovements in, for example, breast cancer survival 23. Clinicians in speciality groups, such as \ngynaecology, need to be aware that contributing study results to an IPD is certainly as important \nas conducting the original research if not more so. Consensus on optimal outcome measures \nwould also be helpful for meta-analysis.  \nAcknowledgements\nWe would like to thank all authors of identified trials, for sending us their trial data, and both the \nBritish and European Societies of Gynaecological Endoscopy for their support and help with the \nreview.\nSupplementary material \nData supplements available at http://www.bmj.com/content/341/bmj.c3929\n*Members of Collaborative group\nThe following authors provided us with individual patient data from their trials \n(name, organisation, country): \nJ Abbott, University of New South Wales, Sydney, Australia; J Barrington, Torbay Hospital, South \nDevon, UK; S Bhattacharya, University of Aberdeen, Aberdeen Maternity Hospital, Aberdeen, \nUK; M.Y. Bongers, Maxima Medical Centre, Veldhoven,The Netherlands; J-L Brun, Hopital Uni-\nversitarie Pellegrin, Bordeaux, France; R Busfield, data supplied by M Sowter, Auckland Obstet-\nric Centre, New Zealand; T.J. Clark, Birmingham Womens Hospital, Birmingham, UK; J Cooper \n(2004 trial), data supplied by Microsulis Medical Ltd, Hampshire, UK; K.G. Cooper, Aberdeen \nRoyal Infirmary, Aberdeen, UK; S.L. Corson (2001 trial), data supplied by Boston Scientific Cor-\nporation, Marlborough, USA; K Dickersin, John Hopkins Bloomberg School of Public Health, \nUSA; N Dwyer, Weston General Hospital, Weston Super Mare, UK; M Gannon, Midland Regional \nHospital, Mullingar, Ireland; J Hawe, Countess of Chester Hospital, Chester, UK; R. Hurskainen, \nUniversity of Helsinki, Finland; W.R. Meyer; data supplied by Ethicon, A Johnson & Johnson \nCompany, New Jersey, USA; H. O’Connor, Coombe Women’s Hospital, Dublin 8, Ireland; S. \nPinion, Aberdeen Royal Infirmary, Aberdeen, UK; A.M. Sambrook, Aberdeen Royal Infirmary, Ab-\nerdeen, UK; W.H. Tam, Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong, \nChina; I.A.A. van Zon-Rabelink, Medical Spectrum Twente, Enschede, The Netherlands, E. Zupi, \nTor Vergata University, Rome, Italy.\nHMB IPD Meta-analysis\n\n134\nReferences\n1 Shapley M, Jordan K, Croft PR. An epidemiological survey of symptoms of menstrual loss \nin the community. Br J Gen Pract 2004;54(502):359-63.\n2 Chapple A. Menorrhagia: women’s perceptions of this condition and its treatment. J Adv \nNurs 1999 June;29(6):1500-6.\n3 Prentice A. Health care implications of dysfunctional uterine bleeding. Baillieres Best Pract \nRes Clin Obstet Gynaecol 1999 June;13(2):181-8.\n4 Lethaby A, Farquhar C. Treatments for heavy menstrual bleeding. BMJ 2003 November \n29;327(7426):1243-4.\n5 Maresh MJA, Metcalfe MA, McPherson K, Overton C, Hall V, Hargreaves J. The VALUE \nnational hysterectomy study: description of the patients and their surgery. British Journal \nof Obstetrics and Gynaecology 2002;109:302-12.\n6 Abbott J, Garry R. The surgical management of menorrhagia. Hum Reprod Update \n2002;8(68):78.\n7 Reid PC, Mukri F. Trends in number of hysterectomies performed in England for men-\norrhagia: examination of health episode statistics, 1989 to 2002-3. BMJ 2005 April \n23;330(7497):938-9.\n8 Reid P .C. Endometrial ablation in England - coming of age? An examination of hospi-\ntal episode statistics 1989/1990 to 2004/2005. Eur J Obstet Gynecol Reprod Biol \n2007;135(2):191-4.\n9 Lethaby AE, Cooke I, Rees M. Progesterone or progestogen-releasing intrauterine sys-\ntems for heavy menstrual bleeding. Cochrane Database Syst Rev 2005;(4):CD002126.\n10 National Collaborating Centre forWomen’s and Children’s Health. Heavy Menstrual Bleed-\ning.  2008. \n11 Lethaby A, Sheppard S, Cooke I, Farquhar C. Endometrial resection and ablation versus \nhysterectomy for heavy menstrual bleeding. Cochrane Database of Systematic Reviews \n1999;(2).\n12 Lethaby A, Hickey M. Endometrial destruction techniques for heavy menstrual bleeding. \nCochrane Database of Systematic Reviews 2005;(4).\n13 Marjoribanks J, Lethaby A, Farquhar C. Surgery versus medical therapy for heavy men-\nstrual bleeding. Cochrane Database of Systematic Reviews 2006;(2).\n14 Stewart LA, Parmar MK. Meta-analysis of the literature or of individual patient data: is \nthere a difference? Lancet 1993 February 13;341(8842):418-22.\n15 Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions Ver-\nsion 5.0.1.  2008.  The Cochrane Collaboration. Ref Type: Online Source\n16 Stewart LA, Tierney JF. To IPD or not to IPD? Advantages and disadvantages of system-\natic reviews using individual patient data. Eval Health Prof 2002 March;25(1):76-97.\nChapter 9\n\n135\n17 Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP et al. The \nPRISMA statement for reporting systematic reviews and meta-analyses of studies that \nevaluate health care interventions: explanation and elaboration. J Clin Epidemiol 2009 \nOctober;62(10):e1-34.\n18 Clark TJ, Khan KS, Foon R, Pattison H, Bryan S, Gupta JK. Quality of life instruments \nin studies of menorrhagia: a systematic review.  Eur J Obstet Gynecol Reprod Biol \n2002;104(2):96-104.\n19 Higham JM, O’Brien PM, Shaw RW. Assessment of menstrual blood loss using a pictorial \nchart. Br J Obstet Gynaecol 1990 August;97(8):734-9.\n20 EuroQol Group. EuroQol - a new facility for the treatment of health-related quality of life. \nHealth Policy 1990;16:199-208.\n21 Ware JE, Sherbourne C.D. The MOS 36-item short-form health survey (SF-36), I: concep-\ntual framework and item selection. Med Care 1992;30:473-83.\n22 Riley RD, Lambert PC, Staessen JA, Wang J, Gueyffier F, Thijs L et al. Meta-analysis of \ncontinuous outcomes combining individual patient data and aggregate data. Stat Med \n2008 May 20;27(11):1870-93.\n23 Early Breast Cancer Trialists’ Collaborative Group. Treatment of Early Breast Cancer: \nWorldwide Evidence, 1985-1990.  Oxford University Press; 1990. \n24 Peto R, Pike MC, Armitage P , Breslow NE, Cox DR, Howard SV et al. Design and analysis \nof randomized clinical trials requiring prolonged observation of each patient. II. analysis \nand examples. Br J Cancer 1977 January;35(1):1-39.\n25 Fleiss JL. The statistical basis of meta-analysis. Statistical Methods in Medical research \n1993;2:121-45.\n26 Glenny AM, Altman DG, Song F, Sakarovitch C, Deeks JJ, D’Amico R et al. Indirect com-\nparisons of competing interventions. Health Technol Assess 2005 July;9(26):1-iv.\n27 Hasselblad V. Meta-analysis of multitreatment studies. Med Decis Making 1998 Janu-\nary;18(1):37-43.\n28 Whitehead A. Meta-Analysis of Controlled Clinical Trials. Wiley, Chichester; 2002.\n29 Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. \nBiometrics 1986 March;42(1):121-30.\n30 Crosignani PG, Vercellini P , Apolone G, De Giorgi O, Cortesi I, Meschia M. Endometrial \nresection versus vaginal hysterectomy for menorrhagia: long-term clinical and quality-of-\nlife outcomes. American Journal of Obstetrics and Gynecology 1997;177(1):95-101.\n31 Dwyer N, Hutton J, Stirrat GM. Randomised controlled trial comparing endometrial re-\nsection with abdominal hysterectomy for the surgical treatment of menorrhagia. British \nJournal of Obstetrics and Gynaecology 1993;100:237-43.\n32 Gannon M, Holt EM, Fairbank J, Fitzgerald M, Milne MA, Crystal A.M. et al. A randomised \ncontrol trial comparing endometrial resection and abdominal hysterectomy for the treat-\nment of menorrhagia. BMJ 1991;303:1362-4.\nHMB IPD Meta-analysis\n\n136\n33 O’Connor H, Broadbent JA, Magos AL, McPherson K. Medical Research Council ran-\ndomised trial of endometrial resection versus hysterectomy in the management of menor-\nrhagia. Lancet 1997;349(9056):891-901.\n34 Pinion SB, Parkin DE, Abramovich DR, Naji A, Alexander DA, Russell IT. Randomised trial \nof hysterectomy, endometrial laser ablation, and transcervical endometrial resection for \ndysfunctional uterine bleeding. BMJ 1994;309(6960):979-83.\n35 Zupi E, Zullo F, Marconi D, Sbracia M, Pellicano M, Solima E. et al. Hysteroscopic endome-\ntrial resection versus laparoscopic supracervical hysterectomy for menorrhagia: a prospec-\ntive randomized trial. American Journal of Obstetrics and Gynecology 2003;188(1):7-12.\n36 Dickersin K, Munro MG, Clark M, Langenberg P , Scherer R, Frick K. et al. Hysterectomy \ncompared with endometrial ablation for dysfunctional uterine bleeding: a randomized \ncontrolled trial. Obstetrics and Gynecology 2007;110(6):1279-89.\n37 Hurskainen R, Teperi J, Rissanen P , Aalto AM, Grenman S, Kivela A. Quality of life and \ncost-effectiveness of levonorgestrel-releasing intrauterine system versus hysterectomy for \ntreatment of menorrhagia: a randomised trial. Lancet 2001;357(9252):273-7.\n38 Hawe J, Abbott J, Hunter D, Phillips G, Garry R. A randomised controlled trial compar -\ning the Cavaterm endometrial ablation system with the Nd:YAG laser for the treatment of \ndysfunctional uterine bleeding. BJOG 2003;110(4):350-7.\n39 van Zon-Rabelink IA, Vleugels MPH, Merkus HMWM, de Graaf R. Efficacy and satisfac-\ntion rate comparing endometrial ablation by rollerball electrocoagulation to uterine bal-\nloon thermal ablation in a randomised controlled trial. Eur J Obstet Gynecol Reprod Biol \n2004;114:97-103.\n40 Soysal ME, Soysal SK, Vicdan K. Thermal balloon ablation in myoma-induced menorrhagia \nunder local anesthesia. Gynecologic and obstetric investigation 2001;51(2):128-33.\n41 Romer T. [Therapy of recurrent menorrhagia--Cavaterm balloon coagulation versus roller-\nball endometrium coagulation--a prospective randomized comparative study]. Zentralbl \nGynakol 1998;120(10):511-4.\n42 Meyer WR, Walsh BW, Grainger DA, Peacock LM, Loffer FD, Steege JF. Thermal Balloon \nand rollerball ablation to treat menorrhagia: a multicenter comparison. Obstetrics and \nGynecology 1998;92:98-103.\n43 Duleba AJ, Heppard MC, Soderstrom RM, Townsend DE. A randomized study compar-\ning endometrial cryoablation and rollerball electroablation for treatment of dysfunctional \nuterine bleeding. Journal of the American Association of Gynecologic Laparoscopists \n2003;10(1):17-26.\n44 Cooper J, Gimpelson R, Laberge P , Galen D, Garza-Leal JG, Scott J. A randomised, mul-\nticenter trial of safety and efficacy of the NovaSure system in the treatment of menorrhagia. \nJournal of the American Association of Gynecologic Laparoscopists 2002;9(4):418-28.\nChapter 9\n\n137\n45 Cooper J, Anderson TL, Fortin CA, Jack SA, Plentl M.B. Microwave endometrial ablation \nvs. rollerball electroablation for menorrhagia: a multicenter randomized trial. Journal of the \nAmerican Association of Gynecologic Laparoscopists 2004;11(3):394-403.\n46 Perino A, Castelli A, Cucinella G, Bionda A, Pane A, Venezia R. A randomized comparison \nof endometrial laser intrauterine thermotherapy and hysteroscopic endometrial resection. \nFertility and Sterility 2004;82(3):731-4.\n47 Cooper KG, Bain C, Parkin DE. Comparison of microwave endometrial ablation and tran-\nscervical resection of the endometrium for treatment of heavy menstrual loss: a ran-\ndomised trial. Lancet 1999;354(9193):1859-63.\n48 Corson SL, Brill AI, Brooks PG, Cooper JM, Indman PD, Liu JH. et al. One-year results \nof the vesta system for endometrial ablation. Journal of the American Association of Gy-\nnecologic Laparoscopists 2000;7(4):489-97.\n49 Corson SL. A multicenter evaluation of endometrial ablation by Hydro ThermAblator and \nrollerball for treatment of menorrhagia. Journal of the American Association of Gyneco-\nlogic Laparoscopists 2001;8(3):359-67.\n50 Pellicano M, Guida M, Acunzo G, Cirillo D, Bifulco G, Nappi C. Hysteroscopic transcer -\nvical endometrial resection versus thermal destruction for menorrhagia: a prospective \nrandomized trial on satisfaction rate. American Journal of Obstetrics and Gynecology \n2002;187(3):545-50.\n51 Brun JL, Burlet G, Galand B, Quereux C, Bernard P . Cavaterm thermal balloon endometrial \nablation versus hysteroscopic endometrial resection to treat menorrhagia: the French, mul-\nticenter, randomized study. Journal of Minimally Invasive Gynecology 2006;13(5):424-30.\n52 Malak KA. Management of menorrhagia with the levonorgestrel intrauterine system ver -\nsus endometrial resection. Gynaecological Surgery 2006;3(4):275-80.\n53 Kittelsen N, Istre O. A randomized study comparing levonorgestrel intrauterine system \n(LNG IUS) and transcervical resection of the endometrium (TCRE) in the treatment of \nmenorrhagia: preliminary results. Gynaecological Endoscopy 1998 April 9;7(2):61-5.\n54 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 \ndysfunctional uterine bleeding. Obstetrics and Gynecology 1997;90(2):257-63.\n55 Soysal M, Soysal S, Ozer S. A randomized controlled trial of levonorgestrel releasing IUD \nand thermal balloon ablation in the treatment of menorrhagia. Zentralbl Gynakol 2002 \nApril;124(4):213-9.\n56 Barrington JW, Arunkalaivanan AS, Abdel-Fattah M. Comparison between the levonorg-\nestrel intrauterine system (LNG-IUS) and thermal balloon ablation in the treatment of men-\norrhagia. Eur J Obstet Gynecol Reprod Biol 2003 May 1;108(1):72-4.\nHMB IPD Meta-analysis\n\n138\n57 Busfield RA, Farquhar CM, Sowter MC, Lethaby A, Sprecher M, Yu Y et al. A randomised \ntrial comparing the levonorgestrel intrauterine system and thermal balloon ablation for \nheavy menstrual bleeding. BJOG 2006 March;113(3):257-63.\n58 Shaw RW, Symonds IM, Tamizian O, Chaplain J, Mukhopadhyay S. Randomised com-\nparative trial of thermal balloon ablation and levonorgestrel intrauterine system in patients \nwith idiopathic menorrhagia. Aust N Z J Obstet Gynaecol 2007 August;47(4):335-40.\n59 Tam WH, Yuen PM, Shan Ng DP , Leung PL, Lok IH, Rogers MS. Health Status Function \nafter treatment with thermal balloon endometrial ablation and levnonorgestrel intrauterine \nsystem for idiopathic menorrhagia: A randomised study. Gynecology and Obstetric Inves-\ntigation 2006;62(2):84-8.\n60 Abbott J, Hawe J, Hunter D, Garry R. A double-blind randomized trial comparing the Ca-\nvaterm and the NovaSure endometrial ablation systems for the treatment of dysfunctional \nuterine bleeding. Fertility and Sterility 2003;80(1):203-8.\n61 Bhattacharya S, Cameron IM, Parkin DE, Abramovich DR, Mollison J, Pinion SB. et al. A \npragmatic randomised comparison of transcervical resection of the endometrium with en-\ndometrial laser ablation for the treatment of menorrhagia. BJOG 1997;104(5):601-7.\n62 Bongers MY, Bourdrez P , Mol BW, Heintz AP , Brolmann HA. Randomised controlled trial \nof bipolar radio-frequency endometrial ablation and balloon endometrial ablation. BJOG \n2004;111(10):1095-102.\n63 Clark TJ. A randomised controlled trial to conmpare the effectiveness of outpatient endo-\nmetrial ablation  techniques (NovaSure versus Thermachoice) in the treatment of menor-\nrhagia. Unpublished 2008.\n64 Sambrook AS, Cooper KG, Campbell MK, Cook JA. Clinical outcomes from a randomised \ncomparison of Microwave Endometrial Ablation with Thermal Balloon endometrial abla-\ntion for the treatment of heavy menstrual bleeding. BJOG 2009;116(8):1038-45.\n65 Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of \nreports of meta-analyses of randomised controlled trials: the QUOROM statement. Qual-\nity of Reporting of Meta-analyses. Lancet 1999;354(9193):1896-900.\n66 Overton C, Hargreaves J, Maresh M. A national survey of the complications of endome-\ntrial destruction for menstrual disorders: the MISTLETOE study. Minimally Invasive Surgi-\ncal Techniques- Laser, EndoThermal or Endoresection. British Journal of Obstetrics and \nGynaecology 1997;104(12):1351-9.\n67 Loffer FD. Three-year comparison of thermal balloon and rollerball ablation in tretament \nof menorrhagia. Journal of the American Association of Gynecologic Laparoscopists \n2001;8(1):48-54.\n68 Loffer FD, Grainger D. Five-year follow-up of patients participating in a randomised trial of \nuterine balloon therapy versus rollerball ablation for treatment of menorrhagia. Journal of \nthe American Association of Gynecologic Laparoscopists 2002;9(4):429-35.\nChapter 9\n\n139\n69 Cooper KG, Bain C, Lawrie L, Parkin DE. A randomised comparison of microwave endo-\nmetrial ablation with transcervical resection of the endometrium; follow up at a minimum \nof five years. British Journal of Obstetrics and Gynaecology 2005;112(4):470-5.\n70 Brown PM, Farquhar CM, Lethaby A, Sadler LC, Johnson NP . Cost-effectiveness analysis \nof levonorgestrel intrauterine system and thermal balloon ablation for heavy menstrual \nbleeding. BJOG 2006;113:797-803.\n71 Cooper KG, Jack SA, Parkin DE, Grant AM. Five-year follow up of women randomised to \nmedical management or transcervical resection of the endometrium as treatment for heavy \nmenses. British Journal of Obstetrics and Gynaecology 2001;108:1222-8.\n72 Sambrook AM, Bain C, Parkin DE, Cooper KG. A randomised comparison of microwave \nendometrial ablation with transcervical resection of the endometrium: follow up at a mini-\nmum of 10 years. BJOG 2009 July;116(8):1033-7.\n73 Stewart LA, Clarke MJ. Practical methodology of meta-analyses (overviews) using \nupdated individual patient data. Cochrane Working Group. Stat Med 1995 October \n15;14(19):2057-79.\n74 Dickersin K, Munro M, Langenberg P , Scherer R, Frick KD, Weber AM. et al. Surgical \nTreatments Outcomes Project for Dysfunctional Uterine Bleeding (STOP-DUB): design \nand methods. Controlled Clinical Trials 2003;24:591-609.\n75 Sculpher MJ, Dwyer N, Byford S, Stirrat GM. Randomised trial comparing hysterectomy \nand transcervical endometrial resection: effect on health related quality of life and costs \ntwo years after surgery. BJOG 1996;103:142-9.\n76 Sculpher M. A cost-utility analysis of abdominal hysterectomy versus transcervical en-\ndometrial resection for the surgical treatment of menorrhagia. International Journal of \nTechnology Assessment in Health Care 1998;14:302-19.\n77 Halmesmaki K, Hurskainen R, Teperi J, Grenman S, Kivela A, Kujansuu E. et al. The \neffect of hysterectomy or levonorgestrel-releasing intrauterine system on sexual func-\ntioning among women with menorrhagia: a 5-year randomised controlled trial. BJOG \n2007;114:563-8.\n78 van Zon-Rabelink IA, Vleugels MP , Merkus HM, de Graaf R. Endometrial ablation by roller-\nball electrocoagulation compared to uterine balloon thermal ablation. Technical and safety \naspects. Eur J Obstet Gynecol Reprod Biol 2003;110(2):220-3.\n79 Bain C, Cooper KG, Parkin DE. Microwave endometrial ablation versus endometrial re-\nsection: a randomized controlled trial. Obstetrics and Gynecology 2002;99(6):983-7.\n80 Rauramo I, Elo I, Istre O. Long-term treatment of menorrhagia with levonorgestrel intrauter-\nine system versus endometrial resection. Obstetrics and Gynecology 2004;104:1341-21.