UvA-DARE (Digital Academic Repository)

Studies on the effectiveness of endoscopic in reproductive

Bosteels, J.J.A.

Publication date 2014 Document Version Final published version

Link to publication

Citation for published version (APA): Bosteels, J. J. A. (2014). Studies on the effectiveness of endoscopic surgery in .

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl) Download date:28 Sep 2021 UNIVERSITY OF AMSTERDAM DISSERTATION

Jan Bosteels Studies on The Efffectiveness of Endoscopic Surgery in Reproductive Medicine

STUDIES ON THE EFFECTIVENESS OF ENDOSCOPIC SURGERY IN REPRODUCTIVE MEDICINE

Jan Bosteels

Studies on the effectiveness of endoscopic surgery in reproductive medicine Thesis, University of Amsterdam, the Netherlands

Paranimfen: Wim Laurens, Philippe Delattin

Lay out: Jan Bosteels Cover design: Coördesign, Leiden Cover illustration: ’ICSI revisited: Independent and Critically Summarized Investigation’

ISBN 978 90 5629 755 8 NUR 870

© J. Bosteels, 2014

All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of both the copyright owner and the author of the book.

STUDIES ON THE EFFECTIVENESS OF ENDOSCOPIC SURGERY IN REPRODUCTIVE MEDICINE

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. D.C. van den Boom ten overstaan van een door het college voor promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op dinsdag 2 september 2014, te 14:00 uur

door

Jan Jeanne Alfons Bosteels

geboren te Dendermonde, België

Promotiecommissie

Promotores: Prof. dr. B.W.J.Mol Prof. dr. F.J.M. Broekmans

Co-promotor: Dr. J.C. Kasius

Overige leden: Prof. dr. F. van der Veen Prof. dr. P.M.M. Bossuyt Prof. dr. P. de Sutter Dr. W.M. Ankum Dr. A. Timmermans

Faculteit der Geneeskunde

“Wie beweert dat iets niet kan, mag degene die ermee bezig is absoluut niet storen.” Romeins gezegde

“Success is not final, failure is not fatal: it is the courage to continue that counts.” Winston Leonard Spencer Churchill

Aan mijn familie, mijn vrienden, mijn collega’s en leermeesters, To all those of the past, the present and the future…

TABLE OF CONTENTS

Chapter 1 Introduction and outline of thesis 9

Chapter 2 The position of 21 in current fertility practice. Hum Reprod Update 2007; 13(5): 477-485

Chapter 3 The effectiveness of hysteroscopy for 49 improving pregnancy rates in subfertile women: a systematic review. Hum Reprod Update 2010; 16(1): 1-11

Chapter 4 Hysteroscopy for treating subfertility 79 associated with suspected major uterine cavity abnormalities. Cochrane Database Syst Rev 2013, Issue 1 Art. No.: CD009461

Chapter 5 Anti-adhesion barrier gels following 115 operative hysteroscopy for treating female subfertility. A systematic review and meta-analysis. Gyn Surg 2014; 11:113-127

Chapter 6 Anti-adhesion following operative 149 hysteroscopy for treating female subfertility. Cochrane Database Syst Rev 2014, Issue 5 Art. No.: CD011110

Chapter 7 Has there been any progress in evidence 167 based practice in ? F V V in Ob/ Gyn 2010; 2(4): 232-252 F V V in Ob/ Gyn 2011; 3(4): 238-244

Chapter 8 Thesis summary and discussion 191 Samenvatting proefschrift en discussie 209

7

Appendices Abbreviations 237 Co- authors and affiliations 239 List of publications 241 PhD Portfolio 245 Dankwoord 249 Curriculum Vitae 253

8

Introduction and outline of thesis

“And while he lives, He wields the boasted prize Whose value all can feel, the weak, the wise, Displays in triumph his distinguish'd boon, The solid honours of the Wooden Spoon”

—"The Wooden Spoon" from the Cambridge Tart (1823) Photo scanned from H P Stokes, Ceremonies of the University of Cambridge, Cambridge University Press 1927. Believed to be in the public domain: the photograph was originally published in the UK in 1909 and reprinted in Stokes' book in 1927.

9

Introduction

INTRODUCTION

Each year at St. John’s College of the University of Cambridge the students at the Mathematics faculty awarded a wooden spoon to their fellow student with the lowest marks in the mathematical tripos exams who still earned a third-class degree. It was given for the last time in 1909 to Cuthbert Lempriere Holthouse, who can be seen in the photograph on the previous page. A wooden spoon is a mock or a real award, given to an individual or a team which has come last in a competition. In 1979 the Scottish and epidemiologist Archibald Leman Cochrane (Kirklands, 12 January 1909-Dorset, 18 June 1988) awarded a figurative wooden spoon to for not having used the design of the randomized controlled trial (RCT) to assess the effectiveness of interventions related to pregnancy and childbirth. A silver spoon was given to Phtisiology (the study of tuberculosis) for the landmark clinical research conducted by Sir Austin Bradford Hill on the use of streptomycin in patients with tuberculosis (Cochrane, 1979). A randomized controlled trial is an experiment in which two or more interventions are compared by being randomly allocated to participants as opposed to an observational study where the investigators simply observe the course of events and do not seek to intervene (Glossary of Cochrane terms). Some time before, ‘Archie’ Cochrane had already criticized the medical profession in general by stating: “we have not organized a critical summary, by specialty or , updated periodically, of all relevant randomized controlled trials” (Cochrane, 1979). Initially perinatal medicine was the first specialty to take up the glove of Cochrane’s provocative challenge by organizing the Oxford Database of Perinatal Trials (Chalmers et al, 1986). The first systematic review of the effectiveness of fertility treatment was published in 1993, hallmarking the introduction of evidence-based practice in the field of reproductive medicine (Vandekerckhove et al, 1993). Evidence-based medicine is the conscientious, explicit, and judicious use of the current best evidence in making decisions about the care of individual patients (Sackett et al, 1996). The practice of evidence-based medicine (EBM) stands for the integration of individual clinical expertise and the patient’s unique values with the best available evidence from systematic research (Strauss et al, 2005). The essentials to put EBM into practice can be summarized in five steps: first of all, to ask the right answerable question; secondly, to find the best level of evidence available; thirdly, to appraise critically the evidence for risk of bias, clinical relevance and applicability; fourthly, to implement the results of the appraisal in every day clinical practice and fifthly to evaluate the

11 Chapter 1

changes in practice (Farquhar et al, 2006). The highest level of evidence is derived from well written critically appraised systematic reviews of RCTs. Why should clinicians expose their clinical practice to the scrutiny of the RCT (Johnson et al, 2008)? The answer is simple: this study design offers the potential for the least biased measurement of the effectiveness of an intervention. Effectiveness is the extent to which an intervention, when used under ordinary circumstances, does what it is intended to do (Glossary of Cochrane terms). Trials that assess effectiveness are called pragmatic or management trials. A trial may have a slightly different purpose: it may aim to measure the extent to which an intervention produces a benefit for fully co-operative participants under ideal conditions. This design is called an explanatory trial; it aims to measure the efficacy of an intervention (Glossary of Cochrane terms). Observational studies are appropriate for detecting rare events e.g. surgical complications but can be misleading if used in searching for treatment benefits that are small to moderate as those observed in health care interventions. The observation of a statistically significant association between an intervention and its effect can be explained by the following mechanisms: chance, bias/confounding or causality (Elwood, 2007). Bias is a systematic error or deviation in results or inferences from the truth as opposed to imprecision, which is random sampling error (Glossary of Cochrane terms). Many types of bias have been described in studies of the effects of health care interventions; they can be classified into information bias, selection bias and confounding (Groenwold, 2013). The risk of bias can be minimized by using a randomized controlled trial as a study design to prevent confounding, by blinding participants and researchers to prevent information bias and by doing the correct analysis taking into account selective loss to follow-up, thereby preventing selection bias. Assessing the risk of bias is very important in the critical appraisal of the effectiveness of an intervention. Different possible scenarios have to be examined. Could there have been a selective error in the measurement of the exposure, the outcome or both? Have all confounders been carefully considered? How has the selection of the comparison groups been determined? Both the likelihood and the extent of bias have to be assessed; measuring the extent of bias is possible by doing several sensitivity analyses (Groenwold, 2013). The higher the risk of bias, the lesser the confidence in the estimation of the true treatment effect even when the confidence intervals are narrow. Systematic reviews comparing observational studies with RCTs of comparable interventions for comparable conditions in identical study populations have demonstrated that the former were clearly unreliable and consistently overestimated the treatment effect (Britton et al, 1998; Kunz et al, 2001). In surgery a non-

12 Introduction evidence-based approach has been the rule by tradition (Johnson et al, 2008). Some authorities consider that a RCT may not be the most appropriate design to measure the true value of a surgical intervention (Black, 1999). By their view factors such as training and expertise of the surgeon are not easy to standardize precluding the possibility to measure the effectiveness of a surgical intervention by a randomized controlled study. In other words, differences in outcomes might be caused by differences in skills or techniques among the participating surgeons rather than to a real treatment effect of the intervention under study. Several authors have reported data on large geographical variations in the rates of specific surgical interventions, e.g. prostatectomy/ hysterectomy/ hemorrhoidectomy (Mc Pherson et al, 1982), tonsillectomy (Mc Pherson et al, 1982; American Child Health Association, 1934) or coronary surgery (Brook et al, 1988). This illustrates that clinical practice research in surgery is determined by differences in the clinical judgment (or beliefs?) of surgeons. The appraisal of the best available evidence based on a rigorously scientific methodology as in basic research may be a stronger base for determining the efficacy and effectiveness of therapeutic interventions (Sackett et al, 1996; Rosenberg and Donald, 1995). There has been a growing recognition that pooling knowledge from high- quality studies provides the best available evidence (Evans et al, 2006). Moreover concerns about spending substantial financial resources for ineffective treatments at the ‘cost’ of effective interventions have put health-care professionals under increasing pressure to work more cost- effectively (Cochrane, 1972; Woolf et al, 1999; Agrawal et al, 2013). This poses a challenge for the clinician who has to struggle with the ever increasing volume and complexity of the published biomedical knowledge (Sackett et al, 1996). Reading a paper on a single study is clearly no longer sufficient to keep clinical knowledge up to date; most single papers do not represent the best available evidence on a particular subject (Ioannidis, 2005). The increasing biomedical knowledge and the need to summarize and critically appraise its findings have led to the development of systematic reviews, to increase the likelihood of the most reliable information being available to guide clinical decision making (Haynes et al, 2006). The Cochrane Collaboration is a global independent network of health practitioners, researchers, patient advocates and others, responding to the challenge of making the vast amounts of evidence generated through research useful for informing decisions about health. Since its formation in 1993, the central mission of this not-for-profit organization with almost 27,000 collaborators from over 120 countries has been to publish systematic reviews of the literature to aid clinicians, patients and

13 Chapter 1

policy makers taking health-care decisions based on the best available evidence from high-quality studies (Farquhar et al, 2013). Given the restricted financial budget of most countries to spend money for care, the question of the effectiveness of treatments is very relevant for the field of subfertility: the costs for treatment with assisted reproductive techniques/technology (ART) are high and the success rates are moderate, at best only 20% live birth rate per started cycle (Macaldowie et al, 2012). Assisted reproductive technology (ART) comprises all treatments or procedures of the in vitro handling of both human oocytes and sperm or of embryos for the purpose of establishing a pregnancy. This includes among others in vitro fertilization and embryo transfer (IVF) (Zegers-Hochschild et al, 2009). It is estimated that >5,000,000 babies have been born around the globe with ART since the birth of Louise Brown in 1978 (ESHRE 2012). Some 1,500,000 ART cycles are performed annually worldwide with an estimated 350,000 babies born. Belgium has a top ranking for consumption of ART; in 2011 the utilization was 2,767 ART cycles per million of the population (cpm) based on data from the Belgian Register for Assisted Procreation (BELRAP) registry (Report 2011 of the College of in Reproductive Medicine) and the web application be.STAT of the Belgian Federal Government. The utilization of ART in Belgium is substantially higher compared with the Netherlands (1,290 cpm in 2008) (Berg Brigham et al, 2013). In Belgium 5% of all babies born are conceived after ART. The total cost of ART in Belgium for the year 2008 has been estimated 99,844,000 € assuming that the direct unsubsidized cost of one ART treatment cycle is 4,000 € (IST).The delivery rate from ART treatment is 22% per oocyte pick-up and 27% cumulative from a single started treatment cycle based on an analysis of world data for 2007 (ESHRE-ART fact sheet). In this thesis we aim to study the effectiveness of reproductive surgery for improving the reproductive outcome in subfertile women by a critical appraisal of the best available evidence using the rigorously scientific methodology developed for writing systematic reviews. A first research question is whether the shift away from reproductive surgery favoring ART for tubal factor infertility is supported by high-quality evidence. A second research question concerns the effectiveness of hysteroscopy in the treatment of female subfertility. A third part of our research aims to investigate if there has been any progress in the evidence-based practice of reproductive surgery. In 1996 the CONSORT statement was published (Begg et al, 1996; Selman et al, 2008). The acronym stands for consolidated standards of reporting trials. Started in 1993 in Ottawa, Canada at a meeting of medical journal editors, clinical trialists,

14 Introduction epidemiologists and methodologists as an initiative to develop a new scale to assess the quality of RCT reports, the CONSORT statement refers to a checklist of 25 items aiming to improve the standards of reporting of RCTs. We aim to study the effects of the publication of the CONSORT statement on the quality of published RCTs on reproductive surgical interventions in subfertile women.

15 Chapter 1

OUTLINE OF THIS THESIS

Chapter 1 predefines the three research questions in the introduction of the present thesis: 1. To examine if there is an evidence-based fundament for the shift from reproductive surgery to ART in women with tubal infertility 2. To appraise the evidence on the effectiveness of hysteroscopy in the treatment of female subfertility 3. To investigate if there has been any progress in the evidence- based practice of reproductive surgery.

In chapter 2 we examine the available evidence on the use of laparoscopy in current fertility practice by a combined narrative and systematic review of the literature. The alternative diagnostic procedures for the evaluation of tubal patency are briefly discussed. We aim to appraise the position of laparoscopy for the diagnosis and treatment of the general population attending the fertility clinic.

In chapter 3 we summarize the findings of a systematic review of the literature on the effectiveness of hysteroscopy for improving pregnancy rates in subfertile women. This review aims to examine the validity of the statements of the Royal College of Obstetricians and Gynecologists (RCOG, 2004) and the European Society for Human Reproduction and Embryology (Crosignani et al, 2000) that hysteroscopy should not be done in all subfertile women but only when indicated.

In chapter 4 the findings of a Cochrane systematic review on the effectiveness of hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities are discussed. Ultrasonography, saline infusion/gel instillation sonography or hysterosalpingography are routinely performed in the initial diagnostic work-up of subfertility. Their use leads to the detection of uterine cavity abnormalities such as endometrial polyps, fibroids, septa or intra-uterine adhesions in 10 to 15% of women seeking treatment for subfertility. This Cochrane review has identified several knowledge gaps that need to be addressed in the near future. This chapter illustrates an often overlooked benefit of rigorous systematic reviews: their capacity to indicate the absence of evidence.

In chapter 5 we discuss the effectiveness of using anti-adhesion barrier gels following operative hysteroscopy for treating female subfertility by a systematic review of the literature and a meta-analysis of the data from

16 Introduction some included studies. We aim to appraise the evidence for the use of any anti-adhesion barrier gel following operative hysteroscopy in subfertile women.

In chapter 6 we present a Cochrane protocol on the broader scope of several different anti-adhesion strategies following operative hysteroscopy for treating female subfertility. The various strategies proposed by observational studies include the insertion of an intrauterine device (IUD) or a Foley catheter balloon, hormone therapy, barrier gels and human amnion membrane grafting. We aim to study the effectiveness of these four anti-adhesion strategies for treating subfertile women.

In chapter 7 we aim to test the hypothesis that the publication of the CONSORT statement in 1996 has had a positive impact on the quality of the published reports of RCTs on the effectiveness of reproductive surgery by summarizing the findings of a systematic review.

In chapter 8 we aim to write a fitting end for the story of the wooden spoon. Firstly, the findings of the studies presented in this thesis are summarized. Secondly, we aim to answer all three research questions of this thesis. Thirdly, after discussing briefly the strengths and weaknesses of RCTs in reproductive surgery we will conclude by ‘random reflections’ (*) on the future of evidence-based practice in reproductive surgery: Is EBM sufficient to give all the answers needed by the women under our care? Should we direct our attention to the emerging “patient-centered” practice or rather stick with “practice-based” evidence? Or is evidence-based practice nothing more than the incense used by yet another institution professing the new dogma of a too mathematical, rigorous and possibly too narrow view on clinical practice?

(*) Reference to Cochrane AL. Effectiveness and Efficiency. Random Reflections on Health Services. London: Nuffield Provincial Hospitals Trust, 1972. (Reprinted in 1999 for Nuffield Trust by the Royal Society of Medicine Press, London).

17 Chapter 1

REFERENCES

1. Agrawal S, Taitsman J, Cassel C. Educating physicians about responsible management of finite resources. JAMA 2013; 309: 1115- 1116. 2. American Child Health Association: Physical defects: The pathway to correction. New York, American Child Health Association, 1934. 3. Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, Pitkin R, Rennie D, Schulz KF, Simel D, Stroup DF. Improving the quality of reporting of randomized controlled trials: the CONSORT statement. JAMA 1996; 276: 637-639. 4. Berg Brigham K, Cadier B, Chevreul K. The diversity of regulation and public financing of IVF in Europe and its impact on utilization. Hum Reprod 2013; 28(3): 666-675. 5. Be.STAT. http://statbel.fgov.be/nl/statistieken/webinterface/beSTAT_home/. 6. Black N. Evidence-based surgery: A passing fad? World J Surg 1999; 23: 789-793. 7. Britton A, McKee M, Black N, McPherson K, Sanderson C, Bain C. Choosing between the randomized and non-randomized studies: a systematic review. Health Technol Assess 1998; 2:13. 8. Brook RH, Kosecoff JB, Park RE, Chassin MR, Winslow CM, Hampton JR. Diagnosis and treatment of coronary disease: comparison of doctors’ attitudes in the USA and the UK. Lancet 1988; 1:750. 9. Chalmers I, Hetherington J, Newdick M, Mutch L, Grant A, Enkin M, Enkin E, Dickersin K. The Oxford database of perinatal trials: Developing a register of published reports of controlled trials. Control Clin Trials. 1986; 7:306–324. 10. Cochrane AL. Medicine for the year 2000. London: Office of Health and Economics; 1979. 1931–1971: A critical review with particular reference to the medical profession. 11. Cochrane AL. Effectiveness and Efficiency: Random reflections on Health Services. 2nd edition, London: Nuffield Provincial Hospitals Trust (published 1989), 1972. 12. Crosignani PG, Rubin BL (2000). Optimal use of infertility diagnostic tests and treatments Hum Reprod 15, 723-732. 13. Elwood JM. Critical appraisal of epidemiological studies and clinical trials. 3rd edition, Oxford University Press, Oxford 2007. 14. ESHRE/ART fact sheet. Available on http://www.eshre.eu/Guidelines-and-Legal/ART-fact-sheet.aspx.

18 Introduction

15. European Society of Human Reproduction and Embryology. The world’s number of IVF and ICSI babies has now reached a calculated total of 5 million. Press release, 2 July 2012. http://www.eshre.eu/press-room/press-releases/press-releases-eshre- 2012/5-million-babies.aspx/1606. 16. Evans I, Thornton H, Chalmers I. Testing treatments: Better research for better healthcare. Reprinted 2010. London: Pinter & Martin, 2006. 17. Farquhar C, Vail A. Pitfalls in systematic reviews. Curr Opin Obstet Gynecol 2006; 18: 433-439. 18. Farquhar C, Moore V, Bhattacharya S, Blake D, Vail A, Thomas J, Cheong Y, Showell M, Nagels H, Marjoribanks J, on behalf of the Cochrane Editorial Board of the Menstrual Disorders and Subfertility Group. Twenty years of Cochrane reviews in menstrual disorders and subfertility. Hum Reprod 2013; 28(11): 2883-2892. 19. Glossary of Cochrane terms. http://www.cochrane.org/glossary. 20. Groenwold RHH. Drie vormen van bias. Vertekening van onderzoeksresultaten en hoe dat te voorkomen. Ned Tijdschr Geneeskd. 2013; 157: A6497. 21. Haynes RB, Cotoi C, Holland J, Walters L, Wilczynski N, Jedraszewski D, McKinlay J, Parrish R, McKibbon KA. Second-order peer review of the medical literature for clinical practitioners. JAMA 2006; 295: 1801-1808. 22. Ioannidis JP. Contradicted and initially stronger effects in highly cited clinical research. JAMA 2005; 294: 218-228. 23. IST-Instituut voor Samenleving en Technologie. Fertiliteitsbehandelingen: de realiteit voorbij de technologie. P 44. 24. Johnson NP, Selman T, Zamora J, Khan KS. Gynecologic surgery from uncertainty to science: evidence-based surgery is no passing fad. Hum Reprod 2008; 23(4): 832-839. 25. Kunz R, Vist G, Oxman AD. Randomization to protect against selection bias in health care trials (Cochrane Methodology Review). Updated 22 August 2001. In: The Cochrane Library, Issue 3. Chichester, UK: John Wiley & Sons, 2004. 26. Macaldowie A, Wang YA, Chambers GM, Sullivan EA. Assisted reproductive technology in Australia and New Zealand 2010. Assisted reproduction technology series no. 16. Cat. No. PER 55. Canberra: AIHW, 2012. 27. Mc Pherson K, Wennberg JE, Hovind OB, Clifford P. Small-area variations in the use of common surgical procedures: an international comparison of New England, England and Norway. N Engl J Med 1982; 307: 1310.

19 Chapter 1

28. Report 2011 of the College of Physicians in Reproductive Medicine. http://www.belrap.be/Documents/Reports/Global/FinalReportV2_IVF 11_10JAN14.pdf. 29. Rosenberg W, Donald A. Evidence-based medicine: an approach to clinical problem-solving. BMJ 1995; 310: 1122. 30. Royal College of Obstetricians and Gynecologists Evidence-based Clinical Guidelines (2004) Guideline: Fertility assessment and treatment for people with fertility problems, RCOG website. 31. Sackett DL, Rosenberg WMC, Gray JAM, Haynes RB, Richardson WS. Evidence-based medicine: what it is and what it isn’t. BMJ 1996; 312:71. 32. Selman TJ, Johnson NP, Zamora J, Khan KS. Gynecologic surgery from uncertainty to science: evolution of randomized controlled trials. Hum Reprod 2008; 23: 827-831. 33. Strauss SE, Richardson WS, Glasziou P, Haynes BR. Evidence-based medicine. How to practice and teach EBM. 3rd edition, Elsevier Churchill Livingstone, Edinburgh 2005. 34. Vandekerckhove P, O'Donovan PA, Lilford RJ, Harada TW. Infertility treatment: from cookery to science. The epidemiology of randomized controlled trials. Br J Obstet Gynaecol 1993; 100: 1005- 1036. 35. Woolf SH, Grol R, Hutchinson A, Eccles M, Grimshaw J. Potential benefits, limitations, and harms of clinical guidelines. Br Med J 1999; 318: 527-530. 36. Zegers-Hochschild F, Adamson GD, de Mouzon J, Ishihara O, Mansour R, Nygren K, Sullivan E, Vanderpoel S; International Committee for Monitoring Assisted Reproductive Technology; World Health Organization. International Committee for Monitoring Assisted Reproductive Technology (ICMART) and the World Health Organization (WHO) revised glossary of ART terminology, 2009. Fertil Steril 2009; 92(5):1520– 1524.

20

The position of laparoscopy in current fertility practice

Jan Bosteels, Bruno van Herendael, Steven Weyers and Thomas M. D’Hooghe

Human Reproduction Update 2007; 13(5): 477-485 http://humupd.oxfordjournals.org/content/13/5/477.full.pdf+html.

21

The position of laparoscopy

ABSTRACT

Background In everyday clinical practice, it is not always clear if and when laparoscopy should be done in the fertility work-up. We aimed to analyze the available evidence on alternative diagnostic methods for detecting tuboperitoneal infertility and on the position of laparoscopy in the fertility clinic.

Methods We searched MEDLINE and The Cochrane Library by using a combination of free-text words and Medical Subject Headings (MeSH) terms for ‘laparoscopy’ and ‘infertility’. We included randomized controlled trials (RCTs) and observational studies.

Results The laparoscopic ablation or excision of minimal/mild endometriotic lesions with or without adhesiolysis increases the outcome ‘ongoing pregnancy at 20 weeks or live birth’ in women with otherwise unexplained subfertility: the odds ratio (OR) was 2.0, 95% confidence interval (CI) was 1.2 to 3.3, P=0.01, 341 women, as demonstrated by a Canadian trial. According to a smaller Italian trial there was no evidence for such a benefit (OR 0.85, 95% CI 0.32 to 2.3, P=0.75, 96 women). Laparoscopic ovarian drilling in women with clomiphene-resistant polycystic ovaries is at least as effective as gonadotrophin treatment- OR was 1.0, 95% CI 0.63 to 1.6, P=0.95, 4 studies, 304 women- but significantly lowers the multiple pregnancy rate (OR 0.13, 95% CI 0.03 to 0.59, P=0.008, 4 studies, 154 women). Doing a laparoscopy before a first treatment cycle or after several failed cycles of intrauterine insemination does not affect the pregnancy rates (OR 0.81, 95% CI 0.43 to 1.5, P=0.52, 1 study, 154 women) or the incidence of pelvic with therapeutic implications (OR 0.72, 95% CI 0.28 to 1.9, P=0.51, 1 study, 89 women). Based on a subgroup analysis of the data of a Cochrane review the laparoscopic removal of a hydrosalpinx visible on ultrasound may increase the delivery rate after in vitro fertilization (IVF)- the risk ratio (RR) was 2.4, 95% CI 1.1 to 5.3, P=0.038, 1 study, 75 women- especially when bilaterally detectable by ultrasound (RR 3.5, 95% CI 1.1 to 11, P=0.019, 1 study, 39 women).

Conclusions In some clinical settings, limited evidence is available to recommend the use of laparoscopy in current fertility practice. Additional high-quality

23 Chapter 2

randomized studies on the effectiveness of laparoscopy for treating female subfertility are needed.

24 The position of laparoscopy

INTRODUCTION

The position of laparoscopy in current fertility practice is under debate. Until recently, laparoscopy was the final procedure of the diagnostic fertility work-up, as recommended by the American Fertility Society (AFS) in 1992 and by the World Health Organization (WHO) guidelines (Rowe et al, 1993). In 1997, Glatstein et al reported that 89% of all reproductive physicians in the USA routinely performed a laparoscopy for the exploration of infertility (Glatstein et al, 1997). However, some investigators argued that diagnostic laparoscopy detected no abnormalities or pathology of questionable prognostic significance, e.g. minimal or mild endometriosis in 40 to 70% of all cases (Forman et al, 1993). Already by the mid 1990s, the ‘test’ laparoscopy was reported to be less than an ideal predictor for infertility (Collins et al, 1995). These findings convinced some authors to challenge the need for this procedure in the routine fertility work-up of all women (Fatum et al, 2002). Worldwide, diagnostic laparoscopy is increasingly bypassed by IVF. Several explanations are possible: cutting costs, faster delivery of treatment to the client or aiming to protect women from complications related to surgery or the use of general anesthesia. Disadvantages of laparoscopy include the need for general anesthesia, operating room facilities and dedicated nursing staff, patient’s anxiety and the possibility of complications related to abdominal entry or the procedure itself. In a large Finnish follow-up study, the complication rate of diagnostic laparoscopy was 6 per 10,000 procedures (Härkki-Sirén et al, 1999). However, the main advantage of a laparoscopy is the “see and treat” principle, the possibility to make a diagnosis and install a therapy at the same time as well as the opportunity to combine the laparoscopy with the hysteroscopic exploration of the uterine cavity with an endometrial biopsy, all as part of day care surgery. In this review paper, we aim to define the position of laparoscopy in current fertility practice. We conducted a literature search of MEDLINE (through Pub Med) and The Cochrane Library, using a combination of free-text words and medical subject headings (MeSH) for the key words ‘laparoscopy’ and ‘infertility’. We included RCTs as well as observational studies. The present review aims to study the effectiveness of laparoscopy for improving pregnancy rates in subfertile women; it should be considered as an intervention and not a diagnostic review. There is however a thin line between diagnostic and operative laparoscopy; therefore we will first address the question if alternative diagnostic procedures to evaluate tuboperitoneal infertility are reliable enough to replace diagnostic

25 Chapter 2

laparoscopy. Subsequently, we will discuss the position of laparoscopy in the context of treatment for ovulation induction, intrauterine insemination (IUI) and IVF based on two questions: (i) Is it necessary to perform a laparoscopy before starting these fertility treatments to detect significant tuboperitoneal pathology with therapeutic consequences and with impact on the treatment’s cumulative ongoing pregnancy rate? (ii) Is it still indicated to perform a laparoscopy after several failed treatment cycles with ovulation induction or IUI/IVF to enhance the couple’s success rate?

ALTERNATIVE PROCEDURES FOR EVALUATING TUBAL PATENCY OR THE PRESENCE OF ENDOMETRIOSIS

The first topic to be discussed is the reliability of alternative diagnostic methods for examining the presence of tuboperitoneal infertility, based on taking medical history, hysterosalpingography (HSG) and serum Chlamydia antibody testing (CAT).

Medical history and tuboperitoneal infertility The positive predictive value of history taking, based on symptoms suggestive for previous pelvic inflammatory disease (PID), a history of abnormal vaginal discharge and a previous diagnosis of a lower genital tract infection was 56, 59 and 35%, respectively, in predicting tuboperitoneal infertility (Hubacher et al, 2004). For endometriosis the predictive value of each symptom or even a combination of symptoms in predicting its presence, remains uncertain. Indeed, these symptoms have a low sensitivity and a significant proportion of women affected by endometriosis, are symptom-free (Kennedy et al, 2005).

HSG and HyCoSy The HSG enables a morphological evaluation of the uterine cavity, the Fallopian tubes and their patency. According to a meta-analysis, HSG has a reasonable specificity (83%) but a low sensitivity (65%) to document patency of the Fallopian tubes (Swart et al, 1995). The fecundability is substantially reduced in cases of bilateral occlusion and/or hydrosalpinx- OR was 0.30 (95% CI 0.13 to 0.71) - but not for a one-sided tubal occlusion or hydrosalpinx (OR 0.81, 95% CI 0.47 to 1.4) (Mol et al, 1997b). Furthermore, an HSG performed with oil-soluble media may be therapeutic in some subfertile women. According to a meta-analysis (Watson et al, 1994), a higher conception rate has been demonstrated in women undergoing an HSG with oil-soluble compared to water-soluble contrast media (OR 1.9, 95% CI 1.3 to 2.7), especially in the subgroup of women with unexplained subfertility (OR 2.7, 95% CI 1.9 to 3.8).

26 The position of laparoscopy

According to another RCT (Ogata et al, 1993), the conception rate was three times higher in subfertile women after having an HSG performed with oil-soluble media versus a control group not undergoing an HSG. However, in daily clinical practice an HSG is performed with water- soluble media to prevent potentially life-threatening allergic reactions despite the fact that RCTs demonstrating any benefit for an increased fecundability after HSG with water-soluble media compared to no flushing are lacking. The technique of HSG has several possible adverse effects. Lower abdominal pain and discomfort are commonly experienced by women undergoing HSG. Many women still remember the HSG for years afterwards as one of the most painful outpatient exams in gynecology. An HSG can induce or exacerbate PID, leading to peritonitis, pelvic abscess and very exceptionally even to death (Chuang et al, 1971). Uterine perforation and post-examination hemorrhage are a possibility. Other complications include granuloma formation and vascular intravasation. Hypersensitivity reactions to iodine exist with any of the HSG media, but allergic reactions are rare. Finally, the ionizing radiation used for HSG can be detrimental to an unsuspected early pregnancy. A multicentre RCT comparing cumulative pregnancy rates (CPR) in a group where HSG was followed by diagnostic laparoscopy within one to two months in case tubal pathology was detected or after 6 months when the HSG was normal and the woman had not yet become pregnant by that time versus a group where diagnostic laparoscopy was performed immediately, showed no significant differences in CPR at 18 months (Perquin et al, 2006). The authors question the added value of HSG performed at an early stage in the fertility work- up prior to laparoscopy and dye, although this conclusion has been criticized by other investigators (Coppus et al, 2006). The prognostic significance of HSG and laparoscopy for fertility outcome was reported in a large prospective cohort study (Mol et al, 1999). Unilateral and bilateral tubal occlusion at HSG and laparoscopy were related to treatment independent pregnancy. The adjusted fecundity rate ratio (FRR) of a one-sided tubal occlusion at HSG was 0.80, whereas two- sided tubal occlusion had a FRR of 0.49. A fecundity rate ratio expresses the probability of a spontaneous intrauterine pregnancy per time unit for women with a particular feature (e.g. one-sided or two-sided occlusion) versus women without that feature (Olsen et al, 1998). In the case of laparoscopy, the adjusted FRRs were 0.51 and 0.15, respectively, for one- sided and two-sided tubal occlusion. A laparoscopy showing two-sided occlusion after a normal or one-sided occluded HSG was found in 5% of the women and the treatment-independent conception rate in this case was

27 Chapter 2

virtually 0%. A normal laparoscopic examination after two-sided occluded HSG was found in 42% of all patients; in these cases fertility prospects were only slightly impaired with a three-year cumulative ongoing intrauterine pregnancy rate of 9%. On the other hand, fertility prospects were strongly impaired in cases where laparoscopy showed one- sided and two-sided occlusions after a two-sided occluded HSG; the adjusted FRR were 0.38 and 0.19, respectively. The authors conclude that a diagnostic laparoscopy should be done after a two-sided occluded HSG because this enables to distinguish two large groups with clinically relevant differences in future fertility prognosis. Furthermore, a diagnostic laparoscopy can be reasonably delayed after a fully normal HSG for at least 10 months due to the very low probability of only 5% that bilateral tubal occlusion may be found. Hysterosalpingo Contrast Sonography (HyCoSy) is an attractive alternative to HSG since the patients are not exposed to X-rays or iodinated contrast media. Fallopian tubal patency is assessed using transvaginal ultrasonography and a galactose micro bubble contrast medium. There were no large differences in the diagnostic performance between HyCoSy and HSG when women were randomly allocated to undergo either procedure ; direct comparison of both procedures demonstrated a poor agreement beyond chance (Dijkman et al, 2000). The authors recommend additional studies on the capacity of HyCoSy to predict fertility. A case-control study has demonstrated that allocation of patients screened as normal with HyCoSy to treatments that rely on an accurate assessment of tubal patency does not change the conception rates (Hamilton et al, 2003).

