{"paper_id":"027e1909-1a60-48f4-a665-9ddd818ffd59","body_text":"Endometriosis is a benign gynecological disorder characterized by the presence and\ngrowth of endometriumlike tissue in sites outside the uterine cavity, primarily on\nthe pelvic peritoneum and ovaries ( Giudice &\nKao, 2004 ;  Bulun, 2009 ). It\naffects 5-10% of women of reproductive age and the main clinical features are\nchronic pelvic pain and infertility ( Gupta\n et al. , 2008 ;  de\nZiegler  et al. , 2010 ;  Carneiro  et al. , 2010 ).\nEndometriosis seemingly causes infertility by interfering with fundamental steps in\nthe reproductive process. The mechanisms of infertility associated with\nendometriosis remain controversial. Current evidence suggests that endometriosis\nadversely affects ovarian and tubal function as well as uterine receptivity\nresulting in female infertility. Endometriosis may also distort pelvic anatomy and\nproduce large ovarian masses ( Gupta  et\nal. , 2008 ;  de Ziegler\n et al. , 2010 ;  Burney\n& Giudice, 2012 ). It is not clear, however, whether the products of\nendometriosis are a cause of infertility or a para-phenomenon.\nApparently, there are three types of endometriosis: superficial endometriosis,\novarian endometrioma, and deeply infiltrating endometriosis (DIE) ( Tosti  et al. , 2015 ). DIE is\nconsidered a specific entity which has been arbitrarily defined in histological\nterms as endometriotic lesions extending more than 5mm underneath the peritoneum\n( Cornillie  et al. , 1990 ;\n Koninckx  et al. , 1991 ).\nDIE is responsible for painful symptoms, whose severity has been strongly correlated\nwith the anatomical location of the DIE lesions ( Ávila  et al. , 2016 ).\nPresent treatment options of endometriosis-associated infertility include surgery,\nsuperovulation with intrauterine insemination, and  in vitro \nfertilization (IVF) ( de Ziegler  et\nal. , 2010 ;  American Society\nfor Reproductive Medicine (ASRM), 2012 ;  Johnson & Hummelshoj, 2013 ). Although many questions remain\nunanswered, there is evidence to support the use of laparoscopic surgery to improve\nfertility. ( Vercellini  et al. ,\n2009 ;  Berlanda  et al. ,\n2013 ).\nIn addition, some studies suggest that the presence of endometriosis per se may\nadversely affect pregnancy and neonatal outcomes ( Brosens  et al. , 2012 ;  Vigano  et al. , 2015 ;  Jacques  et al. , 2016 ), while a recent systematic review\nstated that complications of endometriosis during pregnancy were rare and there was\nno evidence that the disease had a major detrimental effect on pregnancy outcomes\n( Leone Roberti Maggiore  et al. ,\n2016 ). Women with incomplete surgical removal of posterior DIE may\nnevertheless present with a higher of risk complications during pregnancy and\ndelivery. ( Exacoustos  et al. ,\n2016 ).\nThe role of surgical treatment in infertile women with endometriosis remains elusive\n( Vercellini  et al. ,\n2009 ;  Douay-Hauser  et al. ,\n2011 ;  Berlanda  et al. ,\n2013 ). With the exception of peritoneal disease, randomized trials have\nnot looked into the effects of surgery in subfertile women with endometriosis (Dufy\n et al. , 2014). Therefore, it has not been possible to define\nthe absolute benefit of surgery for ovarian and rectovaginal lesions ( Vercellini  et al. , 2012 ;  Berlanda  et al. , 2013 ). The\ndilemma remains when we are faced with an infertile woman suffering from DIE: should\nthey be offered first-line IVF treatment or surgery?\n\nThis paper aimed to review the current literature on the effect of surgery for DIE on\nfertility and IVF outcomes. A search was carried out on the Cochrane Library (July\n2016) and PUBMED (1966 to July 2016) for relevant papers written in English,\nItalian, Spanish or French, which were the languages the authors could read.\nCombinations of medical subject heading terms including \"deep infiltrating\nendometriosis and assisted reproductive technologies,\" \"deep infiltrating\nendometriosis and infertility,\" \" in vitro  fertilization,\"\n\"pregnancy,\" \"intestinal endometriosis,\" and \"urinary endometriosis\" were used. The\nauthors also looked for systematic reviews and Guidelines from the European Society\nof Human Reproduction and Embryology (ESHRE) and the American Society for\nReproductive Medicine (ASRM), and books. The purpose was to find evidence to answer\nthe following clinical questions: how should DIE be diagnosed in infertile women?