\nHMB IPD Meta-analysis\n\n140\n\n\n141\nChapter 10\nSecond generation endometrial ablation \ntechniques for heavy menstrual bleeding: \nnetwork meta-analysis  \n\n142\nChapter 10\nAbstract\nObjective: To determine the relative effectiveness of second generation ablation techniques in \nthe treatment of heavy menstrual bleeding.\nDesign: Network meta-analysis on the primary outcome measures of amenorrhoea, heavy \nbleeding, and patients’ dissatisfaction with treatment.\nData sources: Nineteen randomised controlled trials (involving 3287 women) were identified \nthrough electronic searches of the Cochrane Library, Medline, Embase and PsycINFO databases \nfrom inception to April 2011. The reference lists of known relevant articles were searched for \nfurther articles. Two reviewers independently selected articles without language restrictions.\nEligibility criteria for selecting studies: Randomised controlled trials involving second gen-\neration endometrial destruction techniques for women with heavy menstrual bleeding unrespon-\nsive to medical treatment.\nResults: Of the three most commonly used techniques, network meta-analysis showed that \nbipolar radiofrequency and microwave ablation resulted in higher rates of amenorrhoea than \nthermal balloon ablation at around 12 months (odds ratio 2.51, 95% confidence interval 1.53 to \n4.12, p<0.001; and 1.66, 95% CI 1.01 to 2.71, p=0.05, respectively), but there was no evidence \nof a convincing difference between the three techniques in the number of women dissatisfied \nwith treatment or still experiencing heavy bleeding. Compared with bipolar radio frequency and \nmicrowave devices, an increased number of women still experienced heavy bleeding after free \nfluid ablation (2.19, 95% CI 1.07 to 4.50, p=0.03; and 2.91, 95% CI 1.23 to 6.88, p=0.02, re-\nspectively). Compared with radio frequency ablation, free fluid ablation was associated with re-\nduced rates of amenorrhoea (0.36, 95% CI 0.19 to 0.67, p=0.004) and increased rates of dissat-\nisfaction (4.79, 95% CI 1.07 to 21.5, p=0.04). Of the less commonly used devices, endometrial \nlaser intrauterine thermotherapy was associated with increased rates of amenorrhoea compared \nwith all the other devices, while cryoablation led to a reduced rate compared with bipolar radio \nfrequency and microwave.\nConclusions: Bipolar radio frequency and microwave ablative devices are more effective than \nthermal balloon and free fluid ablation in the treatment of heavy menstrual bleeding with second \ngeneration endometrial ablation devices.\nChapter 10\n\n143\nIntroduction\nHeavy menstrual bleeding is a common incapacitating problem in women of reproductive age, 1 \nwhich can severely affect their quality of life.2,3 When medical treatments fail to provide adequate \nrelief, surgical interventions, including hysterectomy or destruction of the endometrium, can \nbe considered.4 Initially, rollerball ablation, transcervical resection, and laser ablation were the \npredominant endometrial destruction techniques performed under direct hysteroscopic vision. \nThese led to a rapid decrease in the number of hysterectomies performed.\n5\nOver the past decade, a second generation of non-hysteroscopic techniques, which are safer, \ntechnically easier to perform, involve shorter hospital stays, and can be performed under local \nanaesthesia, have become dominant.6,7 These involve devices that are sited and activated to treat \nthe whole endometrial cavity. Destruction is achieved through various methods, the most com-\nmonly used\n8 being high temperature fluids within a balloon (Thermachoice and Cavaterm) 9-11 or \napplication of microwave (Microsulis)12,13 or bipolar radiofrequency electrical energy (Novasure).14 \nLess commonly used ablative techniques include free fluid at high temperature (Hydrotherma-\nblator),\n15 endometrial laser intrauterine thermotherapy (ELITT), 16 and cryoablation (HerOption). 17 \nOther than free fluid thermal ablation, these are “blind” techniques. A recent meta-analysis of in-\ndividual patient data found that second generation techniques were at least as effective, in terms \nof patients’ dissatisfaction, as first generation techniques and, importantly, gave rise to fewer \noperative complications, were considerably quicker, and were less likely to need to be performed \nunder general anaesthesia.\n18 Thus they are clearly preferable to first generation techniques.\nWhen more than two treatments options are available, the use of network meta-analysis, which \nallows us to combine both direct and indirect estimates for treatment effects, has become in-\ncreasingly common.\n19,20 Randomised comparisons within trials are preserved and the main as-\nsumption—that there are no systematic differences between the trials that could bias the indirect \nmeasurements—is similar to those underlying any standard meta-analysis.\nThere have been numerous randomised controlled trials comparing the different second genera-\ntion techniques with each other or with first generation devices, yet with the sheer number of \npotential comparisons there are gaps in direct evidence for efficacy of some techniques, while \nother comparisons have low statistical power because of the small numbers randomised. We \ncarried out a network meta-analysis to assess the relative effectiveness of six second generation \ntechniques to provide both users and recipients of these devices with an improved evidence \nbase to inform their decision making.\nHMB Network Meta-analysis\n\n144\nMethods\nThe systematic review was conducted based on a protocol designed with widely recommended \nmethods21,22 that complied with reporting guidelines for meta-analyses.23\nLiterature search and study selection\nWe carried out an extensive literature search in the Cochrane Library, Medline, Embase, and Psy-\ncINFO from inception to April 2011. Our search term combinations consisted of MeSH subhead-\nings, text words and word variations for “heavy menstrual bleeding,” and the various types of first \nand second generation endometrial destruction techniques. This search strategy was adapted \nto suit each database and was restricted to “humans” and “females”. Bibliographies of known \nrelevant primary articles and reviews were searched by hand to identify articles missed by the \nelectronic searches. Were also contacted experts in the specialty to uncover any grey literature. \nA comprehensive database was constructed with Reference Manager 12.0 to store all identified \nreferences. No language restriction was applied.\nStudies were selected in a two step process. Firstly, we scrutinised the citations identified by \nthe electronic searches and subsequently obtained full manuscripts of citations that met or were \nthought to have met the predetermined inclusion criteria. We included studies if they recruited \nwomen with heavy, abnormal, excessive, or prolonged uterine bleeding that was unresponsive \nto initial medical treatment and the study design was a randomised controlled trial comparing \nendometrial ablation methods.\nData collection and study quality assessment\nTwo reviewers (LJM and RC) independently inspected all the manuscripts to determine if they \nmet the above criteria. When possible we obtained individual patient data from the primary au-\nthor using methods described in the related meta-analysis.\n18 When raw data were not available, \ntwo independent reviewers (RC and LJM) used predesigned forms to extract aggregate data \nfrom manuscripts. Any disagreements were resolved by consensus or arbitration by a third re-\nviewer (JPD).\nOutcome measures used to evaluate the treatments were those most commonly used across \nstudies (Table 1): rate of amenorrhoea (converted from a pictorial bleeding assessment score of \nzero when these data existed, otherwise as reported), rate of heavy bleeding (likewise converted \nfrom a pictorial bleeding assessment score of >10024 when data existed, otherwise as reported), \nand rate of dissatisfaction with treatment (see related paper for a full description of how dissatis-\nfaction was calculated).\n18 We used data at 12 months (the most popular time point) as the focus \nfor analysis or results at two years if 12 month data were not available.\nChapter 10\n\n145\nTable 1:  Characteristics of included trials and event rates for amenorrhoea, heavy bleeding \nand dissatisfaction for the systematic review and network meta-analysis of second generation \nendometrial destruction techniques for heavy menstrual bleeding.\nPrimary \nauthor\nYear Primary outcome \nmeasure \nSecond generation \ntreatment device\nManufacturer/Trade \nname\nAmenorrhoea rate Heavy bleeding rate Dissatisfaction rate\n2nd \ngeneration\n1st \ngeneration\n2nd \ngeneration\n1st \ngeneration\n2nd \ngeneration\n1st \ngeneration\nStudies comparing second generation treatments\nClark31 2011 Amenorrhoea rate\nBipolar radio frequency NovaSure 14/25 - 2/25 - 2/28 -\nThermal balloon Thermachoice 6/26 - 6/26 - 6/32 -\nPenninx32 2010 Amenorrhoea rate\nBipolar radio frequency NovaSure 35/75 - 8/75 - 1/75 -\nFree-fluid BEI Medical Systems 17/71 - 10/71 - 10/71\nSambrook33 2009 Satisfaction rate\nMicrowave Microsulis 61/149 - 6/149 - 35/144 -\nThermal balloon Thermachoice 51/134 - 7/134 - 31/134 -\nBongers34 2004 Amenorrhoea rate\nBiploar radio frequency NovaSure 34/79 - 14/79 - 8/83 -\nThermal balloon Thermachoice 3/38 - 7/38 - 8/43 -\nAbbott35 2003 Amenorrhoea rate\nBiploar radio frequency NovaSure 16/37 - 5/37 - 0/34 -\nThermal balloon Cavaterm 2/17 - 0/17 - 1/15 -\nStudies comparing second generation treatments with first generation hysteroscopic devices\nBrun36 2006 Amenorrhoea rate Thermal balloon Cavaterm 11/30 5/17 2/30 2/17 2/21 3/16\nCooper37 2004 Heavy bleeding rate Microwave Microsulis 120/194 49/96 5/194 5/96 3/196 1/97\nPerino16 2004 Amenorrhoea rate Laser ELITT 35/56 14/55 2/56 4/55 3/56 5/55\nDuleba17 2003 Heavy bleeding rate Cryoablation HerOption 43/156 40/72 19/156 5/72 16/156 10/72\nHawe38 2003 Amenorrhoea rate Thermal balloon Cavaterm 10/30 13/28 1/30 1/28 2/30 1/28\nVan Zon-\nRabelink39\n2003 Heavy bleeding rate Thermal balloon Thermachoice 6/74 4/55 23/74 19/55 15/75* 13/58*\nCooper14 2002 Heavy bleeding rate Biploar radio frequency NovaSure 63/154 29/82 14/154 10/82 11/154 5/82\nPellicano40 2002 Satisfaction rate Thermal balloon Cavaterm nm nm nm nm 7/37 14/38\nCorson15 2001 Heavy bleeding rate Free-fluid BEI Medical Systems 65/172 43/83 31/172 12/83 nm nm\nSoysal41 2001 Heavy bleeding rate Thermal balloon Thermachoice 5/45 8/48 11/45 10/48 15/45 19/48\nCorson42 2000 Heavy bleeding rate Thermal balloon Vesta 38/122 39/112 16/122 19/112 nm nm\nCooper12 1999 Satisfaction rate Microwave Microsulis 46/116 48/124 9/116 11/124 27/116 32/124\nMeyer9 1998 Heavy bleeding rate Thermal balloon Thermachoice 18/122 32/117 18/122 13/117 5/122 1/116\nRomer43 1998 Not stated Thermal balloon Cavaterm 4/10 3/10 0/10 0/10 nm nm\nnm - outcome not measured \n* – indicates data from 24 months post-randomisation but considered as 12 month data. All other data are follow-up data at 12 months post-randomisation.\nHMB Network Meta-analysis\n\n146\nStatistical analysis\nFor all three outcomes we first assessed direct estimates in individual trials by calculating point \nestimates and 95% confidence intervals. We then calculated a traditional weighted average me-\nta-analysis yielding a Mantel-Haenszel odds ratio, with heterogeneity investigated with Cochran’s \nQ\n25 and I2 statistics.26\nWe carried out the network meta-analysis by fitting a linear mixed model to the log odds ratio \nfrom each trial with the GLIMMIX procedure in SAS, using maximum likelihood to estimate treat-\nment effects. Two models were attempted; the first included a fixed treatment effect and allowed \nthe effect of the study to vary randomly. The second model also included a random effect for \nthe difference in treatment in each trial by including a study by treatment interaction parameter.\n27 \nThe inclusion of a random study effect allowed the recovery of any differences between trials, \nwhich has greater importance when not all of the treatments are included in every trial, as is the \ncase here.\n28 We also included a random effect for treatment pair (that is, to separate those trials \ncomparing different treatments types) in both models to allow for any inconsistency or “incoher-\nence” in the model.29 This effect (ω, the standard deviation of the random effect) allows for any \nindirect estimates to be inconsistent with any direct estimates of the same treatments and can be \ncalculated where there any closed “loops” of treatments exist (for instance, A v B, B v C, A v C). \nIncreased incoherence increases standard errors, and hence the size of the confidence interval, \nof differences between treatments. If large levels of incoherence are present it might not be sen-\nsible to combine estimates with this method,\n30 analogous to having high levels of heterogeneity \nin a standard meta-analysis. We used Revman v5.0 (Cochrane Collaboration, Denmark) and SAS \nv9.2 (SAS Institute, Cary, NC) software for analysis.\nResults\nTrials and patients\nWe identified 719 citations through the electronic literature searches (Figure 1) and excluded 672 \nafter screening titles and abstracts. A further 28 were excluded for being duplicate publications \nor not involving direct comparisons of endometrial ablative techniques. After detailed evaluation \nof the citations, 19 primary articles met the selection criteria and were included in the network \nmeta-analysis (see Supplementary data for full details of these 19 studies). These consisted of \nfive head to head comparisons of second generation devices31-35 (744 women) and 14 compar-\ning a second generation device with a first generation (hysteroscopic) device 9,12,14-17,36-43 (2543 \nwomen) (Figure 1). Table 1 shows further descriptions of device type and the frequencies used \nin the analysis. As the fixed and random effect network meta-analysis models produced similar \nresults only those from the latter are described here.\nChapter 10\n\n147\nFigure 1  Study selection process for systematic review and network meta-analysis of second  \n generation endometrial destruction techniques for heavy menstrual bleeding.\nFigure 2  Network of studies evaluating second generation endometrial destruction devices \n for treatment of heavy menstrual bleeding. Number of women randomised to  \n second generation treatment are shown in parentheses.\nCitations identi/f_ied by electronic searches (n=/seven.tf/one.tf/nine.tf)\nCitations retrieved for detailed evaluation (n=/four.tf/seven.tf)\nArticles included in systematic review and meta-analysis (n=/one.tf/nine.tf)\nTrials comparing second\ngeneration devices\nwith /f_irst generation\n(hysteroscopic) devices \n(n=/one.tf/four.tf, /two.tf/five.tf/four.tf/three.tf women)\nTrials comparing two\nsecond generation devices\n(n=/five.tf, /seven.tf/four.tf/four.tf women)\nCitations excluded a/f_ter screening\ntitles and abstracts (n=/six.tf/seven.tf/two.tf)\nExcluded (n=/two.tf/eight.tf):\n  Duplication (n=/two.tf/two.tf)\n  Inappropriate comparator (n=/six.tf)\nFirst generation\nhysteroscopic deviceFree fluid\nOne study /three.tf/two.tf\n(n=/one.tf/six.tf/zero.tf)\nBipolar radio\nfrequency\nThree studies /three.tf/one.tf /three.tf/four.tf /three.tf/five.tf\n(n=/two.tf/six.tf/four.tf)\nThermal\nballoon\nOne study /three.tf/three.tf\n(n=/three.tf/two.tf/zero.tf) Microwave\nLaser\nCryoablation\nOne study /one.tf/four.tf\n(n=/one.tf/seven.tf/five.tf)\nEight studies /nine.tf /three.tf/six.tf /three.tf/eight.tf-/four.tf/three.tf\n(n=/five.tf/one.tf/eight.tf)\nTwo studies /one.tf/two.tf /three.tf/seven.tf\n(n=/three.tf/three.tf/eight.tf)\nOne study /one.tf/six.tf\n(n=/five.tf/six.tf)\nOne study /one.tf/seven.tf\n(n=/one.tf/nine.tf/three.tf)\nOne study /one.tf/five.tf\n(n=/one.tf/eight.tf/seven.tf)\nHMB Network Meta-analysis\n\n148\nEffectiveness in reducing bleeding\nDirect comparisons show an increased rate of amenorrhoea with bipolar radio frequency ablation \ncompared with thermal balloon ablation (odds ratio 4.56, 95% confidence interval 2.24 to 9.26; \np<0.001) (Table 2). Free fluid thermal ablation was associated with reduced rates of amenorrhoea \n(0.36, 95% CI 0.18 to 0.73; p=0.005) and increased rates of heavy bleeding (4.88, 95% CI 1.32 \nto 18.11; p=0.02, Table 3) compared with bipolar radio frequency ablation. We found no other \nconvincing differences. When there was more than one study in a comparison, estimates be-\ntween studies were highly consistent (all I\n2=0% apart from the comparison of amenorrhoea rates \nwith microwave ablation versus first generation techniques where I2=17%).\nResults from the network meta-analysis (Tables 2 and 3 and Figure 3) concurred with the results \nof the direct evidence, with an increased rate of amenorrhea with bipolar radio frequency abla-\ntion compared with thermal balloon ablation (2.51, 95% CI 1.53 to 4.12; p<0.001). Results were \nalso similar for the comparison between free fluid thermal ablation and bipolar radio frequency \nablation for amenorrhoea (0.36, 95% CI 0.19 to 0.67; p=0.004) and heavy bleeding (2.19, 95% \nCI 1.07 to 4.50; p=0.03). There was an increase in the rate of amenorrhoea with radio frequency \nablation compared with cryoablation in the indirect comparison (0.20, 95% CI 0.09 to 0.49; \np=0.002). Microwave ablation was associated with an increased rate of amenorrhoea compared \nwith thermal balloon ablation (1.66, 95% CI 1.01 to 2.71; p=0.05) and cryoablation (0.31, 95% CI \n0.13 to 0.74; p=0.01) and some reduction in the rate of heavy bleeding compared with free fluid \nablation (2.91, 1.23 to 6.88; P=0.02). The endometrial laser intrauterine thermotherapy (ELITT) \nwas associated with increased rates amenorrhoea compared with all the other devices, with \nwide confidence intervals reflecting the small number of women evaluated with this device in ran-\ndomised controlled trials (Table 2). No other compelling differences between devices were noted. \nThere were relatively low but variable rates of heavy bleeding across the studies (such as 0-32% \nfor thermal balloon ablation and 8-18% for bipolar radiofrequency ablation).