Serum CAT The presence of Chlamydia antibodies (by Chlamydia antibody testing or CAT) is indicative of a past infection with Chlamydia trachomatis, the most important cause of PID. The accuracy of serum Chlamydia antibodies in the diagnosis of tubal pathology has been critically appraised in a systematic review and meta-analysis by Mol and co-workers (Mol et al, 1997a). The discriminative capacity of Chlamydia antibody titers by means of enzyme-linked immuno sorbent assay (ELISA), microimmunofluorescence (MIF) or immunofluorescence (IF) assays in the diagnosis of any tubal pathology is comparable with that of HSG in the diagnosis of tubal occlusion or hydrosalpinx; both diagnostic tools had comparable receiver operating characteristics curves (ROC). A summary ROC curve of several diagnostic studies using ELISA or (M)IF demonstrated a better discrimination when compared with studies using immunoperoxidase assay (Mol et al, 1997a). The latter author group

28 The position of laparoscopy published results of a trial aiming to study the cost-effectiveness of HSG, laparoscopy and CAT in >2,000 subfertile couples enrolled in the Canadian Infertility Treatment Evaluation Study-CITES (Mol et al, 2001). The diagnostic work-up to detect tubal pathology in subfertile couples should, according to their findings, begin with CAT in couples with relatively good fertility prospects and immediate HSG in couples with relatively poor fertility prospects. Good fertility prospects were defined by a threshold of having a 3-year chance of conception of >14%, whereas poor fertility prospects were defined by a 3-year chance of conception of <14%. The HSG and CAT screening have no role for detecting endometriosis in subfertile women with patent Fallopian tubes. In summary, both HSG and CAT are reliable procedures as a primary screening tool for the diagnosis of tubal infertility With these data in mind, the discussion if and when a diagnostic laparoscopy should be done in a target population of subfertile women can now be focused to the following clinical settings.

LAPAROSCOPY AND THE TREATMENT OF MINIMAL/MILD ENDOMETRIOSIS

Whether or not minimal or mild endometriosis should be treated in all cases of otherwise unexplained subfertility still remains a never ending discussion. The prevalence of endometriosis in the subfertile population (20 to 68%) is higher than that in the general population of women of reproductive age (2.5 to 3.3%) according to several authors (Houston et al, 1987; Mahmood and Templeton, 1991). Moderate and severe stage endometriosis may lead to a disruption of the normal pelvic anatomy, negatively affecting the reproductive function of the internal genital organs. Minimal and mild stage endometriosis may impair fertility by a variety of hypothetical mechanisms, including toxic factors within the peritoneal fluid, impaired folliculogenesis and luteal function. The monthly fecundity rate is around 7% in r-AFS stages I–II endometriosis and the cumulative live birth rate with expectant management in endometriosis is low (Collins et al, 1995). Endometriosis r-AFS stages I and II are associated with a FRR of 0.52 (Mol et al, 1999). Although the association between minimal and mild endometriosis and subfertility may be co-incidental due to the role of chance, bias or confounding variables, many sound arguments have been presented in the literature that support a causal relationship between endometriosis and subfertility (De Hondt et al, 2006).

29 Chapter 2

According to a meta-analysis (Jacobson et al, 2004b), the ablation or excision of endometriotic lesions with or without adhesiolysis is effective compared to diagnostic laparoscopy alone for improving fertility in women with minimal or mild endometriosis; OR was 1.6 (95% CI 1.1 to 2.6, P=0.03) (Figure 1).

Figure 1 Forest plot of comparison: Laparoscopic surgical treatment versus diagnostic laparoscopy only for minimal/mild endometriosis associated subfertility. Outcome: Ongoing pregnancy at 20 weeks or live birth. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

Operative Diagnostic Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Marcoux 1997 50 172 29 169 70.8% 1.98 [1.18, 3.32] Parazzini 1999 10 51 10 45 29.2% 0.85 [0.32, 2.29]

Total (95% CI) 223 214 100.0% 1.65 [1.05, 2.60] Total events 60 39 Heterogeneity: Chi² = 2.19, df = 1 (P = 0.14); I² = 54% 0.01 0.1 1 10 100 Test for overall effect: Z = 2.16 (P = 0.03) Favours diagnostic Favours operative

This recommendation is based on a meta-analysis of two RCTs conducted in Italy (Parazzini et al, 1999) and in Canada (Marcoux et al, 1997) comparing laparoscopic ablation or excision and adhesiolysis of endometriotic lesions versus diagnostic laparoscopy alone. The fact that these two RCTs have been statistically pooled has often been criticized (De Hondt et al, 2006) because both studies present contradictory results, and by consequence should have led to different conclusions. The Canadian trial indicates a possible benefit favoring the treatment of minimal/mild endometriosis whereas the Italian study demonstrates no evidence for such a beneficial effect. Many possible explanations for this discrepancy have been put forward in the literature: the Italian study included a lower number of participants (n=96), had an unequal number of women in both comparison groups, did not include a power analysis or an outcome analysis on the level of monthly fecundity rate or CPR, and was at high risk of performance bias since a substantial number of women were treated with gonadotrophin releasing hormone (GnRH) agonists after surgery (Parazzini et al, 1999; De Hondt et al., 2006). In the Canadian study (Marcoux et al, 1997) including 341 infertile women aged 20 to 39 years with minimal or mild endometriosis, a higher cumulative probability of ongoing pregnancy after 36 weeks was observed in the surgically treated group (31%) compared with the control group treated by a diagnostic laparoscopy only (18%); the OR was 2.0 (95% CI 1.2 to 3.3, P=0.01, 341 women). A first remark concerns the clinical

30 The position of laparoscopy relevance of the laparoscopic surgical treatment of minimal/mild endometriosis; clearly the intervention did not restore normal fecundability. Indeed, the monthly fecundity rate among women treated by laparoscopic surgery (6.1%), albeit double as high compared to the control group, was still lower than expected compared to normal fertile women (20%). A second remark is that the fertility enhancing effect of the laparoscopic treatment of minimal/mild endometriosis might only be ascribed to the laparoscopic adhesiolysis. However, the authors of the Canadian RCT clearly mention that in the 284 women who did not have adhesions, the destruction of the implants also significantly increased the 36-week cumulative probability of ongoing pregnancy with a cumulative incidence ratio of 1.6 (95% CI 1.1 to 2.5) (Marcoux et al, 1997). The European Society for Human Reproduction and Embryology (ESHRE) Special Interest Group for Endometriosis have recently developed guidelines for the diagnosis and treatment of endometriosis. These guidelines recommend surgical treatment for minimal/mild endometriosis in subfertile women but also report that some members of the guideline developing group questioned the strength of the evidence (Kennedy et al, 2005). The number needed to treat to benefit (NNTB) is 11 (95% CI 6 to 100). One has to adjust for the prevalence of endometriosis in daily clinical practice: if endometriosis is diagnosed in 30% of all women attending a fertility clinic, then the ‘real life’ NNTB is 11/0.3=37 (95% CI 20 to 333), which is less compelling (Practice Committee American Society for Reproductive Medicine-ASRM, 2006). Clearly, there is a need for additional RCTs on this topic. At the same time, it may be hard to convince ethical committees about the need for such studies, and even harder to recruit eligible women, in view of the current level of evidence indicating a likely benefit.

31 Chapter 2

LAPAROSCOPY AND OVULATION INDUCTION TREATMENT

Should a laparoscopy be done routinely in all women before starting ovulation induction? Can a laparoscopy, performed after several failed ovulation induction treatment cycles, reveal significant pathology amenable to surgical treatment with a positive effect on the overall ongoing pregnancy rate? These seemingly straightforward questions are difficult to answer because the limited studies available are non-controlled and retrospective.

Laparoscopy before ovulation induction treatment The available evidence on the role of laparoscopy before ovulation induction merely focuses on the comparison between HSG and laparoscopy findings for the diagnosis of tubal pathology, the diagnosis and treatment of adhesions and the treatment of minimal and mild endometriosis. Concerning the routine use of HSG prior to laparoscopy in the fertility work-up, we refer to the multicentre RCT by Perquin et al (2006). For the CPRs at 18 months, no significant differences were found in 344 women randomized to an intervention group with HSG followed by diagnostic laparoscopy (CPR at 18 months 49%, 95% CI 42% to 57%) or a control group with diagnostic laparoscopy alone (CPR at 18 months 50%, 95% CI 43% to 58%). We also refer to the discussion above on the prospective value of HSG and laparoscopy (Mol et al, 1999). The clinical relevance of treating minimal and mild endometriosis will, as also has been shown above, depend on the prevalence of this disease in the treated population.

Laparoscopy during ovulation induction treatment In a retrospective observational study, Ochoa Capelo and co-workers performed a laparoscopy in 92 subfertile women after four failed cycles of ovulation induction treatment with clomiphene citrate (Ochoa Capelo et al, 2003). All women had at least four ovulatory cycles after treatment, confirmed by basal body temperature and midluteal phase serum progesterone, normal HSG findings and male partners with a normal semen analysis. The incidence of pelvic pathology in this study is summarized below (Table 1). Laparoscopic findings were strictly normal in only 36% of cases, whereas endometriosis and/or pelvic adhesions were observed in 50% and 33% respectively.

32 The position of laparoscopy

Table 1 The incidence of abnormal findings detected at laparoscopy after 4 failed treatment cycles with ovulation induction therapy (adapted from Ochoa Capelo et al., 2003).

Findings at laparoscopy N (*) % (*)

Normal 33/92 36 Minimal endometriosis 21/92 23 Mild endometriosis 6/92 6.5 Moderate endometriosis 8/92 8.7 Severe endometriosis 2/92 2.2 Endometriotic cyst 8/92 8.7 Adhesions 30/92 33 Tubal pathology 1/92 1.1

(*) Several women had more than one abnormal finding at laparoscopy; the sum of all cases exceeds therefore 92 cases or 100%.

The authors concluded that laparoscopy is a useful tool in the fertility work-up despite the lack of data on the pregnancy rates following laparoscopic surgery (Ochoa Capelo et al, 2003). Concerning the effectiveness of the laparoscopic treatment of minimal/mild endometriosis, we refer to the evidence presented above. With regard to laparoscopic adhesiolysis, there is only one clinical controlled study (CCT) by Tulandi and co-workers (1990) that reported higher CPRs of 32% and 45% in 12 and 24 months, respectively, after operative laparoscopy than the 11% and 16% CPRs observed in the control group. To the best of our knowledge, these data have not been confirmed by a randomized trial.

Laparoscopic ovarian drilling in PCOS patients About 20% of all subfertile women diagnosed with polycystic ovaries (PCOS) will not ovulate on the maximum daily dose of 150 mg clomiphene citrate (CC) as the first line of treatment (Imani et al, 1998). Even today, the effective treatment of CC-resistant PCOS remains challenging. More than 20 years ago, Gjonnaess described that the laparoscopic electro coagulation of the ovarian capsule in 62 CC resistant PCOS women resulted in an ovulation rate of 92% and a pregnancy rate of 69% (Gjonnaess 1994). In a recent Cochrane review (Farquhar et al, 2005), the laparoscopic drilling of the ovarian capsule (LOD) by diathermy or laser in CC-resistant PCOS has been randomly compared to

33 Chapter 2

gonadotrophin treatment. The authors retrieved 15 studies on LOD in PCOS women; only six trials were eligible for inclusion. The primary outcomes were live birth rate, ovulation rate and ongoing pregnancy rate. Secondary outcomes were the rate of miscarriage, multiple pregnancies, ovarian hyper stimulation and the total cost of the respective treatments. There were no differences in the live birth or ongoing pregnancy rates between LOD with or without ovulation induction therapy compared to treatment with gonadotrophins (Figure 2) but there was a substantial decrease in the risk for multiple pregnancies after LOD (Figure 3).

Figure 2 Forest plot of comparison: LOD with or without ovulation induction versus gonadotrophin treatment. Outcome: Ongoing pregnancy per woman randomized. M-H: Mantel-Haenszel, fixed- effect model, CI: confidence interval.

LOD Gonadotrophin Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Bayram 2004 56 83 57 85 56.7% 1.02 [0.53, 1.94] Farquhar 2002 5 29 5 21 14.8% 0.67 [0.17, 2.68] Lazoviz 1998 14 29 9 28 14.6% 1.97 [0.67, 5.78] Vegetti 1998 3 16 5 13 13.9% 0.37 [0.07, 1.98]

Total (95% CI) 157 147 100.0% 1.02 [0.63, 1.65] Total events 78 76 Heterogeneity: Chi² = 3.20, df = 3 (P = 0.36); I² = 6% 0.01 0.1 1 10 100 Test for overall effect: Z = 0.06 (P = 0.95) Favours gonadotrophin Favours LOD

Figure 3 Forest plot of comparison: LOD with or without ovulation induction versus gonadotrophin treatment. Outcome: Multiple pregnancy rates per ongoing pregnancy. M-H: Mantel-Haenszel, fixed- effect model, CI: confidence interval.

LOD Gonadotrophin Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Bayram 2004 1 56 9 57 68.9% 0.10 [0.01, 0.79] Farquhar 2002 0 5 0 5 Not estimable Lazoviz 1998 0 14 2 9 22.8% 0.10 [0.00, 2.44] Vegetti 1998 0 3 1 5 8.3% 0.43 [0.01, 14.08]

Total (95% CI) 78 76 100.0% 0.13 [0.03, 0.59] Total events 1 12 Heterogeneity: Chi² = 0.55, df = 2 (P = 0.76); I² = 0% 0.01 0.1 1 10 100 Test for overall effect: Z = 2.64 (P = 0.008) Favours LOD Favours gonadotroph

There were no statistically significant differences in miscarriage rates between both comparison groups (OR 0.8, 95% CI 0.36 to 1.9). Approximately 50% of all treated women will have a live birth and 16%

34 The position of laparoscopy will have a miscarriage with either treatment. The reviewers’ conclusions are firstly that there are no differences in the live birth and the miscarriage rates in women with CC-resistant PCOS undergoing LOD versus gonadotrophin treatment (Farquhar et al, 2005). Secondly, the clinically relevant reduction in multiple pregnancy rates in women undergoing LOD makes this option attractive if a choice has to be made between gonadotrophin treatment and surgery. Disadvantages of the LOD procedure include the risks related to laparoscopic surgery, the need for general anesthesia, the possible risk of thermal damage to adjacent organs and ovarian adhesion formation, and as clearly mentioned in the Cochrane review, the lack of knowledge about possible adverse long-term effects of LOD on the ovarian reproductive function (Farquhar et al, 2005). Moreover it has been reported that the positive effects observed on the ovulation rates are temporary; the signs and symptoms of PCOS may return within months following the LOD (Insler and Lunenfeld, 1993). Advantages of LOD are the opportunity to treat concomitant pelvic pathology such as peritubal adhesions and endometriosis that can be associated with female infertility. Furthermore, during the same endoscopic procedure, tubal patency can be tested, and a hysteroscopy can be performed as part of the fertility work-up. In summary, the position of laparoscopy in the setting of ovulation induction is at present not clear due to the lack of high quality evidence. The routine use of laparoscopy to evaluate all cases of female anovulatory subfertility can therefore not be recommended. In specific clinical conditions laparoscopy is very useful to assess the tuboperitoneal status, to treat pelvic pathology that may limit conception (endometriosis, adhesions), and to perform LOD. LOD is a good option when compared with gonadotrophin treatment in the CC-resistant PCOS patient. Nevertheless gynecological surgeons should counsel women with CC resistant PCOS about the unknown long-term effects on the ovarian function.

35 Chapter 2

LAPAROSCOPY AND IUI

IUI is used in daily practice for treating cervical factor, unexplained and mild male subfertility. Two relevant clinical questions arise in the setting of IUI treatment. Firstly, does laparoscopy significantly change the intended treatment plan in cases where IUI is clinically indicated? Secondly, should laparoscopy be done before IUI or only after several failed IUI cycles?

Laparoscopy before IUI After having conducted a pilot retrospective observational study (Tanahatoe et al, 2003) a Dutch research group published the results of a RCT on the role of laparoscopy in IUI (Tanahatoe et al, 2005). A group of 154 women, eligible for IUI treatment for unexplained, cervical or mild male subfertility with normal HSG was randomly allocated to two comparison groups. In the intervention group (laparoscopy first or LSF; n=77) the participants were planned to undergo a laparoscopy before IUI. Further treatment was discontinued in 13 women, either because of exclusion for not giving informed consent (n=10) or because of pregnancy (n=3). A laparoscopy was therefore performed in the remaining 64 women. Following laparoscopy, IUI treatment was started. Before and during IUI in the LSF group, 11 participants dropped out. Of the 31 women who became pregnant in this group, 9 conceived before or between IUI and 22 conceived due to the IUI treatment. In the control group (IUI first or IUIF; n=77) women were treated with IUI during six treatment cycles. The first three IUI cycles were done without controlled ovarian hyper stimulation (COH). If pregnancy did not occur after three cycles of IUI in the natural cycle, the woman was offered a choice between continuing IUI in the natural cycle and starting IUI with COH with recombinant FSH for a maximum period of three additional cycles. Further treatment in the IUIF group was discontinued in 54 women because of pregnancy (n=38) or due to drop out (n=16). The remaining 23 participants who did not conceive in the IUIF group all underwent a laparoscopy. The main outcomes were the pregnancy rate per woman randomized and the presence of pelvic pathology with therapeutic implications. The results are presented below in figures 4 and 5. The pregnancy rate per woman randomized was 40 to 50% and the presence of pelvic pathology with therapeutic implications was high (48 to 56%) but both outcomes were equally distributed among both comparison groups. Indeed, the at random allocation of subfertile women to the laparoscopy first group versus the IUI first group made no difference for the ongoing pregnancy rates per woman randomized or the presence of

36 The position of laparoscopy pelvic pathology requiring laparoscopic treatment. The respective ORs were 0.81 (95% CI 0.43 to 1.5, P=0.52) for the ongoing pregnancy rate (Figure 4) and 0.72 (95% CI 0.28 to 1.9, P=0.51) for the presence of pelvic pathology with therapeutic implications (Figure 5).

Figure 4 Forest plot of comparison: Laparoscopy first versus IUI first in subfertile women treated by IUI. Outcome: Pregnancy per woman randomized. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

LSF IUIF Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Tanahatoe 2005 34 77 38 77 100.0% 0.81 [0.43, 1.53]

Total (95% CI) 77 77 100.0% 0.81 [0.43, 1.53] Total events 34 38 Heterogeneity: Not applicable 0.01 0.1 1 10 100 Test for overall effect: Z = 0.65 (P = 0.52) Favours IUIF Favours LSF

Figure 5 Forest plot of comparison: Laparoscopy first versus IUI first in subfertile women treated by IUI. Outcome: Presence of pelvic pathology with therapeutic implications per woman randomized. M-H: Mantel- Haenszel, fixed-effect model, CI: confidence interval.

LSF IUIF Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Tanahatoe 2005 31 64 13 23 100.0% 0.72 [0.28, 1.89]

Total (95% CI) 64 23 100.0% 0.72 [0.28, 1.89] Total events 31 13 Heterogeneity: Not applicable 0.01 0.1 1 10 100 Test for overall effect: Z = 0.66 (P = 0.51) Favours IUIF Favours LSF

The authors conclude that additional randomized studies are needed to verify the results of this single trial. They have calculated that at least 1000 women should have been included to detect a difference of 10% in the cumulative ongoing pregnancy rate (Tanahatoe et al., 2005). Furthermore the high spontaneous pregnancy rate in both comparison groups (n=12 in LSF and n=16 in IUIF) is noteworthy.

Laparoscopy after failed IUI There are to our knowledge no published studies on the added value of laparoscopy after several failed cycles of COH and IUI. Referring to the Dutch RCT (Tanahatoe et al, 2005), one may expect to detect pelvic pathology (endometriosis all stages, peritubal adhesions) in at least 50%

37 Chapter 2

of cases. Laparoscopic treatment enhances the chance of spontaneous conception. One may, by extrapolation, expect a higher pregnancy rate after laparoscopic treatment after several failed IUI cycles. In conclusion, the position of laparoscopy for diagnosing and treating endometriosis and/or peritubal adhesions prior to IUI treatment or after several failed IUI cycles seems a matter of debate. Additional RCTs are needed on the position of laparoscopy in IUI.

LAPAROSCOPY AND IVF

Without doubt, the progress in assisted reproductive techniques/ technology (ART) has narrowed the indications for reproductive surgery. Some authors have recommended immediate treatment with ART after a limited and non-invasive fertility work-up in all subfertile women (Speroff et al, 1999). This raises two questions. Firstly, is it always mandatory to do a laparoscopy prior to ART? Secondly, is there any indication for a laparoscopy after several failed ART treatment cycles?

Laparoscopy before IVF On the basis of the studies mentioned above (Mol et al, 2001; Mol et al, 1999; Perquin et al, 2006) one could recommend that a laparoscopy could be avoided in all cases where IVF is clinically judged to be the most appropriate treatment. However, there is a fair degree of consensus that selected adnexal pathology, such as hydrosalpinx and ovarian endometriotic cysts, should still be treated by laparoscopic surgery prior to proceeding to IVF. With respect to hydrosalpinx, two RCTs have demonstrated increased success rates with IVF following salpingectomy for ultrasonically visible hydrosalpinges (Dechaud et al, 1998; Strandell et al, 1999). Both trials have been included in a recent Cochrane review (Johnson et al, 2004). The intention-to-treat analysis for the data of the delivery rates in the Scandinavian trial (Strandell et al, 1999) demonstrated no evidence for statistically significant differences between both comparison groups (27% versus 17%; risk ratio (RR) was 1.6, 95% CI 0.9 to 2.7, P=0.13, 1 study, n=204 women). In a subgroup analysis however, the delivery rates in women treated by IVF after salpingectomy for a hydrosalpinx visible by ultrasound were substantially increased compared to women with the hydrosalpinx left untreated (40% versus 17%; RR 2.4; 95% CI 1.1 to 5.3, P=0.038, 1 study, n=75 women). The highest increase in delivery rate was observed in women with bilateral hydrosalpinx on ultrasound (55% versus 16%; RR 3.5; 95% CI 1.1 to 11, P=0.019, 1 study, n=39 women). Based

38 The position of laparoscopy on the data of the meta-analysis (Johnson et al, 2004), the number needed to treat to benefit (NNTB) is eight (95% CI 5 to 11). There are several hypotheses in the literature to explain the beneficial effect of removing the diseased tubes: a direct toxic effect of tubal fluid on the embryos, a negative effect of tubal fluid on the endometrium by flushing out embryos, dilution of implantation factors and prevention of normal embryonic-endometrial apposition (Erel and Senturk, 2005). Some authors have argued against the indiscriminate removal of all hydrosalpinges prior to ART; they have recommended to consider doing a neosalpingostomy in selected cases identified by concomitant salpingoscopy (Puttemans et al, 1996). A RCT of reconstructive tubal surgery versus salpingectomy prior to proceeding with IVF in women with hydrosalpinx is needed to define the position of both treatment strategies in everyday clinical practice; such a management trial can, reasonably spoken, only be conducted in countries with a high prevalence of PID (Sabatini and Davis, 2005). There are no RCTs or meta-analyses available to answer the question whether surgical treatment of moderate and severe endometriosis enhances the rates of spontaneous conception or pregnancy after IVF (Kennedy et al, 2005). There were no treatment-independent pregnancies among couples with r-AFS stages III or IV endometriosis in an analysis of data from the Canadian CITES study (Mol et al, 1999). Two prospective cohort studies (Nezhat et al, 1989; Vercellini et al, 2006) have reported crude spontaneous pregnancy rates of 57% to 69% (moderate endometriosis) and 52% to 68% (severe endometriosis) following laparoscopic surgery. These numbers are substantially higher than the crude spontaneous pregnancy rates of 33% (moderate endometriosis) and 0% (severe endometriosis) after expectant management reported in a third prospective cohort study (Olive et al, 1985). Based on these observational studies it is generally accepted that in case of subfertility, moderate and severe stage endometriosis should be treated by surgery. There seems to be a negative correlation between the stage of endometriosis and the spontaneous cumulative pregnancy rate after surgical removal of endometriosis based on the evidence of three observational studies (Adamson et al, 1993; Guzick et al, 1997; Osuga et al, 2002), but statistical significance was reached in one study only (Osuga et al., 2002). Recent ESHRE guidelines state that IVF is the appropriate treatment if tubal function is compromised, if there is severe male factor infertility or if other treatments have failed (Kennedy et al., 2005). The IVF pregnancy rates are lower in women with endometriosis than in those with tubal infertility according to a systematic review of 22 non- randomized studies (Barnhart et al, 2002). These authors conclude that

39 Chapter 2

there is an overall 54% reduction in pregnancy rate after IVF in women with endometriosis; the success rates are poorer with advancing severity of the disease, staged according to the r-AFS classification system. In contrast, according to data from large databases namely SART (the Society for Assisted Reproductive Technology) and HFEA (the Human Fertilization and Embryology Authority) endometriosis does not seem to adversely affect the reported pregnancy rates (Templeton et al, 1996). There are no RCTs available that have tested the hypothesis that surgical treatment of endometriosis prior to IVF results in higher pregnancy rates when compared to expectant management of endometriosis. Ovarian endometriotic cysts need extra attention in the context of ART since they can be disadvantageous for IVF treatment: they may interfere with controlled ovarian hyperstimulation (COH), create difficulties in aspirating the ovarian follicles during oocyte retrieval, and be held responsible for producing detrimental substances that are toxic to maturing oocytes, thus impeding embryo cleavage and implantation. Laparoscopic surgery for advanced stage endometriosis can be technically very demanding and time-consuming; moreover it carries a high risk for significant postoperative morbidity and long revalidation. The removal of ovarian endometriomas prior to COH may be associated with significant bleeding and destruction of normal adjacent ovarian tissue, thus diminishing the reproductive ovarian function. There are no RCTs comparing the live birth rates after IVF treatment in women who were surgically treated for endometriotic cysts prior to IVF versus women who were not. According to one retrospective case-control study (Garcia- Velasco et al, 2004) the removal of endometriotic cysts prior to IVF did not improve fertility outcome. In cases of asymptomatic small endometriotic cysts (<3 cm), immediate proceeding to IVF may reduce the time to pregnancy, treatment costs and the possible detrimental effects of inappropriate surgery on the ovarian function. The laparoscopic excision of larger symptomatic endometriotic cysts (>4 cm) increases the chance for spontaneous pregnancy and reduces recurrence compared to simple drainage and coagulation (Beretta et al, 1998; Chapron et al, 2002; Vercellini et al, 2003b).

Laparoscopic treatment of endometriosis after failed IVF Finally, is it worth doing laparoscopic surgery in women with endometriosis after several failed IVF cycles? No RCTs are available but a retrospective cohort study (Littman et al, 2005) deserves a closer look. In a series of 29 women with several failed IVF cycles and endometriosis, a radical treatment of all endometriotic lesions was done by one expert laparoscopic surgeon. After surgery, 22 pregnancies were observed,

40 The position of laparoscopy including 15 spontaneous pregnancies and 7 pregnancies after repeated IVF treatment. The limited evidence from this retrospective observational study highlights the importance of treating women with severe endometriosis in a centre with sufficient surgical expertise (Kennedy et al, 2005).

41 Chapter 2

CONCLUSION

The surgical treatment of minimal/mild endometriosis increases the spontaneous pregnancy rate in women with otherwise unexplained subfertility. The position of laparoscopy for diagnosing and treating endometriosis and/or adhesions prior to IUI treatment or after failed IUI treatment is a matter of debate; additional RCTs are needed to test the hypothesis that this surgical approach might improve the pregnancy rates during IUI treatment. RCTs confirming the benefit of the surgical treatment of moderate and severe endometriosis in subfertile women are lacking, but the value of surgery seems generally accepted. The position and timing of laparoscopy in ovulation induction treatment is difficult to establish due to a lack of RCTs. Laparoscopy is indicated in all cases of bilateral anomalies on HSG: the exclusion of bilateral anatomical tubal pathology by laparoscopy could avoid IVF treatment in some cases. LOD in the treatment of women with CC-resistant PCOS is at least as effective as gonadotrophin treatment, but has a significantly lower risk of multiple pregnancies. Prior to IVF the laparoscopic removal of ultrasonically visible hydrosalpinges may increase the IVF success rates; more research is needed to evaluate the effectiveness of alternative surgical strategies, such as the restoration of a hydrosalpinx with an otherwise healthy endosalpingeal mucosa. Small asymptomatic endometriotic cysts probably need no treatment prior to IVF according to observational evidence in contrast to the larger symptomatic endometriomas. The routine practice of doing a laparoscopy in all subfertile women is questionable and not supported by high-quality evidence; therefore more studies are needed on the add-on value and timing of laparoscopy in current fertility practice.

42 The position of laparoscopy

REFERENCES

1. Adamson GD, Hurd SJ, Pasta DJ, Rodriguez BD. Laparoscopic endometriosis treatment: is it better? Fertil Steril 1993; 59: 35–44. 2. Barnhart K, Dunsmoor-Su R, Coutifaris C. Effect of endometriosis on in vitro fertilization. Fertil Steril 2002; 77: 1148–1155. 3. Bayram N, van Wely M, Kaaijk EM, Bossuyt PMM, van der Veen F. Using an electrocautery strategy or recombinant follicle stimulating hormone to induce ovulation in polycystic ovary syndrome: randomized controlled trial. BMJ 2004; 328:192. 4. Beretta P, Franchi M, Ghezzi F, Busacca M, Zupi E, Bolis P. Randomized clinical trial of two laparoscopic treatments of endometriomas: cystectomy versus drainage and coagulation. Fertil Steril 1998; 70: 1176–1180. 5. Chapron C, Vercellini P, Barakat H, Vieira M, Dubuisson JB. Management of ovarian endometriomas. Hum Reprod Update 2002; 8:6–7. 6. Chuang JT, Hewett WJ, Hreshchyshyn M. Death after hysterosalpingography in choriocarcinoma with pelvic abscess. Report of a patient. Obstet Gynecol 1971; 37:543–545. 7. Collins A, Burrows EA, Willan AR. The prognosis for live birth among untreated infertile couples. Fertil Steril 1995; 64: 22–28. 8. Coppus SFPJ, Opmeer BC, van der Veen F, Bossuyt PMM, Mol BWJ. Routine use of hysterosalpingography prior to diagnostic laparoscopy in the fertility workup. Hum Reprod 2006; 21(10): 2725- 2726. 9. Dechaud H, Daures JP, Arnal F, Humeau C, Hedon B. Does previous salpingectomy improve implantation and pregnancy rates in patients with severe tubal factor infertility who are undergoing in vitro fertilization? A pilot prospective randomized study. Fertil Steril 1998; 69: 1020–1025. 10. De Hondt A, Meuleman C, Tomassetti C, Peeraer K, D’Hooghe TM. Endometriosis and assisted reproduction: the role for reproductive surgery? Curr Opin Obstet Gynecol 2006; 18: 374–379. 11. Dijkman AB, Mol BWJ, van der Veen F, Bossuyt PMM, Hogerzeil HV. Can hysterosalpingocontrast-sonography replace hysterosalpingography in the assessment of tubal subfertility? Eur J Radiol 2000; 35: 44-48. 12. Erel CT, Senturk LM. Is laparoscopy necessary before assisted reproductive technology? Curr Opin Obstet Gynecol 2005; 17: 243– 248.