\nHow might DIE affect fertility and pregnancy? What are the possible benefits of\nsurgery for DIE before IVF?\n\nThe search yielded 1,206 papers. The authors read the abstracts and excluded\nirrelevant papers and repeated titles appearing in more than one of the search\ncombinations used. The full-text papers in the resulting list were obtained and\nread. The final list of relevant publications contained 60 papers.\nComprehensive clinical history is useful to identify patients at risk for\nendometriosis, although establishing the diagnosis of the disease based solely\non risk factors may be misleading, as a large portion of women with\nendometriosis remain completely asymptomatic ( Ballard  et al. , 2010 ;  Carneiro  et al. , 2013 ).\nFinding women at risk may help identify the individuals who might benefit from\ndiagnostic laparoscopy. Clinical history and symptom reports alone cannot be\nrelied on to screen women with chronic pelvic pain. Preoperative questionnaires\nexploring four clinical symptoms of women with DIE associated with endometriomas\nmay help identify high-risk groups ( Lafay Pillet\n et al. , 2014 ). Individuals categorized in these\ngroups should be referred to specialized centers for thorough assessment ( Carneiro  et al. , 2013 ;\n Dunselman  et al. ,\n2014 ). Early diagnosis of DIE reduces care costs and allows the use\nof effective medical and surgical treatments to manage symptoms and improve the\nlong-term outcome for patients. DIE should be considered as a diagnostic\npossibility in women with painful nodules/pain in areas of the rectovaginal wall\nor with visible nodules in the posterior vaginal fornix ( Dunselman  et al. , 2014 ).\nUltrasound has only recently been introduced in the diagnosis of DIE. Several\nstudies provide enough evidence to show that transvaginal ultrasound (TVUS) is\nnot only useful, but a strategic tool in the preoperative mapping of lesions and\nin surgery planning ( Ferreira & Carneiro,\n2010 ;  Dunselman  et\nal. , 2014 ;  Exacoustos\n et al. , 2016 ).\nAs it is a noninvasive, readily available test in most centers, TVUS plays a\npivotal role in the assessment of women with endometriosis along with bimanual\npelvic examination. ( Abrão  et\nal. , 2007 ;  Hudelist\n et al. , 2011 ;). Studies revealed that TVUS, when\nperformed by skilled professionals, clearly enhances diagnostic accuracy,\nespecially in patients with ovarian endometriomas or DIE involving the\nuterosacral ligaments, bladder or rectosigmoid junction; TVUS appears to be\nequally efficient in cases of DIE affecting the vagina and the pouch of Douglas\n( Hudelist  et al. ,\n2011 ; Guerriero  et al. , 2015;  Reid & Condous, 2017 ).\nNisenblat  et al.  (2016) \npublished a Cochrane Review in an attempt to establish the role of imaging in\nthe diagnosis of endometriosis. The authors looked into 49 studies: ten\ninvolving endometriomas, 15 on DIE, and 33 looking at endometriosis at specific\nanatomical sites. Unfortunately, the majority of the studies were of poor\nmethodological quality and none of the imaging methods studied was able to\nreplace surgery in the diagnosis of endometriosis. The ESHRE 2014 Endometriosis\nGuidelines states that imaging may help define the extent of involvement in\nwomen with DIE. In addition to detecting endometriomas, magnetic resonance\nimaging (MRI) is an option when rectosigmoid endometriosis is considered.\nClinical practice still lacks accurate noninvasive diagnostic tests for\nendometriosis. A recent Cochrane review ( Nisenblat  et al. , 2016 ) evaluated possible\nnoninvasive imaging tests and biomarkers for the diagnosis of endometriosis, but\nthe authors were unable to find clinically useful biomarkers, and the evidence\navailable was either insufficient or of poor-quality. Therefore, laparoscopy is\nstill the gold standard for the diagnosis of endometriosis and the use of\nnoninvasive tests either alone or in combination should only be considered in\nresearch settings.\nClinicians should therefore bear in mind that comprehensive clinical history is\nuseful to find patients at risk for endometriosis, although establishing the\ndiagnosis of the disease based solely on risk factors might be misleading, as a\nsignificant portion of women with endometriosis remain completely asymptomatic\n( Carneiro  et al. ,\n2013 ).