\nChapter 10\n\n149\nTable 2  Amenorrhoea rate at 12 months - results from direct comparisons and network  \n meta-analysis \nThermal \nballoon\nBipolar radio \nfrequency\nMicrowave Cryoablation Free-fluid Laser\nFirst \ngeneration \ndevice\nDirect 0.72 \n(0.52,1.01); \np=0.06a\n1.27 \n(0.73,2.20); \np=0.4c\n1.28 \n(0.90,1.83); \np=0.2b\n0.30 \n(0.17,0.55); \np<0.0001c\n0.57 \n(0.33,0.96); \np=0.03c\n4.88 \n(2.17,11.00); \np=0.0001c\nNetwork 0.69 \n(0.49,0.97); \np=0.03\n1.73 \n(1.07,2.78); \np=0.03\n1.14 \n(0.73,1.79); \np=0.5\n0.35 \n(0.17,0.75); \np=0.01\n0.62 \n(0.34,1.13); \np=0.1\n4.36 \n(1.82,10.44); \np=0.002\nThermal \nballoon\nDirect 4.56 \n(2.24,9.26); \np<0.0001a\n1.13 \n(0.70,1.82); \np=0.6c\nNot available Not available Not available\nNetwork 2.51 \n(1.53,4.12); \np=0.0009\n1.66 \n(1.01,2.71); \np=0.05\n0.51 \n(0.23,1.17); \np=0.1\n0.91 \n(0.48,1.73); \np=0.7\n6.34 \n(2.50,16.07); \np=0.0004\nBipolar radio \nfrequency\nDirect Not available Not available 0.36 \n(0.18,0.73); \np=0.005c\nNot available\nNetwork 0.66 \n(0.36,1.21); \np=0.2\n0.20 \n(0.09,0.49); \np=0.002\n0.36 \n(0.19,0.67); \np=0.004\n2.52 \n(0.95,6.71); \np=0.06\nMicrowave Direct Not available Not available Not available\nNetwork 0.31 \n(0.13,0.74); \np=0.01\n0.55 \n(0.27,1.13); \np=0.09\n3.82 \n(1.46,10.01); \np=0.009\nCryoablation Direct Not available Not available\nNetwork 1.77 \n(0.69,4.58); \np=0.2\n12.37 \n(3.96,38.59); \np=0.0002\nFree-fluid Direct Not available \nNetwork 6.98 \n(2.48,19.69); \np=0.0008\nFigures are odds ratios (95% confidence intervals) for device in column compared with device in row. Odds ratio >1 indicate \nincreased rate with device in column. a I2=0%, b I2 =17%, c I2 not applicable, only one study in comparison.\nHMB Network Meta-analysis\n\n150\nTable 3  Heavy bleeding rate at 12 months - results from direct comparisons and network   \n meta-analysis \nThermal \nballoon\nBipolar radio \nfrequency\nMicrowave Cryoablation Free-fluid Laser\nFirst \ngeneration \ndevice\nDirect 0.97 \n(0.66,1.41); \np=0.9a\n0.72 \n(0.30,1.70); \np=0.5b\n0.71 \n(0.34,1.50); \np=0.4a\n1.86 \n(0.67,5.19); \np=0.2b\n1.30 \n(0.63,2.69); \np=0.5b\n0.47 \n(0.08,2.69); \np=0.4b\nNetwork 0.98 \n(0.69,1.39); \np=0.9\n0.95 \n(0.56,1.62); \np=0.8\n0.59 \n(0.32,1.09); \np=0.09\n1.55 \n(0.63,3.81); \np=0.3\n1.37 \n(0.74,2.51); \np=0.3\n0.39 \n(0.07,2.09); \np=0.3\nThermal \nballoon\nDirect 1.02 \n(0.46,2.29); \np=1.0a\n0.76 \n(0.25,2.32); \np=0.6b\nNot available Not available Not available\nNetwork 0.82 \n(0.47,1.44); \np=0.5\n0.62 \n(0.32,1.18); \np=0.1\n1.62 \n(0.62,4.24); \np=0.3\n1.80 \n(0.90,3.57); \np=0.09\n0.41 \n(0.07,2.27); \np=0.3\nBipolar radio \nfrequency\nDirect Not available Not available 4.88 \n(1.32,18.11); \np=0.02b\nNot available\nNetwork 0.75 \n(0.34,1.66); \np=0.5\n1.98 \n(0.70,5.60); \np=0.2\n2.19 \n(1.07,4.50); \np=0.03\n0.50 \n(0.09,2.89); \np=0.4\nMicrowave Direct Not available Not available Not available\nNetwork 2.63 \n(0.89,7.78); \np=0.08\n2.91 \n(1.23,6.88); \np=0.02\n0.66 \n(0.11,3.96); \np=0.6\nCryoablation Direct Not available Not available\nNetwork 1.11 \n(0.37,3.29); \np=0.9\n0.25 \n(0.04,1.70); \np=0.2\nFree-fluid Direct Not available\nNetwork 0.23 \n(0.04,1.36); \np=0.1\nFigures are odds ratios (95% confidence intervals) for device in column compared with device in row. Odds ratio >1 indicate \nincreased rate with device in column. a I2 0% b I2 not applicable, only one study in comparison.\nChapter 10\n\n151\nFigure 3  Amenorrhoea rate at 12 months from network meta-analysis (odds ratio and 95%   \n confidence intervals). Results reflect those in Table 2 (*indicates reversed ratios so all  \n are pointing in direction of increased amenorrhoea rate). Only results with p≤0.05  \n are shown.\nEffectiveness in reducing dissatisfaction with treatment\nDirect comparisons showed some evidence of reduced rates of dissatisfaction with bipolar radio \nfrequency ablation compared with thermal balloon ablation (0.39, 95% CI 0.16 to 0.91; p=0.03) \n(Table 4), though this difference was not as convincing in the network meta-analysis (0.56, 95% \nCI 0.28 to 1.09; p=0.09). Increased dissatisfaction was seen with free fluid thermal ablation com-\npared with bipolar radio frequency (9.40, 95% CI 1.14 to 77.18; p=0.04) and a similar result was \nseen in the network meta-analysis (4.79, 95% CI 1.07 to 21.48; p=0.04). There were no other \nobvious differences. Direct estimates between studies were also consistent for dissatisfaction \nwhere more than one study in a comparison existed (0%<I\n2<4%). Overall rates of dissatisfaction \nwere generally low (such as 0-7% for bipolar radiofrequency ablation).\nMicrowaveCryoablation\nLaserBipolar radio frequency\nThermal balloonFree fluid\n/three.tf./two.tf/three.tf (/one.tf./three.tf/six.tf to /seven.tf./seven.tf/one.tf)*\n/one.tf./six.tf/six.tf (/one.tf./zero.tf/one.tf to /two.tf./seven.tf/one.tf)\n/one.tf/two.tf./three.tf/seven.tf (/three.tf./nine.tf/six.tf to /three.tf/eight.tf./five.tf/nine.tf)\n/six.tf./nine.tf/eight.tf (/two.tf./four.tf/eight.tf to /one.tf/nine.tf./six.tf/nine.tf)\n/two.tf./five.tf/one.tf (/one.tf./five.tf/three.tf to /four.tf./one.tf/two.tf)\n/four.tf./nine.tf/one.tf (/two.tf./zero.tf/four.tf to /one.tf/one.tf./eight.tf/zero.tf)*\n/two.tf./seven.tf/seven.tf (/one.tf./four.tf/nine.tf to /five.tf./one.tf/four.tf)*\n/three.tf./eight.tf/two.tf (/one.tf./four.tf/six.tf to /one.tf/zero.tf./zero.tf/one.tf)\n/six.tf./three.tf/four.tf (/two.tf./five.tf/zero.tf to /one.tf/six.tf./zero.tf/seven.tf)\nHMB Network Meta-analysis\n\n152\nTable 4:  Dissatisfaction rate at 12 months - results from direct comparisons and network \n meta-analysis.\nThermal \nballoon\nBipolar radio \nfrequency\nMicrowave Cryoablation Free-fluid Laser\nFirst \ngeneration \ndevice\nDirect 0.80 (0.50, \n1.27); p=0.3b\n1.18 \n(0.40,3.53); \np=0.8c\n0.90 \n(0.51,1.60); \np=0.7a\n0.71 \n(0.30,1.65); \np=0.4c\nNot available\n0.57 \n(0.13,2.49); \np=0.5c\nNetwork 0.91 \n(0.61,1.36); \np=0.6\n0.51 \n(0.25,1.01); \np=0.05\n0.93 \n(0.59,1.47); \np=0.7\n0.93 \n(0.59,1.47); \np=0.7\n3.14 \n(0.66,14.83); \np=0.1\n0.54 \n(0.12,2.43) \np=0.4\nThermal \nballoon\nDirect 0.39 \n(0.16,0.91); \np=0.03a\n1.07 \n(0.61,1.86); \np=0.8c\nNot available Not available Not available\nNetwork 0.56 \n(0.28,1.09); \np=0.09\n1.02 \n(0.65,1.60); \np=0.9\n0.79 \n(0.31,2.05); \np=0.6\n2.66 \n(0.55,12.96); \np=0.2\n0.59 \n(0.12,2.78); \np=0.5\nBipolar radio \nfrequency\nDirect Not available Not available 9.40 \n(1.14,77.18); \np=0.04c\nNot available\nNetwork 1.83 \n(0.85,3.95); \np=0.1\n1.42 \n(0.47,4.29); \np=0.5\n4.79 \n(1.07,21.48); \np=0.04\n1.06 \n(0.20,5.51); \np=0.9\nMicrowave Direct Not available Not available Not available\nNetwork 0.78 \n(0.29,2.07); \np=0.6\n2.62 \n(0.52,13.23); \np=0.2\n0.58 \n(0.12,2.79); \np=0.5\nCryoablation Direct Not available Not available\nNetwork 3.37 \n(0.56,20.39); \np=0.2\n0.74 \n(0.13,4.25); \np=0.7\nFree-fluid Direct Not available\nNetwork 0.22 \n(0.03,1.95); \np=0.2\nThe device in the column is compared to the device in the row. Increased dissatisfaction rate with column device is indicated \nby OR>1 (for example: increased rate with free-fluid compared to bipolar radio frequency).  Odds ratios, 95% confidence \nintervals and p-values shown. a I2=0%, b I2 =4%, c I2 n/a\nIncoherence\nEstimates of incoherence, ω, from the network model of amenorrhoea rate were relatively high at \n0.23 compared with other studies that have used this technique to evaluate multiple treatments \nsimultaneously.\n19,30,44 When we removed the single trial comparing microwave ablation with ther-\nmal balloon ablation,33 this estimate was reduced to 0.13. This trial’s direct estimate yielded no \nsignificant difference between treatments (1.13, 95% CI 0.70 to 1.82; p=0.6), which conflicted \nChapter 10\n\n153\nwith the estimate expected by any indirect comparison where the network analysis favoured \nmicrowave ablation (1.66, 95% CI 1.01 to 2.71; p=0.05). A sensitivity analysis of the network \nanalysis with this trial excluded produced a higher estimate in favour of microwave ablation (2.23, \n95% CI 1.21 to 4.09; p=0.02), with some other minor adjustments to the other estimates (free \nfluid ablation was now associated with a lower amenorrhoea rate also compared with microwave \nablation: 0.45, 95% CI 0.21 to 0.97; p=0.04). Incoherence parameters for heavy bleeding and \ndissatisfaction, along with the interaction parameter for study by treatment, were estimated to be \nso close to zero that they were negligible and so effectively did not contribute to any estimates.\nDiscussion\nMain findings\nOf the three most popular second generation endometrial ablation devices (thermal balloon, \nbipolar radio frequency, and microwave) examined in this network meta-analysis, bipolar radio \nfrequency and microwave ablation techniques seem to have an advantage over thermal balloon \nablation in terms of increased amenorrhea rate in women with heavy menstrual bleeding. Com-\npared with free fluid ablation, both bipolar radio frequency and microwave ablation techniques \nwere associated with a reduction in the number of women still experiencing heavy bleeding. Ra-\ndio frequency ablation was also associated with an increased amenorrhoea rate and a reduced \nnumber of women dissatisfied with treatment compared with free fluid ablation. Endometrial \nlaser intrauterine thermotherapy (ELITT) was also associated with high rates of amenorrhoea \ncompared with hysteroscopic ablation in the small single study that evaluated this technique. \nNot surprisingly this translated to estimates of amenorrhoea higher than the other devices when \nwe combined results in the network meta-analysis. Conversely, cryoablation was associated \nwith lower rates of amenorrhoea compared with bipolar radio frequency and microwave ablative \ntechniques. Rates of dissatisfaction with treatment or ongoing heavy bleeding were generally low \nacross all studies. Apart from the comparison of microwave ablation versus thermal balloon abla-\ntion, estimates from the network meta-analysis were consistent with available direct estimates \nbetween second generation devices in terms of direction and significance of treatment effect. \nThe implication of this inconsistency from this one comparison is unclear.\nStrengths and limitations of the review\nWe used the relatively new approach of network meta-analysis to synthesise data, and this pro-\nvided effect estimates when direct comparisons were not available. We used optimal methods, \ncomplying with guidelines on reporting of systematic reviews and meta-analyses.\n23 We carried \nout an extensive literature search, with no language restrictions, minimising the risk of missing \ninformation. The collection of individual patient data from some studies improved estimates.\nThe review was not intended to be a comprehensive review of all facets of these devices, and we \nfocused only on outcomes of bleeding after treatment reported by patients and patients’ dissat-\nHMB Network Meta-analysis\n\n154\nisfaction with treatment. Many other factors, such as ease of use and safety of device, are taken \ninto account in the decision regarding a particular device. For example, the time to complete \ntreatment is significantly shorter with bipolar radiofrequency ablation, and it results in greater \ncoverage of the endometrial surface than thermal balloon ablation, in an out-patient setting.31\nBecause of the limited amount of data (several comparisons consisted of only one trial) and the \nfact that the traditional meta-analysis did not show any evidence of heterogeneity between esti-\nmates (I\n2 was generally 0% and never greater than 17%) we did not attempt to further subgroup \nresults by manufacturer or type (Thermachoice, Cavaterm, and the discontinued Vesta all are \ntypes of thermal balloon ablative device). This consistency could have occurred by chance, and \nthere could be some differences because of the design of similar products. This could be reason-\nably evaluated only should more data become available. We considered that the first generation \nablative techniques of transcervical endometrial resection and rollerball electrocoagulation were \nequivalent so we used them as a generic intermediate comparator.\n45,46\nEstimates of incoherence were relatively high for the analysis of amenorrhoea rate (less when we \nomitted one of the studies in a sensitivity analysis), though how this impacts on adjusted esti-\nmates is debatable, given that confidence intervals of estimates were increased as a result. Like \nall meta-analysis, we must consider limitations with combining estimates from different studies \nbecause of clinical heterogeneity such as differences in populations of patients, even with statisti-\ncal adjustments in the analysis. For example, the small study that compared endometrial laser \nintrauterine thermotherapy (ELITT) with first generation techniques seemed to show a large treat-\nment effect. Given that none of the other studies has shown a comparable effect, there could be \nsomething particular about this study, such as the lower rate of amenorrhoea for the first genera-\ntion comparator, that separates it from the rest, but in the absence of raw data, we were unable \nto investigate this further. Conversely, the absence of a significant difference does not necessarily \nmean that techniques are comparable but could be a consequence of small underpowered trial \npopulations. This is reflected in wide confidence intervals for some of the estimates. We did not \ntake into account quality of study in these analyses; the only requirement was that the study \nneeded to be randomised in a specified population of patients.\nThe usefulness of the network meta-analysis in this setting was highlighted during the course of \nthis review when we updated the analysis to include the results of the most recent study of bipolar \nradio frequency and free fluid endometrial ablation\n32; the first to compare these two treatments di-\nrectly. Results from this study for the primary outcome measure of amenorrhoea rate at 12 months \nsuggested improvements with radio frequency, with an odds ratio of 0.36 (0.17 to 0.73). Before \ninclusion of this later study, our original estimated odds ratio from the network meta-analysis was \n0.33, which is similar albeit with a slightly wider confidence interval (0.11 to 0.98). Such an accurate \nestimate could of course occur by chance, but it does show how useful this analytical method \ncould be where direct head to head assessments of treatments are not available.\nChapter 10\n\n155\nInterpretation\nRates of dissatisfaction and heavy bleeding are consistently low for second generation tech-\nniques, which, as a group, represent an excellent conservative alternative to hysterectomy. It is \nclear from the results of our review that bipolar radiofrequency and microwave ablation technique \nare associated with higher amenorrhea rates than other methods, although it should be reiterat-\ned that only hysterectomy can guarantee complete cessation of menstrual bleeding. Endometrial \nlaser intrauterine thermotherapy (ELITT) was associated with highly significant improvements in \namenorrhea rates compared with all other methods, albeit from one small study, but this device \nis no longer marketed (C Dunn, Karl Storz, personal communication).\nData from UK hospital episode statistics show a significant increase in the overall number of \ninpatient endometrial ablative techniques (Figure 4), though radiofrequency ablation techniques \nare recently the most rapidly rising techniques used. These data do not include ambulatory pro-\ncedures, which are also increasing rapidly but not adequately coded in the hospital episode data. \nMore recently, the microwave endometrial ablation system from Microsulis has been withdrawn \nfrom the European market after Hologic acquired the intellectual property for the device (Micro-\nsulis Medical, press release, 2011). Ambiguous categorisation of first generation techniques in \nthe episode data precludes comparison of relative rates, but consensus is that such techniques \nhave rapidly been superseded.\nFigure 4  Trends in types of endometrial ablation procedures performed in England, 2004-11  \n (data from hospital episode statistics)\nYears\nNo of inpatient procedures\nperformed in England (000s)\n/two.tf/zero.tf/zero.tf/four.tf-/five.tf /two.tf/zero.tf/zero.tf/five.tf-/six.tf /two.tf/zero.tf/zero.tf/six.tf-/seven.tf /two.tf/zero.tf/zero.tf/seven.tf-/eight.tf /two.tf/zero.tf/zero.tf/eight.tf-/nine.tf /two.tf/zero.tf/zero.tf/one.tf/zero.tf-/one.tf/one.tf/two.tf/zero.tf/zero.tf/nine.tf-/one.tf/zero.tf\n/zero.tf\n/one.tf\n/two.tf\n/three.tf\n/four.tf\n/six.tf\n/five.tf\nThermal balloon\nEndometrial ablation\nMicrowave\nFree fluid\nBipolar radio frequency\nHMB Network Meta-analysis\n\n156\nWith respect to costs and cost effectiveness, one study has directly compared thermal balloon \nand microwave ablative techniques.47 The mean total health service cost per patient was £181 \n(£70 to £434) more (equivalent to about €219, $287) for the thermal balloon system than for the \nmicrowave ablative device (at 2010 costs), though the study found no significant differences \nbetween the total non-health costs and health benefits of the two devices. The incremental cost \neffectiveness ratio showed microwave ablation, on average, dominated thermal balloon ablation \nat one year after the procedure.\nOur previous review indicated that levels of dissatisfaction with second generation techniques, \nas a class of procedure, were slightly lower than those associated with first generation tech-\nniques. Second generation techniques were also quicker, associated with faster recovery times \nand fewer adverse procedural events, and could be carried out under local anaesthesia.\n18 Thus \nthey should be offered when no structural or histological abnormality is present and the woman \nwants to retain her uterus. While chances of pregnancy are reduced after endometrial ablation, \nthey can occur and such pregnancies have a high risk of adverse outcomes, hence continued \ncontraception is strongly advised.48\nAs to the question of the best second generation technique, bipolar radiofrequency ablation \nseems to offer benefits over thermal balloon in terms of amenorrhea rates and is procedurally \nquicker than thermal balloon and microwave ablation.31,33-35 Any impact on other outcomes was \nnot significant in either direct comparisons or network meta-analysis. Only two studies have \nevaluated the use of these three devices under local anaesthesia,\n31,33 so while out-patient hyster-\noscopic treatment is feasible, it would be premature to state any advantage for either anaesthetic \napproach. The decision should be left to the women’s preference and the operator’s experience, \nthough the withdrawal of the microwave ablation device reduces available choices.\nFurther large scale rigorous randomised trials to compare existing and emerging ablative tech-\nniques should use meaningful and standardised measures of satisfaction and menstrual bleeding \nand be run independently from the manufacturers of the devices. Trials incorporating different \nanaesthetic approaches will also help to refine the evidence available to women and gynaecolo-\ngists. Concerns with regard to the safety of the “blind” techniques with respect to extrauterine \ninjury should lead to further evaluation of visual ablative techniques such as the next generation of \nfree fluid ablation. Adequate assessment of relatively rare adverse events, however, would require \nlarge cohorts similar to the MISTLETOE study.