43 Chapter 2

13. Farquhar C, Lilford RJ, Marjoribanks J, Vandekerckhove P. Laparoscopic drilling by diathermy or laser for ovulation induction in anovulatory polycystic ovary syndrome. Cochrane Database Syst Rev 2005; 3: CD001122. 14. Fatum M, Laufer N, Simon A. Investigation of the infertile couple: should diagnostic laparoscopy be performed after normal hysterosalpingography in treating infertility suspected to be of unknown origin? Hum Reprod 2002; 17:1–3. 15. Forman RG, Robinson JN, Mehta Z, Barlow DH. Patient history as a simple predictor of pelvic pathology in subfertile women. Hum Reprod 1993; 8: 53–55. 16. Garcia-Velasco JA, Mahutte NG, Corona J, Zuniga V, Giles J, Arici A, Pellicer A. Removal of endometriomas before in vitro fertilization does not improve fertility outcomes: a matched case-control study. Fertil Steril 2004; 81: 1194–1197. 17. Gjonnaess H. Polycystic ovarian syndrome treated by ovarian electrocautery through the laparoscope. Fertil Steril 1994; 41: 20–25. 18. Glatstein IZ, Harlow BL, Hornstein MD. Practice patterns among reproductive endocrinologists: the infertility evaluation. Fertil Steril 1997; 67: 443–451. 19. Guzick DS, Silliman NP, Adamson GD, Buttram VC, Canis M, Malinak LR, Schenken RS. Prediction of pregnancy in infertile women based on the American Society for Reproductive Medicine’s revised classification of endometriosis. Fertil Steril 1997; 67: 822– 829. 20. Hamilton J, Latarche E, Gillott C, Lower A, Grudzinskas JG. Intrauterine insemination results are not affected if Hysterosalpingo Contrast Sonography is used as the sole test of tubal patency. Fertil Steril 2003; 80: 165–171. 21. Härkki-Sirén P, Sjöberg J, Kurki T. Major complications of laparoscopy: a follow-up Finnish study. Obstet Gynecol 1999; 94: 94– 98. 22. Houston DE, Noller KL, Melton LJ, Selwyn BJ, Hardy RJ. Incidence of pelvic endometriosis in Rochester, Minnesota, 1970–1979. Epidemiol 1987; 125: 959–969. 23. Hubacher D, Grimes D, Lara-Ricalde R, de la Jara J, Garcia-Luna A. The limited clinical usefulness of taking a history in the evaluation of women with tubal factor infertility. Fertil Steril 2004; 81:6–10. 24. Imani B, Eijkemans MJ, te Velde ER, Habbema JD, Fauser BC. Predictors of patients remaining anovulatory during clomiphene citrate induction of ovulation in normo gonadotropic

44 The position of laparoscopy

oligoamenorrheic infertility. J Clin Endocrinol Metab 1998; 83: 2361- 2365. 25. Insler V, Lunenfeld B. Polycystic ovarian disease, pp. 661–678. In: Infertility, Male and Female, 2nd edn. Edinburgh, New York: Churchill Livingstone, 1993. 26. Jacobson TZ, Barlow DH, Koninckx PR, Olive D, Farquhar C. Laparoscopic surgery for subfertility associated with endometriosis (cochrane review). In: The Cochrane Library, Issue 3. Chichester, UK: John Wiley & Sons LtdChichester, UK, 2004b. 27. Johnson NP, Mak W, Sowter MC. Surgical treatment for tubal disease in women due to undergo in vitro fertilization. Cochrane Database Syst Rev 2004; 3: CD002125. 28. Kennedy S, Bergqvist A, Chapron C, D’Hooghe T, Dunselman G, Greb R, Hummelshoj L, Prentice A, Saridogan E. ESHRE guidelines for the diagnosis and treatment of endometriosis. Hum Reprod 2005; 20: 2698–2704. 29. Lazovic G, Milacic D, Terzic M, Spremovic S, Mitijasevic S. Medicaments or surgical therapy of PCOS. Fertil Steril 1998; 70; S472. 30. Littman E, Giudice L, Lathi R, Berker B, Milki A, Nezhat C. Role of laparoscopic treatment of endometriosis in patients with failed in vitro fertilization cycles. Fertil Steril 2005; 84; 1574–1578. 31. Mahmood TA, Templeton A. Prevalence and genesis of endometriosis. Hum Reprod 1991; 6: 544–549. 32. Marcoux S, Maheux R, Bérubé S, and the Canadian Collaborative Group on Endometriosis. Laparoscopic surgery in infertile women with minimal or mild endometriosis. N Engl J Med 1997; 337: 217– 222. 33. Mol BWJ, Collins JA, Burrows EA, van der Veen F, Bossuyt PM. Comparison of hysterosalpingography and laparoscopy in predicting fertility outcome. Hum Reprod 1999; 14: 1237–1242. 34. Mol BWJ, Collins JA, van der Veen F, Bossuyt PMM. Cost- effectiveness of hysterosalpingography, laparoscopy and Chlamydia antibody testing in subfertile couples. Fertil Steril 2001; 75: 571–580. 35. Mol BWJ, Dijkman B, Wertheim P, Lijmer J, van der Veen F, Bossuyt PMM. The accuracy of serum chlamydial antibodies in the diagnosis of tubal pathology: a meta-analysis. Fertil Steril 1997a; 67: 1031–1037. 36. Mol BWJ, Swart P, Bossuyt PMM, van der Veen F. Is hysterosalpingography an important tool in predicting fertility outcome? Fertil Steril 1997b; 67: 663–669.

45 Chapter 2

37. Nezhat C, Crowgey S, Nezhat F. Videolaseroscopy for the treatment of endometriosis associated with infertility. Fertil Steril 1989; 51: 237-240. 38. Ochoa Capelo F, Kumar A, Steinkampf MP, Azziz R. Laparoscopic evaluation following failure to achieve pregnancy after ovulation induction with clomiphene citrate. Fertil Steril 2003; 80: 1450–1453. 39. Ogata R, Nakamura G, Uchiumi Y, Yokoyama M. Therapeutic efficacy of hysterosalpingography (HSG) in a prospective, randomized, clinical study. Jpn J Fertil Steril 1993; 38: 91–94. 40. Olive DL, Stohs GF, Metzger DA, Franklin RR. Expectant management and hydrotubations in the treatment of endometriosis- associated infertility. Fertil Steril 1985; 44: 35-41. 41. Olsen J, Juul S, Basso O. Measuring time to pregnancy: methodological issues to consider. Hum Reprod 1998; 13: 1751-1753. 42. Osuga Y, Koga K, Tsutsumi O, Yano T, Maruyama M, Kugu K, Momoeda M, Taketani Y. Role of laparoscopy in the treatment of endometriosis-associated infertility. Gynecol Obstet Invest 2002; 53(Suppl): 33–39. 43. Parazzini F. Ablation of lesions or no treatment in minimal-mild endometriosis in infertile women: a randomized trial. Gruppo Italiano per lo Studio dell’Endometriosi. Hum Reprod 1999; 14: 1332–1334. 44. Perquin DAM, Dörr PJ, de Craen AJM, Helmerhorst FM. Routine use of hysterosalpingography prior to laparoscopy in the fertility work-up: a multicentre randomized controlled trial. Hum Reprod 2006; 21: 1127–1231. 45. Practice Committee of ASRM. Interpretation of clinical trial results. Fertil Steril 2006; 86(Suppl 5): S161–167. 46. Puttemans PJ, Brosens IA. Salpingectomy improves in vitro fertilization outcome in patients with a hydrosalpinx: blind victimization of the fallopian tube? Hum Reprod 1996; 11: 2079– 2081. 47. Rowe PJ, Comhaire FH, Hargreave TB, Mahmoud AMA. WHO manual for the standardized investigation of the infertile couple. Cambridge, UK: Cambridge University Press Cambridge, UK, 1993. 48. Sabatini L, Davis C. The management of hydrosalpinges: tubal surgery or salpingectomy? Curr Opin Obstet Gynecol 2005; 17: 323– 328. 49. Speroff L, Glass RH, Kase NG. Female infertility. In: Speroff L, Glass RH, Kase NG (Eds). Clinical Gynaecologic and Infertility, 6th edn. Philadelphia, PA: Lippincott Williams & Wilkins Philadelphia, PA, 1999.

46 The position of laparoscopy

50. Strandell A, Lindhard A, Waldenstrom U, Thorburn J, Janson PO, Hamberger L. Hydrosalpinx and IVF outcome: a prospective, randomized multicentre trial in Scandinavia on salpingectomy prior to IVF. Hum Reprod 1999; 14: 2762–2769. 51. Swart P, Mol BWJ, van der Veen F, van Beurden M, Redekop WK, Bossuyt PMM. The value of hysterosalpingography in the diagnosis of tubal pathology, a meta-analysis. Fertil Steril 1995; 64: 486–491. 52. Tanahatoe SJ, Hompes PGA, Lambalk CB. Accuracy of diagnostic laparoscopy in the infertility work-up before intrauterine insemination. Fertil Steril 2003; 79: 361–366. 53. Tanahatoe SJ, Lambalk CB, Hompes PGA. The role of laparoscopy in intrauterine insemination: a prospective randomized reallocation study. Hum Reprod 2005; 20: 3225–3230. 54. Templeton A, Morris JK, Parslow W. Factors that affect outcome of in vitro fertilization treatment. Lancet 1996; 348:1402–1406. 55. Tulandi T, Collins JA, Burrows E, Jarrell JF, McInnes RA, Wrixon W. Treatment-dependent and treatment-independent pregnancy among women with periadnexal adhesions. Am J Obstet Gynecol 1990; 162: 354–357. 56. Vegetti W, Ragni G, Baroni E, Testa G, Marsico S, Riccaboni A et al. Laparoscopic ovarian drilling versus low-dose pure FSH in anovulatory clomiphene-resistant patients with polycystic ovary syndrome: randomized prospective study. Hum Reprod 1998; 13: S120. 57. Vercellini P, Chapron C, De Giorgi O, Consonni D, Frontino G, Crosignani PG. Coagulation or excision of ovarian endometriomas? Am J Obstet Gynecol 2003b; 188: 606–610. 58. Vercellini P, Fedele L, Aimi G, De Giorgi O, Consonni D, Crosignani PG. Reproductive performance, pain recurrence and disease relapse after conservative surgical treatment for endometriosis: the predictive value of the current classification system. Hum Reprod 2006; 21: 2679-2685. 59. Watson A, Vandekerckhove P, Lilford R, Vail A, Brosens I, Hughes E. A meta-analysis of the therapeutic role of oil soluble contrast media at hysterosalpingography: a surprising result? Fertil Steril 1994; 61: 470–477.

47

The effectiveness of hysteroscopy for improving pregnancy rates in subfertile women: a systematic review

Jan Bosteels, Steven Weyers, Patrick Puttemans, Costas Panayotidis, Bruno van Herendael, Victor Gomel, Ben Willem Mol, Chantal Mathieu, Thomas M. D’Hooghe

Human Reproduction Update 2010; 16(1): 1-11 http://humupd.oxfordjournals.org/content/16/1/1.full.pdf+html

49

Hysteroscopy for improving pregnancy rates in subfertile women

ABSTRACT

Background Hysteroscopy is frequently used for treating subfertile women, but a systematic review of the evidence on this subject is lacking.

Methods We summarized and appraised the evidence for the benefit yielded by this procedure. Our systematic search was limited to randomized controlled trials (RCTs). The QUOROM (Quality of Reporting of Meta-analyses) standards were followed. Language restrictions were not applied.

Results We retrieved five studies. The hysteroscopic removal of endometrial polyps with a mean diameter of 16mm detected by ultrasound doubles the pregnancy rate compared to diagnostic hysteroscopy and polyp biopsy in women undergoing intrauterine insemination: the risk ratio (RR) was 2.2, 95% confidence interval (CI) 1.6 to 3.1, P<0.00001. In subfertile women with one knot and/or one fibroid smaller than 4cm, there was a trend in favor of hysteroscopic myomectomy compared to expectant management; the differences between both comparison groups were not statistically significant (RR 1.9, 95% CI 0.97 to 3.7, P=0.06). Hysteroscopic metroplasty for septate uterus resulted in fewer pregnancies in subfertile women compared to those with recurrent pregnancy loss (RR 0.7, 95% CI 0.5 to 0.9, P=0.01). There are no RCTs on the effectiveness of the hysteroscopic treatment of intrauterine adhesions. Hysteroscopy in the cycle preceding a subsequent in vitro fertilization (IVF) attempt can nearly double the pregnancy rate in women suffering from primary subfertility with at least two failed IVF attempts compared to starting a next IVF cycle without prior hysteroscopy (RR 1.6, 95% CI 1.3 to 1.9, P<0.00001).

Conclusions Scarce evidence on the effectiveness of hysteroscopic surgery in subfertile women with polyps, fibroids, septate uterus or intrauterine adhesions indicates a potential benefit.

51 Chapter 3

INTRODUCTION

The position of hysteroscopy in current fertility practice is unclear (Shushan and Rojansky, 1999) despite several randomized controlled trials (RCTs) on its technical feasibility and high patient compliance (Kremer et al, 2000; Soriano et al, 2000; Unfried et al, 2001; De Angelis et al, 2003; Guida et al, 2003; Litta et al, 2003; Pellicano et al, 2003; Marsh et al, 2004; Shankar et al, 2004; Campo et al, 2005; Sharma et al, 2005; Garbin et al, 2006; Guida et al, 2006; Sagiv et al, 2006; De Placido et al, 2007; Kabli and Tulandi, 2008). There seems to be fewer consensuses on the effectiveness of hysteroscopy for improving the fertility prognosis of subfertile women. The Royal College of Obstetricians and Gynecologists (RCOG) do not recommend hysteroscopy as an initial investigation in the fertility work- up unless clinically indicated. In its evidence-based guidelines on fertility assessment and treatment, the statement for not recommending hysteroscopy as a procedure to be routinely done in all subfertile women has received a grade B (RCOG, 2004). It is argued that the effectiveness of hysteroscopy in the general subfertile population has not yet been proven by high-quality evidence. The European Society for Human Reproduction and Embryology (ESHRE) has adopted a similar viewpoint (Crosignani and Rubin, 2000). This systematic review (SR) aims to examine the effectiveness of the hysteroscopic removal of endometrial polyps, submucous fibroids, uterine septa or intrauterine adhesions in subfertile women for improving their chance to become pregnant. We also aimed to study the effectiveness of hysteroscopy in women treated by IVF or intrauterine insemination (IUI).

52 Hysteroscopy for improving pregnancy rates in subfertile women

METHODS

Literature search methodology We aimed to identify RCTs on the hysteroscopic treatment of endometrial polyps, submucous fibroids, septate uterus or intrauterine adhesions in subfertile women or prior to IVF/ICSI (intracytoplasmic sperm insertion) or IUI. Two review authors conducted independently a systematic literature search using Medical Subject Headings (MeSH) terms for “hysteroscopy, polyps, fibroids, congenital anomalies, Asherman’s syndrome, adhesions and assisted reproductive techniques” in MEDLINE through Pub Med (1966-November 2008), EMBASE (1974-November 2008), CINAHL (1981-November 2008), The Cochrane Library (1970-November 2008) and DARE (Database of Abstracts of Reviews of Effects) for relevant studies. We also searched clinical trial databases namely ISRCTN (International Standard Randomized Controlled Trial Number) register and Meta-register). We classified the retrieved studies in two main categories: studies on operative hysteroscopy in subfertile women with polyps, fibroids, septate uterus and intrauterine adhesions and studies on diagnostic or operative hysteroscopy in women treated by IVF/ICSI or IUI. The full-text manuscripts of all citations that possibly matched the predefined selection criteria were examined by four authors independently for final inclusion or exclusion; any disagreement about inclusion was resolved by consensus or after consultation with an independent fifth author.

Study selection RCTs were included if they dealt with diagnostic or operative hysteroscopy as the study intervention and with pregnancy and/or live birth as one of the main outcomes. The study population included women with polyps, fibroids, septate uterus and intrauterine adhesions suffering from subfertility as the main clinical problem or treated by IVF or IUI. Trials on diagnostic accuracy, patient compliance and cost-effectiveness were excluded. Language restrictions were not applied.

Data extraction We assessed the selected studies for methodological quality by using check-lists available at the Dutch Cochrane Centre website. The QUOROM checklist was followed (Moher et al, 1999). We contacted the primary study authors to obtain missing data or clarification if necessary.

53 Chapter 3

Statistical analysis Statistical analysis was done using Review Manager 5. Dichotomous data were extracted in 2x2 tables. Results were expressed as risk ratios (RR) with 95% confidence intervals (CI) using a fixed-effect model (Mantel- Haenszel). Statistical heterogeneity was assessed with the I² statistic.

RESULTS

The process of the literature search and selection is shown in figure 1: we included five RCTs.

Figure 1 Flowchart for systematic review (SR) of hysteroscopy in fertility practice.

Total number of citations: n= 819

Citations excluded after Screening titles/abstracts: n=641

Full manuscripts retrieved for evaluation: n=178 Excluded with reason: n=148 RCT: n=3 SR of RCTs: n=1 Non- controlled: n=2 Review articles: n=50 Retrospective: n=56 Case control: n=1 Case series: n=14 Potentially eligible for inclusion n=30 Case report: n=7 Clinical article: n=14

Articles excluded with reason: n=25 RCTS: n=3 • outcome:2 • excluded fibroids:1 RCTs included in the present SR n=5 Observational: n=22

54 Hysteroscopy for improving pregnancy rates in subfertile women

We included three trials on the effectiveness of operative hysteroscopy in subfertile women with endometrial polyps, fibroids or septate uterus. A summary of the study characteristics of these three RCTs is presented in table 1. We included two RCTs on the effectiveness of diagnostic or operative hysteroscopy in women undergoing IVF: both were included in a SR with meta-analysis (El-Toukhy et al, 2008). See table 2.

Table 1 The effectiveness of operative hysteroscopy for polyps, fibroids and septate uterus on outcome (pregnancy); study characteristics of the three included RCTs.

Pathology Polyps Fibroids Septate uterus Reference Pérez-Medina 2005 Casini 2006 Colacurci 2007 Method of randomization computer generated random table computer generated list list Allocation concealment yes no no Blinding no no no Groups comparable unclear yes yes Intention-to-treat analysis no yes no Follow-up rate analysis > 95% > 95% < 85% Power calculation yes no no N of included women 215 94 160 Intervention hysteroscopic hysteroscopy/ 5 mm hysteroscopy polypectomy laparotomy with Versapoint N=107 N=52 N=80 Control group diagnostic hysteroscopy no surgery 8 mm resectoscopy N=108 N=42 N=80 Outcome measure pregnancy rate after 4 clinical pregnancy clinical pregnancy rate cycles/time to rate at 12 months pregnancy

55 Chapter 3

Table 2 Effectiveness of outpatient hysteroscopy in recurrent IVF failure; study characteristics of the two included RCTs.

Reference Demirol and Gurgan, 2004 Rama Raju et al, 2006 Method of randomization computer generated list computer generated list Concealment yes not clear Blinding no yes Groups comparable yes yes Intention-to-treat analysis yes yes Follow-up rate analysis > 95% > 95% Number of included women 421 520 Type of infertility primary primary Previous investigations HSG HSG IVF history • 2 failed cycles • 2 failed cycles Timing of hysteroscopy follicular phase follicular phase Distension medium saline glycine % abnormal findings 26% 37% Intervention 5 mm hysteroscopy 5 mm hysteroscopy n=210 n=265 Control no hysteroscopy no hysteroscopy n=211 n=255 Outcome measure clinical pregnancy rate, clinical pregnancy rate, miscarriage rate miscarriage rate, live birth rate

We excluded three RCTs: the conception rate was not studied as the main outcome in two studies (Muzii et al, 2007; Vercellini et al, 1993) and one trial on fibroids (Seracchioli et al, 2000) published in a Cochrane review (Griffiths et al, 2006) excluded submucous fibroids.

Does operative hysteroscopy increase the pregnancy rate in subfertile women with a specified uterine pathology?

Hysteroscopic polypectomy We found two RCTs on the hysteroscopic removal of endometrial polyps (Pérez-Medina et al, 2005; Muzii et al, 2007); only one trial studied the reproductive outcome after hysteroscopic polypectomy in subfertile women with endometrial polyps undergoing IUI (Pérez- Medina et al, 2005). The study characteristics are presented in table 1. The study population consisted of subfertile women with a sonographic diagnosis of endometrial polyps trying to conceive for at least 24 months and planned for IUI. An endometrial polyp was suspected when a hyperechogenic image with regular contour was demonstrated on transvaginal ultrasound (TVUS) with the presence of a vascular stalk on color Doppler map. The

56 Hysteroscopy for improving pregnancy rates in subfertile women polyps were detected in 452 out of 2800 consecutive women scheduled for IUI that attended the fertility clinic of a university hospital during a 50-month period. Of these 452 women, 215 gave informed consent for participation in the study. Randomization was done by a computer generated list using sequentially numbered opaque sealed envelopes. Women allocated to the intervention group (n=107) were treated by polypectomy with a 5.5 mm continuous flow office hysteroscope whereas in the control group (n=108) women underwent a diagnostic hysteroscopy with polyp biopsy. Participants in both comparison groups were subsequently treated with four cycles of IUI with ovarian stimulation with recombinant FSH starting three months after hysteroscopy. Clinical pregnancy was defined as a rising level of ß-hCG combined with TVUS visualization of a gestational sac. The clinical pregnancy rate after 4 IUI cycles was 63% in the polypectomy group compared to 28% in the control group (RR 2.2, 95% CI 1.6 to 3.1, P<0.00001) (Figure 2). The number needed to treat to benefit (NNTB) is 3 (95% CI 2 to 5). Interestingly, 65% of all pregnancies in the polypectomy group occurred before the IUI was started, resulting in a spontaneous pregnancy rate of 29% in the polypectomy group versus 3% in the control group (RR 10, 95% CI 3 to 30, P<0.00001). Data on live birth rates were not available from this trial. There were no statistically significant differences between the groups according to the size of the polyps (P>0.05). There were no data on the number or the location of the polyps.

Figure 2 Forest plot of comparison: Hysteroscopic polypectomy versus hysteroscopy and biopsy in subfertile women with ultrasonographically detected endometrial polyps undergoing intrauterine insemination. Outcome: Clinical pregnancy after 4 IUI cycles. M-H: Mantel-Haenszel, fixed- effect model, CI: confidence interval.

Polypectomy Diagnostic hysteroscopy Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Pérez Medina 2005 64 107 29 108 100.0% 2.23 [1.57, 3.15]

Total (95% CI) 107 108 100.0% 2.23 [1.57, 3.15] Total events 64 29 Heterogeneity: Not applicable 0.01 0.1 1 10 100 Test for overall effect: Z = 4.51 (P < 0.00001) Favours diagnostic only Favours polyp removal

57 Chapter 3

Hysteroscopic myomectomy We found one RCT on reproductive outcome in subfertile women treated with myomectomy versus expectant management (Casini et al, 2006). From an undetermined source population of women referred to a university fertility centre between January 1998 and April 2005, 193 women with unexplained subfertility and with an ultrasonographic diagnosis of fibroids were included in the study. All study participants were younger than 35 years and had one ‘knot’ and/or fibroid smaller than 4cm diagnosed by TVUS; we could not obtain any further clarification by the primary study authors on the definition of ‘knot’. The Italian study authors randomized 181 women to surgical treatment (laparotomy and/or operative hysteroscopy) or expectant management. Women in the intervention group were asked to abstain from having sexual intercourse for 3 months after the surgical procedure; those in the control group were allowed to have timed intercourse immediately after randomization. We analyzed only the data of women with submucous fibroids with or without intramural fibroids in the present systematic review. Data on the study characteristics are presented in table 1. It is unclear whether all women had a diagnostic hysteroscopy to confirm or exclude the presence of submucous fibroids. It is also not clear whether only intramural fibroids with uterine cavity deformation were included or not. Pregnancy was defined as the visualization of an embryo with cardiac activity at 6 to 7 weeks of pregnancy. At 12 months after randomization pregnancy rates were almost doubled in the intervention group compared to the control group; the differences between both comparison groups were not statistically significant according to our computation (RR 1.9, 95% CI 0.97 to 3.7, P=0.06) as opposed to the findings reported by the authors in the primary study publication which claim statistical significance (Figure 3). The NNTB is 5 with a wide confidence interval (95% CI 3 to 100). No significant differences in pregnancy rates were found between the two comparison groups for a subgroup analysis including subfertile women with only submucous fibroids (RR 1.6, 95% CI 0.72 to 3.5, P=0.25). Our computation yielded different P values than those reported in the primary study publication.

58 Hysteroscopy for improving pregnancy rates in subfertile women

Figure 3 Forest plot of comparison: Myomectomy versus expectant management in subfertile women with one knot* and/or one submucous fibroid with or without intramural fibroid smaller than 4 cm. Outcome: Clinical pregnancy. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

Myomectomy Expectant Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Casini 2005 21 52 9 42 100.0% 1.88 [0.97, 3.67]

Total (95% CI) 52 42 100.0% 1.88 [0.97, 3.67] Total events 21 9 Heterogeneity: Not applicable 0.01 0.1 1 10 100 Test for overall effect: Z = 1.86 (P = 0.06) Favours expectant Favours myomectomy

* The definition of knot is unclear; no further clarification could be obtained from the authors of the primary study.

Hysteroscopic metroplasty We found two RCTs on hysteroscopic metroplasty; we excluded one study because there were no published data on the reproductive outcome (Vercellini et al, 1993). The other study randomly compared two methods of hysteroscopic treatment of uterine septa (resectoscopy versus Versapoint electrode) in a mixed population of 160 women with subfertility or recurrent pregnancy loss (Colacurci et al, 2007). The study characteristics are presented in table 1. There were no statistically significant differences in clinical pregnancy rates after hysteroscopic septoplasty with the Versapoint electrode versus the classical resectoscope (RR 1.1, 95% CI 0.90 to 1.3, P=0.36). Clinical pregnancy rate was defined by rising ß-hCG levels and visualization of a gestational sac. We calculated the data on the reproductive outcome in the subgroup of women suffering from primary subfertility compared to those with recurrent pregnancy loss. There were significantly fewer pregnancies after hysteroscopic metroplasty in the subgroup of women with primary subfertility versus those with recurrent pregnancy loss (RR 0.7, 95% CI 0.5 to 0.9, P=0.01) (Figure 4). We could not study interaction (effect modification) formally because it was unknown which individual women with recurrent pregnancy loss also suffered from primary subfertility.

59 Chapter 3

Figure 4 Forest plot of comparison: Primary subfertility versus recurrent pregnancy loss. Outcome: Pregnancy. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

Primary subfertility Recurrent pregnancy loss Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Colacurci 2007 26 45 73 90 100.0% 0.71 [0.54, 0.93]

Total (95% CI) 45 90 100.0% 0.71 [0.54, 0.93] Total events 26 73 Heterogeneity: Not applicable 0.01 0.1 1 10 100 Test for overall effect: Z = 2.47 (P = 0.01) Favours subfertility Favours RPL

Hysteroscopic synechiolysis No RCTs were retrieved on pregnancy rates after hysteroscopic synechiolysis in subfertile women with intrauterine adhesions.

60 Hysteroscopy for improving pregnancy rates in subfertile women

Does diagnostic or operative hysteroscopy increase the pregnancy rate in subfertile women undergoing IVF?

There are two RCTs on the effectiveness of hysteroscopy in women treated by IVF (Demirol and Gurgan, 2004; Rama Raju et al, 2006). The quality of these studies, their characteristics, and the participant and intervention characteristics are summarized in table 2. Both studies included 941 women with at least two failed IVF attempts in two tertiary non-university IVF clinics. All women suffered from primary subfertility and had a normal uterine cavity on hysterosalpingography (HSG). They were randomly allocated to two comparison groups using computer generated random numbers. Group I (n=476) did not undergo an office hysteroscopy prior to a subsequent IVF treatment cycle. In group II (n=465) office hysteroscopy was performed with a 5mm 30° hysteroscope. Group II was subdivided into group IIa (normal findings at office hysteroscopy; n=314) and group IIb (uterine pathology detected and treated by office hysteroscopy; n=151). Clinical pregnancy was defined by the visualization of fetal heart pulsation at 6 weeks. Office hysteroscopy before a next IVF cycle significantly increased the clinical pregnancy rate compared to women where IVF was started immediately without prior hysteroscopy (RR 1.6, 95% CI 1.3 to 1.9, P<0.00001) (Figure 5). There was no statistical heterogeneity beyond chance (Chi²=0.16, df=1(P=0.69); I²=0%). The NNTB is 7 (95% CI 5 to 12). In the intervention group, there were no statistically significant differences for the main outcome between women with normal findings (n=314) and women with uterine pathology (n=151) (RR 0.91, 95% CI 0.71 to 1.2, P=0.48) (Figure 6).

Figure 5 Forest plot of comparison: Office hysteroscopy versus no hysteroscopy prior to subsequent IVF. Outcome: Clinical pregnancy. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

Prior hysteroscopy Immediate IVF attempt Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Demirol and Gurgan 2004 67 210 45 211 39.9% 1.50 [1.08, 2.07] Rama Raju 2006 108 255 69 265 60.1% 1.63 [1.27, 2.09]

Total (95% CI) 465 476 100.0% 1.57 [1.29, 1.92] Total events 175 114 Heterogeneity: Chi² = 0.16, df = 1 (P = 0.69); I² = 0% 0.01 0.1 1 10 100 Test for overall effect: Z = 4.50 (P < 0.00001) IVF without hysteroscopy Hysteroscopy prior to IVF

61 Chapter 3

Figure 6: Forest plot of comparison: Pathology present versus normal. Outcome: Clinical pregnancy. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

Abnormal Normal Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI Demirol and Gurgan 2004 17 56 50 154 33.5% 0.94 [0.59, 1.48] Rama Raju 2006 38 95 71 160 66.5% 0.90 [0.67, 1.22]

Total (95% CI) 151 314 100.0% 0.91 [0.71, 1.17] Total events 55 121 Heterogeneity: Chi² = 0.02, df = 1 (P = 0.90); I² = 0% 0.01 0.1 1 10 100 Test for overall effect: Z = 0.71 (P = 0.48) Favours normal Favours abnormal

62 Hysteroscopy for improving pregnancy rates in subfertile women

DISCUSSION

Polyps The randomized trial on the effectiveness of hysteroscopic polypectomy in subfertile women with endometrial polyps treated subsequently with IUI (Perez-Medina et al, 2005) gives apparently straightforward results. The evidence is necessary but by itself not sufficient to establish a cause- effect relationship between all endometrial polyps and subfertility. Moreover, the absence of blinding of women in this single study could have led to different sexual behavior among the study participants; therefore performance bias may be an alternative explanation for the findings of higher pregnancy rates after polypectomy, especially those observed in women who conceived spontaneously prior to IUI (Ankum, 2005). Observational studies on hysteroscopic polypectomy in subfertile women fail to present consistent results (Valle, 1984; Hereter et al, 1998; Varasteh et al, 1999; Spiewankiewicz et al, 2003; Batioglu and Kaymak, 2005; Annan and Amoah, 2006; Kassab et al, 2007). Therefore we should refrain from presenting recommendations based on the results of this single randomized trial. Several observational studies report higher pregnancy rates after the removal of tubocornual polyps suggesting that endometrial polyps in this region may have a substantial effect on reproductive function (Venturini et al, 1987; Brooks et al, 1990; Lee et al, 1997; Shokeir et al, 2004; Yanaihara et al, 2008). Other investigators have suggested that the observed association between endometrial polyps and subfertility may be alternatively explained by a confounding variable, namely concomitant endometriosis (Mc Bean et al, 1996; Kim et al, 2003).

Fibroids The impact of fibroids on fertility remains controversial (Pritts, 2001; Lefebvre et al, 2003; Vilos, 2003; Griffiths et al, 2006; Somigliana et al, 2007; Vimercati et al, 2007; Somigliana et al, 2008; Klatsky et al, 2008; Pritts et al, 2009). The only randomized trial comparing myomectomy to expectant management in women with fibroids and subfertility (Casini et al, 2006) was underpowered and not blinded. It is unclear whether hysteroscopy was used to screen systematically all the study participants. The inclusion criterion ‘knot’ is not defined properly, which is troublesome. More important, our own recalculation of the available data fails to demonstrate statistically significant differences for the pregnancy rates between both

63 Chapter 3

comparison groups. This statistical error raises some concern about the validity of the published primary data. Observational epidemiological studies aimed at elucidating the causal relationship between fibroids and subfertility are equally not conclusive (Seoud et al, 1992; Narayan and Goswamy, 1994; Farhi et al, 1995; Lumbiganon et al, 1996; Eldar-Geva et al, 1998; Marshall et al, 1998; Ramzy et al, 1998; Stovall et al, 1998; Bulletti et al, 1999; Bajekal and Li, 2000; Dietterich et al, 2000; Healy, 2000; Hart et al, 2001; Jun et al, 2001; Surrey et al, 2001; Wang et al, 2001; Check et al, 2002; Donnez and Jadoul, 2002; Ng and Ho, 2002; Yarali and Bukulmez, 2002; Bulletti et al, 2004; Manyonda et al, 2004; Oliveira et al, 2004; Parazzini et al, 2004; Wang and Check, 2004; Wise et al, 2004; Benecke et al, 2005; Gianaroli et al, 2005; Ng et al, 2005; Rackow and Arici, 2005; Surrey et al, 2005; Khalaf et al, 2006; Klatsky et al, 2007). Overall the majority of the published studies are observational, often non-controlled and retrospective, with small sample sizes and a lack of adjustment of the results for important confounding variables such as maternal age. No data on a size effect relationship were retrieved. Non-controlled studies have suggested that the number, size and distortion effect of fibroids on the uterine cavity may be important (Bulletti et al, 1999; Varasteh et al, 1999; Bernard et al, 2000; Fernandez et al, 2001; Oliveira et al, 2005, Khalaf et al, 2006; Mukhopadhaya et al, 2007). The mechanism by which fibroids interfere with human reproduction is still unknown. Fibroids are believed to interfere with sperm migration, ovum transport and embryo implantation (Richards et al, 1998). This may be caused by altered contours of the uterine cavity resulting in altered mechanical pressure or abnormal uterine contractility (Bettocchi et al, 2002; Farrugia et al, 2002; Oliveira et al, 2004). Local inflammation caused by the presence of submucous fibroids may cause focal endometrial vascular disturbances, chronic endometritis or secretion of vasoactive substances. Even an enhanced intrauterine androgen environment impairing gamete or zygote transport and embryo implantation has been suggested (Richards et al, 1998). It is biologically plausible that fibroids localized near the cervix may interfere with sperm transport whereas tubocornual fibroids may preferentially impair oocyte/embryo transport (Oliveira et al, 2004).