\nDespite its low sensitivity and specificity, vaginal examination and evaluation\nof specific symptoms should not be completely ruled out in the basic diagnosis\nof endometriosis or in the planning of further therapeutic interventions ( Carneiro  et al. , 2013 ;\n Ávila  et al. ,\n2016 ).\nTVUS is a reproducible method used in the assessment of pelvic endometriosis\nseverity that offers good agreement with laparoscopy findings. Other imaging\ntechniques such as MRI are suitable for the diagnosis of endometriosis, but it\nis our belief that TVUS is the preferred initial imaging mode, since it is\nreadily available in most centers and offers an easy access, affordable and\naccurate means of diagnosing patients with ovarian endometriomas and DIE ( Benacerraf & Groszmann, 2012 ;  Dunselman  et al. ,\n2014 ).\nThe decision to perform surgery for deep endometriosis is mainly clinical. TVUS\nand other imaging techniques such as MRI may be useful in the preoperative\nestimation of lesion size and lateral extension, and play a vital role in\nsurgical planning and choice of approach. It remains unclear, however, the\nextent to which preoperative ultrasonography or MRI should influence the\ndecision to perform surgery, or the choice of procedure to treat deep\nendometriosis ( Carneiro  et al. ,\n2013 ).\nAlthough DIE has been frequently associated with infertility, there is little\nevidence connecting the disease and infertility. Studies suggest that\ninfertility in women with DIE is probably related to the strong link between DIE\nand adhesions, superficial endometriotic implants, ovarian endometriomas, and\nadenomyosis ( Somigliana & Garcia-Velasco,\n2015 ).\nSeveral anomalies are thought to contribute to DIE-related infertility: hormonal\nfunction (estrogen and progesterone receptors) and immunological factors, such\nas peritoneal macrophages, natural killer cells, and lymphocytes ( Gupta  et al. , 2008 ;  de Ziegler  et al. , 2010 ;\n Burney & Giudice, 2012 ).\nUnfortunately, the relationship between DIE and infertility is rather complex and\na final conclusion on the matter yet to be produced ( Somigliana & Garcia-Velasco, 2015 ). When considering\ninfertility in women with DIE, clinicians must bear in mind a variety of\nconfounding factors that might interfere with the interpretation of studies.\nFirstly, other types of endometriosis, such as peritoneal and ovarian, may\naccompany DIE and affect fertility on their own ( de Ziegler  et al. , 2010 ). Isolated DIE is rarely\nencountered, whereas associations with other forms of endometriosis are\nfrequently seen ( Somigliana  et\nal. , 2004 ). Secondly, superficial peritoneal lesions may\nproduce inflammatory cytokines and chemokines, and thus produce, in an altered\nhormonal milieu, increased oxidative stress and impaired sperm and tubal\nfunction ( Gupta  et al. ,\n2008 ;  de Ziegler  et\nal. , 2010 ). In addition, endometriomas may interfere\nwith folliculogenesis and result in poor oocyte and embryo quality and impair\novarian response and pregnancy rates in IVF ( Ballester  et al. , 2012 ;  Yang  et al. , 2015 ). Adenomyosis, a finding\nmore frequent in women with DIE than other forms of endometriosis, might also\nsignificantly reduce (68%) the likelihood of pregnancy in women attempting\nconception after surgery for rectovaginal or colorectal endometriosis ( Di Donato  et al. , 2014 ;\n Vercellini  et al. ,\n2014 ). And last but not least, published data suggests the\nendometrium itself harbors several anomalies in women with endometriosis, which\nresult in reduced endometrial receptivity and decreased implantation rates\n( Gupta  et al. , 2008 ;\n de Ziegler  et al.  ,\n2010 ).\nDIE is a rather heterogeneous disease. Analysis of the anatomical location of the\nlesions revealed a multifocal pattern, as 61% of the women with DIE had more\nthan one site simultaneously affected ( Ávila  et al. , 2016 ). The multifocal\ndistribution pattern observed in women with DIE makes it more difficult to study\npossible relationships between pain and anatomical location, or to establish its\nrole on fertility ( Ávila  et\nal. , 2016 ). Besides, there is a paucity of data from\nrandomized controlled trials, as the available published studies are case series\nor uncontrolled studies with many confounding factors such as use of ART and\npresence of other infertility factors.