\n49\nChapter 10\n\n157\nWhat is already known on this topic\nSecond generation endometrial ablation techniques are preferable to first generation techniques \nin the treatment of heavy menstrual bleeding. \nBipolar radiofrequency, thermal balloon, and microwave ablation techniques are the most com-\nmonly used and can safely be offered under local anaesthesia but there are insufficient data to \nrecommend any particular technique.\nWhat this study adds\nBipolar radiofrequency endometrial ablation shows significant benefit in terms of amenorrhea \nrates compared with thermal balloon ablation, as does microwave ablation, but there is no evi-\ndence of any benefit of bipolar radiofrequency over microwave ablation.\nThere is no evidence of greater dissatisfaction with any treatment in comparisons of bipolar radi-\nofrequency, thermal balloon, or microwave ablation.\nOther and emerging methods of second generation endometrial ablation require more rigorous \nevaluation.\nSupplementary Material\nData supplements available at http://dx.doi.org/10.1136/bmj.e2564\nAcknowledgments\nWe thank all authors of identified trials who sent data through the previous related review.18\nThe following authors provided us with individual patient data from their trials for this review: J \nAbbott, University of New South Wales, Sydney, Australia; M Y Bongers, Maxima Medical Centre, \nVeldhoven, Netherlands; J-L Brun, Hopital Universitaire Pellegrin, Bordeaux, France; T J Clark, \nBirmingham Women’s Hospital, Birmingham, UK; J Cooper (2004 trial), data supplied by Micro-\nsulis Medical, Hampshire, UK; K G Cooper, Aberdeen Royal Infirmary, Aberdeen, UK; S L Corson \n(2001 trial), data supplied by Boston Scientific Corporation, Marlborough, US; J Hawe, Countess \nof Chester Hospital, Chester, UK; W R Meyer; data supplied by Ethicon, NJ, US; J P M Penninx, \nMaxima Medical Centre, Veldhoven, Netherlands; A M Sambrook, Aberdeen Royal Infirmary, \nAberdeen, UK; I A van Zon-Rabelink, Medical Spectrum Twente, Enschede, Netherlands.\nHMB Network Meta-analysis\n\n158\nReferences\n1 Shapley M, Jordan K, Croft PR. An epidemiological survey of symptoms of menstrual loss \nin the community. Br J Gen Pract 2004;54:359-63.\n2 Chapple A. Menorrhagia: women’s perceptions of this condition and its treatment. J Adv \nNurs 1999;29:1500-6.\n3 Prentice A. Health care implications of dysfunctional uterine bleeding. Baillieres Best Pract \nRes Clin Obstet Gynaecol 1999;13:181-8.\n4 Lethaby A, Farquhar C. Treatments for heavy menstrual bleeding. BMJ 2003;327:1243-4.\n5 Reid PC, Mukri F. Trends in number of hysterectomies performed in England for menor -\nrhagia: examination of health episode statistics, 1989 to 2002-3. BMJ 2005;330:938-9.\n6 Madhu CK, Nattey J, Naeem T. Second generation endometrial ablation techniques: an \naudit of clinical practice. Arch Gynaecol Obstet 2009;280:599-602.\n7 Reid PC. Endometrial ablation in England—coming of age? An examination of hospi-\ntal episode statistics 1989/1990 to 2004/2005. Eur J Obstet Gynecol Reprod Biol \n2007;135:191-4.\n8 NHS Information Centre. Hospital episode statistics 2011. NHS, 2011.\n9 Meyer WR, Walsh BW, Grainger DA, Peacock LM, Loffer FD, Steege JF. Thermal balloon \nand rollerball ablation to treat menorrhagia: a multicenter comparison. Obstet Gynecol \n1998;92:98-103.\n10 Loffer FD. Three-year comparison of thermal balloon and rollerball ablation in treatment of \nmenorrhagia. J Am Assoc Gynecol Laparoscop 2001;8:48-54.\n11 Loffer FD, Grainger D. Five-year follow-up of patients participating in a randomised trial \nof uterine balloon therapy versus rollerball ablation for treatment of menorrhagia. J Am \nAssoc Gynecol Laparoscop 2002;9:429-35.\n12 Cooper KG, Bain C, Parkin DE. Comparison of microwave endometrial ablation and tran-\nscervical resection of the endometrium for treatment of heavy menstrual loss: a ran-\ndomised trial. Lancet 1999;354:1859-63.\n13 Cooper KG, Bain C, Lawrie L, Parkin DE. A randomised comparison of microwave endo-\nmetrial ablation with transcervical resection of the endometrium: follow up at a minimum \nof five years. BJOG 2005;112:470-5.\n14 Cooper J, Gimpelson R, Laberge P , Galen D, Garza-Leal JG, Scott J. A randomised, \nmulticenter trial of safety and efficacy of the NovaSure system in the treatment of menor-\nrhagia. J Am Assoc Gynecol Laparoscop 2002;9:418-28.\n15 Corson SL. A multicenter evaluation of endometrial ablation by Hydro ThermAblator and \nrollerball for treatment of menorrhagia. J Am Assoc Gynecol Laparoscop 2001;8:359-67.\n16 Perino A, Castelli A, Cucinella G, Bionda A, Pane A, Venezia R. A randomized comparison \nof endometrial laser intrauterine thermotherapy and hysteroscopic endometrial resection. \nFert Steril 2004;82:731-4.\nChapter 10\n\n159\n17 Duleba AJ, Heppard MC, Soderstrom RM, Townsend DE. A randomized study compar-\ning endometrial cryoablation and rollerball electroablation for treatment of dysfunctional \nuterine bleeding. J Am Assoc Gynecol Laparoscop 2003;10:17-26.\n18 Middleton LJ, Champaneria R, Daniels JP , Bhattacharya S, Cooper KG, Hilken NH, et al. \nHysterectomy, endometrial destruction, and levonorgestrel releasing intrauterine system \n(Mirena) for heavy menstrual bleeding: systematic review and meta-analysis of data from \nindividual patients. BMJ 2010;341:c3929.\n19 Thijs V, Lemmens R, Fieuws S. Network meta-analysis: simultaneous meta-analysis of \ncommon antiplatelet regimens after transient ischaemic attack or stroke. Eur Heart J \n2008;29:1086-92.\n20 Caldwell DM, Ades AE, Higgins JP . Simultaneous comparison of multiple treatments: \ncombining direct and indirect evidence. BMJ 2005;331:897-900.\n21 Higgins JPT, Green S. Cochrane handbook for systematic reviews of interventions, Ver -\nsion 5.0.1. Cochrane Collaboration, 2008.\n22 Stewart LA, Tierney JF. To IPD or not to IPD? Advantages and disadvantages of system-\natic reviews using individual patient data. Eval Health Prof 2002;25:76-97.\n23 Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP , et al. The PRISMA \nstatement for reporting systematic reviews and meta-analyses of studies that evaluate \nhealth care interventions: explanation and elaboration. J Clin Epidemiol 2009;62:e1-34.\n24 Higham JM, O’Brien PM, Shaw RW. Assessment of menstrual blood loss using a pictorial \nchart. Br J Obstet Gynaecol 1990;97:734-9.\n25 Early Breast Cancer Trialists’ Collaborative Group. Treatment of early breast cancer: \nworldwide evidence, 1985-1990. Oxford University Press, 1990.\n26 Higgins JP , Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-\nanalysis. BMJ 2003;327:560.\n27 Glenny AM, Altman DG, Song F, Sakarovitch C, Deeks JJ, D’Amico R, et al. Indirect com-\nparisons of competing interventions. Health Technol Assess 2005;9:1-iv.\n28 Whitehead A. Meta-analysis of controlled clinical trials. Wiley, 2002.\n29 Lumley T. Network meta-analysis for indirect treatment comparisons. Stat Med \n2002;21:2313-24.\n30 Van der Valk R, Webers CA, Lumley T, Hendrikse F, Prins MH, Schouten JS. A network \nmeta-analysis combined direct and indirect comparisons between glaucoma drugs to \nrank effectiveness in lowering intraocular pressure. J Clin Epidemiol 2009;62:1279-83.\n31 Clark TJ, Samuel N, Malick S, Middleton LJ, Daniels J, Gupta JK. Bipolar radiofrequency \ncompared with thermal balloon endometrial ablation in the office: a randomized controlled \ntrial. Obstet Gynecol 2011;117:109-18.\nHMB Network Meta-analysis\n\n160\n32 Penninx JP , Mol BWJ, Engels R, van Rumste MM, Kleijn C, Coks CA, et al. Bipolar ra-\ndiofrequency endometrial ablation compared with hydrothermablation for dysfunctional \nuterine bleeding: a randomized controlled trial. Obstet Gynecol 2010;116:819-26.\n33 Sambrook AS, Cooper KG, Campbell MK, Cook JA. Clinical outcomes from a randomised \ncomparison of microwave endometrial ablation with thermal balloon endometrial ablation \nfor the treatment of heavy menstrual bleeding. BJOG 2009;116:1038-45.\n34 Bongers MY, Bourdrez P , Mol BW, Heintz AP , Brolmann HA. Randomised controlled trial \nof bipolar radio-frequency endometrial ablation and balloon endometrial ablation. BJOG \n2004;111:1095-102.\n35 Abbott J, Hawe J, Hunter D, Garry R. A double-blind randomized trial comparing the Ca-\nvaterm and the NovaSure endometrial ablation systems for the treatment of dysfunctional \nuterine bleeding. Fertil Steril 2003;80:203-8.\n36 Brun JL, Burlet G, Galand B, Quereux C, Bernard P . Cavaterm thermal balloon endometri-\nal ablation versus hysteroscopic endometrial resection to treat menorrhagia: the French, \nmulticenter, randomized study. J Min Invasive Gynecol 2006;13:424-30.\n37 Cooper J, Anderson TL, Fortin CA, Jack SA, Plentl MB. Microwave endometrial ablation \nvs rollerball electroablation for menorrhagia: a multicenter randomized trial. J Am Assoc \nGynecol Laparoscop 2004;11:394-403.\n38 Hawe J, Abbott J, Hunter D, Phillips G, Garry R. A randomised controlled trial compar -\ning the Cavaterm endometrial ablation system with the Nd:YAG laser for the treatment of \ndysfunctional uterine bleeding. BJOG 2003;110:350-7.\n39 Van Zon-Rabelink IA, Vleugels MPH, Merkus HMWM, de Graaf R. Efficacy and satisfac-\ntion rate comparing endometrial ablation by rollerball electrocoagulation to uterine bal-\nloon thermal ablation in a randomised controlled trial. Eur J Obstet Gynecol Reprod Biol \n2004;114:97-103.\n40 Pellicano M, Guida M, Acunzo G, Cirillo D, Bifulco G, Nappi C. Hysteroscopic transcer -\nvical endometrial resection versus thermal destruction for menorrhagia: a prospective \nrandomized trial on satisfaction rate. Am J Obstet Gynecol 2002;187:545-50.\n41 Soysal ME, Soysal SK, Vicdan K. Thermal balloon ablation in myoma-induced menor -\nrhagia under local anesthesia. Gynecol Obstet Investigation 2001;51:128-33.\n42 Corson SL, Brill AI, Brooks PG, Cooper JM, Indman PD, Liu JH, et al. One-year results of \nthe vesta system for endometrial ablation. J Am Assoc Gynecol Laparoscop 2000;7:489-\n97.\n43 Romer T. Therapy of recurrent menorrhagia—Cavaterm balloon coagulation versus roller-\nball endometrium coagulation—a prospective randomized comparative study. Zentralbl \nGynakol 1998;120:511-4.\nChapter 10\n\n161\n44 Psaty BM, Lumley T, Furberg CD, Schellenbaum G, Pahor M, Alderman MH, et al. Health \noutcomes associated with various antihypertensive therapies used as first-line agents: a \nnetwork meta-analysis. JAMA 2003;289:2534-44.\n45 Boujida VH, Philipsen T, Pelle J, Joergensen JC. Five-year follow-up of endometrial ablation: \nendometrial coagulation versus endometrial resection. Obstet Gynecol 2002;99:988-92.\n46 Lethaby A, Hickey M. Endometrial destruction techniques for heavy menstrual bleeding. \nCochrane Database Syst Rev 2005;4.\n47 Kilonzo MM, Sambrook AM, Cook JA, Campbell MK, Cooper KG. A cost-utility analysis \nof microwave endometrial ablation versus thermal balloon endometrial ablation. Value \nHealth 2010;13:528-34.\n48 Yin C-S. Pregnancy after hysteroscopic endometrial ablation without endometrial \npreparation: a report of five cases and a literature review. Taiwan J Obstet Gynaecol \n2010;49:311-9.\n49 Overton C, Hargreaves J, Maresh M. A national survey of the complications of endometrial \ndestruction for menstrual disorders: the MISTLETOE study. Minimally invasive surgical tech-\nniques—laser, endothermal or endoresection. BJOG 1997;104:1351-9.\nHMB Network Meta-analysis\n\n162\n\n\n163\nChapter 11\nThe measurement properties of the \nmenorrhagia multi-attribute quality-of-life scale: \na psychometric analysis\n \n\n164\nChapter 11\nIntroduction\nMenorrhagia (heavy blood loss during menstruation) is a common gynaecological condition in \nthe UK that has a significant impact on the wellbeing of many women. The number and cost of \nconsultations and treatments impose substantial demands on the NHS. It has been estimated \nthat 6.5% of women aged 12–51 years experience heavy menstrual bleeding (HMB), 1 with 5% \nof women aged 30–49 years consulting their GP each year because of the condition. 2 Women \nare mainly treated on the basis of symptoms, and therefore assessment of the condition and the \neffectiveness of treatment has a large subjective component. There is poor correlation between \nquasi-objective measures of menorrhagia, e.g. self-reports of blood loss and patient-based \nhealth status measures. As the aim of treatment is to improve women’s wellbeing and quality of \nlife (QoL), it is necessary to have valid and reliable instruments to measure this.\nA systematic review of published research assessed the quality of QoL instruments in 19 studies \nof menorrhagia, using a checklist of items for clinical face validity (issues relevant to patients’ ex-\npectations and concerns) and items for measurement properties (including reliability, responsive-\nness, criterion or construct validity and acceptability).3 The generic Short Form 36 Health Survey \nQuestionnaire (SF-36) was the most commonly used (63% of studies). Only two studies used a \ncondition-specific QoL instrument for menorrhagia, and although 90% complied with more than \nhalf of the criteria for measurement properties, only 37% of studies complied with more than half \nthe criteria for face validity. The conclusion of this review, and a review conducted by the UK’s \nNational Institute for Clinical Excellence (NICE), was that there is a need for methodologically \nsound, condition-specific QoL instruments in menorrhagia with clinical face validity to assess \ntreatment outcomes.\nOne disease-specific instrument identified as having high face validity is the menorrhagia multi-\nattribute scale (MMAS) developed by Shaw et al. 4 This was developed in collaboration with \nwomen with menorrhagia using multi-attribute utility method. Although the MMAS was designed \nto have good face and content validity, other psychometric properties of this instrument, i.e. \nreliability, convergent and discriminant validity, and floor and ceiling effects, could not be evalu-\nated in the original study. One study reported a small advantage for the MMAS over SF-36 in \npredicting the management outcome for HMB, 5 but otherwise the measurement properties of \nthe questionnaire when used in a treatment or observational study have not been reported. The \naim of this study was to assess these properties of the MMAS when used within a treatment trial.\nChapter 11\n\n165\nMethods\nParticipants\nAll women who took part in this study were enrolled in the ECLIPSE (Effectiveness and Cost-\neffectiveness of Levonorgestrel-containing Intrauterine system in Primary care against Standard \ntrEatment for menorrhagia) trial, a randomised controlled trial designed to assess the effective-\nness of a levonorgestrel-containing intrauterine system in the treatment of menorrhagia [National \nInstitute for Health Research (NIHR) Health Technology Assessment Programme (HTA) project \nreference 02/06/02]. In order to be invited into the trial, women had to be between the ages of \n25 and 50 years, presenting to GPs with menorrhagia, not intending to become pregnant in the \nnext 5 years and agreeing to be randomised. Data from the first 431 women enrolled in the trial \nwere analysed for this study.\nProcedure\nThe ECLIPSE trial, including the study reported here, received multicentre research ethics com-\nmittee approval. Participants completed the MMAS and a battery of other tests as part of the \nbaseline measures for the trial after ran- domisation, but before treatment. The other tests used \nwere the SF-36 v2, EuroQol EQ-5D and the Sexual Activity Questionnaire. Fifty women were also \ninvited to participate in a test–retest study of the MMAS, and complete the test again within 2 \nweeks of the baseline measurement.\nMMAS\nThe MMAS questionnaire captures the subjective consequences of menorrhagia on six domains: \npractical difficul- ties; social life; psychological wellbeing; physical health; work routine; and fam-\nily life. Each of the six domains has four statements that represent four levels of response. Re-\nspondents indicate the statement that best matches their feelings for each domain. The state-\nment scores derive from a weighting of the domains and a weighting of the statements in level of \nseverity by women in the original study. Scores range from 0 (worst possible state in all domains) \nto 100 (best possible state in all domains).\nInstruments used in the validation of the MMAS\n1 -  Short Form 36. The SF-36 v2 is a 36-item short-form survey that measures general health-\nrelated QoL. The SF-36 is a practical and reliable way to obtain important health outcomes data \nin a variety of settings, measuring eight domains of health: physical functioning; role limitations \nbecause of physical health; bodily pain; general health perceptions; vitality; social functioning; \nrole limitations because of emotional problems; and mental health. Respondents are asked to re-\ncall how they have felt over the previous 4 weeks. It is commonly used in studies of menorrhagia.\n5\nMeasurement properties of MMAS\n\n166\n2 - EuroQol EQ-5D. EQ-5D is a standardised instrument for use as a measure of health out-\ncome, and is widely used in economic evaluations of medical interventions. Applicable to a wide \nrange of health conditions and treatments, it provides a simple descriptive profile and a single \nindex value for health status.\n3 - The Sexual Activity Questionnaire. Sexual activity is an important dimension of QoL, \nwhich may be specifically affected by menorrhagia. The Sexual Activity Questionnaire (SAQ) was \ndeveloped as a self-report questionnaire for use in gynaecological clinical trials, designed to be \nquick to complete and acceptable to the majority of women. 6 Three dimensions of perceptions \nof sexual activity are measured: pleasure; discomfort; habit.\nPsychometric analysis\nThe measurement properties of the MMAS were assessed using traditional psychometric proce-\ndures. Estimates of reliability and validity were made using intra- and inter-test correlations, with \nan additional comparison of means for the subscales of the questionnaire. Inspection of the data \nsupported the use of Pearson’s product-moment correlation coefficient (r) for bivariate correla-\ntions and analysis of variance (anova) for comparison of means. Floor and ceiling effects were \nalso assessed using frequency data on high and low scores.\nResults\nThe mean age of the participants was 41.25 years (SD 5.3 years). The ethnic mix of the sample \nwas representative of the patient population for the area, with 93% of women categorising them-\nselves as from the two main ethnic groups: white British (83%) and South Asian (10%).