Septate uterus Hysteroscopic metroplasty is frequently performed in women with recurrent miscarriage because a uterine septum is associated with an adverse pregnancy outcome. However, the effectiveness of hysteroscopic metroplasty in reducing miscarriage in women with septate uterus is at the

64 Hysteroscopy for improving pregnancy rates in subfertile women present not demonstrated by RCTs. A review of non-controlled studies suggests that the pregnancy outcome is more favorable after hysteroscopic metroplasty (Homer et al, 2000). However, the results of this review are biased because the reproductive outcomes were compared before and after metroplasty using women as their own controls; doing a before-after comparison will exaggerate the treatment effect of the intervention because the reason for the intervention has been the poor reproductive outcome in the index pregnancy (Christiansen et al, 2005). Although it could be considered unethical to design a RCT on the basis that a hysteroscopic metroplasty is technically easy to perform, a randomized trial is needed, and in fact is currently underway (http://www.studies- obsgyn.nl/home/page.asp?page_id=596). The mechanism by which a septum may cause subfertility is not fully understood. It is biologically plausible that the endometrium of the septum may be unsuitable for blastocyst implantation. One author observed that the morphological development of endometrial septal specimens is suboptimal, using scanning electron microscopy (Fedele et al, 1996). The association between septate uterus and endometriosis, as reported in some non- controlled studies, may explain the subfertility of at least some women with septate uterus but requires further research (Fayez, 1986; Fedele et al, 1993; Grimbizis et al, 1998; Nawroth et al, 2006).

Intrauterine adhesions Randomized or clinical controlled studies on reproductive outcome after hysteroscopic synechiolysis are absent. The overall quality of the bulk of available observational studies on intrauterine adhesions is very poor (Sugimoto, 1978; Schenker and Margalioth, 1982; Fedele et al, 1986; Friedman et al, 1986; Parent et al, 1988; Valle and Sciarra, 1988; Pistofidis et al, 1996; Mc Comb and Wagner, 1997; Pabuçcu et al, 1997; Roge et al, 1997; Protopapas et al, 1998; Capella-Allouc et al, 1999; Feng et al, 1999; Preutthipan and Linasmita, 2000; Zikopoulos et al, 2004; Kodaman and Arici, 2007;Yu et al, 2008). Moreover, the results cannot be directly compared since different non-validated classification systems of the severity of disease are used (American Fertility Society, 1988; Wamsteker et al, 1998; Nasr et al, 2000). Subfertility in women with intrauterine adhesions may be caused by complete or partial occlusion of the tubal ostia, uterine cavity or the cervical canal, preventing the migration of sperm or the implantation of the embryo. Severe destruction of the endometrium may also lead to defective or absent implantation.

65 Chapter 3

IVF The higher pregnancy rates in a subsequent IVF cycle after hysteroscopy in women suffering from primary subfertility with at least two failed IVF attempts even in the absence of intrauterine pathology is a somewhat unexpected but nevertheless biologically plausible observation. It is acceptable that cervical dilatation and/or direct hysteroscopic visualization of the uterine cavity may facilitate embryo transfer in this population (Mc Manus et al, 2000; Mansour and Aboulghar, 2002), Moreover, an immunological mechanism triggered by the hysteroscopic manipulation or by the effect of the distension medium on the endometrium, similar to the increased odds of spontaneous pregnancy after hysterosalpingography (Luttjeboer et al, 2007), might play a role. A new randomized trial on this subject is ongoing (Geslevich et al, 2006).

66 Hysteroscopy for improving pregnancy rates in subfertile women

SUMMARY CONCLUSIONS

Implications for clinical practice In women with primary subfertility and at least two failed IVF or ICSI attempts, diagnostic or operative hysteroscopy before a subsequent IVF or ICSI treatment is thought to improve reproductive outcome. Scarce evidence on the effectiveness of hysteroscopic surgery in subfertile women with polyps, fibroids, septate uterus or intrauterine adhesions suggests a potential benefit, but it is clear that more randomized trials are needed before evidence-based recommendations for the general subfertile target population can be given.

Implications for further research Given the high costs of Assisted Reproductive Techniques (ART) it is imperative to study whether there is a benefit for doing a hysteroscopy in all women undergoing IVF or whether this benefit is limited to specific subgroups e. g. women with primary subfertility and at least two or more failed IVF treatment cycles.

67 Chapter 3

REFERENCES

1. American Fertility Society. The American Fertility Society classifications of adnexal adhesions, distal tubal occlusion, tubal occlusion secondary to tubal ligation, tubal pregnancies, müllerian anomalies and intrauterine adhesions. Fertil Steril 1988; 49: 944-955. 2. Ankum WM. Hysteroscopic removal of endometrial polyps more than doubled the pregnancy rate after intrauterine insemination. Evidence- based Obstet Gynecol 2005; 7: 193-194. 3. Annan H, Amoah C. Successful pregnancy after recurrent hysteroscopic polypectomy Gynecol Surg 2006; 3: 144-145. 4. Bajekal N, Li TC. Fibroids, infertility and pregnancy wastage. Hum Reprod Update 2000; 6: 614-620. 5. Batioglu S, Kaymak O. Does hysteroscopic polypectomy without cycle cancellation affect IVF? Reprod Biomed Online 2005; 10: 767- 769. 6. Benecke C, Kruger TF, Siebert TI, Van der Merwe JP, Steyn DW. Effect of fibroids on fertility in patients undergoing assisted reproduction: A structured literature review. Gynecol Obstet Invest 2005; 59: 225-230. 7. Bernard G, Darai E, Poncelet C, Benifla JL, Madelenat P. Fertility after hysteroscopic myomectomy: effect of intramural fibroids associated. Eur J Obstet Gynecol Reprod Biol 2000; 88: 85-90. 8. Bettocchi S, Ceci O, Di Venere R, Pansini MV, Pellegrino A, Marello F, Nappi L. Advanced operative office hysteroscopy without anaesthesia: analysis of 501 cases treated with a 5 Fr bipolar electrode. Hum Reprod 2002; 17: 2435-2438. 9. Brooks PG, Kerin JF, Daykhovsky L. Removal of cornual polyps with flexible hysteroscopy and a wire basket: A case report. J Reprod Med Obstet Gynecol 1990; 35: 613-615. 10. Bulletti C, De Ziegler D, Polli V, Diotallevi L, del Ferro E, Flamigni C. The role of leiomyomas in infertility. J Am Assoc Gynecol Laparosc 1999; 6: 441-445. 11. Bulletti C, De Ziegler D, Setti PL, Cicinelli E, Polli V, Stefanetti M. Myomas, pregnancy outcome, and in vitro fertilization. Ann N Y Acad Sci 2004; 1034: 84-92. 12. Campo R, Molinas CR, Rombauts L, Mestdagh G, Lauwers M, Braekmans P, Brosens I, Van Belle Y, Gordts S. Prospective multicentre randomized controlled trial to evaluate factors influencing the success rate of office diagnostic hysteroscopy. Hum Reprod 2005; 20: 258-263.

68 Hysteroscopy for improving pregnancy rates in subfertile women

13. Capella-Allouc S, Morsad F, Rongieres-Bertrand C, Taylor S, Fernandez H. Hysteroscopic treatment of severe Asherman’s syndrome and subsequent fertility Hum Reprod 1999; 14, 1230- 1233. 14. Casini ML, Rossi F, Agostini R, Unfer V. Effect of the position of fibroids on fertility. Gynecol Endocrinol 2006; 22: 106-109. 15. Check JH, Chloe JK, Lee G, Dietterich C. The effect on IVF outcome of small intramural fibroids not compressing the uterine cavity as determined by a prospective matched control study. Hum Reprod 2002; 17: 1244-1248. 16. Christiansen OB, Nybo Andersen AM, Bosch E, Daya S, Delves PJ, Hviid TV, Kutteh W, Laird SM, Li TC, van der Ven K. Evidence- based investigations and treatments of recurrent pregnancy loss. Fertil Steril 2005; 83: 821-839. 17. Colacurci N, De Franciscis P, Mollo A, Litta P, Perino A, Cobellis L, De Placido G. Small-diameter hysteroscopy with Versapoint versus resectoscopy with a unipolar knife for the treatment of septate uterus: a prospective randomized study. J Minim Invas Gynecol 2007; 14: 622-627. 18. Crosignani PG, Rubin BL. Optimal use of infertility diagnostic tests and treatments. Hum Reprod 2000; 15, 723-732. 19. De Angelis C, Santoro G, Re ME, Nofroni I. Office hysteroscopy and compliance: mini-hysteroscopy versus traditional hysteroscopy in a randomized trial. Hum Reprod 2003; 18: 2441-2445. 20. Demirol A, Gurgan T. Effect of treatment of intrauterine with office hysteroscopy in patients with recurrent IVF failure. Reprod Biomed Online 2004; 8: 590-594. 21. De Placido G, Clarizia R, Cadente C, Castaldo G, Romano C, Mollo A, Alviggi C, Conforti S. Compliance and diagnostic efficacy of mini- hysteroscopy versus traditional hysteroscopy in infertility investigation. Eur J Obstet Gynecol 2007; 135: 83-87. 22. Dietterich C, Check JH, Choe JK, Nazari A, Fox F. The presence of small uterine fibroids not distorting the endometrial cavity does not adversely affect conception outcome following embryo transfer in older recipients. Clin Exp Obstet Gynecol 2000; 27: 168-170. 23. Donnez J, Jadoul P. What are the implications of myomas in fertility? A need for a debate? Hum Reprod 2002; 17: 424-1430. 24. Eldar-Geva T, Meagher S, Healy DL, MacLachlan V, Breheny S, Wood C. Effect of intramural, subserosal, and submucosal uterine fibroids on the outcome of assisted reproductive technology treatment. Fertil Steril 1998; 70: 687-691.

69 Chapter 3

25. El-Toukhy T, Sunkara SK, Coomarasamy A, Grace J, Khalaf Y. Outpatient hysteroscopy and subsequent IVF cycle outcome: a systematic review and meta-analysis. RBM Online 2008; 16: 712-719. 26. Farhi J, Ashkenazi J, Feldberg D, Dicker D, Orvieto R, Ben Rafael Z. Effect of uterine leiomyomata on the results of in-vitro fertilisation treatment. Hum Reprod 1995; 10: 2576-2578. 27. Farrugia M, McGurgan P, McMillan DL, O’Donovan PJ. Recent advances in electrosurgery 2002; Chap 18: In: O’Donovan PJ, Dowens E (Eds) Advances in Gynecological surgery, 1st edn. London: Greenwich Medical Media. 28. Fayez AJ. Comparison between abdominal and hysteroscopic metroplasty. Obstet Gynecol 1986; 68: 399- 403. 29. Fedele L, Vercellini P, Viezzoli T, Ricciardiello O, Zamberletti D. Intrauterine adhesions: current diagnostic and therapeutic trends. Acta Eur Fertil 1986; 17: 31-37. 30. Fedele L, Arcaini L, Parazzini F, Vercellini P, di Nola G. Reproductive prognosis after hysteroscopic metroplasty in 102 women: life-table analysis. Fertil Steril 1993; 59: 768-772. 31. Fedele L, Bianchi S, Marchini M, Franchi D, Tozzi L, Dorta M. Ultrastructural aspects of endometrium in infertile women with septate uterus. Fertil Steril 1996; 65: 750-752. 32. Feng ZC, Yang B, Shao J, Liu S. Diagnostic and therapeutic hysteroscopy for traumatic intrauterine adhesions after induced abortions: clinical analysis of 365 cases. Gynaecol Endosc 1999; 8: 95-98. 33. Fernandez H, Sefrioui O, Virelizier C, Gervaise A, Gomel V, Frydman R. Hysteroscopic resection of submucosal fibroids in patients with infertility. Hum Reprod 2001; 16: 1489-1492. 34. Friedman A, DeFazio J, DeCherney A. Severe obstetric complications after aggressive treatment of Asherman syndrome. Obstet Gynecol 1986; 67: 864-867. 35. Garbin O, Kutnahorsky R, Göllner JL, Vayssiere C. Vaginoscopic versus conventional approaches to outpatient diagnostic hysteroscopy: a two-centre randomized prospective study. Hum Reprod 2006; 21: 2996-3000. 36. Geslevich J, Weiss A, Romano S, Goldman S, Shalev E. Endometrial curettage before embryo transfer 2006 (NCT00367367) http://www.clinicaltrials.gov/ct2/show/NCT00367367. 37. Gianaroli L, Gordts S, D’Angelo A, Magli MC, Brosens I, Cetera C, Campo R, Ferraretti AP. Effect of inner myometrium fibroid on reproductive outcome after IVF. Reprod Biomed Online 2005; 10: 473-477.

70 Hysteroscopy for improving pregnancy rates in subfertile women

38. Griffiths A, D’Angelo A, Amso N. Surgical treatment of fibroids for subfertility. Cochrane Database Syst Rev 2006; 3; CD 003857. 39. Grimbizis G, Camus M, Clasen K, Tournaye H, De Munck L, Devroey P. Hysteroscopic septum resection in patients with recurrent abortions and infertility. Hum Reprod 1998; 13: 1188-1193. 40. Guida M, Pellicano M, Zullo F, Acunzo G, Lavitola G, Palomba S, Nappi C. Outpatient operative hysteroscopy with bipolar electrode: a prospective multicentre randomised study between local anaesthesia and conscious sedation. Hum Reprod 2003; 18: 840-843. 41. Guida M, Di Spiezio Sardo A, Acunzo G, Sparice S, Bramante S, Piccoli R, Bifulco G, Cirillo D, Pellicano M, Nappi C. Vaginoscopic versus traditional office hysteroscopy: a randomized controlled study. Hum Reprod 2006; 21: 3253-3257. 42. Hart R, Khalaf Y, Yeong CT, Seed P, Taylor A, Braude P. A prospective controlled study of the effect of intramural uterine fibroids on the outcome of assisted conception. Hum Reprod 2001; 16: 2411-2417. 43. Healy DL. Impact of uterine fibroids on ART outcome. Environ Health Perspect 2000; 108 (suppl 5): 845-847. 44. Hereter L, Carreras O, Pasciial MA. Repercusion de la presencia de polipos endometriales en un cicio de FIV. Prog Obstet Ginecol 1998; 41: 5-7. 45. Homer HA, Li TC, Cooke ID. The septate uterus: a review of management and reproductive outcome. Fertil Steril 2000; 73: 1-14. 46. Jun SH, Ginsburg ES, Racowsky C, Wise LA, Hornstein MD. Uterine leiomyomas and their effect on in vitro fertilisation outcome: a retrospective study. J Assist Reprod Genet 2001; 18: 139-143. 47. Kabli N, Tulandi T. A randomized trial of outpatient hysteroscopy with and without intrauterine anesthesia. J Minim Invasive Gynecol 2008; 15: 308-310. 48. Kassab AS, Zosmer A, Gillott C, Tozer A, Al-Shawaf T. Successful pregnancy and delivery following IVF treatment in a patient with endometrial polyp diagnosed by saline infusion sonohysterography at oocyte retrieval. Middle East Fertil Soc J 2007; 12: 66-68. 49. Khalaf Y, Ross C, El-Toukhy T, Hart R, Seed P, Braude P. The effect of small intramural uterine fibroids on the cumulative outcome of assisted conception. Hum Reprod 2006; 21: 2640-2644. 50. Kim MR, Kim YA, Jo MY, Hwang KJ, Ryu HS. High frequency of endometrial polyps in endometriosis. J Am Assoc Gynecol Laparosc 2003; 10: 46-48.

71 Chapter 3

51. Klatsky PC, Lane DE, Ryan IP, Fujimoto VY. The effect of fibroids without cavity involvement on ART outcomes independent of ovarian age. Hum Reprod 2007; 22: 521-526. 52. Klatsky PC, Tran ND, Caughey AB, Fujimoto VY. Fibroids and reproductive outcomes: a systematic literature review from conception to delivery. Am J Obstet Gynecol 2008; 198: 357-366. 53. Kodaman PH, Arici A. Intra-uterine adhesions and fertility outcome: how to optimise success? Curr Opin Obstet Gynecol 2007; 19: 207- 214. 54. Kremer C, Duffy S, Moroney M. Patient satisfaction with outpatient hysteroscopy versus day case hysteroscopy: randomized controlled trial. BMJ 2000; 320: 279-282. 55. Lee A, Ying YK, Novy MJ. Hysteroscopy, hysterosalpingography and tubal ostial polyps in infertility patients. J Reprod Med Obstet Gynecol 1997; 42: 337-341. 56. Lefebvre G, Vilos G, Allaire C, Jeffrey J. The management of uterine leiomyomas. J Obstet Gynecol Can 2003; 25: 396-405. 57. Litta P, Bonora M, Pozzan C, Merlin F, Sacco G, Fracas M, Capobianco G, Dessole S. Carbon dioxide versus normal saline in outpatient hysteroscopy. Hum Reprod 2003; 18: 2446-2449. 58. Lumbiganon P, Rugpao S, Phandhu- fung S, Laopaiboon M, Vudhikamraksa N, Werawatakul Y. Protective effect of depot- medroxyprogesterone acetate on surgically treated uterine leiomyomas: a multicentre case-control study. Br J Obstet Gynaecol 1996; 103: 909-914. 59. Luttjeboer F, Harada T, Hughes E, Johnson N, Lilford R, Mol BW. Tubal flushing for subfertility. Cochrane Database Syst Rev 2007; CD 003718. 60. Mansour R, Aboulghar M. Optimizing the embryo transfer technique. Hum Reprod 2002; 17: 1149-1153. 61. Manyonda I, Sinthamoney E, Belli M. Controversies and challenges in the modern management of uterine fibroids. BJOG 2004; 111: 95- 102. 62. Marsh F, Kremer C, Duffy S. Delivering an effective outpatient service in gynecology: A randomized controlled trial analyzing the cost of outpatient versus day case hysteroscopy. BJOG 2004; 111: 243-248. 63. Marshall LM, Spiegelman D, Goldman MB, Manson JE, Colditz GA, Barbieri RL, Stampfer MJ, Hunter DJ. A prospective study of reproductive factors and oral contraceptive use in relation to the risk of uterine leiomyomata. Fertil Steril 1998; 70: 432-439.

72 Hysteroscopy for improving pregnancy rates in subfertile women

64. McBean JH, Gibson M, Brumsted JR. The association of intrauterine filling defects on hysterosalpingogram with endometriosis. Fertil Steril 1996; 66: 522-526. 65. Mc Comb PF, Wagner BL. Simplified therapy for Asherman’s syndrome. Fertil Steril 1997; 68: 1047-1050. 66. McManus J, McClure N, Traub AI. The effect of cervical dilatation in patients with previous difficult embryo transfer. Fertil Steril 2000; 74: 159. 67. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. The Lancet 1999; 354: 1896-1900. 68. Mukhopadhaya N, Pokuah Asante G, Manyonda IT. Uterine fibroids: impact on fertility and pregnancy loss. Obstet Gynecol Reprod Med 2007; 17: 311-317. 69. Muzii L, Bellati F, Pernice M, Manci N, Angioli R, Panici PB. Resectoscopic versus bipolar electrode excision of endometrial polyps: a randomized study. Fertil Steril 2007; 87: 909-917. 70. Narayan R, Goswamy RK. Treatment of submucous fibroids, and outcome of assisted reproduction. J Am Assoc Gynecol Laparosc 1994; 1: 307-311. 71. Nasr AL, Al-Inany HG, Thabet SM, Aboulghar M. A clinicohysteroscopic scoring system of intrauterine adhesions. Gynecol Obstet Invest 2000; 50, 178-181. 72. Nawroth F, Rahimi G, Nawroth C, Foth D, Ludwig M, Schmidt T. Is there an association between septate uterus and endometriosis? Hum Reprod 2006; 21: 542-544. 73. Ng EH, Ho PC. Doppler ultrasound examination of uterine arteries on the day of oocyte retrieval in patients with uterine fibroids undergoing IVF. Hum Reprod 2002; 17: 765-770. 74. Ng EH, Chan CC, Tang OS, Yeung WS, Ho PC. Endometrial and subendometrial blood flow measured by three-dimensional power Doppler ultrasound in patients with small intramural uterine fibroids during IVF treatment. Hum Reprod 2005; 20: 501-506. 75. Oliveira FG, Abdelmassih VG, Diamond MP, Dozortsev D, Melo NR, Abdelmassih R. Impact of subserosal and intramural uterine fibroids that do not distort the endometrial cavity on the outcome of in vitro fertilisation-intracytoplasmic sperm injection. Fertil Steril 2004; 81: 582-587. 76. Oliveira FG, Abdelmassih VG, Diamond MP, Dozortsev D, Melo NR, Abdelmassih R, Pritts EA. Small uterine fibroids did not negatively

73 Chapter 3

affect the success of infertility treatment with intracytoplasmic sperm injection. Evidence-based Obstet Gynecol 2005; 7: 31-32. 77. Pabuçcu R, Atay V, Orhon E, Urman B, Ergun A. Hysteroscopic treatment of intrauterine adhesions is safe and effective in the restoration of normal menstruation and fertility. Fertil Steril 1997; 68: 1141-1143. 78. Parazzini F, Chiaffarino F, Polverino G, Chiantera V, Surace M, La Vecchia C. Uterine fibroids risk and history of selected medical conditions linked with female hormones. Eur J Epidemiol 2004; 19: 249-253. 79. Parent B, Barbot J, Dubuisson JB. Synéchies uterine encyclopédie medico-chirurgicale Gynécologie 1988; 140A: 10. 80. Pellicano M, Guida M, Zullo F, Lavitola G, Cirillo D, Nappi C. Carbon dioxide versus normal saline as a uterine distension medium for diagnostic vaginoscopic hysteroscopy in infertile patients: a prospective, randomized, multicenter study. Fertil Steril 2003; 79: 418-421. 81. Pérez-Medina T, Bajo-Arenas J, Salazar F, Redondo T, Sanfrutos L, Alvarez P, Engels V. Endometrial polyps and their implication in the pregnancy rates of patients undergoing intrauterine insemination: a prospective randomized study. Hum Reprod 2005; 20: 1632-1635. 82. Pistofidis GA, Dimitropoulos K, Mastrominas M. Comparison of operative and fertility outcome between groups of women with intrauterine adhesions after adhesiolysis. J Am Assoc Gynecol Laparosc 1996; 3: S 40. 83. Preutthipan S, Linasmita V. Reproductive outcome following hysteroscopic lysis of intrauterine adhesions: a result of 65 cases at Ramathibodi Hospital. J Med Assoc Thai 2000; 83: 42-46. 84. Pritts EA. Fibroids and infertility: a systematic review of the evidence. Obstet Gynecol Surv 2001; 56: 483-491 85. Pritts EA, Parker WH, Olive DL. Fibroids and infertility: an updated systematic review of the evidence. Fertil Steril 2009; 91: 1215-1223. 86. Protopapas A, Shushan A, Magos A. Myometrial scoring: a new technique for the management of severe Asherman’s syndrome. Fertil Steril 1998; 69: 860-864. 87. Rackow BW, Arici A. Fibroids and in-vitro fertilisation: which comes first? Curr Opin Obstet Gynecol 2005; 15: 225-231. 88. Rama Raju GA, Shashi Kumari G, Krishna KM, Prakash GJ, Madan K. Assessment of uterine cavity by hysteroscopy in assisted reproduction programme and its influence on pregnancy outcome. Arch Gynecol Obstet 2006; 274: 160-164.

74 Hysteroscopy for improving pregnancy rates in subfertile women

89. Ramzy AM, Sattar M, Amin Y, Mansour RT, Serour GI, Aboulghar MA. Uterine myomata and outcome of assisted reproduction. Hum Reprod 1998; 13: 198-202. 90. Richards PA, Richards PD, Tiltman AJ. The ultrastructure of fibromyomatous myometrium and its relationship to infertility. Hum Reprod Update 1998; 4: 520-525. 91. Roge P, Cravello L, D’Ercole C, Brousse M, Boubli L, Blanc B. Intrauterine adhesions and fertility: results of hysteroscopic treatment. Gynaecol Endosc 1997; 6: 225-228. 92. Royal College of Obstetricians and Gynecologists Evidence-based Clinical Guidelines. Fertility assessment and treatment for people with fertility problems 2004, RCOG website. 93. Sagiv R, Sadan O, Boaz M, Dishi M, Schechter E, Golan A. A new approach to office hysteroscopy compared with traditional hysteroscopy. Obstet Gynecol 2006; 108: 387-392. 94. Schenker JG, Margalioth EJ. Intrauterine adhesions: an updated appraisal. Fertil Steril 1982; 37: 593-610. 95. Seracchioli R, Rossi S, Govoni F, Rossi E, Venturoli S, Bulletti C, Flamigni C. Fertility and obstetric outcome after laparoscopic myomectomy of large myomata: a randomised comparison with abdominal myomectomy. Hum Reprod 2000; 15: 2663-2668. 96. Seoud M, Patterson R, Muasher S, Coddington C. Effects of myomas or prior myomectomy on in vitro fertilisation (IVF) programme. J Assist Reprod genet 1992; 9: 217-221. 97. Shankar M, Davidson A, Taub N, Habiba M. Randomised comparison of distension media for outpatient hysteroscopy. BJOG 2004; 111: 57- 62. 98. Sharma M, Taylor A, Di Spiezio Sardo A, Buck L, Mastrogamvrakis G, Kosmas I, Tsirkas P, Magos A. Outpatient hysteroscopy: traditional versus the ‘no-touch’ technique. BJOG 2005; 112: 963- 967. 99. Shokeir TA, Shalan HM, El-Shafei MM. Significance of endometrial polyps detected hysteroscopically in eumenorrheic infertile women. J Obstet Gynaecol Res 2004; 30: 84-89. 100. Shushan A, Rojansky N. Should hysteroscopy be part of the basic infertility work-up? Hum Reprod 1999; 14: 1923-1924. 101. Somigliana E, Vercellini P, Daguati R, Pasin R, De Giorgi O, Crosignani PG. Fibroids and female reproduction: a critical analysis of the evidence. Hum Reprod Update 2007; 13: 465-476. 102. Somigliana E, Vercellini P, Benaglia L, Abbiati A, Barbara G, Fedele L. The role of myomectomy in fertility enhancement. Curr Opin Obstet Gynecol 2008; 20: 379-385.

75 Chapter 3

103. Soriano D, Ajaj S, Chuong T, Deval B, Fauconnier A, Daraï E. Lidocaine spray and outpatient hysteroscopy: Randomised placebo- controlled trial. Obstet Gynecol 2000; 96: 661-664. 104. Spiewankiewicz B, Stelmachów J, Sawicki W, Cendrowski K, Wypych P, Swiderska K. The effectiveness of hysteroscopic polypectomy in cases of female infertility. Clin Exp Obstet Gynecol 2003; 30: 23-25. 105. Stovall DW, Parrish SB, Van Voorhis BJ, Hahn SJ, Sparks AE, Syrop CH. Uterine leiomyomas reduce the efficacy of assisted reproduction cycles: results of a matched follow-up study. Hum Reprod 1998; 13: 192-197. 106. Sugimoto O. Diagnostic and therapeutic hysteroscopy for traumatic intrauterine adhesions. Am J Obstet Gynecol 1978; 131: 539-547. 107. Surrey ES, Lietz AK, Schoolcraft WB. Impact of intramural leiomyomata in patients with a normal endometrial cavity on in vitro fertilisation-embryo transfer cycle outcome. Fertil Steril 2001; 75: 405-410. 108. Surrey ES, Minjarez DA, Stevens JM, Schoolcraft WB. Effect of myomectomy on the outcome of assisted reproductive technologies. Fertil Steril 2005; 83: 1473-1479. 109. Unfried G, Wieser F, Albrecht A, Kaider A, Nagele F. Flexible versus rigid endoscopes for outpatient hysteroscopy: a prospective randomized clinical trial. Hum Reprod 2001; 16: 168-171. 110. Valle . Therapeutic hysteroscopy in infertility. Int J Fertil 1984; 29:143-148. 111. Valle RF, Sciarra JJ. Intrauterine adhesions: hysteroscopic diagnosis, classification, treatment, and reproductive outcome. Am J Obstet Gynecol 1988; 158: 1459-1470. 112. Varasteh NN, Neuwirth RS, Levin B, Keltz MD. Pregnancy rates after hysteroscopic polypectomy and myomectomy in infertile women. Obstet Gynecol 1999; 94: 168-171. 113. Venturini N, Tantini C, Bargilli G. Hysteroscopy for evaluation of tubal ostium pathology. Acta Europea Fertilitatis 1987; 18: 61-62. 114. Vercellini P, Vendola N, Colombo A, Passadore C, Trespidi L, Fedele L. Hysteroscopic metroplasty with resectoscope or microscissors for the correction of septate uterus. Surg Gynecol Obstet 1993; 176: 439-442. 115. Vilos G. Uterine fibroids: relationship to reproduction. Minerva Ginecol 2003; 55: 417-423.

76 Hysteroscopy for improving pregnancy rates in subfertile women

116. Vimercati A, Scioscia M, Lorusso F, Laera AF, Lamanna G, Coluccia A, Bettocchi S, Selvaggi L, Depalo R. Do uterine fibroids affect IVF outcomes? Reprod Biomed Online 2007; 15: 686-691. 117. Wamsteker K, De Block S. Diagnostic hysteroscopy: technique and documentation In: Sutton C, Diamond M, 1998 Eds. Endoscopic Surgery for Gynecologists London: WB Saunders, 511-524. 118. Wang W, Check JH, Dietterich C, Lurie D. Effect of fibroids on cumulative probability of pregnancy in women taking follicle maturing drugs without assisted reproductive technology. Clin Exp Obstet Gynecol 2001; 28: 86-88. 119. Wang W, Check JH. Effect of corporal fibroids on outcome following embryo transfer in donor-oocyte recipients. Clin Exp Obstet Gynecol 2004; 31: 263-264. 120. Wise LA, Palmer JR, Harlow BL, Spiegelman D, Stewart EA, Adams-Campbell LL, Rosenberg L. Reproductive factors, hormonal contraception, and risk of uterine leiomyomata in African-American women: a prospective study. Am J Epidemiol 2004; 159: 113-123. 121. Yanaihara A, Yorimitsu T, Motoyama H, Iwasaki S, Kawamura T. Location of endometrial polyp and pregnancy rate in infertility patients. Fertil Steril 2008; 90: 180-182. 122. Yarali H, Bukulmez O. The effect of intramural and subserous uterine fibroids on implantation and clinical pregnancy rates in patients having intracytoplasmic sperm injection. Arch Gynecol Obstet 2002; 266: 30-33. 123. Yu D, Wong YM, Cheong Y, Xia E, Li TC. Asherman syndrome- one century later. Fertil Steril 2008; 89: 759-779. 124. Zikopoulos KA, Kolibianakis EM, Platteau P, de Munck L, Tournaye H, Devroey P, Camus M. Live delivery rates in subfertile women with Asherman’s syndrome after hysteroscopic adhesiolysis using the resectoscope or the Versapoint. Reprod Biomed Online 2004; 8: 720-725.

77

Hysteroscopy for treating subfertility associated with suspected major uterine cavity anomalies

Jan Bosteels, Jenneke C. Kasius, Steven Weyers, Frank J. M. Broekmans, Ben Willem Mol, Thomas M. D’Hooghe

The Cochrane Library, Cochrane Database of Systematic Reviews 2013, Issue 1. Art.No.:CD009461 http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD009461.pub2/pdf

79

Hysteroscopy for suspected major uterine cavity abnormalities

ABSTRACT

Background Observational studies suggest higher pregnancy rates after the hysteroscopic removal of endometrial polyps, submucous fibroids, uterine septum or intrauterine adhesions.

Objectives To assess the effects of the hysteroscopic removal of endometrial polyps, submucous fibroids, uterine septum or intrauterine adhesions suspected on ultrasound, hysterosalpingography, diagnostic hysteroscopy or any combination of these methods in women with otherwise unexplained subfertility or prior to intrauterine insemination (IUI), in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI).