\nNatural fecundity in women with endometriosis is rarely evaluated, and except for\nperitoneal disease, no randomized trials have been carried out to assess the\neffect of surgery in this setting ( Somigliana\n& Garcia-Velasco, 2015 ).\nBrown & Farquhar (2014)  compiled\nevidence from 17 Cochrane Systematic Reviews on treatment options for women with\npain or subfertility associated with endometriosis using live birth, clinical\npregnancy, ongoing pregnancy, miscarriage and adverse events as primary\noutcomes. Seven reviews concerned infertility: two presented ART-related\noutcomes ( Sallam  et al. ,\n2006  and  Benschop  et\nal. , 2010 ), while the remaining described spontaneous\npregnancy. Post-surgical medical treatment resulted in no benefits in terms of\npregnancy rates, but three months of treatment with GnRH agonists improved\npregnancy rates in women with endometriosis undergoing IVF. Excisional surgery\nresulted in better spontaneous pregnancy rates in the 9-12 month period after\nsurgery when compared to ablative surgery. Laparoscopic surgery improved live\nbirth and pregnancy rates when compared to diagnostic laparoscopy alone. Medical\ntreatment apparently did not improve clinical pregnancy rates.\nDuffy  et al.  (2014)  found\nthat laparoscopic surgery was associated with an increased live birth or ongoing\npregnancy and clinical pregnancy rates in comparison to diagnostic laparoscopy.\nNo solid conclusions of safety were drawn, as there was insufficient evidence on\nadverse events.\nThe two randomized controlled trials reported by  Hart  et al.  (2008)  suggested a benefit of\nexcisional surgery over drainage or ablation of endometriomata to achieve\npregnancy.  Benschop  et al. \n(2010)  evaluated surgical interventions for endometrioma before IVF.\nThe authors did not find evidence of a difference in clinical pregnancy rates\nbetween surgery (aspiration or cystectomy) for endometrioma prior to ART and\nexpectant management.\nWhile some advocate complete surgical removal of endometriotic lesions to improve\nfertility ( Daraï  et al. ,\n2005 ;  Ferrero  et\nal. , 2009 ), others affirm that extensive surgery for\nperitoneal endometriosis and DIE in infertile women does not improve global\nfertility prognosis and may be associated with higher complication rates ( Vercellini  et al. , 2006 ;\n Douay-Hauser  et al. ,\n2011 ;  Vercellini  et\nal. , 2012 ).  Vercellini\n et al.  (2012)  pinpointed in their literature\nreview that women should be carefully counseled on the real chances of getting\npregnant after surgery. The authors found that pregnancy rates decreased by 15%\nin individuals who sought spontaneous conception after surgery versus women\noffered IVF (39% vs. 24%). Apparently, time to conception after surgery is also\na matter to consider, as delays have been associated with lower pregnancy and\nhigher relapse rates ( Somigliana  et\nal. , 2010 ).\nIn short, the effect of surgery on the fertility of women with DIE remains\nunanswered due to the heterogeneous nature of the disease and the lack of\nadequate trials with enough power and follow-up to study the matter. Surgery\ncannot be recommended when the main goal is to treat infertility, as the\nevidence to support such approach is still scant. Decisions should be tailored\naccording to the individual needs of each woman after they are provided with\ninformation on the potential benefits, harm, and costs of each treatment\nalternative ( Vercellini, 2015 ).\nAccording to the guidelines of the European Society of Human Reproduction and\nEmbryology (ESHRE) ( Dunselman  et\nal. , 2014 ), there is limited evidence for performing\nsurgery with the sole objective of increasing live birth rates. IVF is\nrecommended in cases of infertility associated with endometriosis, if pelvic\nanatomy is distorted and tubal function impaired, or in cases of male factor\ninfertility and/or other treatments have failed. The main purpose of surgery for\nwomen suffering from endometriosis-related infertility ideally revolves around\nthe restoration of normal pelvic anatomical relationships and preservation of\nthe function of pelvic organs.