\nReliability\nForty-nine of the 50 women asked to take the questionnaire again within 2 weeks returned \ncompleted questionnaires. The test–retest correlation for the total MMAS score was acceptable \n(r = 0.836), suggesting that it is stable over time. Cronbach’s alpha was calculated as a measure \nof the internal consistency of the six items of the MMAS. This was 0.82, which, being within the \nrange 0.7 and 0.9, suggests that the items are all measuring aspects of the same construct. \nEvi- dence of the homogeneity of the MMAS was also provided by the item–total correlations \n(the correlation between each item and the total score minus that item), which range between \n0.44 and 0.68. Only one such correlation fell below 0.6: that for ‘practical difficulties’ caused by \nmenorrhagia.\nChapter 11\n\n167\nValidity\nThe MMAS has already been shown to have good face validity, so this analysis concentrated on \nconstruct validity: i.e. the extent to which the instrument measures what it is intended to meas-\nure. An important method of demonstrating construct validity is by showing that an instrument \ncorrelates with other instruments measuring similar constructs, e.g. generic QoL scales (con-\nvergent validity), and does not correlate with measures of different constructs, e.g. age or body \nmass index (discriminant validity).\nConvergent validity\nTo show convergent validity, correlations between test scores should be moderate to high: i.e. \nequal to or above 0.4. Table 1 shows the correlations between the MMAS total score, the sub-\nscales of the SF-36 and the EQ-5D (note that whereas all participants completed all items of the \nMMAS at baseline, some missed items in the SF-36). These show mod- erate correlations for \nmost components of the SF-36, but weaker correlations for physical functioning, general health \nperception and for the EQ-5D standardised score.\nTable 1  Correlations between the MMAS total score and the EQ-5D summary and  \n SF-36 subscales.\nThe individual domain scores of the MMAS were compared with the SF-36 subscales. With the \nexception of the practical difficulties domain on the MMAS, which did not significantly correlate \nwith any SF-36 subscale, the highest correlations were between each of the MMAS domains and\nconsistency of the six items of the MMAS. This was 0.82,\nwhich, being within the range 0.7 and 0.9, suggests that the\nitems are all measuring aspects of the same construct. Evi-\ndence of the homogeneity of the MMAS was also provided by\nthe item–total correlations (the correlation between each\nitem and the total score minus that item), which range\nbetween 0.44 and 0.68. Only one such correlation fell below\n0.6: that for ‘practical difficulties’ caused by menorrhagia.\nValidity\nThe MMAS has already been shown to have good face\nvalidity, so this analysis concentrated on construct validity:\ni.e. the extent to which the instrument measures what it is\nintended to measure. An important method of demonstrat-\ning construct validity is by showing that an instrument cor-\nrelates with other instruments measuring similar\nconstructs, e.g. generic QoL scales (convergent validity),\nand does not correlate with measures of different con-\nstructs, e.g. age or body mass index (discriminant validity).\nConvergent validity\nTo show convergent validity, correlations between test scores\nshould be moderate to high: i.e. equal to or above 0.4.\nTable 1 shows the correlations between the MMAS total\nscore, the subscales of the SF-36 and the EQ-5D (note that\nwhereas all participants completed all items of the MMAS at\nbaseline, some missed items in the SF-36). These show mod-\nerate correlations for most components of the SF-36, but\nweaker correlations for physical functioning, general health\nperception and for the EQ-5D standardised score.\nThe individual domain scores of the MMAS were com-\npared with the SF-36 subscales. With the exception of the\npractical difficulties domain on the MMAS, which did not\nsignificantly correlate with any SF-36 subscale, the highest\ncorrelations were between each of the MMAS domains and\nthe SF-36 social functioning scale. The correlation between\nthe psychological health scales of the two instruments is\n0.39, and that between the MMAS relationships scale and\nthe SF-36 emotional role limitations is 0.47. However, these\ncorrelations of subscales should be interpreted with caution\nas there are only four points on the MMAS subscales. So,\nto investigate these relationships further, respondents were\ndivided into four groups for each MMAS domain, and their\nscores on the SF-36 were compared using anova. The\nresults for most subscales and domains were highly signifi-\ncant for each comparison, with high scores on the MMAS\ncorresponding to high scores on the SF-36. The exception\nagain was for comparisons in the practical difficulties\ndomain. Here, three comparisons were not significant: men-\ntal health ( F\n3,422 = 1.66; p = 0.175); pain ( F3,424 = 0.62;\np = 0.60); and general health ( F3,415 = 2.21; p = 0.09).\nThere was a small, non-significant difference in the total\nMMAS scores of women reporting that they were not\nengaging in sexual activity on the SAQ and those who were\n(38.86 and 41.49, respectively). It may be that the MMAS\nrelationship domain reflects a particular problem with sex-\nual relationships. To test this, comparisons were made\nbetween the scores on the pleasure, discomfort and habit\nscales of the SAQ for women scoring at the four different\nlevels of the relationship domain. This revealed a significant\neffect of level on the discomfort scale ( F\n3,321 = 4.05;\np = 0.008), but insignificant effects of pleasure\n(F3,321 = 1.71; p = 0.165) and habit ( F3,326 = 2.54; p =\n0.057).\nDiscriminant validity\nThe MMAS is a condition-specific measure, and so should\nnot be affected by variables that are not directly related to\nvariation within the condition. So the total MMAS score\nwould not be expected to correlate with age or body mass\nindex (BMI). As expected, these correlations are very low:\nage r = 0.086; BMI r = )0.057. As well as providing evi-\ndence for the validity of the MMAS, it suggests that it can\nbe used with all age groups.\nAcceptability\nAll items in the MMAS were completed by all participants,\nin contrast to the other instruments, e.g. SF-36, where par-\nticipants missed various sections of the questionnaire.\nWe would suggest that this indicates that the MMAS was\nacceptable to participants, and that the questions were seen\nas relevant to their condition.\nFloor and ceiling effects\nIn psychometric instruments these have implications for\nthe precision of the instrument and its responsiveness to\nchange, as they reduce the likelihood that the instrument\nwill measure further improvement or deterioration. In this\nTable 1. Correlations between the MMAS total score and the\nEQ-5D summary and SF-36 subscales\nMMAS total\nrn\nSF-36 subscales\nPhysical functioning 0.33 420\nPhysical role limitations 0.45 422\nEmotional role limitations 0.49 421\nSocial functioning 0.54 427\nMental health 0.40 426\nEnergy/vitality 0.42 426\nPain 0.38 428\nGeneral health 0.32 419\nEQ-5D summary score 0.32 428\nPattison et al.\n1530 ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG\nMeasurement properties of MMAS\n\n168\nthe SF-36 social functioning scale. The correlation between the psychological health scales of \nthe two instruments is 0.39, and that between the MMAS relationships scale and the SF-36 \nemotional role limitations is 0.47. However, these correlations of subscales should be interpreted \nwith caution as there are only four points on the MMAS subscales. So, to investigate these rela-\ntionships further, respondents were divided into four groups for each MMAS domain, and their \nscores on the SF-36 were compared using anova. The results for most subscales and domains \nwere highly significant for each comparison, with high scores on the MMAS corresponding to \nhigh scores on the SF-36. The exception again was for comparisons in the practical difficulties \ndomain. Here, three comparisons were not significant: mental health (F\n3,422 = 1.66; p = 0.175); \npain (F3,424 = 0.62; p = 0.60); and general health (F3,415 = 2.21; p = 0.09).\nThere was a small, non-significant difference in the total MMAS scores of women reporting that \nthey were not engaging in sexual activity on the SAQ and those who were (38.86 and 41.49, re-\nspectively). It may be that the MMAS relationship domain reflects a particular problem with sexual \nrelationships. To test this, comparisons were made between the scores on the pleasure, discom-\nfort and habit scales of the SAQ for women scoring at the four different levels of the relationship \ndomain. This revealed a significant effect of level on the discomfort scale (F\n3,321 = 4.05; p = 0.008), \nbut insignificant effects of pleasure (F3,321 = 1.71; p = 0.165) and habit (F3,326 = 2.54; p =0.057).\nDiscriminant validity\nThe MMAS is a condition-specific measure, and so should not be affected by variables that are \nnot directly related to variation within the condition. So the total MMAS score would not be ex-\npected to correlate with age or body mass index (BMI). As expected, these correlations are very \nlow: age r = 0.086; BMI r = 0.057. As well as providing evidence for the validity of the MMAS, it \nsuggests that it can be used with all age groups.\nAcceptability\nAll items in the MMAS were completed by all participants, in contrast to the other instruments, \ne.g. SF-36, where participants missed various sections of the questionnaire. We would suggest \nthat this indicates that the MMAS was acceptable to participants, and that the questions were \nseen as relevant to their condition.\nFloor and ceiling effects\nIn psychometric instruments these have implications for the precision of the instrument and its \nresponsiveness to change, as they reduce the likelihood that the instrument will measure further \nimprovement or deterioration. In this sample five (1.2%) women’s scores were the lowest pos-\nsible and three (0.7%) were the highest possible. The overall distribution of scores was slightly \npositively skewed (skew = 0.41; SE = 0.12), which is what might be expected in a sample of \nChapter 11\n\n169\nwomen just about to embark on treatment for menorrhagia. These statistics indicate that there \nare no floor or ceiling effects.\nDiscussion\nThis study investigated further the psychometric properties of the MMAS, a menorrhagia-specific \nQoL instrument. It has shown that in addition to having high face validity, the MMAS has good \nconvergent and discriminant validity and test–retest reliability. It also has high internal consis- \ntency, with a Cronbach’s alpha score of 0.82, which is further evidence of reliability and indicates \nthat all the items are measuring different aspects of the same construct.\nThe overall MMAS showed moderate correlations with most of the subscales of the SF-36. Low-\ner correlations were found for physical functioning and general health perception. Rather than \nbeing a negative feature these low correlations may reflect the inappropriateness of the SF-36 \nfor measuring QoL in women with heavy menstrual bleeding. The physical functioning subscale \nis heavily weighted towards mobility and self-care, which are not likely to be affected by HMB. \nThe general health subscale and the EQ-5D, which were also relatively poorly correlated with \nthe MMAS, may not be precise enough to measure differences in this patient group, especially \nwhen women are asked to consider their general condition rather than that pertaining to their \nmenstrual cycle.\nConversely, scores on the practical difficulties subscale of the MMAS were not related to other \nQoL measures. This domain of the MMAS is very specific to menorrhagia, relating to sanitary \nprotection, flooding, etc., and may pick up specific important issues not measured by more \ngeneral scales.\nAll women completed the whole of MMAS, which suggests that it is acceptable to women and \nseen to be relevant. It could be argued that this sample was more highly motivated than women \nwith HMB more generally, because they had already agreed to take part in the ECLIPSE trial. \nHowever, there were missing data in other questionnaires administered at the same time.\nWe have not at this stage been able to assess the sensitivity of the MMAS to changes in treat-\nment. However, the findings of the ECLIPSE trial will enable us to report on this in due course.\nConclusion\nThese results suggest that the MMAS has good measurement properties, and is therefore an \nappropriate condition-specific instrument to measure the outcome of treatment for HMB.\nMeasurement properties of MMAS\n\n170\nAcknowledgements\nFirstly we would thank all the women who agreed to participate in this study. We thank all staff \nin primary and secondary care involved in the recruitment of participants to the ECLIPSE study \nand delivery of baseline questionnaires. We also thank Laura Gerrard, the trial co-ordinator, Lisa \nLeighton, the trial administrator, and Lee Middleton of the Birmingham Clinical Trials Unit for pre-\nparing the data for analysis. \nChapter 11\n\n171\nReferences\n1  Harlow SD, Campbell OMR. Epidemiology of menstrual disorders in developing coun-\ntries: a systematic review. BJOG 2004;111:6–16.\n2  Vessey MP , Villardmackintosh L, McPherson K, Coulter A, Yeates D. The Epidemiology of \nHysterectomy—Findings in a large cohort study. BOJG 1992;99:402–7.\n3  Clark TJ, Khan KS, Foon R, Pattison H, Bryan S, Gupta JK. Quality of life instruments \nin studies of menorrhagia: a systematic review. Eur J Obstet Gynecol Reprod Biol \n2002;104:96–104.\n4  Shaw RW, Brickley MR, Evans L, Edwards MJ. Perceptions of women on the im-\npact of menorrhagia on their health using multi-attribute utility assessment. BOJG \n1998;105:1155–9.\n5  Habiba M, Julian S, Taub N, Clark M, Rashid A, Baker R, et al. Limited role of multi-\nattribute utility scale and SF-36 in predicting management outcome of heavy menstrual \nbleeding. Eur J Obstet Gynecol Reprod Biol 2010;148:81–5.\n6  Stead ML, Crocombe WD, Fallowfield LJ, Selby P , Perren TJ, Garry R, et al. Sexual activ-\nity questionnaires in clinical trials: acceptability to patients with gynaecological disorders. \nBJOG 1999;106:50–4.\nMeasurement properties of MMAS\n\n172\n\n\n173\nChapter 12\nSummary and Discussion\n \n\n174\nChapter 12 - Summary and Discussion\nDiscussion\nThe first part of this thesis presents the results of the LUNA trial to determine the efficacy of \nlaparoscopic uterosacral nerve ablation (LUNA) for chronic pelvic pain (CPP), a multicentre trial \ncoordinated by the University of Birmingham Clinical Trials Unit, and the subsequent individual \npatient data (IPD) meta-analysis of this and all other comparable trials. A review of psychological \ntherapies is also presented. The second part of the thesis was undertaken as part of a commis-\nsioned project from the National Institute for Health Research Health Technology Assessment \nprogramme and assessed, by systematic review, the effectiveness of the levonorgestrel releasing \nintra-uterine system (LNG-IUS, Mirena®), hysterectomy and endometrial ablation in the treatment \nof heavy menstrual bleeding (HMB). The measurement properties of a quality of life instrument \nfor heavy menstrual bleeding were also assessed.\nThe introduction (Chapter 1) defines the two benign gynaecological conditions, chronic pelvic \npain and heavy menstrual bleeding, highlighting the burden to women and health care systems \nand describes the symptoms, aetiology and epidemiology for both. \nChapter 2 provides a complete clinical overview of chronic pelvic pain. The importance of tak-\ning a careful and thorough history to prompt further investigations is emphasised, particularly in \nview that an organic cause is not found at laparoscopy in at least a third of women with chronic \npelvic pain. The biopsychosocial model of pain is discussed and consideration of psychological \nsymptoms highlighted. The absence of effective gynaecological treatments is discussed and a \nmultidisciplinary approach is recommended.\nChapter 3 sets the scene for the randomised trial by describing the state of clinical practice in \nthe UK gynaecological community. We undertook a national survey and found around 45% of \nrespondents performed LUNA, of whom three quarters used the procedure for chronic pelvic \npain and just under two-thirds employed it for endometriosis. However, 81% of surgeons were \nprepared to randomise patients into a clinical trial of LUNA versus no intervention, demonstrating \nthat equipoise existed amongst gynaecologists and a clinical trial was warranted.\nThe protocol for the LUNA trial is detailed in Chapter 4. Women with chronic pain of greater \nthan 6 months duration, in whom diagnostic laparoscopy revealed either no pathology or mild \nendometriosis, were eligible. Consenting, eligible women were randomised during laparoscopy \nto undergo immediate uterosacral nerve ablation or no LUNA. Women were blinded to the allo-\nChapter 12\n\n175\ncation by use of a sham incision in the no LUNA group. Pelvic pain was measured using a visual \nanalogue scale, with general quality of life and sexual activity as secondary outcomes. Women \nresponded by postal questionnaire at 3, 6 and 12 months after randomisation. Pre-specified \nsub-groups included the presence or absence of some minimal pathology, the site of pain and \nparity. Analysis was by intention to treat. The sample size required to detect a difference of two \npoints on a ten centimetre scale, allowing for 20% loss to follow-up, was 440 participants.\nChapter 5 reports the results of the LUNA trial. After a median follow up of 69 months, there were \nno significant differences between the visual analogue pain scales for the worst pain reported \n(mean difference (MD) between LUNA group and No LUNA group -0·04cm, 95% confidence \ninterval (CI), -0·33 to 0·25cm; p=0·80), non-cyclical pain (-0·11cm, 95%CI -0·50 to 0·29cm; \np=0·60) dysmenorrhoea (-0·09cm, 95%CI -0·49 to 0·30cm; p=0·60), or dyspareunia (0·18cm,  \n-0·22 to 0·62cm; p=0·40). No differences were observed between the LUNA group and the No \nLUNA group for quality of life. We concluded LUNA did not result in improvements in pain, dys-\nmenorrhea, dyspareunia or quality of life, compared to laparoscopy without pelvic denervation.\nChapter 6 describes the protocol for a meta-analysis collecting IPD from the existing trials, to \nprovide a comprehensive assessment of the effectiveness of LUNA. We updated searches, con-\ntacted the authors of identified trials to obtain IPD and explored any heterogeneity. Pain scores \nwere calibrated to a 10 point scale and were analysed using a multilevel model allowing for \nrepeated measures.