Search strategy We searched the Cochrane Menstrual Disorders and Subfertility Specialized Register (6 August 2012), the Cochrane Central Register of Controlled Trials (The Cochrane Library 2012, Issue 7), MEDLINE (1950 to October 2012), EMBASE (1974 to October 2012), CINAHL (from inception to October 2012) and other electronic sources of trials including trial registers, sources of unpublished literature and reference lists. We handsearched the American Society for Reproductive Medicine (ASRM) conference abstracts and proceedings (from January 2008 to October 2012) and we contacted experts in the field.

Selection criteria Randomized comparisons between operative hysteroscopy versus control in women with otherwise unexplained subfertility or undergoing IUI, IVF or ICSI and suspected major uterine cavity abnormalities. Primary outcomes were live birth and hysteroscopy complications. Secondary outcomes were pregnancy and miscarriage.

Data collection and analysis Two authors independently assessed studies for inclusion and risk of bias, and extracted data. We contacted study authors for additional information.

Main results Two studies met the inclusion criteria. Neither reported the primary outcomes of live birth and complications from the procedure. In women with otherwise unexplained subfertility and submucous fibroids, there is no evidence of benefit with hysteroscopic myomectomy compared to

81 Chapter 4

regular fertility-oriented intercourse during 12 months for clinical pregnancy (OR 2.4, 95% CI 0.97 to 6.2, P=0.06, 94 women) and miscarriage (OR 0.58, 95% CI 0.12 to 2.9, P=0.50, 30 clinical pregnancies, both very low-quality evidence). The hysteroscopic removal of polyps prior to IUI increases the odds of clinical pregnancy: the experimental event rate (EER) was 63% compared to the control event rate (CER) of 28% (OR 4.4, 95% CI 2.5 to 8.0, P< 0.00001, 204 women, high- quality evidence).

Authors’ conclusions Hysteroscopic myomectomy might increase the odds of clinical pregnancy in women with unexplained subfertility and submucous fibroids, but the evidence is at present not conclusive. The hysteroscopic removal of endometrial polyps suspected on ultrasound in women prior to IUI might increase the clinical pregnancy rate. More randomized studies are needed to substantiate the effectiveness of the hysteroscopic removal of suspected endometrial polyps, submucous fibroids, uterine septum or intrauterine adhesions in women with unexplained subfertility or prior to IUI, IVF or ICSI.

82 Hysteroscopy for suspected major uterine cavity abnormalities

BACKGROUND

Description of the condition Subfertility is “a disease of the reproductive system defined by the failure to achieve a clinical pregnancy after 12 months or more of regular unprotected sexual intercourse” according to the International Committee for Monitoring Assisted Reproductive Technology (ICMART) and the World Health Organization (WHO) revised glossary of assisted reproductive technology (ART) (Zegers-Hochschild 2009). It is estimated that 72,400,000 women are subfertile and that 40,500,000 of these are currently seeking fertility treatment (Boivin 2007). Unexplained subfertility usually refers to a diagnosis (or lack of diagnosis) made in couples in whom all the standard investigations such as tests of ovulation, tubal patency and semen analysis are normal: it can be found in as many as 30% to 40% of subfertile couples (Ray 2012). The evaluation of the uterine cavity is a basic step in the investigation of subfertile women since the uterine cavity and its inner layer, the endometrium, are assumed to be important for the implantation of the human blastocyst. Nevertheless, the complex mechanisms leading to successful implantation are still poorly understood (Taylor 2008). Despite the huge investment in research and developments of the technologies and biology involved in medically assisted reproduction (MAR), the maximum implantation rate per embryo transferred still remains only 30% (Andersen 2008). The different phases of the implantation process are established by the complex interplay between the blastocyst and the endometrium (Singh 2011). Major uterine cavity abnormalities can be found in 10% to 15% of women seeking treatment for subfertility; they usually consist of the presence of excessive normal uterine tissue (Wallach 1972). The most common acquired uterine cavity abnormality is an endometrial polyp. This benign, endometrial stalk-like mass protrudes into the uterine cavity and has its own vascular supply. Depending on the population under study and the applied diagnostic test, endometrial polyps can be found in 1% to 41% of the subfertile population (Silberstein 2006). A fibroid is an excessive growth originating from the muscular part of the uterine cavity. Fibroids are present in 2.4% of subfertile women without any other obvious cause of subfertility (Donnez 2002). A submucous fibroid is located underneath the endometrium and is thought to interfere with fertility by deforming the uterine cavity. Intrauterine adhesions are fibrous tissue strings connecting parts of the uterine wall. They are commonly caused by inflammation or iatrogenic tissue damage (for example suction curettage after miscarriage) and are present in 0.3% to 14% of subfertile women (Fatemi 2010). A

83 Chapter 4

septate uterus is a congenital malformation in which the longitudinal band separating the left and right Müllerian ducts, which form the uterus in the human female fetus, has not been entirely resorbed. A uterine septum is present in 1% to 3.6% of women with otherwise unexplained subfertility (Saravelos 2008). Ultrasonography (US), preferably transvaginally (TVUS), is used to screen for possible endometrium or uterine cavity abnormalities in the work-up of subfertile patients. This evaluation can be expanded with hysterosalpingography (HSG), saline infusion/gel instillation sonography (SIS/GIS) and diagnostic hysteroscopy. Diagnostic hysteroscopy is generally considered as being the gold standard procedure for the assessment of the uterine cavity since it enables direct visualization; moreover, treatment of intrauterine pathology can be done in the same setting (Bettocchi 2004). Nevertheless, even for experienced gynecologists the hysteroscopic diagnosis of the major uterine cavity abnormalities may be problematic (Kasius 2011).

Description of the intervention Hysteroscopy is performed for the evaluation, or for the treatment of the uterine cavity, tubal ostia and endocervical canal in women with uterine bleeding disorders, Müllerian tract anomalies, retained intrauterine contraceptive devices or other foreign bodies, retained products of conception, desire for sterilization, recurrent miscarriage and subfertility. Most diagnostic and many operative procedures can be done in an office setting using local anesthesia and fluid distension media, while more complex procedures are generally performed as day surgery under general anesthesia (Clark 2005). Operative hysteroscopic procedures require a complex instrumentation set-up, special training of the surgeon and knowledge and skills in the adequate management of complications (Campo 1999). Complications from hysteroscopy are rare; a multicentre study including 13,600 diagnostic and operative hysteroscopic procedures performed in 82 centers reported a complication rate of 0.28% (Jansen 2000).

How the intervention might work It is plausible that major uterine cavity abnormalities may interfere with the factors that regulate the blastocyst-endometrium interplay, for example hormones and cytokines, precluding the possibility of pregnancy. Many hypotheses have been formulated in the literature of how endometrial polyps (Shokeir 2004; Silberstein 2006; Taylor 2008; Yanaihara 2008), submucous fibroids (Pritts 2001; Somigliana 2007; Taylor 2008), intrauterine adhesions (Yu 2008) and uterine septum

84 Hysteroscopy for suspected major uterine cavity abnormalities

(Fedele 1996) are likely to disturb the implantation of the human embryo; nevertheless, the precise mechanisms of action by which each one of these major uterine cavity abnormalities affects the essential reproductive process of embryo implantation are poorly understood. For endometrial polyps, submucous fibroids, intrauterine adhesions and uterine septum, observational studies have suggested an improvement in the spontaneous pregnancy rate after the hysteroscopic removal of the abnormality (Taylor 2008). The chance for pregnancy is significantly lower in subfertile women with submucous fibroids compared to other causes of subfertility according to a systematic review and meta-analysis of 11 observational studies (Pritts 2001; Pritts 2009). Three observational studies found a major benefit for removing a uterine septum by hysteroscopic metroplasty in subfertile women with a uterine septum (Mollo 2009; Shokeir 2011; Tomaževiþ 2010).

OBJECTIVES

To assess the effects of the hysteroscopic removal of endometrial polyps, submucous fibroids, uterine septum or intrauterine adhesions suspected on ultrasound, hysterosalpingography, diagnostic hysteroscopy or any combination of these methods in women with otherwise unexplained subfertility or prior to intrauterine insemination (IUI), in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI).

METHODS

Criteria for considering studies for this review

Types of studies We included only trials that were either clearly randomized or claimed to be randomized.

Types of participants • Women of reproductive age with otherwise unexplained subfertility and endometrial polyps, submucous fibroids, septate uterus or intrauterine adhesions. • Women of reproductive age with subfertility, undergoing IUI, IVF or ICSI with endometrial polyps, submucous fibroids, septate uterus or intrauterine adhesions.

85 Chapter 4

Types of interventions • Randomized comparison between operative hysteroscopy versus control in women with otherwise unexplained subfertility and suspected major uterine cavity abnormalities. • Randomized comparison between operative hysteroscopy versus control in women undergoing IUI, IVF or ICSI with suspected major uterine cavity abnormalities.

Types of outcome measures

*Primary outcomes

1. Effectiveness: live birth, defined as a delivery of a live fetus after 20 completed weeks of gestational age that resulted in at least one live baby born. The delivery of a singleton, twin or multiple pregnancies was counted as one live birth (Zegers-Hochschild 2009).

2. Adverse events: hysteroscopy complications, defined as any complication due to hysteroscopy.

*Secondary outcomes

3. Pregnancy • Ongoing pregnancy, defined as a pregnancy surpassing the first trimester or 12 weeks of pregnancy. • Clinical pregnancy, defined as a pregnancy diagnosed by US visualization of one or more gestational sacs or definitive clinical signs of pregnancy (Zegers-Hochschild 2009).

4. Adverse events: miscarriage, defined as the spontaneous loss of a clinical pregnancy before 20 completed weeks of gestation, or if gestational age is unknown a fetus with a weight of 500 g or less (Zegers- Hochschild 2009).

Search methods for identification of studies

Electronic searches We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2012, Issue 7), the Menstrual Disorders and Subfertility Group (MDSG) Specialized Register, MEDLINE using Pub Med (1950 to 27 October 2012) and EMBASE using EMBASE.com (1974 to 27 October 2012). The search was not

86 Hysteroscopy for suspected major uterine cavity abnormalities limited by language, year of publication or document format. We also searched other electronic sources of trials in the Cochrane Database of Systematic Reviews (CDSR), the Database of Abstracts of Reviews of Effects (DARE) and the Health Technology Assessment Database (HTA Database) through the Centre for Reviews and Dissemination, National Guideline Clearinghouse, ISI Web of Knowledge, CINAHL and trial registers for ongoing and registered trials (Current Controlled Trials and the World Health Organization International Clinical Trials Registry).

Searching other resources We handsearched the reference lists of reviews, guidelines, included and excluded studies and other related articles for additional eligible studies. We handsearched the American Society for Reproductive Medicine (ASRM) conference abstracts and proceedings. Moreover we contacted European experts in the field of hysteroscopic surgery.

Data collection and analysis

Selection of studies, data extraction and management We scanned titles and abstracts from the searches and obtained the full text of those articles that appeared to be eligible for inclusion. Two review authors independently assessed the studies that appeared to meet the inclusion criteria and extracted data by using a pilot-tested data extraction form. Any disagreements between the review authors were resolved by discussion or by a third review author. We contacted the authors of the primary study reports for clarification when required.

Assessment of risk of bias in included studies Two authors independently assessed the risk of bias of the included studies by using the Cochrane ‘Risk of bias’ tool.

Measures of treatment effect We used the numbers of events in both comparison groups of each study for the dichotomous data for live birth, pregnancy, miscarriage and hysteroscopy complications to calculate Mantel-Haenszel (M-H) odds ratios (OR). We presented 95% confidence intervals (95% CI) for all outcomes.

Unit of analysis issues All primary and secondary outcomes were expressed as per woman randomized; per pregnancy data were included for the outcome

87 Chapter 4

miscarriage. Multiple live births and multiple pregnancies were counted as one live birth or one pregnancy event.

Dealing with missing data We aimed to analyze the data on an intention-to-treat basis. We tried to obtain as much missing data as possible from the original investigators.

Assessment of heterogeneity Due to the limited number of included studies no formal assessment of heterogeneity by using the I² statistic combined with the Q-statistic was done.

Assessment of reporting biases We planned to minimize the potential impact of publication bias, reporting bias and within-study reporting bias by ensuring a comprehensive search for eligible studies and by being alert in identifying duplication of data. Due to the low number of included studies we did not study reporting bias formally by a funnel plot.

Data synthesis One review author carried out the statistical analysis of the data using Review Manager 5. Meta-analysis was not possible due to the limited number of included studies; therefore we presented a narrative overview as prespecified in the protocol of this Cochrane review (Bosteels 2011).

88 Hysteroscopy for suspected major uterine cavity abnormalities

RESULTS

Description of studies

Results of the search See: Figure 1. PRISMA study flow diagram.

89 Chapter 4

Included studies

*Study design and setting

Two parallel-design randomized controlled trials were included in the review. Both were single-centre studies, one conducted in Italy (Casini 2006) and the other in Spain (Pérez-Medina 2005).

*Participants

One study (Casini 2006) included 94 women with submucous fibroids and otherwise unexplained subfertility. The number of study participants was 52 in the intervention and 42 in the control group. The mean age was 31 years (range 29 to 34 years) in the subgroup of women with submucous fibroids only and 32 years (range 30 to 35 years) in the subgroup with mixed intramural-submucous fibroids. All women underwent a complete fertility assessment. Transvaginal ultrasonography was done to detect uterine fibroids. All women with fibroids and no other causes of subfertility were invited to participate in the study. Only women aged”35 years with subfertility for at least one year and one fibroid of diameter”40mm were selected for randomization. The second study (Pérez-Medina 2005) included 215 women with unexplained, male or female factor subfertility for at least 24 months bound to undergo IUI with a sonographic diagnosis of endometrial polyps. The number of women was 101 in the intervention and 103 in the control group; 11 study participants were lost to follow-up, six in the intervention and five in the control group. The mean age was 31 years (range 27 to 35 years). All women had primary subfertility and underwent a complete fertility assessment. The sonographic diagnosis of endometrial polyps was established by the visualization of the vascular stalk of the polyp using color Doppler. Women older than 39 years of age or with anovulation, uncorrected tubal disease, previous unsuccessful use of recombinant FSH or a male partner with azoospermia were excluded.

*Interventions

In one trial (Casini 2006) the intervention group underwent hysteroscopic removal of the fibroids. After three months of sexual abstinence these women were asked to have regular fertility-oriented intercourse. Women in the control group were asked to start having regular fertility oriented intercourse immediately. Both groups were monitored for up to 12 months after study commencement.

90 Hysteroscopy for suspected major uterine cavity abnormalities

In the second trial (Pérez-Medina 2005) all hysteroscopic interventions were done in an outpatient office setting under local anesthesia by one gynecologist. In the intervention group the endometrial polyps suspected on Doppler ultrasound were removed by hysteroscopy and sent for histopathological examination. The women of the intervention group were scheduled to receive four cycles of IUI using subcutaneous injections of recombinant follicle stimulating hormone (FSH) 50 IU daily from the third day of the cycle, starting three cycles after the operative hysteroscopy. In the control group the endometrial polyps suspected on Doppler ultrasound were left in place during diagnostic hysteroscopy; polyp biopsy was taken to establish a histopathological diagnosis. Ovarian stimulation was scheduled three cycles after the diagnostic hysteroscopy following the same regimen as in the intervention group. Four IUI cycles were attempted before finishing the trial.

*Outcomes

Neither of the two included studies reported data on the primary outcomes.

Excluded studies

We excluded 20 trials on hysteroscopic interventions for various reasons: see the PRISMA flow diagram (Figure 1). One study (Shokeir 2010) was retracted at the request of the editor of the publishing journal; another study (Pabuccu 2008) used an inadequate sequence generation based on the order of study entry. We excluded 18 studies for not addressing the prespecified PICO research questions of this Cochrane review (Acunzo 2003; Aghahosseini 2012; Amer 2010; Colacurci 2007; Darwish 2008; De Iaco 2003; Demirol 2004; Di Spiezio Sardo 2011; El-Nashar 2011; Guida 2004; Lieng 2010a; Muzii 2007; Parsanezhad 2006; Rama Raju 2006; Shawki 2010; Tonguc 2008; van Dongen 2008; Vercellini 1993).

Studies awaiting classification

Two trials are awaiting classification (Pansky 2009; Trniniü-Pjeviü 2011).

Ongoing studies

Six trials are ongoing (Broekmans 2010; El-Khayat 2012; El-Toukhy 2009; Maramazi 2012; Revel 2011; Sohrabvand 2012).

91 Chapter 4

Risk of bias in included studies

Figure 2 ‘Risk of bias’ summary: review authors’ judgments about each risk of bias item for each included study.

Random sequence generation bias) (selection generation sequence Random bias) (selection concealment Allocation bias) (performance and personnel Blinding of participants (detection bias) Blinding of outcome assessment bias) (attrition data outcome Incomplete bias) (reporting reporting Selective Other bias

Casini 2006 + ? ? ? + ? ?

Pérez-Medina 2005 + + ? ? + ? +

Figure 3 ’Risk of bias’ graph: review authors’ judgments about each risk of bias item presented as percentages across all included studies.

Random sequence generation (selection bias)

Allocation concealment (selection bias)

Blinding of participants and personnel (performance bias)

Blinding of outcome assessment (detection bias)

Incomplete outcome data (attrition bias)

Selective reporting (reporting bias)

Other bias

0% 25% 50% 75% 100%

Low risk of bias Unclear risk of bias High risk of bias

92 Hysteroscopy for suspected major uterine cavity abnormalities

Random sequence generation We judged both studies (Casini 2006; Pérez-Medina 2005) to be at low risk of selection bias related to random sequence generation: both used computerized random numbers tables.

Allocation concealment We judged one study (Pérez-Medina 2005) to be at low risk for selection bias related to allocation concealment, as sequentially numbered, opaque, sealed envelopes were used. We judged the second trial (Casini 2006) to be at an unclear risk since the used method was not fully described.

Blinding The risk of bias items ’blinding of participants and personnel’ and ’blinding of outcome assessors’ were judged to be at unclear risk of bias for both studies; we were undetermined to score these items either at high risk (complete follow-up with ‘hard’ outcomes) or at low risk of bias (no blinding). The absence of blinding of women in both studies could have led to differences in sexual behavior, especially in the women in the control groups of both studies. We cannot exclude that bias related to absence of blinding might be an alternative explanation for the findings of higher pregnancy rates after myomectomy and polypectomy.

Incomplete outcome data We judged both studies to be at low risk of attrition bias: one study (Casini 2006) reported outcome data of all randomized women and in the second study (Pérez-Medina 2005) the missing outcome data in 5% of the participants were balanced in numbers with similar reasons for missing data.

Selective reporting We judged both studies (Casini 2006; Pérez-Medina 2005) to be at unclear risk of reporting bias because they both failed to include data for live birth: reasonably this primary outcome should have been reported in studies conducted over a seven-year (Casini 2006) and a four-year (Pérez- Medina 2005) period.

Other potential sources of bias We judged one study to be at unclear risk for this item due to potential imbalance in the baseline characteristics between both comparison groups (Casini 2006); we judged the second study (Pérez-Medina 2005) to be at low risk.

93 Chapter 4

Effects of interventions

Table 1 1. Operative hysteroscopy compared with control for unexplained subfertility and suspected major uterine cavity abnormalities-submucous fibroids Patient or population: women with submucous fibroids and otherwise unexplained subfertility Settings: infertility centre in Rome, Italy Intervention: hysteroscopic removal of one submucous fibroid”40 mm Comparison: regular fertility-oriented intercourse Outcomes Illustrative comparative Relative N Quality risks* (95% CI) effect participant of the Assumed Corresponding (95% CI) (studies) evidence risk risk GRADE Control Myomectomy Clinical Study population OR 2.44 94 ْٚٚ pregnancy (0.97 to (1 study) ٚ 1 214 399 per 1000 ultrasound per1000 (209 to 627) 6.17) very Follow-up: 12 low2-4 months Miscarriage Study population OR 1.54 94 ْٚٚ 5 ultrasound 119 172 per 1000 (0.71 to (1 study) ٚ Follow-up: 12 per 1000 (88 to 403) 5.00) very months low2-4 *The basis for the assumed risk is the control event rate in the primary study. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; OR: Odds ratio; GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate. 1 A clinical pregnancy was defined by the visualization of an embryo with cardiac activity at 6-7 weeks of pregnancy 2 Unclear concealment of allocation 3 Wide confidence intervals 4 Unclear selective reporting and unclear whether there was imbalance in the baseline characteristics 5 Miscarriage was defined by the clinical loss of an intrauterine pregnancy between the 7th and 12th weeks of gestation

94 Hysteroscopy for suspected major uterine cavity abnormalities

Endometrial polyps No studies were retrieved.

Submucous fibroids We retrieved one study (Casini 2006). See table 1.

*Primary outcomes

1.1. Live birth There were no data for this primary outcome.

1.2. Adverse events: hysteroscopy complications There were no data for this primary outcome.

*Secondary outcomes

1.3. Clinical pregnancy In women with otherwise unexplained subfertility for at least one year and one submucous fibroid of diameter”40 mm, there is no evidence of benefit with hysteroscopic myomectomy compared to regular fertility- oriented intercourse for the secondary outcome of clinical pregnancy. The odds ratio (OR) was 2.4, 95% confidence interval (CI) 0.97 to 6.2, P=0.06, 1 study, 94 women) (Figure 4).

95 Chapter 4

Figure 4 Forest plot of comparison: Hysteroscopic myomectomy versus regular fertility-oriented intercourse in women with unexplained subfertility and submucous fibroids. Outcome: Clinical pregnancy. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

Operative hysteroscopy Control Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI 1.1.1 Removal of submucous fibroids only vs regular fertility-oriented intercourse Casini 2006 13 30 6 22 66.2% 2.04 [0.62, 6.66] Subtotal (95% CI) 30 22 66.2% 2.04 [0.62, 6.66] Total events 13 6 Heterogeneity: Not applicable Test for overall effect: Z = 1.18 (P = 0.24)

1.1.2 Removal of mixed submucous-intramural fibroids vs regular fertility-oriented intercourse Casini 2006 8 22 3 20 33.8% 3.24 [0.72, 14.57] Subtotal (95% CI) 22 20 33.8% 3.24 [0.72, 14.57] Total events 8 3 Heterogeneity: Not applicable Test for overall effect: Z = 1.53 (P = 0.13)

Total (95% CI) 52 42 100.0% 2.44 [0.97, 6.17] Total events 21 9 Heterogeneity: Chi² = 0.22, df = 1 (P = 0.64); I² = 0% 0.01 0.1 1 10 100 Test for overall effect: Z = 1.89 (P = 0.06) Favours intercourse Favours myomectomy Test for subgroup differences: Chi² = 0.22, df = 1 (P = 0.64), I² = 0%

1.4. Adverse events: miscarriage

There is no evidence of benefit with the surgical intervention compared to regular fertility- oriented intercourse for the secondary outcome of miscarriage (OR 0.58, 95% CI 0.12 to 2.9, P=0.50, 1 study, 30 clinical pregnancies in 94 women) (Figure 5).

96 Hysteroscopy for suspected major uterine cavity abnormalities

Figure 5 Forest plot of comparison: Hysteroscopic myomectomy versus regular fertility-oriented intercourse in women with unexplained subfertility and submucous fibroids. Outcome: Miscarriage. M-H: Mantel- Haenszel, fixed-effect model, CI: confidence interval.

Operative hysteroscopy Control Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI 1.2.1 Removal of submucous fibroids only vs regular fertility-oriented intercourse Casini 2006 5 13 3 6 63.5% 0.63 [0.09, 4.40] Subtotal (95% CI) 13 6 63.5% 0.63 [0.09, 4.40] Total events 5 3 Heterogeneity: Not applicable Test for overall effect: Z = 0.47 (P = 0.64)

1.2.2 Removal of mixed submucous-intramural fibroids vs regular fertility-oriented intercourse Casini 2006 4 8 2 3 36.5% 0.50 [0.03, 7.99] Subtotal (95% CI) 8 3 36.5% 0.50 [0.03, 7.99] Total events 4 2 Heterogeneity: Not applicable Test for overall effect: Z = 0.49 (P = 0.62)

Total (95% CI) 21 9 100.0% 0.58 [0.12, 2.85] Total events 9 5 Heterogeneity: Chi² = 0.02, df = 1 (P = 0.90); I² = 0% 0.01 0.1 1 10 100 Test for overall effect: Z = 0.67 (P = 0.50) Favours intercourse Favours myomectomy Test for subgroup differences: Chi² = 0.02, df = 1 (P = 0.90), I² = 0%

*Subgroup analyses

There is no evidence for a treatment effect favoring the hysteroscopic removal of one submucous fibroid”40 mm in women with otherwise unexplained subfertility compared to regular fertility-oriented intercourse for the secondary outcome clinical pregnancy in the ‘submucous only’ subgroup (OR 2.0, 95% CI 0.62 to 6.7, P=0.24, 1 study 52 women) or the ‘mixed submucous-intramural’ subgroup (OR 3.2, 95% CI 0.72 to 15, P=0.13, 1 study, 42 women); the test for subgroup differences demonstrated no significant difference (P=0.64). There were no differences for the secondary outcome miscarriage between the ‘submucous only’ subgroup (OR 0.63, 95% CI 0.09 to 4.4, P=0.64, 1 study, 19 clinical pregnancies in 52 women) versus the ’mixed submucous-intramural’ subgroup (OR 0.50, 95% CI 0.03 to 8.0, P=0.62, 1 study, 11 clinical pregnancies in 42 women); the tests for subgroup differences demonstrated no statistical difference (P=0.90).

*Sensitivity analyses

Sensitivity analysis comparing the use of the risk ratio rather than the odds ratio as effect measure did not demonstrate an effect on the statistical significance of the main analysis for the secondary outcomes ‘clinical pregnancy’ (P=0.07) and ‘miscarriage’ (P=0.48). No effect was

97 Chapter 4

demonstrated by a sensitivity analysis studying the effect of the choice to include women with mixed intramural-submucous fibroids rather than submucous fibroids only for the secondary outcomes ‘clinical pregnancy’ (P=0.06) and ‘miscarriage’ (P=0.64).

Uterine septum No studies were retrieved.

Intrauterine adhesions No studies were retrieved.

98 Hysteroscopy for suspected major uterine cavity abnormalities

Table 2

2. Operative hysteroscopy compared with control for suspected major uterine cavity abnormalities prior to medically assisted reproduction-polyps prior to IUI Patient or population: women with endometrial polyps diagnosed by ultrasonography prior to treatment with gonadotrophin and intrauterine insemination Settings: infertility unit of a university hospital in the Spanish capital Madrid Intervention: hysteroscopic polypectomy using a 5.5 mm continuous flow office hysteroscope with a 1.5 mm scissors and forceps Comparison: diagnostic hysteroscopy using a 5.5 mm continuous flow office hysteroscope and polyp biopsy

Outcomes Illustrative comparative Relative No of Quality of risks* (95% CI) effect Participants the (95% CI) (studies) evidence Assumed Corresponding (GRADE) risk risk Control Polypectomy ْْْْ Clinical Low risk population 2 OR 4.41 204 pregnancy (2.45 to (1 study) high 5,6 ultrasound1 250 per 595 per 1000 7.96) Follow-up: 1000 (450 to 726) 4 IUI cycles Medium risk population 3 366 per 718 per 1000 1000 (586 to 821) High risk population 4 528 per 831 per 1000 1000 (733 to 899) *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the OR of the intervention (and its 95% CI).

GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

99 Chapter 4

1 Clinical pregnancy was defined by the presence of at least one gestational sac on ultrasound 2 Based on the clinical pregnancy rate per woman after 4 cycles gonadotrophins and IUI for male subfertility based on data from Bensdorp 2007 3 Based on the clinical pregnancy rate per woman after 4 cycles gonadotrophins and IUI for unexplained subfertility based on data from Veltman-Verhulst 2012 4 Based on the clinical pregnancy rate per woman after 4 cycles gonadotropins and IUI for unexplained subfertility based on data from Spiessens 2003 5 There was some potential for reporting bias 6 Large treatment effect in the absence of plausible confounders.

Endometrial polyps prior to intrauterine insemination (IUI) We retrieved one study (Pérez-Medina 2005). See table 2.

*Primary outcomes

2.1. Live birth There were no data for this primary outcome.

2.2. Adverse events: hysteroscopy complications There were no data for this primary outcome.

*Secondary outcomes

2.3. Clinical pregnancy

The hysteroscopic removal of polyps with a mean size of 16mm, detected by Doppler ultrasonography in women with unexplained, male or female factor subfertility for at least 24 months bound to undergo IUI, increases the odds of clinical pregnancy compared to diagnostic hysteroscopy and biopsy only (OR 4.4, 95% CI 2.5 to 8.0, P<0.00001, 1 study, 204 women) (Figure 6). The number needed to treat to benefit (NNTB) is 3 (95% CI 2 to 4). These results are based on an ‘available data’ analysis.

100 Hysteroscopy for suspected major uterine cavity abnormalities

Figure 6 Forest plot of comparison: Hysteroscopic removal of polyps versus diagnostic hysteroscopy and biopsy only prior to IUI. Outcome: Clinical pregnancy. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval.

Operative hysteroscopy Control Odds Ratio Odds Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI 2.1.1 Hysteroscopic polypectomy vs diagnostic hysteroscopy and biopsy only prior to IUI Pérez-Medina 2005 64 101 29 103 100.0% 4.41 [2.45, 7.96] Subtotal (95% CI) 101 103 100.0% 4.41 [2.45, 7.96] Total events 64 29 Heterogeneity: Not applicable Test for overall effect: Z = 4.93 (P < 0.00001)

Total (95% CI) 101 103 100.0% 4.41 [2.45, 7.96] Total events 64 29 Heterogeneity: Not applicable 0.01 0.1 1 10 100 Test for overall effect: Z = 4.93 (P < 0.00001) Favours diagnostic only Favours polypectomy Test for subgroup differences: Not applicable

2.4. Adverse events: miscarriage

There were no data for this secondary outcome.

*Subgroup analyses

A first prespecified subgroup analysis studied the effect of polyp size on the secondary outcome of clinical pregnancy. On histopathological examination the mean size of the polyps removed was 16mm (range 3 to 24mm). In the primary study the effect of the polyp size on the clinical pregnancy rate was studied in the intervention group. The data were analyzed based on the size of the removed polyps, subdivided into four groups based in their quartiles (<5mm, 5 to 10mm, 11 to 20mm and >20mm); the differences between these four subgroups within this study were not statistically significant (P=0.32). There is no evidence of an effect of the polyp size on the outcome of clinical pregnancy, but these results should be interpreted with caution due to the limited numbers in this single study. There were no data on the estimated size of the polyps in the control group. The second subgroup analysis studied the effect of the timing of the IUI treatment after hysteroscopy on the secondary outcome clinical pregnancy. About 29% of women in the polypectomy group, compared to 3% in the diagnostic hysteroscopy group became pregnant in the three- month period after the hysteroscopy before the treatment with gonadotrophin and IUI was started; this was calculated from the Kaplan- Meier survival analysis in the published report of the study (Pérez-Medina 2005). Hysteroscopic polypectomy increases the odds of clinical

101 Chapter 4

pregnancy compared to diagnostic hysteroscopy and polyp biopsy in women waiting to be treated with gonadotrophin and IUI (OR 13, 95%CI 3.9 to 46, P<0.0001, 1 study, 204 women, available data analysis). The number needed to treat to benefit (NNTB) following the hysteroscopic removal of polyps before starting the treatment with gonadotrophin and IUI is 4 (95% CI 3 to 6). In women who started gonadotrophin and IUI treatment the pregnancy rates per woman were 49% and 26% in the intervention and control group respectively. Hysteroscopic polypectomy increases the odds of clinical pregnancy in women who started from three months after the surgical procedure with gonadotrophin and IUI treatment (OR 2.7, 95% CI 1.4 to 5.1, P=0.003, 1 study, 172 women, available data analysis). The NNTB when starting the treatment with gonadotrophin and IUI after a prior hysteroscopic polypectomy is 4 (95% CI 3 to 12). We judged this to be an honest and sensible post hoc analysis. Subgroup analysis according to the indication for IUI (ovulatory disorder, male factor or unexplained subfertility) was not possible because individual patient data linking outcomes to the indication for IUI were not available after contacting the study authors.

*Sensitivity analyses

A sensitivity analysis studying the choice of risk ratio rather than odds ratio as the effect measure demonstrated no effect on the statistical significance of the main analysis for the secondary outcome ’clinical pregnancy’ (P<0.00001). A sensitivity analysis comparing an intention-to- treat analysis assuming that clinical pregnancies would not have occurred in participants with missing data, rather than an ‘available data’ analysis, did not affect the statistical significance of the main analysis for the secondary outcome ‘clinical pregnancy’ (OR 4.0, 95% CI 2.3 to 7.2, P<0.00001, 1 study, 215 women randomized).

Endometrial polyps prior to IVF-ICSI No studies were retrieved.

Submucous fibroids prior to IUI-IVF-ICSI No studies were retrieved.

Uterine septum prior to IUI-IVF-ICSI No studies were retrieved.

Intrauterine adhesions prior to IUI-IVF-ICSI No studies were retrieved.