\nBianchi  et al.  (2009) \npublished the only prospective study available to date. The authors aimed at\ncomparing IVF results in women with DIE-associated infertility submitted to\nextensive laparoscopic excision of endometriosis before IVF to subjects not\noperated on before IVF. This prospective cohort study included 179 women divided\ninto two groups: IVF only (n = 105) and surgical resection of DIE lesions before\nIVF (n=64). The odds of achieving pregnancy were 2.45 times higher in the group\nsubmitted to surgical excision before IVF.\nDIE has been blamed for lowering the pregnancy rates of IVF cycles.  Ballester  et al.  (2012) \ncarried out a retrospective study and looked at 103 women with endometriomas (n\n= 30) and with endometriomas associated with DIE (n = 73). When associated with\nendometrioma, DIE adversely affected cumulative pregnancy rates (82.5% vs.\n69.4%). The authors recommended surgery when pregnancy was not achieved after\nthree attempts at IVF.  Centini  et\nal.  (2016)  also reported that the surgical removal of\nmultiple lesions increased pregnancy and live birth rates both spontaneously and\nafter IVF. On the other hand,  Capelle  et\nal.  (2015)  considered that surgery for DIE before IVF\ndid not result in improved pregnancy and birth rates. Interestingly, infertile\nwomen with DIE trying to have a second child and who had had surgery had high\nlive birth rates (78%) and a spontaneous pregnancy rate of 54% ( Boujenah  et al. ,\n2016 ).\nThe impact of surgery on IVF results remains controversial. While women with\nadvanced-stage endometriosis submitted to surgery before IVF did not respond as\nwell to gonadotropins when compared to women with tubal-factor infertility, the\nimplantation, pregnancy, and delivery rates were nonetheless similar ( Matalliotakis  et al. ,\n2007 ). Complete elimination of all visible endometriotic lesions in women\nwith minimal and mild endometriosis undergoing IVF may result in shorter time to\npregnancy and higher live birth rates ( Opøien  et al. , 2011 ). Other authors have\nsuggested that IVF pregnancy rates of women with DIE may be significantly\nimproved after extensive laparoscopic excision of DIE ( Bianchi  et al. , 2009 ). A combination of\nsurgery with IVF has been suggested as a more effective approach in\nendometriosis-associated infertility ( Coccia\n et al. , 2008 ;  Barri  et al. , 2010 ;  Ballester  et al. , 2012 ;  Capelle  et al. , 2015 ). With the exception\nof patients with endometrioma, infertile women with various stages of\nendometriosis have enjoyed the same success rates with IVF as patients with\ntubal factor infertility ( Opøien\n et al. , 2011 ).\nThe effect of surgery in this setting remains controversial. The available\npublished studies are observational and are not large enough to allow for any\ndefinitive conclusions. Except for peritoneal disease, no randomized trials have\nbeen published to determine the effect of surgery in subfertile women with\nendometriosis ( Brown & Farquhar,\n2014 ). It remains therefore impossible to define the absolute benefit of\nsurgery for ovarian and rectovaginal lesions. The decision to undergo surgery\nfor endometriosis-associated subfertility must be thoroughly debated with the\npatient. Detailed information weighing risks and benefits, and other variables\nsuch as presence of pain, large or complex adnexal masses, bowel or ureteral\nstenosis, and coexisting infertility factors, must be discussed. Specifically in\ncases of recurrent endometriosis, IVF should generally be the first option\n( Vercellini  et al. ,\n2012 ;  ASRM, 2012 ;  Berlanda  et al. , 2013 ).\n\nThe role of surgery in the treatment of infertile women with DIE remains a matter of\nintense debate. Available evidence is poor, as it amounts mostly from case series,\nwhich may bias possible conclusions. Therefore, a clear-cut conclusion cannot be\ndrawn. Women with DIE should be counseled individually taking into consideration\nseveral factors such as presence of pelvic pain and other symptoms, age, lesion\nlocation, previous treatments (surgery and ART), as well as possible pregnancy\ncomplications. In this scenario, management by a multidisciplinary endometriosis\nteam is a key factor in achieving successful outcomes.","source_license":"CC0","license_restricted":false}