\nChapter 7 reports the results of the IPD meta-analysis. Raw data were available from 862 women \nrandomised into five trials. There was no significant difference between LUNA and No LUNA for \nthe worst pain recorded over a 12 month time period (MD: 0.25 points in favour of No LUNA on \na 0-10 point scale, 95% CI: -0.08 to 0.58; p=0.1). This IPD meta-analysis reinforces the conclu-\nsions drawn from our LUNA trial that the procedure is not effective in alleviating pain within one \nyear of treatment.\nThe effectiveness of psychological therapies for CPP was evaluated in a systematic review, de-\nscribed in Chapter 8. The selected four studies involving women with CPP were randomized \ncontrolled trials of psychological therapies compared to no treatment or standard gynecological \ntreatment. Compared to no psychological intervention, therapy produced a standardized mean \ndifference (SMD) of -3.27 (95% CI -4.52 to -2.02) and 1.11 (95% CI -0.05 to 2.27) at 3 months \nand -3.95 (95%CI -5.35 to -2.55) and 0.54 (95%CI -0.78 to 1.86) at 6 months and greater, based \non a visual analog scale score of 0-10. The current evidence does not allow us to conclude \nwhether psychological interventions have an effect on self-reported pain scores in women with \nCPP .\nSummary and Discussion\n\n176\nIn Chapter 9, we undertook an IPD meta-analysis to evaluate the relative effectiveness of hyster-\nectomy, endometrial destruction (both ‘first generation’ hysteroscopic and ‘second generation’ \nnon-hysteroscopic techniques) and LNG-IUS for HMB. Data on 2814 women were available from \n17 of the 30 randomised controlled trials identified. At around 12 months, 7.3% (12.6% v 5.3%) \nmore women were dissatisfied with outcome of first generation hysteroscopic techniques than \nhysterectomy (Odds Ratio (OR): 2.46, 95%CI:1.54-3.93, p=0.0002), but hospital stay (Weighted \nMean Difference (WMD): 3.0 days, 95%CI:2.9-3.1, p<0.00001) and time to resumption of normal \nactivities (WMD: 5.2 days, 95%CI:4.7-5.7, p<0.00001) were longer for hysterectomy.  Unsat-\nisfactory outcomes were comparable with first and second generation techniques (OR: 1.20, \n95%CI:0.88-1.62, p=0.2), although second generation techniques were quicker (WMD:14.5 \nminutes, 95%CI:13.7-15.3, p<0.00001) and women recovered sooner (WMD: 0.48 days, \n95%CI:0.20-0.75, p=0.0008) with fewer procedural complications. More women are dissatis-\nfied following endometrial destruction than hysterectomy. However, dissatisfaction rates are low \nafter all treatments and hysterectomy is associated with increased hospital stay and recovery \nperiod.  Definitive evidence on effectiveness of LNG-IUS compared to more invasive procedures \nis lacking.\nIn chapter 10, we extended the previous review to determine the relative effectiveness of second \ngeneration ablation techniques in the treatment of heavy menstrual bleeding.  Nineteen ran-\ndomised controlled trials (involving 3287 women) were identified. Of the three most commonly \nused techniques, network meta-analysis showed that bipolar radiofrequency and microwave \nablation resulted in higher rates of amenorrhoea than thermal balloon ablation at around 12 \nmonths (OR: 2.51, 95%CI 1.53 to 4.12, p<0.001; and 1.66, 1.01 to 2.71, p=0.05, respectively), \nbut there was no evidence of a convincing difference between the three techniques in the num-\nber of women dissatisfied with treatment or still experiencing heavy bleeding. Compared with \nbipolar radio frequency and microwave devices, an increased number of women still experienced \nheavy bleeding after free fluid ablation (OR: 2.19, 95%CI 1.07 to 4.50, p=0.03; and 2.91, 1.23 \nto 6.88, p=0.02, respectively). Compared with radio frequency ablation, free fluid ablation was \nassociated with reduced rates of amenorrhoea (OR: 0.36, 95%CI 0.19 to 0.67, p=0.004) and \nincreased rates of dissatisfaction (OR: 4.79, 95%CI 1.07 to 21.5, p=0.04). We concluded bipolar \nradio frequency and microwave ablative devices are more effective than thermal balloon and free \nfluid ablation in the treatment of heavy menstrual bleeding with second generation endometrial \nablation devices.\nFinally, in Chapter 11, the reliability and validity of the Menorrhagia Multi-Attribute Scale (MMAS), \na menorrhagia-specific quality of life instrument, was investigated. The MMAS was found to have \ngood measurement properties and is therefore an appropriate condition-specific instrument to \nmeasure the outcome of treatment for HMB.\nChapter 12\n\n177\nChronic Pelvic Pain\nCPP represents a significant disease burden in both primary and secondary care 1. A diagnosis \nof idiopathic CPP is arrived at once all other organic causes of pain are excluded by diagnostic \nlaparoscopy.  That is not to say that idiopathic CPP is a psychogenic condition. Endometriosis, \npelvic adhesions, chronic pelvic inflammatory disease, and ovarian cysts were the pathologies \nmost frequently observed at laparoscopy in women with CPP in the LUNA Trial. 2 Moreover, the \nseverity of pain is not related to severity of underlying pathology, as illustrated by endometriosis, \nwhere stage of disease is poorly correlated with reported pain.\n3 CPP is seldom caused by a sin-\ngle factor alone and psychological symptoms may be both causative and associative.\nA troublesome clinical issue is the lack of an accurate tool to efficiently diagnose and direct cases \nof patients.4 The Royal College of Obstetricians and Gynaecologists (RCOG) guidelines provide a \nnumber of suggested initial investigations, including history, microbiological screening and vagi-\nnal examination, all with weak evidence for utility.\n5 If no cause of the pain is found, the first port \nof call would usully be to perform a diagnostic laparoscopy, although the RCOG guidelines now \nsuggest it may be better seen as a second line investigation if other therapeutic options have \nfailed.\n5\nThe parasympathetic ganglia and nerve plexuses in the uterosacral ligaments are believed to \ncarry the somatic pain from the pelvis to the pain, thus theoretically, ablation of these nerve trunks \nshould offer pain relief.  Neuroablation at the time of diagnostic laparoscopy was widely practiced \nin the UK,\n6 but its efficacy was never reliably demonstrated prior to our work. Our IPD meta-\nanalysis reinforced the conclusions drawn from the LUNA trial that the procedure is not effective \nin alleviating pain within one year of treatment.  Subgroup analysis did suggest a benefit of LUNA \nin those without visible pathology, consistent with the Cochrane review, although this effect was \nonly seen at 12 months and was of borderline statistical significance.7 Conversely, the IPD meta-\nanalysis of those with visible pathology showed greater decrease in pain in the No LUNA group, \ncontradicting the Cochrane review. These subgroup effects lack biological plausibility and are \nlikely to have arisen by chance.\nThe failure to demonstrate an improvement in pain relief or quality of life in women who had the \nLUNA procedure illustrates the importance of rigorous assessment in the pursuit of evidence \nbased medicine. Negative results are equally important as data supporting a procedure. LUNA \nis not without risks and costs and the findings of our research here should discourage the use of \nan ineffective procedure. Conversely, there is still a lack of evidence regarding the effectiveness \nof adhesionolysis in reduction of CPP .\n8 There is insufficient evidence to comment on pain relief \nafter hysterectomy, and reflecting similar guidelines for heavy menstrual bleeding, such a major \noperation perhaps should not be considered as a first line solution to pelvic pain.\nSummary and Discussion\n\n178\nWe postulate that the initial general improvement seen over the first 6 months is attributable to \nthe reassurance derived from being told that no serious cause had been found for their pain. \nA study identified in our review of psychological therapies found no difference in pain between \nwomen debriefed after laparoscopy with a photograph of their pelvis, compared with those that \ndid not receive a photograph.\n9 Grace, in her reflections from interviews with women with CPP , \nobserves that women expect the source of pain should be “visible”, in the form of pathology, and \nin the absence of a laparoscopic diagnosis, tend to normalise the pain. Yet women continue to \nrefer in their thoughts about the cause of pain to biological mechanisms, for example hormonal \nimbalance or pelvic trauma due to childbirth.10 Living with chronic pelvic pain requires women to \ndevelop coping strategies to maintain their daily family and working lives. Cognitive behavioural \ntherapy (CBT) may help with this. CBT involves encourages women to first assess the thoughts \nassociated with pain, and the extent and consequences of such thinking. The goal then is to \ndevelop adaptive coping behaviour that seeks to achieve positive goals. CBT has demonstrated \nto be effective in improving mood and may be useful in reducing pain and disability in a review of \nmany musculoskeletal chronic pain,\n11 but the effectiveness for CPP is unproven.\nFuture research in the management of CPP\nCPP can be viewed in a ‘biopsychosocial’ model, in which underlying ‘disease’ states interact \nwith psychological and social factors to produce the symptoms experienced.\n12 This approach \ndemands a multidisciplinary approach to management and there is some evidence that this \nresults in improved quality of life, if not reduced pain, 13 but further health service research is re-\nquired. Such multidisciplinary care should be considered as a complex intervention, built up from \na number of components, each of which may act both independently and inter-dependently. \nThe components usually include treatment activities e.g. psychological therapy, parameters of \nthese activities e.g. frequency, timing in relation to other aspects of management and methods \nof organising those activities e.g. group sessions or CBT by telephone. Sometimes is difficult to \nidentify the most efficacious components within a multidisciplinary care approach and it is not \nnecessarily enough to conduct randomised trials of individual components and assume that put-\nting together those activities that are of benefit will produce a synergistic effect. Optimisation of \na multidisciplinary approach will require a coordinated programme of research, modelled upon \nthe Medical Research Council’s Framework for the development and evaluation of complex in-\nterventions.\n14\nFuture trials should also consider the recommendations of the IMMPACT initiative and employ \noutcome measures that not only capture the intensity of pain, but the effect on physical and emo-\ntional functioning.15  If pharmacological treatments are to be assessed, the use of titrated dosing \nand a minimum treatment period of 3 months are recommended in order to identify respond-\ners.16 New trials should have larger numbers,17,18 aiming to recruit 350 or more patients to detect \nChapter 12\n\n179\na more realistic, but worthwhile, effect of any intervention. To explore when the benefit emerges \nand whether it is maintained, longer follow-up of participants is recommended. In chronic condi-\ntions where outcomes may vary over time, repeated measures analyses will maximize statistical \npower from the trial sample. Finally, to truly capture the impact of CPP on a woman’s life, a dis-\nease specific instrument, developed and validated by rigorous methods, and driven by accounts \nof symptoms and burden from women with CPP , needs to be develop in order to assess the \neffectiveness of future treatments.\n \nHeavy Menstrual Bleeding\nHeavy menstrual bleeding (HMB) is a common problem that can significantly impact on women’s \nlives, and burden individuals and healthcare systems.19,20 A range of non-hormonal and hormonal \nmedical treatments are available as first line therapy for women presenting with HMB in primary \ncare. ECLIPSE, a recent large, pragmatic randomized trial found both LNG-IUS and usual medi-\ncal treatments reduce the adverse impact of HMB on women’s lives over two years, but LNG-IUS \nis the more effective.21 Discontinuation rates for LNG-IUS in our review were 28% by two years, \ncomparable to ECLIPSE where 31% of those who had the system fitted had it removed, pre-\ndominantly for perceived lack of effectiveness or irregular or prolonged bleeding.\nWhen medical treatments fail to provide adequate relief, surgical interventions, including hys-\nterectomy or endometrium destruction, can be considered.\n22 A rapid decrease in the number \nof hysterectomies has been observed in the UK, 23 with a corresponding rise in the number of \nendometrial ablations, 24 but overall, surgical procedures are decreasing, which can perhaps \nbe ascribed to the increasing use of LNG-IUS over the last decade. This is reflected in the \nECLIPSE trial, where 6% (32 of 571 women randomised) had hysterectomies within 2 years, yet \n5% (28/571) had endometrial ablation. Our review found that both first and second generation \nendometrial destruction techniques are associated with higher dissatisfaction rates than hyster-\nectomy, although all three procedures provided high levels of satisfaction, with small absolute \ndifferences. The predicted dominance of endometrial destruction is neither being witnessed nor \nmay be warranted.\n24 The results of an current randomised trial of LNG-IUS compared with a \nsecond generation endometrial device also help resolve the question as to first line treatment.25\nTechnological developments may influence the further uptake of endometrial destruction. Whilst \nour review found no statistically significant difference in amenorrhea rates between first and \nsecond generation techniques, the latter were clearly quicker to perform, and enabled women to \nreturn to work and normal activities more quickly.\n26 Comparing second generation techniques, \nbipolar radiofrequency ablation seems to offer benefits over thermal balloon in terms of amenor-\nrhea rates and is procedurally quicker than thermal balloon and microwave ablation.27 Again, this \nSummary and Discussion\n\n180\nis reflected in UK usage data, with radiofrequency ablation techniques being the most rapidly ris-\ning procedure. Whether this market share is sustained will depend on the emergence of simpler, \nnext generation devices, the replacement chosen by users of the microwave ablation system and \nthe increasing pace of out-patient based procedures. \nPain, from cervical dilation and vasovagal reactions are limiting factors for the wider use of out-\npatient procedures. Only two studies have evaluated the use of any endometrial device under lo-\ncal anaesthesia,28,29 both of which employed intracervical anaesthesia. The efficacy of intra- and \nparacervical anaesthesia in reducing pain during diagnostic hysteroscopy has been demonstrat-\ned,\n30 but data on the most effective anaesthetic for interventional procedures is lacking. Other \naspects, such as the distension media, diameter of the ablative device, and the use of speculum \nand other cervical instruments may also influence patient discomfort with the procedure. Ulti-\nmately, the decision should be left to the women’s preference and the operator’s experience.\nThere has been a long term interest in progesterone antagonists and selective progesterone re-\nceptor modulators for treatment for idiopathic HMB and HMB associated with uterine fibroids.\n31 \nRecently, a pair of randomised trials has recently demonstrated the effectiveness of the ulipristal \nacetate in reducing fibroid size and quickly inducing amenorrhea.\n32,33 However, reversible chang-\nes to the endometrium have been witnessed,34 limiting their indication to short-term pre-operative \nuse. There is potential for use of ulipristal as a first line, long term treatment for heavy menstrual \nbleeding associated with uterine fibroids, but not without further data on the long-term safety in \nthis population.\nFuture Research in Heavy Menstrual Bleeding\nAny further developments in endometrial ablation devices should be compared against the bipo-\nlar radiofrequency procedure before widespread adoption. New thermal balloon devices are now \non the market which are suitable for the out-patient setting and have shorter treatment cycles. \nEvidence-based improvements in pain management for diagnostic hysteroscopy should be ex-\nplored in interventional procedures.\nThe UK guidelines for heavy menstrual bleeding recommend that for women with a normal uter-\nus, endometrial ablation should be considered preferable to hysterectomy,\n35 whilst our review in-\ndicates higher levels of dissatisfaction with endometrial ablation.26 It has been suggested that this \ndifference might be amplified if laparoscopic sub-total hysterectomy was performed, in contrast \nto the total hysterectomies predominant in the contributing studies. However, a Cochrane review \nfailed to corroborate anecdotal evidence of improved sexual, bowel or urinary outcomes after \nsub-total hysterectomy.36 A randomized trial of sub-total hysterectomy compared with second \ngeneration ablative techniques would address this uncertainty.\nChapter 12\n\n181\nThe current epidemiological data on the prevalence of heavy menstrual bleeding are derived \nfrom regional surveys and are out-of-date.37,38 It is also likely that those presenting in primary care \nrepresent only a minority of women experiencing difficulties with their periods. A shift in empha-\nsis away from menstrual blood loss to a more patient-centred definition of HMB that interferes \nwith women’s physical, emotional and social life, is advocated.39 Thus, the priority for healthcare \nproviders should be to respond to the impact of HMB on women’s day-to-day lives, 40 basing \ntreatment decisions on evidence using patient reported outcomes, such as the validated Multi-\nattribute Menorrhagia Assessment Scale.35,41,42 \nSummary and Discussion\n\n182\nNederlandse samenvatting\nIn het eerste deel van dit proefschrift wordt de LUNA studie over de effectiviteit van laparoscopis-\nche uterosacrale zenuw ablatie (LUNA) voor chronische pelviene pijn besproken. De LUNA studie \nis een multicenter onderzoek gecoördineerd door de universiteit van Birmingham Clinical Trials \nUnit. Daarnaast bevat het eerste deel de daaropvolgende meta-analyse met individuele patiënt-\ngegevens (IPD) van deze en alle andere vergelijkbare studies. Ook wordt een overzicht van psy-\nchologische therapieën wordt ook gepresenteerd. \nHet tweede deel van het proefschrift werd verricht in het kader van een project in het kader van \nhet programma van het nationale instituut voor Health Research Health Technology Assessment. \nBinnen dit project werd via een systematische review verricht naar de effectiviteit van de lev-\nonorgestrel houdend intra-uteriene spiraal (LNG-IUS), hysterectomie en endometrium ablatie in \nde behandeling van hevige menstruele bloedingen (HMB). Ook werden de meet eigenschappen \nvan een kwaliteit van leven instrument voor hevige menstruele bloedingen beoordeeld.