102 Hysteroscopy for suspected major uterine cavity abnormalities

DISCUSSION

Summary of main results

This Cochrane systematic review aimed to investigate whether the hysteroscopic treatment of suspected major uterine cavity abnormalities made a difference to the main outcomes of live birth or pregnancy and the adverse events - hysteroscopy complications and miscarriage - in women with otherwise unexplained subfertility or before IUI-IVF-ICSI. The first randomized comparison was operative hysteroscopy versus control in women with otherwise unexplained subfertility and suspected major uterine cavity abnormalities - stratified into endometrial polyps, submucous fibroids, intrauterine adhesions or septate uterus. The second randomized comparison was operative hysteroscopy versus control in subfertile women undergoing medically assisted reproduction(MAR) – categorized as IUI, IVF or ICSI - with suspected major uterine cavity abnormalities - stratified into endometrial polyps, submucous fibroids, intrauterine adhesions or septate uterus. We critically appraised one trial (Casini 2006) comparing hysteroscopic removal of one submucous fibroid with a diameter”40 mm in women aged”35 years with otherwise unexplained subfertility versus regular fertility-oriented intercourse for a period of 12 months. There is no evidence of a benefit with removing submucous fibroids compared to control for the secondary outcome of clinical pregnancy. We similarly retrieved one single trial (Pérez-Medina 2005) for the second category of randomized comparisons. There is a clinically relevant and statistically significant increase in the odds of clinical pregnancy favoring the hysteroscopic removal of polyps with a mean size of 16mm (range 3 to 24mm) versus diagnostic hysteroscopy and biopsy in subfertile women with suspected endometrial polyps bound to undergo IUI. There were no data for the primary outcomes of live birth and hysteroscopy complications and the secondary outcome of miscarriage. The increase in clinical pregnancies after hysteroscopic polypectomy might be mainly due to a higher proportion of spontaneous conceptions before starting IUI and to a lesser- but still clinically relevant and statistically significant- extent to higher odds of conceiving after starting gonadotrophin treatment and IUI. The results of this sensible ‘post hoc’ subgroup analysis should nevertheless be interpreted with caution. There is no evidence for a dose- effect relationship between the size of the polyps and the increase in the outcome clinical pregnancy, but no definitive conclusions can be drawn based on this subgroup analysis.

103 Chapter 4

Overall completeness and applicability of the evidence

Evidence on the effectiveness of treating suspected major uterine cavity abnormalities by operative hysteroscopy compared to a control intervention in women with otherwise unexplained subfertility is very limited. We found no RCTs on the hysteroscopic treatment of endometrial polyps, intrauterine adhesions or septa in women with otherwise unexplained subfertility. The only included study in this category fails to report on the primary outcomes for this review. Evidence on the effectiveness of operative hysteroscopy compared to control in subfertile women with associated major uterine cavity abnormalities prior to medically assisted reproduction is incomplete; our search retrieved only data for subfertile women with suspected endometrial polyps prior to IUI. No data were retrieved on the effectiveness of operative hysteroscopy in subfertile women with submucous fibroids, intrauterine adhesions or septa prior to IUI or other techniques such as IVF or ICSI for all outcomes. Moreover for the randomized comparison hysteroscopic polypectomy versus diagnostic hysteroscopy prior to IUI no data are available for the primary outcomes. The evidence retrieved is by consequence insufficient to address all the objectives of the present Cochrane review. The absence of evidence favoring the hysteroscopic removal of submucous fibroids in women with otherwise unexplained subfertility is in contrast with daily clinical practice: it is generally accepted that submucous fibroids are very likely to interfere with normal fertility (Pritts 2001; Pritts 2009). In everyday practice most hysteroscopists will counsel subfertile women with submucous fibroids to undergo a hysteroscopic myomectomy since it does not seem reasonable to counsel timed intercourse with or without ovarian stimulation or proceed to IUI or IVF/ICSI without removing the fibroids; this clinical practice is based on observational evidence only and is therefore at risk of bias. The results of the trial on hysteroscopic polypectomy (Pérez-Medina 2005) are very relevant for everyday practice. In this study one-third of the randomized women treated by IUI suffered from an isolated ovulatory disorder which is by itself not an indication for IUI in everyday clinical practice as opposed to mild male factor (Bensdorp 2007) or unexplained subfertility (Veltman-Verhulst 2012).

104 Hysteroscopy for suspected major uterine cavity abnormalities

Quality of the evidence

The present review included only two trials; neither reported the primary outcomes live birth or hysteroscopy complications. By using the GRADE tool (Grading of Recommendations Assessment, Development and Evaluation) we assessed the overall quality of the evidence of the first trial on hysteroscopic myomectomy (Casini 2006) as ‘very low’ and that of the second study on hysteroscopic polypectomy (Pérez-Medina 2005) as ‘high’.

Agreements and disagreements with other studies or reviews

There are two systematic reviews on fibroids and subfertility (Pritts 2001; Pritts 2009). We refer to the data and conclusions reported in the most recent review because the MOOSE (Meta-analysis of Observational Studies in Epidemiology) guidelines for systematic reviews of observational studies were strictly followed (Pritts 2009). The review authors concluded that the fertility outcomes are decreased in women with submucous fibroids, and that the removal of the fibroids is likely to benefit the reproductive outcome. The results of the trial on the effectiveness of hysteroscopic polypectomy prior to IUI are consistent with the findings of two recently published observational studies. The first study planned to measure the effect of the presence of endometrial polyps on pregnancy rates and how polypectomy could affect pregnancy rates in 171 women scheduled for IUI (Kalampokas 2012). The authors concluded that hysteroscopic polypectomy appears to improve fertility in women with otherwise unexplained subfertility. The second study, a prospective clinical controlled trial (CCT) including 120 women with endometrial polyps, aimed to measure whether polypectomy before IUI achieved better pregnancy outcomes than no intervention (Shohayeb 2011). The authors of this trial concluded that persistent endometrial polyps are likely to impair reproductive performance and that hysteroscopic polypectomy before IUI could be considered as an effective intervention. A systematic review (Lieng 2010b) included 11 studies in 935 subfertile women with endometrial polyps: one RCT (Pérez-Medina 2005), three CCTs and seven observational studies (three retrospective, one prospective and three undetermined). There was no evidence of a benefit with hysteroscopic polypectomy on the IVF outcomes according to two smaller observational studies but the authors nevertheless concluded that the limited evidence suggests a favorable outcome on pregnancy rates in subfertile women with endometrial polyps. The conclusion of this review should be interpreted with caution because there was substantial clinical diversity precluding

105 Chapter 4

formal meta-analysis; moreover there are several methodological concerns (no study of the effect of confounders, no formal assessment of publication bias).

106 Hysteroscopy for suspected major uterine cavity abnormalities

AUTHORS’CONCLUSIONS

Implications for practice

We cannot exclude nor confirm a benefit in favor of removing submucous fibroids in women with otherwise unexplained subfertility: the differences in clinical pregnancy rates between both comparison groups in the single retrieved RCT nearly reached statistical significance. The level of the evidence on this subject was very low. Before treating subfertile women with endometrial polyps with gonadotrophins combined with IUI for unexplained, mild male or female factor subfertility for at least 24 months, it may be recommended to do a hysteroscopic polypectomy to improve the chances of conceiving either spontaneously during a waiting period of three months or after starting IUI combined with gonadotrophins; the level of evidence of this study was high.

Implications for research

More well-designed RCTs are needed to assess whether the hysteroscopic removal of endometrial polyps, submucous fibroids, septa or intrauterine adhesions is likely to benefit women with otherwise unexplained subfertility or before treatment with IUI, IVF or ICSI. There are knowledge gaps on the effects of the number, size or extent and the localization of these four uterine cavity abnormalities on the main outcomes in women with otherwise unexplained subfertility or prior to MAR as well as on the time-effect relationship between the hysteroscopic intervention and subsequent IUI, IVF or ICSI treatment. Therefore more RCTs on these topics are needed to extend our present knowledge of the potential benefits of hysteroscopy for treating female subfertility.

107 Chapter 4

REFERENCES

References to included studies

1. Casini ML, Rossi F, Agostini R, Unfer V. Effects of the position of fibroids on fertility. Gynecol Endocrinol 2006; 22(2):106-109. 2. Pérez-Medina T, Bajo-Arenas J, Salazar F, Redondo T, Sanfrutos L, Alvarez P, Engels V. Endometrial polyps and their implication in the pregnancy rates of patients undergoing intrauterine insemination: a prospective, randomized study. Hum Reprod 2005; 20(6):1632–1635.

References to excluded studies

1. Acunzo G, Guida M, Pellicano M, Tommaselli GA, Di Spiezio Sardo A, Bifulco G, Cirillo D, Taylor A, Nappi C. Effectiveness of auto- crosslinked gel in the prevention of intrauterine adhesions after hysteroscopic adhesiolysis: a prospective, randomized, controlled study. Hum Reprod 2003; 18(9):1918–1921. 2. Aghahosseini M, Ebrahimi N, Mahdavi A, Aleyasin A, Safdarian L, Sina S. Hysteroscopy prior to assisted reproductive technique in women with recurrent implantation failure improves pregnancy likelihood. Fertil Steril 2012; 98(3 Suppl):S4, O-13. 3. Amer MI, Abd-El-Maeboud KHI, Abdelfatah I, Salama FA, Abdallah AS. Human amnion as a temporary biologic barrier after hysteroscopic lysis of severe intrauterine adhesions: pilot study. J Minim Invasive Gynecol 2010; 17(5):605–611. 4. Colacurci N, De Franciscis P, Mollo A, Litta P, Perino A, Cobellis L, De Placido G. Small-diameter hysteroscopy with Versapoint versus resectoscopy with a unipolar knife for the treatment of septate uterus: a prospective randomized study. J Minim Invasive Gynecol 2007; 14:622-627. 5. Darwish AM, Elsaman AM. Extended resectoscopic versus sequential cold knife- resectoscopic excision of the unclassified complete uterocervicovaginal septum: a randomised trial. Fertil Steril 2009; 92(2):722–726. 6. De Iaco PA,Muzzupapa G, Bovicelli A,Marconi S, Bitti SR, Sansovini M, Bovicelli L. Hyaluronan derivative gel (Hyalobarrier gel®) in intrauterine adhesion prevention after operative hysteroscopy. Ellipse 2003; 19(1):15–18. 7. Demirol A, Gurgan T. Effect of treatment of intrauterine pathologies with office hysteroscopy in patients with recurrent IVF failure. Reprod Biomed Online 2004; 8(5):590–594.

108 Hysteroscopy for suspected major uterine cavity abnormalities

8. Di Spiezio Sardo A, Spinelli M, Bramante S, Scognamiglio M, Greco E, Guida M, Cela V, Nappi C. Efficacy of a polyethylene oxide- sodium carboxymethylcellulose gel in prevention of intrauterine adhesions after hysteroscopic surgery. J Minim Invasive Gynecol 2011; 18:462–469. 9. El- Nashar IH, Nasr A. The role of hysteroscopy before intracytoplasmic sperm injection (ICSI): a randomized controlled trial. Fertil Steril 2011; 96(3 Suppl): S266. 10. Guida M, Acunzo G, Di Spiezio Sardo A, Bifulco G, Piccoli R, Pellicano M, Cerrota G, Cirillo D, Nappi C. Effectiveness of auto- crosslinked hyaluronic acid gel in the prevention of intrauterine adhesions after hysteroscopic surgery: a prospective, randomized, controlled study. Hum Reprod 2004; 19(6):1461–1464. 11. Lieng M, Istre O, Sandvik L, Engh V, Qvigstad E. Clinical effectiveness of transcervical polyp resection in women with endometrial polyps: randomized controlled trial. J Minim Invasive Gynecol 2010; 17(3):351–357. 12. Muzii L, Bellati F, Pernice M, Manci N, Angioli R, Panici PB. Resectoscopic versus bipolar electrode excision of endometrial polyps: a randomized study. Fertil Steril 2007; 87(4):909–917. 13. Pabuccu R, Onalan G, Kaya C, Selam B, Ceyhan T, Ornek T, Kuzudisli E. Efficiency and pregnancy outcome of serial intrauterine device-guided hysteroscopic adhesiolysis of intrauterine synechiae. Fertil Steril 2008; 90(5): 1973–1977. 14. Parsanezhad ME, Alborzi S, Zarei A, Dehbashi S, Shirazi LG, Rajaeefard A, Schmidt EH. Hysteroscopic metroplasty of the complete uterine septum, duplicate cervix, and vaginal septum. Fertil Steril 2006; 85(5):1473–1477. 15. Rama Raju GA, Shashi Kumari G, Krishna KM, Prakash GJ, Madan K. Assessment of uterine cavity by hysteroscopy in assisted reproduction programme and its influence on pregnancy outcome. Arch Gynecol Obstet 2006; 274(3):160–164. 16. Shawki HE, Elmorsy M, Eissa MK. Routine office hysteroscopy prior to ICSI and its impact on assisted reproduction program outcome: a randomized controlled trial. Middle East Fertility Society Journal 2012; 17(1):14–21. 17. Shokeir T, El-Shafei M, Yousef H, Allam AF, Sadek E. Submucous myomas and their implications in the pregnancy rates of patients with otherwise unexplained primary infertility undergoing hysteroscopic myomectomy: a randomized matched control study. Fertil Steril 2010; 94(2):724–729 (this publication has been retracted).

109 Chapter 4

18. Tonguc EA, Var T, Yilmaz N, Batioglu S. Intrauterine device or estrogen treatment after hysteroscopic uterine septum resection. Int J Gynaecol Obstet 2010; 109(3):226–229. 19. van Dongen H, Emanuel MH, Wolterbeek R, Trimbos JB, Jansen FW. Hysteroscopic morcellator for removal of intrauterine polyps and myomas: a randomized controlled pilot study among residents in training. J Minim Invasive Gynecol 2008; 15(4):466–471. 20. Vercellini P, Vendola N, Colombo A, Passadore C, Trespidi L, Fedele L. Hysteroscopic metroplasty with resectoscope or microscissors for the correction of septate uterus. Surg Gynecol Obstet 1993; 176(5): 439–442.

References to studies awaiting assessment

1. Pansky M. Efficiency of Intercoat (Oxiplex/AP Gel) in decreasing intrauterine adhesions [Intercoat (Oxiplex/ AP Gel) for preventing intrauterine adhesions following operative hysteroscopy for suspected retained products of conception – a prospective randomized pilot study].ClinicalTrials.gov 2012. [NCT: 1377779]. 2. Trniniü-Pjeviü A, Kopitoviü V, Pop-Trajkoviü S, Bjelica A, Bujas I, Tabs D, Iliü D, Stajiü D. Effect of hysteroscopic examination on the outcome of in vitro fertilization. Vojnosanit Pregl. 2011; 68(6):476– 480.

110 Hysteroscopy for suspected major uterine cavity abnormalities

References to ongoing studies

1. Smit JG, Kasius JC, Eijkemans MJ, Koks CA, Van Golde R, Oosterhuis JG, Nap AW, Scheffer GJ, Manger PA, Hoek A, Kaplan M, Schoot DB, van Heusden AM, Kuchenbecker WK, Perquin DA, Fleischer K, Kaaijk EM, Sluijmer A, Friederich J, Laven JS, van Hooff M, Louwe LA, Kwee J, Boomgaard JJ, de Koning CH, Janssen IC, Mol F, Mol BW, Torrance HL, Broekmans FJ. The inSIGHT study: costs and effects of routine hysteroscopy prior to a first IVF treatment cycle. A randomised controlled trial. BMC Women’s Health 2012; 12:22. 2. El-Khayat W. Does office hysteroscopy and endometrial snip improve IUI outcome? A randomized controlled trial. International Clinical Trials Registry Platform Search Portal 2012. [NCT: 01544426]. 3. El-Toukhy T, Campo R, Sunkara SK, Khalaf Y, Coomarasamy A. A multi-centre randomised controlled study of pre-IVF outpatient hysteroscopy in women with recurrent IVF implantation failure: Trial of Outpatient Hysteroscopy - [TROPHY] in IVF. Reproductive Health 2009; 6:20. 4. Maramazi F.Effect of hysteroscopy before intra uterine insemination on fertility in infertile couples referred to Imam Khomeini Hospital, Ahwaz IVF. International Clinical Trials Registry Platform Search Portal 2012. [IRCT: 201201308867N1]. 5. Revel A. Safety study of use of hyaluronic acid gel to prevent intrauterine adhesions in hysteroscopic surgery. ClinicalTrials.gov 2011. [NCT: 01464528]. 6. Sohrabvand F. Evaluation of diagnostic hysteroscopy findings in patients, candidate for ART (IVF, ICSI) and its effect on pregnancy rate compared to control group. International Clinical Trials Registry Platform Search Portal 2012. [IRCT: 201208152565N6].

Additional references

1. Andersen AN, Goossens V, Ferraretti AP, Bhattacharya S, Felberbaum R, de Mouzon J, Nygren KG; European IVF-monitoring (EIM) Consortium; European Society of Human Reproduction and Embryology (ESHRE). Assisted reproductive technology in Europe, 2004: results generated from European registers by ESHRE. Hum Reprod 2008; 23:756–771. 2. Bensdorp A, Cohlen BJ, Heineman MJ, Vanderkerchove P. Intra- uterine insemination for male subfertility. Cochrane Database Syst Rev 2007, Issue 4, CD000360.

111 Chapter 4

3. Bettocchi S, Nappi L, Ceci O, Selvaggi L. Office hysteroscopy. Obstet Gynecol Clin North Am 2004; 31: 641–654. 4. Boivin J, Bunting L, Collins JA, Nygren KG. International estimates of infertility prevalence and treatment-seeking: potential need and demand for infertility medical care. Hum Reprod 2007; 22(6):1506– 1512. 5. Bosteels J, Kasius J, Weyers S, Broekmans FJ, Mol BWJ, D’Hooghe TM. Hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities. Cochrane Database Syst Rev 2011, Issue 11, CD009461. 6. Clark TJ, Gupta JK. Handbook of Outpatient Hysteroscopy. A Complete Guide to Diagnosis and Therapy. Hodder Education, 2005. 7. Donnez J, Jadoul P. What are the implications of myomas on fertility? A need for debate? Hum Reprod 2002; 17:1424–1430. 8. Fatemi HM, Kasius JC, Timmermans A, van Disseldorp J, Fauser BC, Devroey P, Broekmans FJ. Prevalence of unsuspected uterine cavity abnormalities diagnosed by office hysteroscopy prior to in vitro fertilization. Hum Reprod 2010; 25(8):1959–1965. 9. Fedele L, Bianchi S, MarchiniM, Franchi D, Tozzi L, Dorta M. Ultrastructural aspects of endometrium in infertile women with septate uterus. Fertil Steril 1996; 65: 750–752. 10. Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. 11. Jansen FW, Vredevoogd CB, van Ulzen K, Hermans J, Trimbos JB, Trimbos-Kemper TC. Complications of hysteroscopy: a prospective, multicenter study. Obstet Gynecol 2000; 96(2):266–270. 12. Kalampokas T, Tzanakaki D, Konidaris S, Iavazzo C, Kalampokas E, Gregoriou O. Endometrial polyps and their relationship in the pregnancy rates of patients undergoing intrauterine insemination. Clin Exp Obstet Gynecol 2012; 39(3):299–302. 13. Kasius JC, Broekmans FJ, Veersema S, Eijkemans MJ, van Santbrink EJ, Devroey P, Fauser BC, Fatemi HM. Observer agreement in the evaluation of the uterine cavity by hysteroscopy prior to in vitro fertilization. Hum Reprod 2011; 26(4):801–807. 14. Lieng M, Istre O, Qvigstad E. Treatment of endometrial polyps: a systematic review. Acta Obstet Gynecol Scand. 2010; 89:992–1002. 15. Mollo A, De Franciscis P, Colacurci N, Cobellis L, Perino A, Venezia R, Alviggi C, De Placido G. Hysteroscopic resection of the septum improves the pregnancy rate of women with unexplained infertility: a prospective controlled trial. Fertil Steril 2009; 91:2628–2631.

112 Hysteroscopy for suspected major uterine cavity abnormalities

16. Pritts EA. Fibroids and infertility: a systematic review of the evidence. Obstet Gynecol Surv 2001; 56 (8):483-491. 17. Pritts EA, Parker WH, Olive DL. Fibroids and infertility: an updated systematic review of the evidence. Fertil Steril 2009; 91:1215–1223. 18. Ray A, Shah A, Gudi A, Homburg R. Unexplained infertility: an update and review of practice. Reprod Biomed Online 2012; 24:591– 602. 19. Rogers PA, Milne BJ, Trounson AO. A model to show human uterine receptivity and embryo viability following ovarian stimulation for in vitro fertilization. J In Vitro Fert Embryo Transf 1986; 3: 93–98. 20. Saravelos SH, Cocksedge KA, Li TC. Prevalence and diagnosis of congenital uterine anomalies in women with reproductive failure: a critical appraisal. Hum Reprod Update 2008; 14:415–429. 21. Shohayeb A, Shaltout A. Persistent endometrial polyps may affect the pregnancy rate in patients undergoing intrauterine insemination. Middle East Fertility Society Journal 2011; 16(4):259–264. 22. Shokeir TA, Shalan HM, El-Shafei MM. Significance of endometrial polyps detected hysteroscopically in eumenorrheic infertile women. J Obstet Gynaecol Res 2004; 30:84–89. 23. Shokeir T, Abdelshaheed M, El-Shafei M, Sherif L, Badawy A. Determinants of fertility and reproductive success after hysteroscopic septoplasty for women with unexplained primary infertility: a prospective analysis of 88 cases. Eur J Obstet Gynecol Reprod Biol 2011; 155:54–57. 24. Silberstein T, Saphier O, van Voorhis BJ, Plosker SM. Endometrial polyps in reproductive-age fertile and infertile women. Isr Med Assoc J 2006; 8:192– 195. 25. Singh M, Chaudhry P, Asselin E. Bridging endometrial receptivity and implantation; network of hormones, cytokines, and growth factors. J Endocrinol 26. 2011; 210:5–14. 27. Somigliana E, Vercellini P, Daguati R, Pasin R, De Giorgi O, Crosignani PG. Fibroids and female reproduction: a critical analysis of the evidence. Hum Reprod Update 2007; 13:465–476. 28. Spiessens C, Vanderschueren D, Meuleman C, D’Hooghe T. Isolated teratozoospermia and intrauterine insemination. Fertil Steril 2003; 80(5):1185–1189. 29. Taylor E, Gomel V. The uterus and fertility. Fertil Steril 2008; 89(1):1–15. 30. Tomaževiþ T, Ban-Frangež H, Virant-Klun I, Verdenik I, Požlep B, Vrtaþnik-Bokal E. Septate, subseptate and arcuate uterus decrease

113 Chapter 4

pregnancy and live birth rates in IVF/ ICSI. Reprod Biomed Online 2010; 21(5):700–705. 31. Veltman-Verhulst SM, Cohlen BJ, Hughes E, Heineman MJ. Intra- uterine insemination for unexplained subfertility. Cochrane Database Syst Rev 2012, Issue 9, CD001838. 32. Wallach EE. The uterine factor in infertility. Fertil Steril 1972; 23(2):138– 158. 33. Wamsteker K, De Block S. Diagnostic hysteroscopy: technique and documentation. Endoscopic Surgery for Gynecologists. London: Saunders, 1998:511– 24. 34. Yanaihara A, Yorimitsu T, Motoyama H, Iwasaki S, Kawamura T. Location of endometrial polyp and pregnancy rate in infertility patients. Fertil Steril 2008; 90: 180– 182. 35. Yu D, Li TC, Xia E, Huang X, Liu Y, Peng X. Factors affecting reproductive outcome of hysteroscopic adhesiolysis for Asherman’s syndrome. Fertil Steril 2008; 89(3): 715– 722. 36. Zegers-Hochschild F, Adamson GD, de Mouzon J, Ishihara O, Mansour R, Nygren K, Sullivan E, Vanderpoel S; International Committee for Monitoring Assisted Reproductive Technology; World Health Organization. International Committee for Monitoring Assisted Reproductive Technology (ICMART) and the World Health Organization (WHO) revised glossary of ART terminology, 2009. Fertil Steril 2009; 92(5):1520– 1524.

114

Anti-adhesion barrier gels following operative hysteroscopy for treating female subfertility: a systematic review and meta- analysis

Jan Bosteels, Steven Weyers, Ben Willem Mol, Thomas M. D’Hooghe

Gynecological Surgery 2014; 11: 113-127 http://link.springer.com/article/10.1007/s10397-014-0832-x

115

Anti-adhesion barriers gels following operative hysteroscopy

ABSTRACT

Study question To assess the effects of any anti-adhesion barrier gel used after hysteroscopy for treating subfertility.

Summary answer Gynecologists might use any barrier gel following operative hysteroscopy in subfertile women for decreasing adhesion formation; the use of any barrier gel is associated with less severe adhesions and lower mean adhesion scores. Nevertheless subfertile women should be counseled that there is at the present no evidence for higher live birth or pregnancy rates. Data for the outcome miscarriage are absent.

What is already known Preclinical studies suggest that the use of biodegradable surgical barriers decreases the incidence of postsurgical adhesions but observational studies in the human report conflicting results.

Study design, size, duration We searched the Cochrane Menstrual Disorders and Subfertility Specialized Register (10 April 2013), the Cochrane Central Register of Controlled Trials (The Cochrane Library 2013, Issue 1), MEDLINE (1950 to 4 April 2013), EMBASE (1974 to 4 April 2013) and other electronic databases of trials including trial registers, sources of unpublished literature and reference lists. We handsearched the Journal of Minimally Invasive Gynecology (from 1 January 1992 to 13 April 2013); we also contacted experts in the field.

Participants/materials, setting, methods Randomized comparisons between any anti-adhesion barrier gel versus another barrier gel, placebo or no adjunctive therapy following operative hysteroscopy. Primary outcomes were live birth rates and adhesion formation at second-look hysteroscopy. Secondary outcomes were pregnancy and miscarriage rates, mean adhesion scores and severity of adhesions at second-look hysteroscopy. Two authors independently assessed eligible studies for inclusion and risk of bias, and extracted data. We contacted primary study authors for additional information or other clarification.

117 Chapter 5

Main results and the role of chance Five trials met the inclusion criteria. There is no evidence of benefit with the use of any barrier gel following operative hysteroscopy for the key outcome clinical pregnancy: the risk ratio (RR) was 3.0, 95% confidence interval (CI) was 0.35 to 26, P=0.32, 1 study, 30 women (very low quality evidence). There were no data for the outcomes live birth or miscarriage. The use of any gel following operative hysteroscopy decreases the incidence of adhesions at second-look hysteroscopy (RR 0.65, 95% CI 0.45 to 0.93, P=0.02, 5 studies, 372 women, very low quality evidence). The number needed to treat to benefit (NNTB) is 9 (95% CI 5 to 33). The use of auto-cross-linked hyaluronic acid gel in women undergoing operative hysteroscopy for fibroids, endometrial polyps or uterine septa is associated with a lower mean adhesion score at second-look hysteroscopy at 3 months: the mean difference (MD) was -1.4, 95% CI -1.8 to -1.0, P<0.00001, 1 study, 24 women. The decrease in adhesion scores is greater in the subgroup of women treated for intrauterine adhesions (MD -3.3, 95% CI -3.4 to -3.2, P<0.00001, 1 study, 19 women). When adhesions occur after hysteroscopic surgery the use of any gel following operative hysteroscopy is associated with more mild adhesions (RR 2.8, 95% CI 1.1 to 7.0, P=0.03, 4 studies, 79 women) and less ’moderate or severe adhesions’ at second-look hysteroscopy (RR 0.25, 95% CI 0.10 to 0.67, P=0.006, 4 studies, 79 women). The NNTB is 2 with a 95% CI 1 to 2 for the former and 95% CI 1 to 4 for the latter outcome (all very low quality evidence).

Limitations, reason for caution Only two trials included subfertile women, the proportion of participants with subfertility in the remaining three studies was unclear. Only one study reported on the effects of anti-adhesion barrier gels for the key outcome of pregnancy but the length of follow up was not reported.

Wider implications of the findings More RCTs are needed to assess whether the use of any anti-adhesion gel affects the key reproductive outcomes in a target population of subfertile women.

118 Anti-adhesion barrier gels following operative hysteroscopy

BACKGROUND

Intrauterine adhesions (IUAs) are fibrous strings at opposing walls of the uterus. The causes of IUAs are multifactorial; nearly 90% of cases are associated with postpartum or post abortion dilatation and curettage. The role of infection in the development of IUAs is undetermined with the exception of genital tuberculosis (Deans and Abbott, 2010). The pathophysiology and the mechanisms of tissue repair in the endometrium are poorly understood despite several theories on the source of cells involved in this process (Okulicz 2002). Intrauterine adhesion (IUA) formation is the major long-term complication of operative hysteroscopy in women of reproductive age. The incidence of postsurgical IUAs at second-look hysteroscopy varies according to the type of surgery as demonstrated by a RCT: 3.6% after polyp removal, 6.7% after resection of uterine septa, 31% after removal of a single fibroid and 45% after resection of multiple fibroids (Taskin et al, 2000). The investigators of a prospective cohort study in 163 women undergoing operative hysteroscopy observed significantly more IUAs after septoplasty (14 of 16 women or 88 %) or adhesiolysis (34 of 45 women or 76 %) compared to the removal of submucous fibroids (26 of 65 women or 40 %) or endometrial polyps (0 of 37 women or 0%). They concluded that the full recovery of the endometrium may vary from one month after the removal of polyps to between two to three months following hysteroscopic myomectomy (Yang et al, 2013). Intrauterine adhesions may cause poor reproductive outcome. Firstly, IUAs have a negative impact on fertility as demonstrated by large review of observational studies: 922 of 2151 women with IUAs or 43% suffered from subfertility (Schenker and Margalioth, 1982). The hypothetical underlying mechanisms for subfertility due to IUAs are obstruction of sperm transport into the cervix, impaired embryo migration within the uterine cavity or failure of embryo implantation due to endometrial insufficiency (Deans and Abbott, 2010). Secondly, IUAs are often associated with recurrent miscarriage; the prevalence of IUAs in women suffering from this reproductive health problem ranges from 5 to 39% according to a narrative review of observational studies (Kodaman and Arici, 2007). Thirdly, the successful hysteroscopic treatment of severe IUAs may cause long term major obstetrical complications, such as placenta accreta/ increta and higher risks for preterm delivery, uterine rupture and peripartum hysterectomy (Deans and Abbott, 2010). Hyaluronic acid or hyaluronan (HA) is a water-soluble polysaccharide consisting of multiple disaccharide units of glucuronic acid and N- acetylglucosamine, bound together by a ß1-3-type glycoside bond.

119 Chapter 5

Solutions of HA have interesting visco-elastic properties which have led to the development of applications of HA in surgical procedures, for example in . HA is not an ideal substance for all procedures, due to its limited residence time when applied to a surgical site. It quickly enters the systemic circulation and is cleared rapidly by catabolic pathways. Attempts to use hyaluronan for preventing postsurgical adhesions have therefore been met with variable success. Several chemically modified derivatives of HA have been developed to circumvent the disadvantages of HA. One such derivative is auto-cross- linked polysaccharide (ACP). It is formed by cross-linking hyaluronan, by direct formation of covalent ester bonds between hydroxyl and carboxyl groups of the hyaluronan molecule. ACP can be prepared with various degrees of cross-linking, which allows tailoring of the viscosity properties of ACP gels (Renier et al, 2005). Carboxymethylcellulose (CMC) is a high molecular weight polysaccharide that has a viscosity greater than dextran 70. CMC can be used for adhesion prevention as a membrane barrier or a gel as a mixture of chemically derivative sodium hyaluronate and carboxymethylcellulose gel (HA-CMC) (Leach et al, 1998). The ideal anti-adhesion barrier for hysteroscopic surgery would be the application of a biologically active mechanical separator that achieves the suppression of intrauterine adhesion formation and promotes the healing of the endometrial tissue. The use of the biodegradable gel surgical barriers is based on the principle of keeping the adjacent wound surfaces as mechanically separate (Renier et al, 2005). Several preclinical studies in various animal models have reported the effectiveness of both ACP (Belluco et al, 2001; Binda et al, 2007; Binda et al, 2009; Binda et al, 2010; De Iaco et al, 1998; Koçak et al, 1999; Shamiyeh et al, 2007; Wallwiener et al, 2006) and HA-CMC gels (Leach et al, 1998; Schonman et al, 2008) or HA-CMC membranes (Kelekci et al, 2004; Rajab et al, 2010) for preventing postsurgical adhesions. Other preclinical studies in animal models suggest that HA gel remains in situ for more than 5 to 6 days (Laurent and Fraser, 1992; Nimrod et al, 1992). Similarly, animal studies have demonstrated the persistence of HA-CMC for about 7 days after its application (Diamond et al, 1988). However, most of these studies were done in rodent models, and not in non-human primate models with reproductive anatomy similar to humans, like the baboon, a validated model for endometriosis research (D’Hooghe et al, 2009). The exact mechanisms by which ACP and HA-CMC are able to reduce adhesion reformation are not well known, but may be related to "hydro flotation" or "siliconizing" effects. One French clinical controlled trial (N=54 women) studied the effectiveness of the application of ACP gel (n=30) versus no gel (n=24) at the end of an operative hysteroscopic

120 Anti-adhesion barrier gels following operative hysteroscopy procedure for treating fibroids, polyps, uterine septa or IUAs; there were no statistically significant differences for the rate of adhesion formation, the mean adhesion scores or the severity of the adhesions between both comparison groups (Ducarme et al, 2006). No data were available for the reproductive outcome. The health burden associated with subfertility, abdominal pain, or bowel obstruction due to adhesions is substantial (Renier et al, 2005; DeCherney et al, 1997; DiZerega, 1994); the total cost of adhesion-related morbidity in the US Health Care system exceeds $ 1 billion annually (Baakdah and Tulandi, 2005). To the best of our knowledge no economical studies on adhesion prevention after operative hysteroscopy have been conducted in a subfertile population. Postoperative adhesion formation is a determining factor influencing endometrial wound healing (Yang et al, 2013). At the present there is uncertainty whether the use of anti-adhesion barrier gels following operative hysteroscopy for preventing de novo adhesions or reformation of existing adhesions affects the pregnancy or live birth rates; this is the main objective of the present systematic review.