\nDe inleiding (hoofdstuk 1) definieert de twee goedaardige gynaecologische aandoeningen, chro-\nnische bekken pijn in het en menorrhagie (hevige menstruatie), de symptomen, etiologie en epi-\ndemiologie en beschrijft de impact op vrouwen en op het zorgsysteem. \nHoofdstuk 2 geeft een compleet klinisch overzicht van het ziektebeeld chronische pelviene pijn. \nHet belang van het afnemen van een zorgvuldige en grondige anamnese wordt benadrukt, met \nname gezien het feit dat dat een organische oorzaak in ten minste een derde van de vrouwen \nmet chronische pelviene pijn niet wordt gevonden bij laparoscopie. Het biopsychosociale model \nvan pijn wordt besproken en betekenis van psychische klachten wordt gemarkeerd. Vervolgens \nwordt het ontbreken van effectieve gynaecologische behandelingen besproken en wordt een \nmultidisciplinaire aanpak wordt aanbevolen.\nHoofdstuk 3 beschrijft de noodzaak voor een gerandomiseerde studie aan de hand van een be-\nschrijving van de klinische praktijk in het Verenigd Koninkrijk. We vonden in een nationaal onder-\nzoek dat ongeveer 45% van de respondenten LUNA gebruikt, van wie driekwart de procedure \nvoor chronische pelviene pijn gebruikte en tweederde voor endometriose. Het bleek dat 81% \nvan de gynaecologen bereid was om patiënten the randomiseren in de LUNA trial. Gezien deze \n‘equipoise’  bleek een klinisch onderzoek gerechtvaardigd was.\nHet protocol voor de LUNA trial is beschreven in hoofdstuk 4. Vrouwen met langer dan 6 maanden \nbestaande pelviene pijn, die bij diagnostische laparoscopie geen pathologie of milde endome-\nChapter 12\n\n183\ntriose hadden, kwamen in aanmerking. Na geïnformeerde toestemming, werden in aanmerking \nkomende vrouwen tijdens laparoscopie gerandomiseerd voor directe uterosacrale zenuw ablatie \nof geen interventie. Vrouwen bleven geblindeerd voor de toewijzing door het gebruik van een \nschijnincisie in de controle groep. \nVervolgens werd pelviene pijn gemeten met behulp van een visuele analoge schaal, met een \nalgemene kwaliteit van leven vragenlijst en seksuele activiteit als secundaire uitkomsten. Deelne-\nmers vulden vragenlijsten in op 3, 6 en 12 maanden na randomisatie. Vooraf gespecificeerde \nsubgroepen omvatten de aanwezigheid of afwezigheid van een minimale pathologie, de plaats \nvan de pijn en pariteit. Analyse gebeurde volgens intention to treat. De steekproefgrootte nodig \nom een verschil van twee punten te kunnen detecteren op een tien centimeter schaal, gecor -\nrigeerd voor 20% verlies voor follow-up, was 440.\nHoofdstuk 5 beschrijft de resultaten van de LUNA trial. Na een mediane follow-up van 69 \nmaanden waren er geen significante verschillen tussen de visuele analoge pijnschalen voor de \ngerapporteerde ergste pijn (gemiddeld verschil (MD) tussen LUNA groep en geen LUNA groep \n-0.04cm, 95% betrouwbaarheidsinterval (BI) , -0.33 tot 0.25cm, p = 0.80), niet-cyclische pijn \n(-0.11cm, 95% CI -0.50 tot 0.29cm, p = 0.60), dysmenorroe (-0.09cm, 95% CI -0.49 tot 0.30cm, \np = 0.60), of dyspareunie (0.18cm, -0.22 tot 0.62cm, p = 0.40). Er werden geen verschillen \nwaargenomen tussen de LUNA groep en de controle-groep voor kwaliteit van het leven. Wij \nconcludeerden dat LUNA niet resulteert in verbetering van de pijn, dysmenorroe, dyspareunie en \nkwaliteit van leven, vergeleken met laparoscopie zonder LUNA.\nHoofdstuk 6 beschrijft het protocol voor een meta-analyse op basis van individuele patiënten \ndata van de bestaande studies naar de effectiviteit van LUNA. We hernieuwden zoekopdrachten, \nzochten contact met de auteurs van geïdentificeerde studies voor het verkrijgen van individuele \npatiënten data en onderzochten welke heterogeniteit. Pijnscores werden gekalibreerd op een \n10-puntsschaal en werden geanalyseerd met behulp van een multilevel-model waardoor her -\nhaalde metingen.\nHoofdstuk 7 rapporteert de resultaten van de IPD meta-analyse. Ruwe data waren beschikbaar \nvan 862 vrouwen gerandomiseerd in vijf studies. Er was geen significant verschil tussen LUNA \nen geen LUNA voor de ergste pijn tijdens een 12 maanden durende periode (MD: 0.25 punten in \nhet voordeel van geen LUNA op een 0-10 schaal van, 95% BI: -0.08 tot 0.58, p = 0.1). Uit deze \nIPD meta-analyse blijkt dat LUNA niet effectief is in het reduceren van pelviene pijn.\nDe effectiviteit van psychologische therapieën voor chronische pelviene pijn werd geëvalueerd \nin een systematische review  beschreven in hoofdstuk 8. De vier geselecteerde studies over \nvrouwen met  chronische pelviene pijn waren gerandomiseerde gecontroleerde studies waarin \nSummary and Discussion\n\n184\npsychologische therapieën werden vergeleken met geen behandeling of standaard gynaecolo-\ngische zorg. In vergelijking met geen psychologische interventie, resulteerde de therapie na 3 \nmaanden in een gestandaardiseerd gemiddeld verschil (SMD) van -3,27 (95% CI -4.52 tot -2.02) \nen 1.11 (95% CI -0.05 tot 2.27) en na 6 maanden of meer in -3.95 (95% CI – 5.35 tot -2.55) \nen 0.54 (95% CI -0.78 tot 1.86), gebaseerd op een visuele analoge schaal score van 0-10. De \nhuidige gegevens maken een conclusie over de effectiviteit van psychologische interventies op \nzelf-gerapporteerde pijnscores bij vrouwen met  chronische pelviene pijn niet mogelijk.\nIn hoofdstuk 9 ondernamen we een IPD meta-analyse naar de relatieve effectiviteit van hyster -\nectomie, endometrium ablatie (zowel ‘eerste generatie’ hysteroscopische en ‘tweede generatie’ \nniet-hysteroscopische technieken) en LNG-IUS in de behandeling van heftig menstrueel bloed-\nverlies. Wij verkregen gegevens van 2814 vrouwen afkomstig van 17 van de 30 gerandomis-\neerde gecontroleerde studies. Na 12 maanden, bleek 7.3% (12.6% v 5.3%) meer vrouwen waren \nontevreden met het resultaat van de eerste generatie hysteroscopische ablatie dan hysterecto-\nmie (Odds Ratio (OR): 2.46, 95% CI :1.54 tot 3.93, p = 0.0002), maar verblijf in het ziekenhuis \n(Gewogen gemiddeld verschil (GGV): 3.0 dagen, 95% CI :2.9 tot 3.1, p <0.00001) en de tijd tot \nhervatting van de normale activiteiten (GGV: 5.2 dagen, 95% CI :4.7 tot 5.7, p <0.00001) waren \nlanger voor hysterectomie. \nOntevredenheid met het behandelresultaat trad in vergelijkbare mate op tussen eerste en tweede \ngeneratie technieken (OR: 1.20, 95% CI :0.88 tot 1.62, p = 0.2), hoewel de tweede generatie \ntechnieken minder tijd kosten (GGV: 14.5 minuten, 95% CI :13.7 tot 15.3, p < 0.00001) en \nvrouwen sneller herstelden (GGV: 0.48 dagen, 95% CI :0.20 tot 0.75, p = 0.0008) en minder \ncomplicaties ondervonden. Meer vrouwen waren ontevreden na endometrium ablatie dan na \nhysterectomie. Echter, het percentage vrouwen dat ontevreden is, is laag na alle behandelingen, \nen hysterectomie wordt geassocieerd met een verhoogd verblijf in het ziekenhuis en herstelp-\neriode. Definitief bewijs voor de effectiviteit van de LNG-IUS in vergelijking met meer invasieve \nprocedures ontbreekt vooralsnog.\nIn hoofdstuk 10 breidden we het review uit en evalueerden ook de relatieve effectiviteit van de \ntweede generatie ablatie technieken bij de behandeling van hevige menstruaties. Wij identifi-\nceerden 19 gerandomiseerde gecontroleerde studies met 3287 vrouwen. Netwerk meta-analyse \nliet zien dat na 12 maanden bipolaire radiofrequente ablatie en microwave ablatie resulteerden in \nhogere amenorroe percentages dan thermische ballon ablatie (OR: 2.51, 95% CI 1.53 tot 4.12, p \n<0.001, en 1.66, 1.01 tot 2.71, p = 0.05, respectievelijk). Toch was er geen overtuigend verschil \ntussen de drie technieken in het aantal vrouwen dat nog steeds ontevreden was over de be-\nhandeling of nog steeds last had van hevig menstrueel bloedverlies. In vergelijking met bipolaire \nradiofrequente ablatie en microwave ablatie, was het aantal vrouwen dat met hevig menstrueel \nChapter 12\n\n185\nbloedverlies na vrij vocht ablatie verhoogd (OR: 2.19, 95% CI 1.07 tot 4.50, p = 0.03, en 2.91, \n1.23 tot 6.88, p = 0.02, respectievelijk ). In vergelijking met bipolaire radiofrequente ablatie, was \nvrij vocht ablatie geassocieerd met een verminderde kans op amenorroe (OR: 0.36, 95% CI 0.19 \ntot 0.67, p = 0.004) en een verhoogde mate van ontevredenheid (OR: 4.79, 95% CI 1.07 tot \n21.5, p = 0.04). Wij concludeerden dat bipolaire radiofrequente ablatie en microwave ablatie ef-\nfectiever zijn dan thermische ballon ablatie in de behandeling van heftig menstrueel bloedverlies \nmet tweede generatie ablatie apparaten.\nTen slotte, werd in hoofdstuk 11 de betrouwbaarheid en validiteit van de Menorragie Multi-At-\ntribute Scale (MMAS), een menorragie-specifieke kwaliteit van leven instrument, onderzocht. De \nMMAS bleek goede meet eigenschappen te hebben en is daarom een geschikt conditie-speci-\nfiek instrument om de uitkomst van de behandeling van heftig menstrueel bloedverlies te meten.\nSummary and Discussion\n\n186\nReferences /  Referentielijst\n1  Latthe P , Latthe M, Say L, Gulmezoglu M, Khan KS. WHO systematic review of prev-\nalence of chronic pelvic pain: a neglected reproductive health morbidity. BMC Public \nHealth 2006;6:177.\n2 Howard F.M. The role of laparoscopy in chronic pelvic pain: promise and pitfalls. Obstetri-\ncal & Gynecological Survey 1993;48:357-87.\n3 Gruppo Italiano per lo Studio dell’Endometriosi. Relationship between stage, site and \nmorphological characteristics of pelvic endometriosis and pain. Hum Reprod 2001 Dec \n1;16(12):2668-71.\n4 Prentice A. Medical management of chronic pelvic pain. Baillieres Best Practice & Re-\nsearch Clinical Obstetrics and Gynaecology 2000;14(3):495-9.\n5 Royal College of Obstetrics and Gynaecology. Initial management of chronic pelvic pain. \nLondon: Royal College of Obstetrics and Gynaecology; 2012. Report No.: Greentop \nguidelines 41, 2nd edition.\n6 Daniels J, Gray R, Khan K, Gupta J. Laparoscopic uterine nerve ablation: A survey of \ngynaecological practice in the UK. Gynaecol Endosc 2000;9(3):157-9.\n7 Proctor M.L., Latthe P .M., Farquahr C.M., Khan K.S., Johnson N.P . Surgical interruption \nof pelvic nerve pathways for primary and secondary dysmenorrhoea. Cochrane Database \nSyst Rev 2005;4.\n8 Stones W, Cheong YC, Howard FM. Interventions for treating chronic pelvic pain in wom-\nen. Cochrane Database of Systematic Reviews 2005;2.\n9 Onwude JL, Thornton JG, Morely S, Lilleyman J, Currie I, Lilford R. A randomised trial of \nphotographic reinforcement during post-operative counselling after diagnostic laparos-\ncopy for plevic pain. BJOG 2004;112:89-94.\n10 Grace VM, MacBride-Stewart S. ‘Women get this’: gendered meanings of chronic pelvic \npain. Health (London) 2007;11(1):47-67.\n11  Eccleston C, Williams A, Morley S. Psychological therapies for the management of chron-\nic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2009.\n12 Chapman RC, Nakamura Y, Flores LY. Chronic pelvic pain and consciousness: a con-\nstructivist perspective. In: Gatchel RJ, Turk DC, editors. Psychosocial factors in pain: \ncritical perspectives.New York: Guilford Press; 1999. p. 35-55.\n13 Peters AA, van DE, Jellis B, van ZE, Hermans J, Trimbos JB. A randomized clinical trial \nto compare two different approaches in women with chronic pelvic pain. Obstet Gynecol \n1991;77(5):740-4.\n14 Medical Research Council. Developing and evaluating complex interventions: new guid-\nance. London: Medical Research Council; 2006. \n15 Dworkin R.H., Turk D.C., Farrar J.T., Haythornthwaite J.A., et al. Core outcomes meas-\nures for chronic pain clinical trials: IMMPACT guidelines. Pain 2005;113:9-19.\nChapter 12\n\n187\n16 Dworkin RH, Turk DC, Peirce-Sandner S, Baron R, Bellamy N, Burke LB, et al. Research \ndesign considerations for confirmatory chronic pain clinical trials: IMMPACT recommen-\ndations. Pain 2010;149(2):177-93.\n17 Hills R.K., Daniels J.P . Assessing new interventions in women’s health. Best Pract Res \nClin Obstet Gynaecol 2006;20(5):713-28.\n18 Khan K.S., Selman T., Daniels J. Gynecology Randomized Control Trials. Handbook of \nClinical Trials. Wiley and Sons, Inc. New Jersey, USA; 2009. p. 519-30.\n19 Frick KD, Clark MA, Steinwachs DM, Langenberg P , Stovall D, Munro MG, et al. Financial \nand quality-of-life burden of dysfunctional uterine bleeding among women agreeing to \nobtain surgical treatment. Womens Health Issues 2009;19(1):70-8.\n20 Roberts TE, Tsourapas A, Middleton LJ, Champaneria R, Daniels JP , Cooper KG, et \nal. Hysterectomy, endometrial ablation, and levonorgestrel releasing intrauterine system \n(Mirena) for treatment of heavy menstrual bleeding: cost effectiveness analysis. BMJ \n2011;342:d2202.\n21 Gupta J.K., Kai J, Middleton LJ, Pattison H, Gray R., Daniels JP , et al. Levonorgestrel \nintrauterine system versus medical therapy for menorrhagia. NEJM 2013; 368:128-137\n22 Lethaby A, Farquhar C. Treatments for heavy menstrual bleeding. BMJ \n2003;327(7426):1243-4.\n23 Reid PC, Mukri F. Trends in number of hysterectomies performed in England for menorrha-\ngia: examination of health episode statistics, 1989 to 2002-3. BMJ 2005;330(7497):938-9.\n24 Reid P .C. Endometrial ablation in England - coming of age? An examination of hospi-\ntal episode statistics 1989/1990 to 2004/2005. Eur J Obstet Gynecol Reprod Biol \n2007;135(2):191-4.\n25 Netherlands Consortium for Women’s health and Reproductivity studies. Mira Trial.  2012. \n15-11-0012. \n26 Middleton LJ, Champaneria R, Daniels JP , Bhattacharya S, Cooper KG, Hilken NH, et al. \nHysterectomy, endometrial destruction, and levonorgestrel releasing intrauterine system \n(Mirena) for heavy menstrual bleeding: systematic review and meta-analysis of data from \nindividual patients. BMJ 2010;341:c3929.\n27 Daniels JP , Middleton LJ, Champaneria R, Khan KS, Cooper K, Mol BW, et al. Second \ngeneration endometrial ablation techniques for heavy menstrual bleeding: network meta-\nanalysis. BMJ 2012;344:e2564.\n28 Sambrook A.S., Cooper K.G., Campbell M.K., Cook J.A. Clinical outcomes from a ran-\ndomised comparison of Microwave Endometrial Ablation with Thermal Balloon endome-\ntrial ablation for the treatment of heavy menstrual bleeding. BJOG 2009;116(8):1038-45.\n29 Clark TJ, Samuel N, Malick S, Middleton LJ, Daniels J, Gupta JK. Bipolar radiofrequency \ncompared with thermal balloon endometrial ablation in the office: a randomized controlled \ntrial. Obstet Gynecol 2011;117(1):109-18.\nSummary and Discussion\n\n188\n30 Cooper NA, Khan KS, Clark TJ. Local anaesthesia for pain control during outpatient hys-\nteroscopy: systematic review and meta-analysis. BMJ 2010;340:c1130.\n31 Heikinheimo O, Vani S, Carpen O, Tapper A, arkki PÃ, utanen EM, et al. Intrauterine re-\nlease of progesterone antagonist ZK230211 is feasible and results in novel endometrial \neffects: a pilot study. Human Reproduction 2007;22(9):2515-22.\n32 Donnez J, Tatarchuk TF, Bouchard P , Puscasiu L, Zakharenko NF, Ivanova T, et al. Ul-\nipristal acetate versus placebo for fibroid treatment before surgery. N Engl J Med \n2012;366(5):409-20.\n33 Donnez J, Tomaszewski J, Vazquez F, Bouchard P , Lemieszczuk B, Baro F, et al. Ulipristal \nacetate versus leuprolide acetate for uterine fibroids. N Engl J Med 2012;366(5):421-32.\n34 Williams AR, Bergeron C, Barlow DH, Ferenczy A. Endometrial morphology after treat-\nment of uterine fibroids with the selective progesterone receptor modulator, ulipristal ac-\netate. Int J Gynecol Pathol 2012;31(6):556-69.\n35 National Collaborating Centre for Women’s and Children’s Health. Heavy Menstrual Bleed-\ning. London: Royal College of Obstetricians and Gynaecologists; 2007. Report No.: CG44.\n36 Lethaby A, Mukhopadhyay A, Naik R. Total versus subtotal hysterectomy for benign gy-\nnaecological conditions. Cochrane Database of Systematic Reviews 2012.\n37 Shapley M, Jordan K, Croft PR. An epidemiological survey of symptoms of menstrual loss \nin the community. Brit Gen Pract 2004;54(502):359-63.\n38 Office of Population Census and Surveys. Morbidity Statistics from General Practice. Lon-\ndon: HMSO; 1995. Report No.: Fourth National Morbidity Study 1991-1992.\n39 National Collaborating Centre for Women’s and Children’s Health. Heavy Menstrual \nBleeding.  2008. \n40 Santer M, Wyke S, Warner P . What aspects of periods are most bothersome for women \nreporting heavy menstrual bleeding? Community survey and qualitative study. BMC Wo-\nmens Health 2007;7:8.\n41 Pattison H, Daniels JP , Kai J, Gupta JK. The measurement properties of the menorrhagia \nmulti-attribute quality-of-life scale: a psychometric analysis. BJOG 2011;118(12):1528-31.\n42 Shaw RW, Brickley MR, Evans L, Edwards MJ. Perceptions of women on the im-\npact of menorrhagia on their health using multi-attribute utility assessment. BJOG \n1998;105(11):1155-9.\nChapter 12\n\n189\nBibliography\nChronic Pelvic Pain\nDaniels, J, Khan, K.S. BMJ 2010;341:c4834\nLaparoscopic uterine nerve ablation: a survey of gynaecological practice in the UK\nDaniels J, Gray R, Khan KS, Gupta JK. Laparoscopic uterine nerve ablation: A survey of gynae-\ncological practice in the UK. Gynaecol Endoscopy 2000; 9:157-159.\nA randomised controlled trial to assess the efficacy of Laparoscopic Uterosacral Nerve \nAblation in the treatment of chronic pelvic pain: the trial protocol\nA randomised controlled trial to assess the efficacy of Laparoscopic Uterosacral Nerve Ablation \n(LUNA) in the treatment of chronic pelvic pain: The trial protocol [ISRCTN41196151]. BMC Wo-\nmens Health 2003;3(1):6.\nA randomized controlled trial to assess the effectiveness of Laparoscopic Uterosacral \nNerve Ablation for alleviating chronic pelvic pain\nDaniels J, Gray R, Hills RK, Buckley L, Gupta JK, Latthe P , et al on behalf of the LUNA Trial Col-\nlaboration A Randomized Controlled Trial to Assess the Effectiveness of Laparoscopic Uterosa-\ncral Nerve Ablation (LUNA) for alleviating Chronic Pelvic Pain JAMA 2009;302(9):955-961\nMeta-analysis using individual patient data from randomised trials to assess the effec-\ntiveness of Laparoscopic Uterosacral Nerve Ablation in the treatment of chronic pelvic \npain: a proposed protocol.\nXiong T, Daniels J, Middleton L, Champaneria R, Khan KS, Gray R, et al International LUNA IPD \nMeta-analysis Collaborative Group Meta-analysis using individual patient data from randomised \ntrials to assess the effectiveness of laparoscopic uterosacral nerve ablation in the treatment of \nchronic pelvic pain: a proposed protocol. BJOG. 2007 Dec;114(12):1580, e1-7\nIndividual patient data meta-analysis to assess the effectiveness of laparoscopic uter-\nosacral nerve ablation in the treatment of chronic pelvic pain.\nDaniels JP , Middleton L, Xiong X, Champaneria R, Johnson NP , Lichten EM, Sutton C, Vercellini \nP ,, Gray R, Hills RK, Jones KD, Aimi G, Khan KS. Individual patient data meta-analysis to assess \nthe effectiveness of laparoscopic uterosacral nerve ablation in the treatment of chronic pelvic \npain. Hum Reprod Update doi: 10.1093/humupd/dmq031\nBibliography\n\n190\nPsychological therapies for chronic pelvic pain: systematic review of randomised con-\ntrolled trials\nChampaneria R, Daniels JP , Raza A, Pattison HM, Khan KS. Psychological Therapies for Chron-\nic Pelvic Pain: Systematic review of Randomised Controlled Trials. Acta Obs Gynae Scand \n2012;91(3):281-6\n \nHysterectomy, endometrial destruction and Mirena\n® for heavy menstrual bleeding: a \nsystematic review and individual patient data meta-analysis\nMiddleton L, Champaneria R, Daniels J, Bhattachayra S, Cooper K, Khan KS Hysterectomy, \nendometrial destruction and Mirena® for heavy menstrual bleeding: a systematic review and \nindividual patient data meta-analysis. BMJ 2010; 341:c3929\nHysterectomy, endometrial ablation, and levonorgestrel releasing intrauterine system \n(Mirena) for treatment of heavy menstrual bleeding: cost effectiveness analysis\nRoberts TE, Tsourapas A, Middleton LJ, Champaneria R, Daniels JP , Cooper KG, Bhattacha-\nrya S, Barton PM. Hysterectomy, endometrial ablation, and levonorgestrel releasing intrauterine \nsystem (Mirena) for treatment of heavy menstrual bleeding: cost effectiveness analysis. BMJ. \n2011;342:d2202\nSecond generation endometrial ablation techniques for heavy menstrual bleeding: a \nnetwork meta-analysis.\nDaniels JP , Middleton LJ, Champaneria R, Khan KS, Cooper K, Moll BW, Bhattacharya S. Sec-\nond generation endometrial ablation techniques for heavy menstrual bleeding: a network meta-\nanalysis. BMJ. 2012 Apr 23;344:e2564\nThe measurement properties of the menorrhagia multi-attribute quality-of-life scale. a \npsychometric analysis.\nPattison H, Daniels J, Kai J, Gupta JK. BJOG 2011;118:1528–1531.\nOther publications\n1. Gupta J.K., Kai J, Middleton LJ, Pattison H, Gray R., Daniels JP , on behalf of the ECLIPSE \nCollaborative Group. Levonorgestrel intrauterine system versus medical therapy for men-\norrhagia. NEJM 2013; 368:128-137 \n2. Champaneria R, Shah L, Geoghegan J, Gupta JK, Daniels JP . Analgesic effectiveness of \ntransversus abdominis plane blocks after hysterectomy: a meta-analysis. Eur J Obstet \nGynecol Reprod Biol. 2012 Oct 4. doi: 10.1016/j.ejogrb.2012.09.012.\nBibliography\n\n191\n3. Gray JW, Milner PJ, Edwards EH, Daniels JP , Khan KS. Feasibility of using microbiology \ndiagnostic tests of moderate or high complexity at the point - of - care in a delivery suite. \nJ Obstet Gynaecol. 2012;32(5):458-60.\n4. Powell R, Pattison H, Bhoyar A, Furmston AT, Middleton LJ, Daniels JP , Ewer A. Pulse \noximetry screening for congenital heart defects in newborn infants: an evaluation of ac-\nceptability to mothers. Arch Dis Child Fetal Neonatal Ed. 2013;98: F59-63.\n5. Ewer AK, Middleton LJ, Furmston AT, Bhoyar A, Daniels JP , Thangaratinam S, Deeks JJ, \nKhan KS; PulseOx Study Group. Pulse oximetry screening for congenital heart defects in \nnewborn infants (PulseOx): a test accuracy study. Lancet. 2011 Aug 27;378(9793):785-94\n6 Bhattacharya S, Middleton L, Tsourapas A, Lee A, Champaneria R, Daniels J, Roberts T, \nHilken N, Barton P , Gray R, Khan K, Chien P , O’Donovan P , Cooper K; the International \nHeavy Menstrual Bleeding Individual Patient Data Meta-analysis Collaborative Group. \nHysterectomy, endometrial ablation and Mirena® for heavy menstrual bleeding: a sys-\ntematic review of clinical effectiveness and cost-effectiveness analysis. Health Technol \nAssess. 2011 Apr;15(19):1-25.\n7. Roberts TE, Tsourapas A, Middleton LJ, Champaneria R, Daniels JP , Cooper KG, Bhat-\ntacharya S, Barton PM. Hysterectomy, endometrial ablation, and levonorgestrel releasing \nintrauterine system (Mirena) for treatment of heavy menstrual bleeding: cost effectiveness \nanalysis. BMJ. 2011;342:d2202.\n8. Clark TJ, Samuel N, Malick S, Middleton LJ, ,Daniels JP , Gupta JK Bipolar radiofrequency \ncompared with thermal balloon endometrial ablation in the office: a randomized controlled \ntrial Obstet Gynaecol  2011;117(1):109-18 .\n9. Daniels JP , Gray J, Pattison H, Gray R, Hills RK, Khan KS Intrapartum tests for Group B \nStreptococcus: Accuracy and acceptability of screening. BJOG 2011;118(2):257-65.\n10. Kaambwa B; Bryan B; Gray J; Milner P; Daniels J; Khan KS; Roberts T. Cost-Effective-\nness Of Rapid Tests And Other Existing Strategies For Screening And Management Of \nEarly Onset Group B Streptococcus During Labour. BJOG 2010;117(13):1616-2.\n11. Champaneria R, Abedin P , Daniels J, Balogun M, Khan KS Magnetic resonance imaging \nand ultrasound scan for the diagnosis of adenomyosis: Systematic review of test accu-\nracy Acta Scand Obs Gynae 2010. 89(11):1374-84.\n12. Brown C, Morris RK, Daniels J, Khan KS, Lilford RJ, Kilby MD.Effectiveness of percutane-\nous vesico-amniotic shunting in congenital lower urinary tract obstruction: divergence in \nprior beliefs among specialist groups.Eur J Obstet Gynecol Reprod Biol. 2010;152(1):25-9.\n13. Williams A, Gill S, Varma T, Jenkinson C, Quinn N, Mitchell R, Scoot, R, Ives N, Rick C, \nDaniels J, Patel S, Wheatley K on behalf of the PD SURG Collaborative Group Deep brain \nstimulation plus best medical therapy versus best medical therapy alone for advanced \nParkinson’s disease (PD SURG trial): a randomised, open-label trial. Lancet Neurol.  2010 \n9(6):581-91.\nBibliography\n\n192\n14. Thomas KS on behalf of the UK Dermatology Clinical Trials Network’s PATCH study \ngroup. Studying a disease with no home--lessons in trial recruitment from the PATCH II \nstudy. Trials. 2010;11:22.\n15. ASTRAL Collaboration Revascularisation versus medical therapy for renal-artery stenosis. \nN Eng J Med 2009;361(20)1951-62.\n16. Daniels J; Gray J, Pattison H, Roberts T, Edwards E; Milner P et al Rapid testing for GBS \nduring labour: a test accuracy study with evaluation of acceptability and cost-effective-\nness. Health Technol Assess 2009;13(42).\n17. Geoghegan J, Daniels JP , Moore PAS, Thompson PJ, Khan KS, Gulmezoglu AM Cell \nSalvage at Caesarean Section: The need for an evidence based approach. BJOG 2009 \n116(6):743-7.\n18. Ainsworth R,  Dziedzic K, Hiller L, Daniels J, Bruton A, Broadfield J A prospective double \nblind placebo-controlled randomised trial of ultrasound in the physiotherapy treatment of \nshoulder pain Rheumatology  2007; 46: 815-820.\n19. Thangaratinam S, Daniels J, Ewer AK, Zamora J, Khan KS. The accuracy of pulse oxi-\nmetry in screening for congenital heart disease in asymptomatic newborns: A systematic \nreview. Arch Dis Child.  2007; 92:F176-F180.\n20. Kilby MD; Daniels JP; Khan K Congenital lower urinary tract obstruction: To shunt or not \nto shunt? Brit J Urol 2006; 97(1):6-8.\n21. Dziedzic K, Hill J, Lewis M, Sim J, Daniels J, Hay EM. Effectiveness of manual therapy \nor pulsed shortwave diathermy in addition to advice and exercise for neck disorders: a \npragmatic randomized controlled trial in physical therapy clinics. Arthritis Rheum 2005; \n53(2):214-222.\n22. Latthe PM, Powell RJ, Daniels J, Hills RK, Gray R, Gupta JK et al. Variation in practice \nof laparoscopic uterosacral nerve ablation: a European survey. Journal of Obstetrics and \nGynaecology 2004; 24(5):547-51.\n23. Stowe RL, Wheatley K, Clarke CE, Ives NJ, Hills RK, Williams AC et al. Surgery for Parkin-\nson’s disease: lack of reliable clinical trial evidence. J Neurol Neurosurg Psychiatry 2003; \n74(4):519-521.\n24. Clark TJ, Daniels J, Khan KS, Gupta JK. Hysterectomy with bilateral salpingo-oopho-\nrectomy: A survey of gynaecological practice in the UK. Acta Obs Gynae Scand 2001; \n80(1):62-64.\n25. Adams CE, Gray R, Daniels J, Thornley B, Philpott H, Wahlbeck K. Are UK consultants \ninterested in pragmatic trials for those with schizophrenia? Schizophrenia Research 41, \n233. 2000. \nBibliography\n\n193\nAuthors’ affiliations\nElizabeth Adey  Research Manager, Birmingham Heartlands Hospital, UK\n   Formerly, Birmingham Clinical Trials Unit, University of Birmingham, UK\nGiorgio Aimi  Gynaecologist, University of Milan, Italy\nSiladitya Bhattacharya Professor of Obstetrics and Gynaecology, University of Aberdeen, UK\nLaura Buckley  Medical Statistician, CRUK Clinical Trials Unit, University of Birmingham, UK\n   Formerly, Birmingham Clinical Trials Unit, University of Birmingham, UK\nRita Champaneria  Research Fellow, Birmingham Clinical Trials Unit, University of Birmingham, UK\nFP Chen   Gynaecologist, Keelung Chang Gung Memorial Hospital, Keeling, Taiwan\nKevin Cooper  Consultant in Obstetrics and Gynaecology, Aberdeen Royal Infirmary, UK\nMichael Gannon  Consultant in Obstetrics and Gynaecology, Midland Regional Hospital, \n   Mulligar, Republic of Ireland\nRichard Gray  Professor of Medical Statistics, Clinical Trial Service Unit, University of Oxford, UK\n   Formerly, Birmingham Clinical Trials Unit, University of Birmingham, UK\nJanesh Gupta  Professor of Obstetrics and Gynaecology, University of Birmingham, UK\nNicholas Hilken  IT Coordinator, Birmingham Clinical Trials Unit, University of Birmingham, UK\nRobert K Hills  Senior Lecturer in Translational Statistics, Department of Haematology, \n   Cardiff University, UK\n   Formerly, Birmingham Clinical Trials Unit, University of Birmingham, UK\nNeil Johnson  Associate Professor, University of Auckland, New Zealand\nKevin Jones  Consultant Gynaecologist, Great Western Hospital, Swindon, UK\nJoe Kai   Professor of Primary Care, University of Nottingham, UK\nKhalid S Khan  Professor of Women’s Health, Blizzard Institute, \n   Queen Mary University London, UK\n   Formerly, Academic Department of Obstetrics and Gynaecology, \n   University of Birmingham, UK\nAuthor’s affiliations\n\n194\nPallavi Latthe  Consultant Urogynaecologist, Birmingham Women’s Hospital, UK\nEdward Lichten  Gynaecologist, Birmingham, Michigan, USA\nRichard Lilford  Professor of Clinical Epidemiology, University of Birmingham, UK\nWM Lui   Gynaecologist, Taipei Medical University Hospital, Taiwan\nLee Middleton  Senior Medical Statistician, Birmingham Clinical Trials Unit, \n   University of Birmingham, UK\nBen Willem Mol  Professor of Obstetric, Gynaecology and Clinical Epidemiology, \n   Academisch Medisch Centrum, Amsterdam, Netherlands\nPeter O’Donovan  Professor of Medical Innovation, University of Bradford, UK\nHelen Pattison  Professor of Health Psychology, Aston University, Birmingham, UK\nAmer Raza  Consultant in Obstetrics and Gynaecology, \n   Chelsea and Westminster Hospital, London, UK\nTara Selman  Specialist Registrar, Birmingham Women’s Hospital, UK\nChristopher Sutton  Consultant Gynaecologist, Guildford Nuffield Hospital, UK\nPaolo Vercellini  Associate Professor, University of Milan, Italy\nTengbin Xiong  Research Fellow, University of East Anglia, UK\n   Formerly, Department of Public Health, University of Birmingham, UK \nAuthor’s affiliations\n\n195\nDankwoord (Acknowledgements)\nThose that know me will know that it has taken a long time and a roundabout route to get to the \nend of this thesis and the opportunity to defend it. This is the end of a journey started many years \nago, with several detours and dead-ends, but that has ultimately brought me great satisfaction. \nI am privileged to have a career that can improve the lives of women worldwide, and that brings \nme such satisfaction. There are a number of people who have helped and supported me along \nthe way, and I want to thank them all now, and hope I have not forgotten anyone.\nFirst, my supervisor Professor Khalid Khan. I sincerely believe most of the best opportunities I \nhave had in my career in clinical research have been due to Khalid. We both arrived in Birming-\nham at about the same time: he with a trial in need of trials unit support, me embarking on a \nchange of direction from laboratory research. Together we learnt the art and science of clinical \ntrial research from Professor Richard Gray (more of whom later) in the completion of the LUNA \nTrial. In the following years, Khalid has helped broaden my methodological repertoire, my network \nof collaborators and my way of thinking about science and society. I thank you for your patience, \nwisdom, generosity, general chilled-outness. More than anything else though, you gave me the \nbelief that I could finally achieve this PhD and for that I am so grateful.\nVery many thanks go to Professor Doctor Ben Willem Mol, my Dutch promoter. I was introduced \nto Ben Willem by Khalid over an expensive meal and believe this was one of the most fortuitous \nmeetings in my life. Although I have not been one of your “traditional” PhD students, I have learnt \na lot from your approach to clinical research. I am grateful for the opportunity to work in the de-\npartment and the seminars you made me present, and wish that I could have spent more time \nin Amsterdam. Your research consortium has established the Netherlands as a major player in \nwomen’s health research and Holland’s loss will be Australia’s gain.\nThe other major influence in my career has been Professor Richard Gray, formerly Director of the \nBirmingham Clinical Trials Unit (BCTU), to whom I will be forever grateful. From you, I learnt that \ntrial should be large and simple, systematic reviews are great for quickly getting to know a sub-\nject and meta-analyses are never as easy as they sound. I intend to maintain the high standard \nof trials research you established at Birmingham and create a legacy for research at Birmingham \nWomen’s Hospital.\nI would like to remember the LUNA Trial team, including Khalid, Richard, Liz Adey, Pallavi Latthe, \nTara Selman, Janesh Gupta, Robert Hills, Laura Buckley, Becky Gair, Becky Powell and Leanne \nLynch amongst others. LUNA was a great trial run on a shoe-string, compared to more recent \nDankwoord (Acknowledgements)\n\n196\nprojects, so it shows what great teamwork can do. LUNA also proves that surgeons can ran-\ndomise and surgical trials can be blinded, and undoubtedly laid the foundations for other BCTU \ngynaecological studies. I thank the clinical investigators who recruited their patients and all the \nwomen, who in looking for a solution to their pain, agree to participate.\nAnother big thank-you goes to the Heavy Menstrual Bleeding Individual Patient Data Meta-Analy-\nsis Collaborative group, including Bhatty, Khalid, Richard, Kevin Cooper, Rita Champaneria, Nick \nHilken, Lee Middleton, Tracy Roberts and all the primary study authors. Never let it be said that \nIPD meta-analyses are easy, but working together with this team made it enjoyable and interest-\ning, and started a great partnership with University of Aberdeen.\nI’d also like to acknowledge the ECLIPSE team, including Janesh Gupta, Joe Kai, Laura Gennard \nand Lisa Leighton but particularly Helen Pattison. It is important to know how to ask the right \nquestions, if we are to get the right answers, and together we pondered over the MMAS scale \nand the definition of a psychological therapy.\nI thank the Promotion Committee, who have graciously allowed me to proceed to defend my \nthesis in public and who will come from Aberdeen and Veldhoven and all points in-between to the \ndefence ceremony. My thanks go to Professor Dr P .M.M Bossuyt, Dr J.P .W.R. Roovers, Dr M.Y. \nBongers, Professor Dr R.J. Lilford, Dr W.M. Ankum and Professor Dr S. Bhattachayra.\nHuge thanks go to my paranymphs Merel Breijer, who is teaching me the etiquette of Dutch \ndefence ceremonies and Lee Middleton, who doesn’t yet know what he has let himself in for.\nI would like to mention all the staff of the Birmingham Clinical Trials Unit, who continue to work \nextremely hard to design and deliver clinical trials to the highest standard, and in particular the \npast and present members of the Women’s Health team. The BCTU is not only responsible for \nproducing much important data which is changing people’s lives, but is also a great place to \nwork. I am so proud of the work we all do.\nI really want to thank my old friends Harry, Shona, Laura, Donna and Jon who have seen me \nthrough the years and waited so patiently for the post-PhD party; Mum and Dad, who although \nwill not be there to see the ceremony, I know are suitably proud of their daughter, and last but not \nleast, my husband Pete. I simply could not have done this without your love. \nDankwoord (Acknowledgements)\n\n197\nCurriculum vitae\nName PhD student:  Jane Daniels\nPhD period:  \n 2011 -2013\nName PhD supervisor:  prof dr B.W.J. Mol, prof dr K.S. Khan\n1. PhD training\nYear Workload\n(Hours)\nGeneral courses \n- None   \nSpecific courses \n- None   \nSeminars, workshops and master classes\n- Departmental seminars at AMC\n- Internal seminars at the University of Birmingham Clinical Trials Unit, \nDepartments of Public Health and Obstetrics and Gynaecology\n2011\n2011\n2012\n1.5\n4\n4\nPresentations\n- Internal seminars at University of Birmingham Clinical Trials Unit\n- Trial specific presentations at trial collaborators’ meetings\n2011, 2012\n2011, 2012\n2\n1\n1\n2\n(Inter)national conferences\n- Joint British and European Society of Gynaecological Endoscopy, \nLondon. Invited oral presentation\n- Cochrane Colloquium, Madrid. Poster presentation\n- British Society of Gynaecological Endoscopy, Cardiff. Selected oral \npresentation\n- European Society of Gynaecological Endoscopy, Paris. Invited oral \npresentation\n- GLOW Conference, Liverpool. Invited oral presentation\n- British Society of Gynaecological Endoscopy, Brighton. Invited oral \npresentation\n2011\n2011\n2012\n2012\n2012\n2013\n30\n38\n30\n30\n22\n30\nOther\nCurriculum vitae\n\n198\n2. Teaching\nYear Workload \n(Hours/\nECTS)\nLecturing\n- Research Methods for Clinical Trials, University of Birmingham (triannual)\n- \n- MSc Public Health, University of Birmingham\n2011\n2012 2013\n2011\n2012\n3\n3\n0.5\n3\n3\nTutoring, Mentoring\n- MBChB Personal Mentor, 1st and 2nd year undergraduates (6 students) 2011\n2012\n2013\n15\n15\n5\nSupervising \n- Team leader for 12 members of staff within University of Birmingham, \nClinical Trials Unit.\nOngoing\nOther\n- Data Monitoring Committee member: EASEBACK trial 2012 3\n3. Parameters of Esteem\nYear\nGrants\n1. National Institute for Health Research, Health Technology \nAssesment Programme. The relationship between refluxing pelvic \nveins and chronic pelvic pain in women. \n2. Medical Research Council/ Department for International \nDevelopment/Wellcome Trust. AIMS: The Antibiotics in Miscarriage \nSurgery Trial.\n3. National Institute for Health Research, Health Technology \nAssesment Programme. Preventing Recurrence of Endometriosis by \nMeans of long acting Protestogen Therapy: PREEMPT Trial\n2011\n2012\n2012\nAwards and Prizes\n-   \nCurriculum vitae\n\n199","source_license":"CC0","license_restricted":false}