METHODS

Two review authors independently searched the Cochrane Menstrual Disorders and Subfertility Specialized Register (10 April 2013), the Cochrane Central Register of Controlled Trials (The Cochrane Library 2013, Issue 1), MEDLINE (1950 to 4 April 2013), EMBASE (1974 to 4 April 2013) using a combination of both index and free-text terms. We used no language restrictions. We searched other electronic databases including trial registers, sources of unpublished literature and reference lists. We handsearched the Journal of Minimally Invasive Gynecology (from 1 January 1992 to 13 April 2013) and contacted experts in the field.

We included only studies that were clearly randomized or claimed to be randomized. Studies were selected if the source population included women of reproductive age suffering from subfertility, bound to undergo operative hysteroscopy before any infertility treatment or expectant management. Subfertility was defined as “a disease of the reproductive system defined by the failure to achieve a clinical pregnancy after 12 months or more of regular unprotected sexual intercourse” (Zegers- Hochschild et al, 2009). Studies were excluded if subfertility was explicitly reported among the exclusion criteria. We included the following types of randomized comparisons: any anti-adhesion barrier gel

121 Chapter 5

versus placebo, no barrier gel or another type of barrier gel following operative hysteroscopy.

We selected live birth and adhesion formation at second-look hysteroscopy as primary outcomes. Live birth was defined as a delivery of a live fetus after 20 completed weeks of gestational age that resulted in at least one live baby born. The delivery of a singleton, twin or multiple pregnancies was counted as one live birth (Zegers-Hochschild et al, 2009). Ongoing or clinical pregnancy, miscarriage and mean adhesion scores or severity of adhesions at second-look hysteroscopy were secondary outcomes. Ongoing pregnancy was defined as a pregnancy surpassing the first trimester or 12 weeks of pregnancy; clinical pregnancy was defined as a pregnancy diagnosed by the ultrasound visualization of one or more gestational sacs or definitive clinical signs of pregnancy (Zegers- Hochschild et al, 2009). There are at the present seven reported classification systems for scoring the extent or severity of intrauterine adhesions (Deans and Abbott, 2010). Some classification systems have incorporated menstrual and obstetric history (Wamsteker and De Block, 1998; AFS, 1988; Nasr et al, 2000); others rely exclusively on the hysteroscopic evaluation of the uterine cavity (March et al, 1978; Hamou et al, 1983; Valle and Sciarra, 1988; Donnez and Nisolle, 1994). None of these systems has been validated and therefore universally accepted (Deans and Abbott, 2010). We avoided pooling data from studies using different scoring systems. After screening the titles and abstracts, removing duplicates and after linking multiple reports of the same study together, two authors independently assessed the studies by examining the full text reports. We contacted the authors of the primary study report whenever additional information was required Two review authors independently assessed the risk of bias by using the Cochrane ‘Risk of bias’ tool assessing all six items. Any disagreements between the review authors for the selection, data extraction or risk of bias assessment were resolved by discussion or by a third author; any residual disagreement was reported in the final review. We used the numbers of events in both comparison groups of each included study to calculate the Mantel-Haenszel (M-H) risk ratios (RR) for the binary data for all the main outcomes; for the secondary outcome ‘adhesion scores’ the mean values and the standard deviations (SD) were used to calculate the inverse variance (IV) mean differences (MD) and the 95% confidence intervals (CI). We used Review Manager 5 for all the computations including the 95% CI.

122 Anti-adhesion barrier gels following operative hysteroscopy

All main outcomes were expressed as per woman randomized. Multiple live births and multiple pregnancies were counted as one event. We aimed to analyze the data on an intention-to-treat basis (ITT). We tried to obtain missing data after contacting the primary study authors. If missing data could not be obtained, we undertook imputation of individual values for the primary outcomes only by assuming that live births or adhesions would not have occurred in participants without a reported primary outcome. For all other main outcomes we used an available data analysis. We subjected any imputation of missing data for the primary outcomes to sensitivity analyses; any substantial difference in the imputed ITT analyses compared to available data analyses was incorporated in the interpretation of the study findings and the discussion. Meta-analysis was done to provide a meaningful summary when enough studies were available that were sufficiently similar with respect to the clinical and methodological characteristics. A formal assessment of statistical heterogeneity was done by using the Q-statistic and the I² statistic. If there was evidence of substantial heterogeneity beyond chance, we aimed to explore possible explanations for this heterogeneity instead of pooling studies that were too clinically diverse. We aimed to do the search for eligible studies as comprehensively as possible and by being alert in identifying duplicated reports of trials in order to minimize the potential impact of reporting and publication bias. Since we retrieved only a small number of studies, we did not study publication bias or other forms of small study effects by creating a funnel plot. One review author entered the study data and carried out the statistical analysis using Review Manager 5. We considered the outcomes live birth and pregnancy to be positive outcomes of effectiveness and by consequence higher numbers of these events as a benefit. The outcomes miscarriage, adhesion formation and adhesion scores were on the contrary considered as negative outcomes and higher numbers as harmful. We aimed to combine data from primary studies in a meta-analysis with Review Manager 5 using the risk ratio as a summary outcome measure using a random-effects model (RE) if enough studies were retrieved and after significant clinical diversity and substantial statistical heterogeneity were confidently ruled out. We aimed to carry out subgroup analyses according to the extent or the severity of the uterine abnormality treated and for studies that reported both ‘live birth’ and ‘pregnancy’ in order to assess any overestimation of the treatment effect. We planned to do sensitivity analyses for the primary outcomes to investigate whether the results and conclusions were robust to arbitrary decisions regarding the eligibility and analysis. These sensitivity

123 Chapter 5

analyses included consideration whether conclusions would have differed if the eligibility was restricted to studies without high risk of bias versus all studies or if alternative imputation strategies were adopted, e.g. using odds ratio rather than risk ratio for the summary effect measure or a fixed- effect rather than a random-effects as the analysis model.

124 Anti-adhesion barrier gels following operative hysteroscopy

FINDINGS

Description of studies

Results of the search

The search and selection process is presented in figure 1: we retrieved 14 potentially eligible randomized studies; we included five trials, six trials were excluded and three are ongoing.

Figure 1.PRISMA flow chart of the search and selection process.

125 Chapter 5

Included studies

*Study design and setting

Five single-centre parallel group RCTs were included in the present systematic review: four were conducted in Italy (Acunzo et al, 2003; De Iaco et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004) and one in Israel (Pansky et al, 2011). All five trials used two comparison groups. Only one trial (Di Spiezio Sardo et al, 2011) reported a statistical power calculation for one of the primary outcomes (incidence of adhesion formation). The protocol of all included trials was approved by the Institutional Review Board (IRB). The protocol of the trial from Israel (Pansky et al, 2011) was registered in a clinical trial registry (See NCT01377779 in Clinical Trials.gov). None of the trials reported on funding or other potential conflicts of interest.

*Participants

Two of the four Italian trials included subfertile women: 34 women out of 92 participants (Acunzo et al, 2003) and 21 out of 110 participants (Di Spiezio Sardo et al, 2011). The characteristics and data from these subfertile women were not available for individual patient data (IPD) meta-analysis. In the remaining two studies including 60 women (De Iaco et al, 2003) and 138 women (Guida et al, 2004) it is not clear whether and how many participants suffered from subfertility. We could not obtain further clarification from the study authors. The study from Israel included 30 women who were trying to conceive after miscarriage. The proportion of women suffering from subfertility was not reported; this could not be clarified (Pansky et al, 2011). Three of the four Italian trials (Acunzo et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004) were conducted in the same university hospital; several co-authors participated in the clinical research of all three trials. The in- and exclusion criteria were similar in these three studies; one trial included only women with intrauterine adhesions (Acunzo et al, 2003), the other trials included women with fibroids, polyps or uterine septa (Guida et al, 2004) or women with single or multiple intrauterine lesions except intrauterine adhesions or suffering from dysfunctional uterine bleeding (Di Spiezio Sardo et al, 2011). The description of the source population was not adequate in the fourth Italian study (De Iaco et al, 2003). The fifth included study was conducted in a source population of women with retained products of conception after miscarriage (Pansky et al, 2011). The mean age of the participants was below 35 years in one study

126 Anti-adhesion barrier gels following operative hysteroscopy

(Acunzo et al, 2003). In two trials the mean patient age in both comparison groups was above 35 years (Di Spiezio Sardo et al, 2011; Guida et al, 2004). The other two studies reported a range between 18 to 65 years (De Iaco et al, 2003) or 18 to 50 years (Pansky et al, 2011) without reporting data on the mean ages and SD in both comparison groups.

*Interventions

All trials randomly compared the intrauterine application of an anti- adhesion gel versus no gel following operative hysteroscopy. In three studies ACP gel was used (Acunzo et al, 2003; De Iaco et al, 2003; Guida et al, 2004); the other two (Di Spiezio Sardo et al, 2011; Pansky et al, 2011) used polyethylene oxide-sodium CMC gel for the intervention. In three trials the gel was administered into the uterine cavity through one of the flow channels of the resectoscope; the procedure was judged to be adequate when under hysteroscopic control the gel seemed to have replaced the liquid medium, filling the cavity from the fundus to the internal ostium of the cervix (Acunzo et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004). In one of these three studies (Acunzo et al, 2003) ultrasonographic data demonstrated that the anti-adhesive gel was able to keep the uterine walls separated for at least 72 hours. In one study (De Iaco et al, 2003) the gel was applied using the cannula in a blind way without using hysteroscopic visualization; for another trial (Pansky et al, 2011) the method of application of the anti-adhesion gel is not clear.

*Outcomes

The primary outcome of live birth was reported in none of the included studies; the incidence of adhesions was reported in all five studies. The following secondary outcomes were reported as follows: clinical pregnancy (Pansky et al, 2011), mean adhesion scores (Acunzo et al, 2003; Guida et al, 2004) and severity of the adhesions (Acunzo et al, 2003; De Iaco et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004; Pansky et al, 2011). The definition of pregnancy and the time period during which this secondary outcome was assessed in one trial (Pansky et al, 2011) was not described. Four studies (Acunzo et al, 2003 Di Spiezio Sardo et al, 2011; Guida et al, 2004; Pansky et al, 2011) used the 1988 AFS classification system for scoring intrauterine adhesions at second- look hysteroscopy; one trial (De Iaco et al, 2003) used the ASRM modified scoring system. In all five studies the incidence and the severity of adhesion formation was measured at one time point only, ranging from

127 Chapter 5

4 to 12 weeks after the operative hysteroscopy (Acunzo et al, 2003; De Iaco et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004; Pansky et al, 2011).

Risk of bias in included studies

Allocation (selection bias) We judged four of the five trials to be at low risk for selection bias related to random sequence generation (Acunzo et al, 2003; De Iaco et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004). One trial (Pansky et al, 2011) did not describe the method of random sequence generation; no further clarification could be obtained. We judged all five studies to be at unclear risk for selection bias related to allocation since they did not adequately describe the method of allocation concealment (Acunzo et al, 2003; De Iaco et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004; Pansky et al, 2011).

Blinding (performance bias and detection bias) In all five trials the method of blinding of the outcome assessors was not described (Acunzo et al, 2003; De Iaco et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004; Pansky et al, 2011).We judged this risk of bias item to be important for the outcomes ‘incidence of adhesions’, ‘mean adhesion scores’ and ‘severity of adhesions’ but less relevant for the outcomes of ‘live birth’, ‘ongoing’ or ‘clinical pregnancy’ and ‘miscarriage’ unless the follow up period was not long enough.

Incomplete outcome data (attrition bias) We judged four trials to be at low risk for attrition bias (Acunzo et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004; Pansky et al, 2011).We judged one study to be at high risk for attrition bias related to incomplete outcome data; the loss to follow up in this study of 33% (20 out of 60 enrolled women) is high and thus very likely to cause substantial attrition bias (De Iaco et al, 2003).

Selective outcome reporting (reporting bias) We judged all trials to be at low risk of reporting bias; no evidence for selective outcome reporting was retrieved in any of the included studies when comparing abstract, methods and results section.

128 Anti-adhesion barrier gels following operative hysteroscopy

Other potential sources of bias We judged three studies to be at low risk for other potential sources of bias (Acunzo et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004). We judged one study to be at an unclear risk for other potential sources of bias (Pansky et al, 2011); the other study (De Iaco et al, 2003) was judged to be at high risk of bias due to likely imbalance of patient characteristics, imbalanced distribution of co-treatment and other methodological study flaws.

Figure 2 Risk of bias summary: review authors' judgments about each risk of bias item for each included study.

Random sequence generation (selection (selection bias) Random sequence generation bias) (selection Allocation concealment Blinding of outcome (detection assessment bias): outcome: live birth or pregnancy Blinding of outcome (detection bias): outcome: adhesions assessment bias) Incomplete outcome data (attrition bias) (reporting reporting Selective Other bias

Acunzo 2003 + ? + ? + + +

De Iaco 2003 + ? + ? – + –

Di Spiezio Sardo 2011 + ? + ? + + +

Guida 2004 + ? + ? + + +

Pansky 2011 ? ? + ? + + ?

129 Chapter 5

Figure 3 Risk of bias graph: review authors' judgments about each risk of bias item presented as percentages across all included studies.

Random sequence generation (selection bias)

Allocation concealment (selection bias)

Blinding of outcome assessment (detection bias): outcome: live birth or pregnancy

Blinding of outcome assessment (detection bias): outcome: adhesions

Incomplete outcome data (attrition bias)

Selective reporting (reporting bias)

Other bias

0% 25% 50% 75% 100%

Low risk of bias Unclear risk of bias High risk of bias

Effects of interventions

Any gel versus no gel

*Primary outcomes

1. Live birth

There were no data for this primary outcome.

2. Incidence of adhesion formation at second-look hysteroscopy

The use of any gel after operative hysteroscopy decreases the incidence of adhesions (RR 0.65, 95% CI 0.45 to 0.93, P=0.02, 5 studies, 372 women) The NNTB is 9 (95% CI 5 to 33) (Figure 4).

130 Anti-adhesion barrier gels following operative hysteroscopy

Figure 4 Forest plot of comparison: Any anti-adhesion gel versus no gel. Outcome 2: Incidence of adhesions at second-look hysteroscopy. M-H: Mantel-Haenszel, random-effects model, CI: confidence interval. Any gel No gel Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI 4.1.1 Myomas Di Spiezio Sardo 2011 1 16 3 15 2.7% 0.31 [0.04, 2.68] Guida 2004 4 25 8 24 10.9% 0.48 [0.17, 1.39] Subtotal (95% CI) 41 39 13.7% 0.44 [0.17, 1.14] Total events 5 11 Heterogeneity: Tau² = 0.00; Chi² = 0.12, df = 1 (P = 0.72); I² = 0% Test for overall effect: Z = 1.68 (P = 0.09)

4.1.2 Polyps Di Spiezio Sardo 2011 0 22 0 20 Not estimable Guida 2004 2 34 6 33 5.4% 0.32 [0.07, 1.49] Subtotal (95% CI) 56 53 5.4% 0.32 [0.07, 1.49] Total events 2 6 Heterogeneity: Not applicable Test for overall effect: Z = 1.45 (P = 0.15)

4.1.3 Septa Di Spiezio Sardo 2011 0 6 3 7 1.6% 0.16 [0.01, 2.64] Guida 2004 1 8 3 8 3.1% 0.33 [0.04, 2.56] Subtotal (95% CI) 14 15 4.7% 0.26 [0.05, 1.35] Total events 1 6 Heterogeneity: Tau² = 0.00; Chi² = 0.17, df = 1 (P = 0.68); I² = 0% Test for overall effect: Z = 1.61 (P = 0.11)

4.1.4 Adhesions Acunzo 2003 6 43 13 41 16.0% 0.44 [0.18, 1.05] Subtotal (95% CI) 43 41 16.0% 0.44 [0.18, 1.05] Total events 6 13 Heterogeneity: Not applicable Test for overall effect: Z = 1.85 (P = 0.06)

4.1.5 Retained products of conception Pansky 2011 10 15 11 15 47.1% 0.91 [0.57, 1.45] Subtotal (95% CI) 15 15 47.1% 0.91 [0.57, 1.45] Total events 10 11 Heterogeneity: Not applicable Test for overall effect: Z = 0.40 (P = 0.69)

4.1.6 Not specified De Iaco 2003 5 18 7 22 13.1% 0.87 [0.33, 2.29] Subtotal (95% CI) 18 22 13.1% 0.87 [0.33, 2.29] Total events 5 7 Heterogeneity: Not applicable Test for overall effect: Z = 0.28 (P = 0.78)

Total (95% CI) 187 185 100.0% 0.65 [0.45, 0.93] Total events 29 54 Heterogeneity: Tau² = 0.01; Chi² = 7.31, df = 7 (P = 0.40); I² = 4% 0.001 0.1 1 10 1000 Test for overall effect: Z = 2.35 (P = 0.02) Favours any gel Favours no gel Test for subgroup differences: Chi² = 5.69, df = 5 (P = 0.34), I² = 12.2%

131 Chapter 5

*Secondary outcomes

3. Pregnancy

There is no evidence for a benefit with the use of polyethylene oxide- sodium CMC gel following operative hysteroscopy for suspected retained products of conception for the outcome of clinical pregnancy (RR 3.0, 95% CI 0.35 to 26, P=0.32, 1 study, 30 women) (Figure 5).

Figure 5 Forest plot of comparison: Any anti-adhesion gel versus no gel. Outcome 3: Pregnancy. M-H: Mantel-Haenszel, random-effects model, CI: confidence interval.

Any gel No gel Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI Pansky 2011 3 15 1 15 100.0% 3.00 [0.35, 25.68]

Total (95% CI) 15 15 100.0% 3.00 [0.35, 25.68] Total events 3 1 Heterogeneity: Not applicable 0.01 0.1 1 10 100 Test for overall effect: Z = 1.00 (P = 0.32) Favours no gel Favours any gel

4. Miscarriage

There were no data for this secondary outcome.

132 Anti-adhesion barrier gels following operative hysteroscopy

*Other secondary outcomes

5.1. Mean adhesion score at 3 months in women with fibroids, polyps or uterine septa

The use of ACP gel in women undergoing operative hysteroscopy for fibroids, endometrial polyps or uterine septa is associated with a lower mean adhesion score at second-look hysteroscopy at 3 months (MD -1.4, 95% CI -1.8 to -1.0, P<0.00001, 1 study, 24 women) (Figure 6).

Figure 6 Forest plot of comparison: Auto-cross-linked hyaluronic acid gel versus no gel. Outcome 5: mean adhesion score AFS 1988 at 3 months in women with myomas, polyps or uterine septa. IV: inverse variance, random-effects model, CI: confidence interval.

ACP gel No gel Mean Difference Mean Difference Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI 1.3.1 Myomas Guida 2004 2.25 0.5 4 3.5 1.19 8 16.7% -1.25 [-2.21, -0.29] Subtotal (95% CI) 4 8 16.7% -1.25 [-2.21, -0.29] Heterogeneity: Not applicable Test for overall effect: Z = 2.55 (P = 0.01)

1.3.2 Polyps Guida 2004 2 0.1 2 3.5 0.54 6 74.5% -1.50 [-1.95, -1.05] Subtotal (95% CI) 2 6 74.5% -1.50 [-1.95, -1.05] Heterogeneity: Not applicable Test for overall effect: Z = 6.48 (P < 0.00001)

1.3.3 Septa Guida 2004 4 0.1 1 5.33 1.15 3 8.9% -1.33 [-2.65, -0.01] Subtotal (95% CI) 1 3 8.9% -1.33 [-2.65, -0.01] Heterogeneity: Not applicable Test for overall effect: Z = 1.98 (P = 0.05)

Total (95% CI) 7 17 100.0% -1.44 [-1.83, -1.05] Heterogeneity: Tau² = 0.00; Chi² = 0.24, df = 2 (P = 0.88); I² = 0% -4 -2 0 2 4 Test for overall effect: Z = 7.22 (P < 0.00001) Favours ACP gel Favours no gel Test for subgroup differences: Chi² = 0.24, df = 2 (P = 0.88), I² = 0%

5.2. Mean adhesion score at 3 months in women with intrauterine adhesions

There are statistically significant differences in the mean adhesion scores at second-look hysteroscopy at 3 months after the use of ACP gel compared to operative hysteroscopy only in women undergoing operative

133 Chapter 5

hysteroscopy for intrauterine adhesions (MD -3.3, 95% CI -3.4 to -3.2, P<0.00001, 1 study, 19 women) (Figure 7).

Figure 7 Forest plot of comparison: Auto-cross-linked hyaluronic acid gel versus no gel. Outcome 5 mean adhesion score AFS 1988 at 3 months in women with intrauterine adhesions. IV: inverse variance, random-effects model, CI: confidence interval.

ACP gel No gel Mean Difference Mean Difference Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI Acunzo 2003 2 0.1 6 5.3 0.2 13 100.0% -3.30 [-3.43, -3.17]

Total (95% CI) 6 13 100.0% -3.30 [-3.43, -3.17] Heterogeneity: Not applicable -10 -5 0 5 10 Test for overall effect: Z = 47.91 (P < 0.00001) Favours ACP gel Favours no gel

5.3. Severity of adhesions at second-look hysteroscopy when using any gel

At second-look hysteroscopy there are more mild adhesions when using any gel following operative hysteroscopy (RR 2.8, 95% CI 1.1 to 7.0, P=0.03, 4 studies, 79 women) (Figure 8). Statistical heterogeneity beyond chance was high (Chi2=12.30, df=3 (P=0.006); I2=76%). The NNTB is 2 (95% CI 1 to 4).

Figure 8 Forest plot of comparison: Any anti-adhesion gel versus no gel. Outcome 5: AFS 1988 stage I (mild) adhesions at second- look hysteroscopy. M-H: Mantel-Haenszel, random-effects model, CI: confidence interval.

Any gel No gel Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI Acunzo 2003 6 6 3 13 26.4% 3.71 [1.47, 9.42] Di Spiezio Sardo 2011 2 3 1 12 12.9% 8.00 [1.04, 61.52] Guida 2004 6 7 4 17 26.8% 3.64 [1.47, 9.04] Pansky 2011 9 10 8 11 33.9% 1.24 [0.82, 1.88]

Total (95% CI) 26 53 100.0% 2.81 [1.13, 7.01] Total events 23 16 Heterogeneity: Tau² = 0.60; Chi² = 12.30, df = 3 (P = 0.006); I² = 76% 0.001 0.1 1 10 1000 Test for overall effect: Z = 2.22 (P = 0.03) Favours any gel Favours no gel

134 Anti-adhesion barrier gels following operative hysteroscopy

There is an effect favoring the use of any gel following operative hysteroscopy for the outcome of AFS 1988 stage II (moderate) adhesions at second-look hysteroscopy (RR 0.26, 0.09 to 0.80, P=0.02, 3 studies, 58 women) (Figure 9). The NNTB is 2 (95% CI 1 to 2).

Figure 9 Forest plot of comparison: Any anti-adhesion gel versus no gel. Outcome 5: AFS 1988 stage II (moderate) adhesions at second-look hysteroscopy. M-H: Mantel-Haenszel, random-effects model, CI: confidence interval.

Any gel No gel Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI Acunzo 2003 0 6 10 13 17.2% 0.10 [0.01, 1.40] Di Spiezio Sardo 2011 1 3 8 12 45.7% 0.50 [0.10, 2.60] Guida 2004 1 7 13 17 37.0% 0.19 [0.03, 1.17]

Total (95% CI) 16 42 100.0% 0.26 [0.09, 0.80] Total events 2 31 Heterogeneity: Tau² = 0.00; Chi² = 1.43, df = 2 (P = 0.49); I² = 0% 0.001 0.1 1 10 1000 Test for overall effect: Z = 2.36 (P = 0.02) Favours any gel Favours no gel

There is no evidence for an effect in favor of any gel versus no gel following operative hysteroscopy for the outcome of AFS 1988 stage III (severe) adhesions at second-look hysteroscopy (RR 0.46, 95% CI 0.03 to 7.2, P=0.58, 3 studies, 58 women) (Figure 10).

Figure 10 Forest plot of comparison: Any anti-adhesion gel versus no gel. Outcome 5: AFS 1988 stage III (severe) adhesions at second-look hysteroscopy. M-H: Mantel-Haenszel, random-effects model, CI: confidence interval.

Any gel No gel Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI Acunzo 2003 0 6 0 13 Not estimable Di Spiezio Sardo 2011 0 3 3 12 100.0% 0.46 [0.03, 7.21] Guida 2004 0 7 0 17 Not estimable

Total (95% CI) 16 42 100.0% 0.46 [0.03, 7.21] Total events 0 3 Heterogeneity: Not applicable 0.001 0.1 1 10 1000 Test for overall effect: Z = 0.55 (P = 0.58) Favours any gel Favours no gel

135 Chapter 5

For the composite outcome ’moderate or severe adhesions’ there are statistically significant differences favoring the use of any gel following operative hysteroscopy (RR 0.25, 95% CI 0.10 to 0.67, P=0.006, 4 studies, 79 women) (Figure 11). The NNTB is 2 (95% CI 1 to 4).

Figure 11 Forest plot of comparison: Any anti-adhesion gel versus no gel. Outcome 5: AFS 1988 stage II (moderate) or stage III (severe) adhesions at second-look hysteroscopy. M-H: Mantel-Haenszel, random-effects model, CI: confidence interval.

Any gel No gel Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI Acunzo 2003 0 6 10 13 13.2% 0.10 [0.01, 1.40] Di Spiezio Sardo 2011 1 3 11 12 36.8% 0.36 [0.07, 1.82] Guida 2004 1 7 13 17 28.3% 0.19 [0.03, 1.17] Pansky 2011 1 10 3 11 21.7% 0.37 [0.05, 2.98]

Total (95% CI) 26 53 100.0% 0.25 [0.10, 0.67] Total events 3 37 Heterogeneity: Tau² = 0.00; Chi² = 1.02, df = 3 (P = 0.80); I² = 0% 0.01 0.1 1 10 100 Test for overall effect: Z = 2.76 (P = 0.006) Favours any gel Favours no gel

*Subgroup analyses

We conducted a subgroup analysis according to the type of pathology treated by operative hysteroscopy. The use of any gel following operative hysteroscopy for fibroids (RR 0.44, 95% CI 0.17 to 1.1, P=0.09, 2 studies, 80 women), endometrial polyps (RR 0.32, 95% CI 0.07 to 1.5, P=0.15, 2 studies, 109 women), uterine septa (RR 0.26, 95% CI 0.05 to 1.3, P=0.11, 2 studies, 29 women), intrauterine adhesions (RR 0.44, 95% CI 0.18 to 1.0, P=0.06, 1 study, 84 women), retained products of conception (RR 0.91, 95 % CI 0.57 to 1.4, P=0.69, 1 study, 30 women) or 'not specified' (RR 0.87, 95% CI 0.33 to 2.3, P=0.78, 1 study, 40 women) consistently tends to decrease the incidence of adhesions at second-look hysteroscopy. The fact that the differences between both comparison groups within each subgroup were not statistically significant is a common problem with subgroup analyses due to smaller numbers. There is no evidence for substantial subgroup differences (Chi²=5.7, df=5 (P=0.34), I²=12%) as shown in figure 4. The data of this sensible predefined subgroup analysis should not be over interpreted.

136 Anti-adhesion barrier gels following operative hysteroscopy

*Sensitivity analyses

A sensitivity analysis was done to study the impact of the study quality on the treatment effect for the primary outcomes. If we excluded the single study at high risk of bias (De Iaco et al, 2003) in a sensitivity analysis, the use of any gel following operative hysteroscopy was still beneficial for decreasing the incidence of adhesions but the treatment effect was larger (RR 0.56, 95% CI 0.35 to 0.90, P=0.02, 4 studies, 332 women). A second sensitivity analysis to study the influence of the choice for the analysis model (fixed-effect rather than random-effects model) did not demonstrate any impact on the treatment effect (RR 0.55, 95% CI 0.38 to 0.79, P=0.001, 5 studies, 372 women). The choice of the summary effect measure (RR rather than OR) did not have a substantial effect on the magnitude or the direction of the estimated treatment effect nor on the statistical significance of the differences between the comparison groups (OR 0.40, 95% CI 0.23 to 0.70, P=0.001, 5 studies, 372 women).

137 Chapter 5

SUMMARY OF MAIN RESULTS

Table 1

Any anti-adhesive barrier gel following operative hysteroscopy for treating female subfertility

Outcomes No of Quality of the Relative Anticipated absolute Participants evidence effect effects (studies) (GRADE) (95% Risk Risk difference Follow up CI) with with Anti- Control adhesion barrier gels (95% CI) Incidence of 372 ْٚٚٚ RR 0.65 Study population adhesions (5 studies) VERY (0.45 to 1,2 292 per 102 fewer per hysteroscopy 1 to 3 months LOW 0.93) 1000 1000 due to risk of (from 20 fewer bias, to 161 fewer) indirectness Medium risk 510 per 179 fewer per 1000 1000 (from 36 fewer to 280 fewer) High risk 880 per 308 fewer per 1000 1000 (from 62 fewer to 484 fewer) Clinical 30 ْٚٚٚ RR 3.00 67 per 133 more per pregnancy (1 study) VERY (0.35 to 1000 1000 LOW3,4 26) (from 43 fewer unknown due to to 1000 more) indirectness, imprecision *The basis for the assumed medium risk for the outcome ‘incidence of adhesions’ is the average risk (51%)of adhesions for the different pathologies reported in the prospective cohort study by Yang et al,2013. The assumed high risk is based on the incidence of adhesions following septum resection (88%) reported by Yang et al, 2013. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the risk ratio of the intervention (and its 95% CI).CI: Confidence interval; RR: Risk ratio.

138 Anti-adhesion barrier gels following operative hysteroscopy

GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

1 The study by De Iaco 2003 was at high risk for attrition bias 2 In two of the four included trials 19% (Di Spiezio Sardo 2011) and 37% (Acunzo 2003) of the participants suffered from subfertility; for a third Italian trial (Guida 2004) it is unclear whether subfertile women were included. The proportion of women suffering from subfertility in the Israeli study (Pansky 2011) is not reported. 3 The number of women suffering from subfertility was not reported 4 The 95% confidence interval is very wide

139 Chapter 5

DISCUSSION

This systematic review aimed to appraise critically whether the use of any anti-adhesion barrier gel following operative hysteroscopy in subfertile women made a difference for the main outcomes of live birth, incidence of adhesion formation, pregnancy, miscarriage, mean adhesions scores or severity of adhesions at second-look hysteroscopy. We searched for RCTs on anti-adhesion barrier gels versus other barrier gels, placebo or no anti- adhesion barrier gels following operative hysteroscopy. We critically appraised five studies randomly comparing the use of any anti-adhesion gel versus no gel in women of reproductive age treated by operative hysteroscopy for fibroids, polyps, septa, adhesions or retained products of conception (Acunzo et al, 2003; De Iaco et al, 2003; Di Spiezio Sardo et al, 2011; Guida et al, 2004; Pansky et al, 2011). We judged a statistical pooling of the results of these five studies to be sensible given that substantial clinical diversity and statistical heterogeneity could confidently be ruled out. We refer to table 1 on the previous page for a summary of findings for the key outcomes clinical pregnancy and incidence of adhesions at second-look hysteroscopy (Table 1).

According to our meta-analysis there is evidence for a benefit with the use of any anti-adhesive gel following operative hysteroscopy for decreasing the incidence of adhesions at second-look hysteroscopy. The benefit for this outcome tends to be consistently present across different subgroups according to the type of pathology treated. The treatment effect for decreasing adhesions at second-look hysteroscopy is robust; the direction of the treatment effect and the statistical significance of the differences between the comparison groups are independent from the study quality and the choices made for the analysis model and the summary effect measure.

The use of ACP gel in women undergoing operative hysteroscopy for fibroids, endometrial polyps, uterine septa or intrauterine adhesions is associated with a lower mean adhesion score at second-look hysteroscopy at 3 months; the clinical relevance of this finding is unclear since only non-validated classification systems for scoring IUAs are used at the present. When adhesion formation is observed at second-look hysteroscopy, there are more mild adhesions and less moderate or severe adhesions by using any anti-adhesion gel after operative hysteroscopy. There is no evidence for a benefit with the use of polyethylene oxide- sodium CMC gel versus no gel in women treated by operative hysteroscopy for suspected retained products of conception for the

140 Anti-adhesion barrier gels following operative hysteroscopy outcome of pregnancy (Pansky et al, 2011). Although there was a beneficial trend when using the anti-adhesion gel, the differences between both comparison groups were not statistically significant. This may be a type II error: to detect a difference between both comparison groups of 13% in the clinical pregnancy rate with a statistical power of 80% at a confidence level of 95% (Į=0.05 and ȕ=0.20) and under the assumption of a loss to follow up of 10%, a sample size of 120 would be required; meaning 240 rather than 30 participants as in this single centre study.

Overall completeness and applicability of evidence

The evidence of the effectiveness of using any anti-adhesion gel versus no gel in women of reproductive age treated by operative hysteroscopy for fibroids, polyps, septa or intrauterine adhesions is limited: no data on live birth, pregnancy or miscarriage rate were retrieved. In women of reproductive age treated by operative hysteroscopy for retained products of conception, the use of polyethylene oxide-sodium CMC gel tends to increase the clinical pregnancy rate; the differences between both comparison groups were not statistically significant due to the inadequate statistical power of the trial. Moreover, the proportion of the women suffering from subfertility in both comparison groups was not reported and the trial was at unclear risk of bias for several items.

There are at the present two ongoing trials on the use of anti-adhesion barrier gels after operative hysteroscopy. The first is a parallel group randomized study on the effectiveness of applying Oxiplex/AP Gel (Intercoat) for preventing intrauterine adhesions in women aged 18 to 50 years following hysteroscopic surgery. This study is at the present not yet recruiting (NCT01637974). The second trial will address the effectiveness of hyaluronic acid gel in women older than 18 years following hysteroscopic surgery; this trial will not answer the research question in the present review since the primary and only outcome measured is the patient satisfaction rate two months after the gel application (NCT01464528).

There is some reason for concern about the applicability of the evidence retrieved: most trials were conducted in a target population including -but not limited to- women suffering from subfertility: two trials (Acunzo et al, 2003; Di Spiezio Sardo et al, 2011) included variable proportions of women suffering from subfertility, miscarriage or risk of preterm delivery, for two studies ( De Iaco et al, 2003; Guida et al, 2004) it is unclear whether and how many participants suffered from subfertility while the

141 Chapter 5

final fifth included study (Pansky et al, 2011) included a source population of women with proven fertility trying to conceive after miscarriage. It is unlikely that the mechanisms of adhesion formation might be different in subfertile compared to a general or fertile target population; nevertheless, we judge the overall applicability of the retrieved best available evidence in a more general source population to a target population of subfertile women to be limited.

Quality of the evidence

We graded the evidence for the randomized comparison between any anti- adhesion barrier gel versus no gel following operative hysteroscopy for the outcomes of pregnancy and incidence of adhesions as very low.

Agreements and disagreements with other studies or reviews

Two reviews support the use of anti-adhesive gel for reducing adhesion formation following operative hysteroscopy. The first review (Deans and Abbott, 2010) is a narrative review reporting the results and conclusions of one RCT included in the present systematic review (Acunzo et al, 2003). The second review (Mais et al, 2012) is a systematic review and meta-analysis studying the effectiveness of auto-cross-linked hyaluronan gel for adhesion prevention in laparoscopic and hysteroscopic surgery. The data of three RCTs included in the present systematic review (Acunzo et al, 2003; De Iaco et al, 2003; Guida et al, 2004) were pooled: the proportion of women with adhesions at second-look was significantly lower in women who received ACP gel than in the control group of women undergoing operative hysteroscopy without ACP gel (RR 0.50, 95% CI 0.31 to 0.85, P=0.009, 3 studies, 256 women). The authors used an older methodological tool (the Jadad scale) for assessing the validity of the included trials. The “scale” methodology is at the present no longer supported by the Cochrane Collaboration which recommends using a more formal assessment by means of the ‘Risk of bias’ tool. This different methodology explains the discrepancy between the statement of Mais et al that all the included trials in their systematic review were judged to be of a high quality which contrasts with our judgment of “very low quality evidence” for the outcomes of pregnancy and incidence of de novo adhesions.

One French small comparative study (n=54 women with uterine pathology) has studied the efficacy of auto-cross-linked hyaluronic acid gel in the prevention of adhesions following operative hysteroscopy

142 Anti-adhesion barrier gels following operative hysteroscopy

(Ducarme et al, 2006). The authors concluded that the use of auto-cross- linked hyaluronic acid gel does not reduce the incidence and the severity of intrauterine adhesions after hysteroscopic surgery. According to a more recent review of the literature (Revaux et al, 2008) -with the first author of the French comparative study (Ducarme et al, 2006) as co-author- the majority of the limited published studies until 2008 only evaluated the anatomic efficiency of anti-adhesion agents after hysteroscopic surgery. The authors concluded that the available data for the key reproductive outcomes were not convincing to promote the widespread clinical use of anti-adhesive barrier agents as an effective treatment strategy for infertile women treated by operative hysteroscopy; hence their conclusion that additional RCTs are needed.

143 Chapter 5

AUTHORS’ CONCLUSIONS

Implications for practice

Gynecologists should counsel their patients that intrauterine adhesion formation is the major long-term complication of operative hysteroscopy in women of reproductive age. The use of any barrier gel following operative hysteroscopy for suspected uterine cavity abnormalities in subfertile women might be considered: its use may decrease adhesion formation (very low quality evidence). If adhesion formation occurs, there are less moderate or severe adhesions and more mild adhesions by using any anti-adhesion gel; the mean adhesion scores at second-look hysteroscopy are lower after using ACP gel. Subfertile women nevertheless should be counseled that there is no evidence for higher live birth or pregnancy rates by using any barrier gel following operative hysteroscopy (very low quality evidence). There are no data at the present of the effects on the miscarriage rates.

Implications for research

The very low quality evidence retrieved from the small number of RCTs in a general source population including, but not restricted to subfertile women, is at the present not sufficient to draw robust conclusions in favor of any barrier as an adjunctive therapy following operative hysteroscopy for the key reproductive outcomes; more adequately powered and well- designed management RCTs are needed to assess whether the use of any anti-adhesion gel affects the live birth, the pregnancy and miscarriage rates in a target population of subfertile women. There are no data on a dose-response relationship between the size, the number or the severity of the treated pathology and the corresponding magnitude of the increase in effectiveness or decrease in the adverse outcomes that were defined in the present systematic review. A universally accepted classification system for scoring intrauterine adhesions validated for the prediction of spontaneous pregnancy is needed; this is a research priority that should be addressed by inception cohort studies.

144 Anti-adhesion barrier gels following operative hysteroscopy

REFERENCES

1. Acunzo G, Guida M, Pellicano M,Tommaselli GA, Di Spiezio Sardo A, Bifulco G,Cirillo D, Taylor A, Nappi C. Effectiveness of auto- cross-linked hyaluronic acid gel in the prevention of intrauterine adhesions after hysteroscopic adhesiolysis: a prospective, randomized, controlled study. Hum Reprod 2003; 18(9):1918-1921. 2. AFS: The American Fertility Society classifications of adnexal adhesions, distal tubal occlusion, tubal occlusion secondary to tubal ligation, tubal pregnancies, Müllerian anomalies and intrauterine adhesions. Fertil Steril 1988; 49: 944–955. 3. Baakdah H, Tulandi T. Adhesion in gynecology complication, cost, and prevention: a review. Surg Technol Int 2005; 14:185-190. 4. Belluco C, Meggiolaro F, Pressato D, Pavesio A, Bigon E, Dona M , Forlin M, Nitti D, Lise M. Prevention of Postsurgical Adhesions with an Autocrosslinked Hyaluronan Derivative Gel. J Surg Res 2001; 100:217–221. 5. Binda MM, Molinas CR, Bastidas A, Jansen M, Koninckx PR. Efficacy of barriers and hypoxia-inducible factor inhibitors to prevent CO2 pneumoperitoneum-enhanced adhesions in a laparoscopic mouse model. J Minim Invasive Gynecol 2007; 14(5): 591-599. 6. Binda MM, Koninckx PR. Prevention of adhesion formation in a laparoscopic mouse model should combine local treatment with peritoneal cavity conditioning. Hum Reprod 2009; 24(6):1473-1479. 7. Binda MM, Koninckx PR. Hyperoxia and prevention of adhesion formation: a laparoscopic mouse model for open surgery. BJOG 2010; 117(3):331-339. 8. Bosteels J, Kasius J, Weyers S, Broekmans FJ, Mol BWJ, D'Hooghe TM. Hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities. Cochrane Database Syst Rev 2013, Issue 1. Art. No.: CD009461. 9. Deans R, Abbott J. Review of Intrauterine Adhesions. J Minim Invasive Gynecol 2010; 17: 555-569. 10. DeCherney AH, diZerega GS. Clinical problem of intraperitoneal postsurgical adhesion formation following and the use of adhesion prevention barriers. Surg Clin North Am1997; 77: 671– 688. 11. De Iaco PA, Stefanetti M, Pressato D, Piana S, Donà M, Pavesio A, Bovicelli L. A novel hyaluronan-based gel in laparoscopic adhesion prevention: preclinical evaluation in an animal model. Fertil Steril 1998; 69:318-323.

145 Chapter 5

12. De Iaco PA, Muzzupapa G, Bovicelli A, Marconi S, Bitti SR, Sansovini M, Bovicelli L. Hyaluronan derivative gel (Hyalobarrier gel) in intrauterine adhesion (IUA) prevention after operative hysteroscopy. Ellipse 2003; 19(1):15-18. 13. D'Hooghe TM, Kyama CM, Chai D, Fassbender A, Vodolazkaia A, Bokor A, Mwenda JM. Nonhuman primate models for translational research in endometriosis. Reprod Sci 2009 Feb; 16 (2):152-161. 14. Diamond MP, DeCherney AH, Linsky CB, Cunningham T, Constantine B. Adhesion re-formation in the rabbit uterine horn model: I. Reduction with carboxymethylcellulose. Int J Fertil 1988; 33:372-375. 15. Di Spiezio Sardo A, Spinelli M, Bramante S, Scognamiglio M, Greco E, Guida M, Cela V, Nappi C. Efficacy of a Polyethylene Oxide– Sodium Carboxymethylcellulose Gel in Prevention of Intrauterine Adhesions After Hysteroscopic Surgery. J Minim Invasive Gynecol 2011; 18(4):462–469. 16. DiZerega GS. Contemporary adhesion prevention. Fertil Steril 1994; 61:219-235. 17. Donnez J, Nisolle M. Hysteroscopic adhesiolysis of intrauterine adhesions (Asherman syndrome). In: Donnez J, editor. Atlas of Laser Operative Laparoscopy and Hysteroscopy. London, England: Parthenon Publishing Group; 1994. p. 305–322. 18. Ducarme G, Davitian C, Zarrouk S, Uzan M, Poncelet C. Interest of auto-cross-linked hyaluronic acid gel in the prevention of intrauterine adhesions after hysteroscopic surgery: a case-control study. J Gynecol Obstet Biol Reprod 2006; 35(7):691-695. 19. Guida M, Acunzo G, Di Spiezio Sardo A, Bifulco G, Piccoli R, Pellicano M, Cerrota G, Cirillo D, Nappi C. Effectiveness of auto- cross-linked hyaluronic acid gel in the prevention of intrauterine adhesions after hysteroscopic surgery: a prospective, randomized, controlled study. Hum Reprod 2004; 19(6):1461-1464. 20. Hamou J, Salat-Baroux J, Siegler A. Diagnosis and treatment of intrauterine adhesions by microhysteroscopy. Fertil Steril 1983; 39: 321–326. 21. Higgins JPT, Green S (editors), Cochrane Handbook for Systematic Reviews of Interventions. Chichester (UK): John Wiley & Sons, 2008. 22. Efficiency of INTERCOAT (Oxiplex/AP Gel) in Preventing Intrauterine Adhesion Formation in Hysteroscopic Surgery - a Prospective Double Blind Randomized Study. ClinicalTrials.gov: NCT01637974.

146 Anti-adhesion barrier gels following operative hysteroscopy

23. Kelekci S, Yilmaz B, Oguz S, Zergero÷lu S, Inan I, Tokuco÷lu S. The efficacy of a hyaluronate/carboxymethylcellulose membrane in prevention of postoperative adhesion in a rat uterine horn model. Tohoku J Exp Med 2004; 204:189-194. 24. Koçak I, Unlü C, Akçan Y, Yakin K. Reduction of adhesion formation with cross-linked hyaluronic acid after peritoneal surgery in rats. Fertil Steril 1999; 72: 873-878. 25. Kodaman PH, Arici A. Intra-uterine adhesions and fertility outcome: how to optimize success? Curr Opin Obstet Gynecol 2007; 19(3):207- 214. 26. Laurent TC, Fraser JRE. Hyaluronan. FASEB J 1992; 6:2397-2404. 27. Leach RE, Burns JW, Dawe EJ, SmithBarbour MD, Diamond MP. Reduction of postsurgical adhesion formation in the rabbit uterine horn model with use of hyaluronate/carboxymethylcellulose gel. Fertil Steril 1998; 69(3):415-417. 28. Mais V, Cirronis MG, Peiretti M, Ferrucci G, Cossu E, Melis GB. Efficacy of auto-cross-linked hyaluronan gel for adhesion prevention in laparoscopy and hysteroscopy; a systematic review and meta- analysis of randomized controlled trials. Eur J Obstet Gynecol Reprod Biol 2012; 160:1-5. 29. March C, Israel R, March A. Hysteroscopic management of intrauterine adhesions. Am J Obstet Gynecol 1978; 130: 653–657. 30. Nasr A, Al-Inany H, Thabet S, Aboulghar M. A clinicohysteroscopic scoring system of intrauterine adhesions. Gynecol Obstet Invest 2000; 50:178–181. 31. Nimrod A, Ezra E, Ezov N, Nachum G, Parisada B. Absorption, distribution, metabolism and excretion of bacteria-derived hyaluronic acid in rats and rabbits. J Ocul Pharmacol 1992; 8:161-172. 32. Okulicz WC. Regeneration. In: Glasser SR, Aplin JD, Giudice LC, Tabibzadeh S, editor(s). The Endometrium. London: Taylor and Francis, 2002:110-120. 33. Pansky M, Fuchs N, Ami I Ben, Tovbin Y, Halperin R, Vaknin Z, Smorgick N. Intercoat (Oxiplex/AP Gel) for Preventing Intrauterine Adhesions Following Operative Hysteroscopy for Suspected Retained Products of Conception – A Pilot Study. J Minim Invasive Gynecol 2011; 18(S21):68. 34. Rajab TK, Wallwiener M, Planck C, Brochhausen C, Kraemer B, Wallwiener CW. A Direct Comparison of Seprafilm, Adept, Intercoat, and Spraygel for Adhesion Prophylaxis. J Surg Res 2010; 161:246– 249. 35. Renier D, Bellato PA, Bellini D, Pavesio A, Pressato D, Borrione A. Pharmacokinetic behaviour of ACP gel, an auto-cross-linked

147 Chapter 5

hyaluronan derivative, after intraperitoneal administration. Biomaterials 2005; 26(26):5368–5374. 36. Revaux A, Ducarme G, Luton D. Prevention of intrauterine adhesions after hysteroscopic surgery. Gynecol Obstet Fertil 2008; 36(3):311- 317. 37. Schenker JG, Margalioth EJ. Intrauterine adhesions: an updated appraisal. Fertil Steril 1982; 37:593-610. 38. Schonman R, Corona R, Bastidas A, De Cicco C, Mailova K, Koninckx PR. Intercoat Gel (Oxiplex): Efficacy, Safety, and Tissue Response in a Laparoscopic Mouse Model. J Minim Invasive Gynecol 2008; 16(2):188-194. 39. Shamiyeh A, Danis J, Benkö L, Vattay P, Röth E, Tulipan L, Shebl O, Wayand W. Effect of hyaluron derivate gel in prevention of postsurgical peritoneal adhesions--an experimental study in pigs. Hepatogastroenterology 2007; 54(76):1121-1124. 40. Taskin O, Sadik S, Onoglu A, Gokdeniz R, Erturan E, Burak F, Wheeler JM. Role of endometrial suppression on the frequency of intrauterine adhesions after resectoscopic surgery. J Am Assoc Gynecol Laparosc 2000; 7(3):351-354. 41. Use of hyaluronic acid gel to prevent intrauterine adhesions in hysteroscopic surgery. ClinicalTrials.gov: NCT01464528. 42. Valle RF, Sciarra JJ. Intrauterine adhesions: hysteroscopic diagnosis, classification, treatment, and reproductive outcome. Am J Obstet Gynecol 1988; 158: 1459–1470. 43. Wallwiener M, Brucker S, Hierlemann H, Brochhausen C, Solomayer E, Wallwiener C. Innovative barriers for peritoneal adhesion prevention: liquid or solid? A rat uterine horn model. Fertil Steril 2006; 86 (4 Suppl):1266-1276. 44. Wamsteker K, De Block S. Diagnostic hysteroscopy: technique and documentation. In: Sutton C, Diamond M, editors. Endoscopic surgery for gynecologists. London: WB Saunders; 1998. p. 511–524. 45. Yang JH, Chen MJ, Chen CD, Chen SU, Ho HN, Yang YS. Optimal waiting period for subsequent fertility treatment after various hysteroscopic . Fertil Steril 2013 Jun;99(7):2092-2096. 46. Zegers-Hochschild F, Adamson GD, de Mouzon J, Ishihara O, Mansour R, Nygren K, Sullivan E, Vanderpoel S; International Committee for Monitoring Assisted Reproductive Technology; World Health Organization. International Committee for Monitoring Assisted Reproductive Technology (ICMART) and the World Health Organization (WHO) revised glossary of ART terminology, 2009. Fertil Steril 2009; 92(5):1520–1524.

148

Anti-adhesion therapy following operative hysteroscopy for treating female subfertility

Jan Bosteels, Jenneke C. Kasius, Steven Weyers, Frank J. M. Broekmans, Ben Willem Mol, Thomas M. D’Hooghe

The Cochrane Library, Cochrane Database of Systematic Reviews 2014, Issue 5. Art.No.:CD011110 http://onlinelibrary.wiley.com /doi/10.1002/14651858.CD011110/pdf

149

Anti-adhesion therapy following operative hysteroscopy

BACKGROUND

Description of the interventions for preventing intrauterine adhesions Observational studies have suggested that several different anti-adhesion strategies might be used for preventing intrauterine adhesions following operative hysteroscopy.

IUD or Foley catheter balloon An intrauterine device (IUD) may provide a physical barrier between the uterine walls, separating the endometrial layers after lysis of IUAs. Its insertion as an adjunctive therapy has been recommended in at least 13 observational studies (Deans 2010). The use of a Foley catheter balloon has been reported as an alternative for similar purposes in eight observational studies (Deans 2010).

Hormone therapy In 1964, Wood and Pena suggested to prescribe estrogen therapy to stimulate the regeneration of the endometrium after the surgical treatment of IUAs (Wood 1964).

Barrier gels Hyaluronic acid or hyaluronan (HA), is a water soluble polysaccharide: it consists of multiple disaccharide units of glucuronic acid and N-acetyl glucosamine, bound together by a ß1-3-type glycoside bond. Solutions of HA have visco-elastic properties which have led to interests in developing applications of HA in surgical procedures, for example in ocular surgery and prevention of postsurgical adhesions. However, HA may not be the ideal substance for all procedures, due to its limited residence time when applied to a surgical site. It quickly enters the systemic circulation and is then cleared rapidly by catabolic pathways. Attempts to use hyaluronan for preventing postsurgical adhesions have therefore been met with variable success. Chemically modified derivatives of HA have been developed to circumvent the disadvantages of HA. One such derivative is auto-cross-linked polysaccharide (ACP). It is formed by cross linking hyaluronan, via direct formation of covalent ester bonds between hydroxyl and carboxyl groups of the hyaluronan molecule. ACP can be prepared with various degrees of cross linking, which allows tailoring of the viscosity properties of ACP gels (Renier 2005). Carboxymethylcellulose (CMC) is a high-molecular-weight polysaccharide that has a viscosity greater than Dextran 70. CMC can be used for adhesion prevention as a membrane barrier or a gel as a mixture of chemically derivative sodium hyaluronate and carboxymethylcellulose gel (HA-CMC) (Leach 1998).

151 Chapter 6

Human amnion membrane grafting Over the last three decades, the surgical community has become aware of the increasing potential of human amnion membrane (HAM) as an adjunctive anti-adhesion intervention. The use of whole human fetal membranes or amnion alone in surgery has primarily developed to aid the repair of surface epithelial defects in the skin, eye, abdominal wall and peritoneum. HAM grafting has not been very popular in the field of obstetrics and gynecology: its clinical use is limited to the use as a graft in forming an artificial vagina or as a barrier to prevent postoperative intra- abdominal adhesion formation or finally as a biological dressing following radical vulvectomies and groin dissections (Amer 2006).

How these interventions might work The hypothetical underlying mechanisms of subfertility associated with IUAs are obstruction of sperm transport into the cervix, impaired embryo migration within the uterine cavity or failure of embryo implantation due to endometrial insufficiency (Deans 2010). The ideal anti-adhesion adjunctive therapy following operative hysteroscopy would be the application of a biologically active mechanical separator that achieves the suppression of intrauterine adhesion formation and promotes the healing of the endometrium. The bulk of evidence on how the different interventions might work is derived from animal studies-largely in rodents and not in validated animal models for the study of human reproduction- or observational studies.

IUD or Foley catheter balloon The use of an IUD (13 observational studies) or a Foley catheter balloon (8 observational studies) (Deans 2010) is often recommended following the hysteroscopic treatment of IUAs or septoplasty to act as a physical barrier separating the opposing walls of the uterine cavity. The type of IUD may be important; copper-containing IUDs provoke an inflammatory reaction with probably detrimental effects whereas T-shaped IUDs might have too small a surface area to be truly effective in providing an efficient physical barrier. The loop IUD (e.g. Lippes loop) is generally considered the IUD of choice when treating IUAs; it is however no longer available in many countries (Kodaman 2007). One controlled clinical trial (CCT) (Orhue 2003) compared the use of a Foley catheter balloon for 10 days (N=59) versus the insertion of an IUD during three months (N=51); the fertility rates were equally poor in both the IUD group (20/59 or 34%) and the Foley catheter group (14/51 or 28%)(P=0.47).

152 Anti-adhesion therapy following operative hysteroscopy

Hormone therapy Many studies recommend the use of a cyclical estrogen and progestogen treatment regimen following the hysteroscopic treatment of IUAs to promote the regeneration of the endometrium (Deans 2010). Various regimens have been proposed consisting of estrogen (e.g. a typical daily dose of 2.5 mg of conjugated equine estrogen twice daily for 30 days) with or without a progestin (e.g. 10 mg medroxyprogesterone acetate for 10 days) (Kodaman 2007); no comparative studies have been performed on dosage, administration, or combination of hormones (Deans 2010). In a randomized study (Farhi 1993) 60 women undergoing dilation and curettage during the first trimester of pregnancy were allocated to receive estrogen and progestin or no treatment. Women in the intervention group had a significantly thicker endometrium (8.4 vs. 6.7mm, P=0.02) compared with the control group. The authors concluded that postoperative hormone treatment may be beneficial for intrauterine adhesion prevention following surgical trauma to the uterine cavity. Nevertheless, no data were available on pregnancy outcome or intrauterine adhesion formation or recurrence (Farhi 1993). A systematic review of observational studies concluded that hormonal therapy, particularly estrogen therapy, may be beneficial to women with IUAs but as an adjunctive therapy combined with other anti-adhesion strategies (Johary 2013).

Barrier gels The use of the biodegradable gel surgical barriers is based on the principle of keeping the adjacent wound surfaces as mechanically separate (Renier 2005). Several preclinical studies in various animal models have demonstrated the effectiveness of both ACP (Belluco 2001; Binda 2007; Binda 2009; Binda 2010; De Iaco 1998; Koçak 1999; Shamiyeh 2007; Wallwiener 2006) and HA-CMC gels (Leach 1998; Schonman 2008) or HA-CMC membranes (Kelekci 2004; Rajab 2010) for preventing postsurgical adhesions. Other preclinical studies in animal models suggest that HA gel remains in situ for more than 5 to 6 days (Laurent 1992; Nimrod 1992). Similarly, animal studies demonstrate the persistence of HA-CMC for about seven days after its application (Diamond 1988). The precise mechanisms by which ACP and HA-CMC are able to reduce adhesion reformation are not well known, but may be related to "hydro flotation" or "siliconizing" effects. One French CCT (N=54 women) compares the application of ACP gel (n=30) versus no gel (n=24) at the end of an operative hysteroscopic procedure for treating myomas, polyps, uterine septa or IUAs; there are no statistically significant differences for the rate of adhesion formation between both comparison groups nor for

153 Chapter 6

the mean adhesion scores or the severity of the adhesions (Ducarme 2006). There were no data on the reproductive outcome.

Human amnion membrane grafting The preclinical data on the effectiveness of HAM grafting in different animal models present conflicting results: one trial (Szabo 2002) demonstrates a beneficial effect in preventing de novo adhesions whereas in two other animal studies (Arora 1994; Badawy 1989) HAM grafting fails to prevent de novo adhesion formation. One observational study reports on the use of a fresh amnion graft over an inflated Foley catheter to prevent recurrence of intrauterine adhesions after hysteroscopic lysis in 25 women with moderate to severe Asherman syndrome: minimal adhesion reformation was demonstrated in 48% of study participants with severe adhesions. The authors conclude that HAM grafting might be promising as an adjunctive therapy following hysteroscopic adhesiolysis; it acts as a biologically active mechanical barrier suppressing adhesion formation and promoting endometrial healing (Amer 2006). Fresh HAM graft preserves its viability for 21 days following its application in the pelvic cavity (Trelford Sauder 1977). In addition to being an anatomical barrier HAM may promote the regeneration of epithelium by acting as a basement membrane substrate; HAM may facilitate the migration of epithelial cells, reinforce the adhesion of the basal epithelium, promote epithelial cell differentiation (Meller 1999) and prevent cellular apoptosis (Hori 2006). Human amnion epithelial cells produce factors or create a microenvironment for effective tissue repair and endometrial regeneration, possibly by stimulating endogenous stem cells (Padykula 1991).

154 Anti-adhesion therapy following operative hysteroscopy

OBJECTIVES

To assess the effectiveness of anti-adhesion therapy versus placebo, no therapy or an alternative anti-adhesion therapy following operative hysteroscopy for the treatment of female subfertility.

METHODS

Criteria for considering studies for this review

Types of studies Published and unpublished parallel group randomized controlled trials (RCTs) will be eligible for inclusion. We will exclude non-randomized studies (e.g. studies with evidence of inadequate sequence generation such as alternate days, patient numbers) as they are associated with a high risk of bias. We will include crossover trials if individually randomized women are the unit of analysis; only data from the first phase will be included in meta-analyses, as the crossover is not a valid study design in the context of subfertility.

Types of participants Women of reproductive age undergoing operative hysteroscopy for subfertility associated with suspected or unsuspected intrauterine pathology before spontaneous conception or any subfertility treatment. Studies excluding women wishing to conceive will not be eligible.

Types of interventions We will include the following randomized comparisons: • Anti-adhesion therapy versus placebo or no active anti- adhesion therapy following operative hysteroscopy. • Anti-adhesion therapy A versus anti-adhesion therapy B following operative hysteroscopy.

Types of outcome measures

*Primary outcomes

1. Effectiveness: live birth, defined as a delivery of at least one live fetus after 20 weeks of gestational age that resulted in at least one live baby born; we will count the delivery of singleton, twin or multiple pregnancies as one live birth.

155 Chapter 6

2. Adverse event: incidence of adhesion formation at second-look hysteroscopy.

*Secondary outcomes

3. Effectiveness: clinical pregnancy, defined as a pregnancy diagnosed by ultrasonographic visualization of one or more gestational sacs or definitive clinical signs of pregnancy; it includes ectopic pregnancy. We will count multiple gestational sacs as one clinical pregnancy.

4. Adverse event: miscarriage; mean adhesion scores and severity of adhesions at second look hysteroscopy. A miscarriage is the spontaneous loss of a clinical pregnancy that occurs before 20 completed weeks of gestational age (18 weeks post fertilization) or, if gestational age is unknown, the loss of an embryo/fetus of less than 400 grams.

We will avoid excluding studies on the basis of their reported outcome measures. Eligible studies that could have measured the outcomes of interest will be reviewed; we will report any lack of data for the key outcomes in the final review. We aim to follow the ICMART terminology (International Committee for Monitoring Assisted Reproductive Technologies) for the key reproductive outcomes (live birth, pregnancy and miscarriage) as much as possible (Zegers-Hochschild 2009); we will contact the primary study authors for clarification in case of unclear definitions. We will report any discrepancies in the final review. There are at the present seven reported classification systems for scoring the extent or severity of intrauterine adhesions. None of these systems has been validated or universally accepted (Deans 2010). We will therefore avoid pooling data from studies using different scoring systems; we will ask clarification from the primary study authors if necessary.

156 Anti-adhesion therapy following operative hysteroscopy

Search methods for identification of studies

We will search for all published and unpublished RCTs of anti-adhesion following operative hysteroscopy in subfertile women, without language restriction and in consultation with the Menstrual Disorders and Subfertility Group (MDSG) Trials Search Co-ordinator.

Electronic searches We will search the following electronic databases, trial registers and web sites using the search strategies in the appropriate appendices: the Cochrane Central Register of Controlled Trials (CENTRAL), the Menstrual Disorders and Subfertility Group (MDSG) Specialized Register, MEDLINE using OVID and EMBASE using EMBASE.com from inception till the present. The search strategy will combine both index and free-text terms. • Cochrane Database of Systematic Reviews (CDSR). • Database of Abstracts of Reviews of Effects (DARE) and the Health Technology Assessment Database (HTA Database) through the Centre for Reviews and Dissemination. • National Guideline Clearinghouse. • BIOSIS previews through ISI Web of Knowledge. • Trial registers for ongoing and registered trials: 'Current Controlled Trials' and the World Health Organization International Clinical Trials Registry. • Citation indexes: Science Citation Index through Web of Science. • Conference abstracts and proceedings on the ISI Web of Knowledge. • LILACS database, which is a source of trials from the Spanish and Portuguese speaking world. • European grey literature through Open Grey database. • General search engines: Turning Research into Practice (TRIP) database, Google Scholar and Scirus.

Searching other resources Two review authors (JB and JK) will handsearch reference lists of articles retrieved by the search and contact experts in the field to obtain additional data. We will contact the first or corresponding authors of included studies to ascertain if they are aware of any ongoing or unpublished trials. We will also handsearch relevant journals and conference abstracts that are not covered in the MDSG register, in liaison with the Trials Search Co- ordinator. The search process will be reported in a PRISMA (Preferred

157 Chapter 6

Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram in the review.

Data collection and analysis

Selection of studies After an initial screen of titles and abstracts retrieved by the search, conducted by JB, the full texts of all potentially eligible studies will be retrieved. Two review authors (FB and TD) will independently examine these full-text articles for compliance with the inclusion criteria and select studies eligible for inclusion in the review. We will correspond with study investigators as required, to clarify study eligibility. Disagreements as to study eligibility will be resolved by discussion or by a third review author (BWM). We will classify the study as ’awaiting classification’ if disagreements between review authors cannot be resolved and will report the disagreement in the final review.

Data extraction and management Two review authors (JB and SW) will independently extract data from eligible studies using a data extraction form designed and pilot-tested by the authors. Any disagreements will be resolved by discussion or by a third review author (BWM). Data extracted will include study characteristics and outcome data. Where studies have multiple publications, the main trial report will be used as the reference and additional details derived from secondary papers. We will correspond with study investigators for further data on methods and/or results, as required. We will include studies irrespective of whether outcomes are reported in a “usable” way. In multi-arm studies, data from arms that do not meet eligibility criteria will be excluded.

Assessment of risk of bias in included studies Two reviewers (JB and SW) will independently assess the included studies for risk of bias using the ‘Risk of bias’ tool. We will resolve disagreements by discussion or by arbitration from a third review author (BWM). We will describe all judgments fully and present the conclusions in the ‘Risk of bias’ table, which will be incorporated into the interpretation of review findings by means of sensitivity analyses. Selective reporting is a type of reporting bias that affects the internal validity of an individual study. It refers to the selective reporting of some outcomes (e.g. positive outcomes) and the failure to report others (e.g. adverse events). We will take care to search for within-trial selective reporting, such as trials failing to report obvious outcomes, or reporting

158 Anti-adhesion therapy following operative hysteroscopy them in insufficient detail to allow inclusion. We will seek published protocols and compare the outcomes between the protocol and the final published study. Where identified studies fail to report the primary outcome of live birth, but do report interim outcomes such as pregnancy, we will undertake informal assessment as to whether the interim values (e.g. pregnancy rates) are similar to those reported in studies that also report live birth. If there are outcomes defined in the protocol or the study report with insufficient data to allow inclusion, the review will indicate this lack of data and suggest that further clinical trials need to be conducted to clarify these knowledge gaps.

Measures of treatment effect For dichotomous data (e.g. live birth or clinical pregnancy rates) we will use the numbers of events in the two comparison groups of each study to calculate Mantel-Haenszel odds ratios (ORs). We will treat ordinal data (e.g. adhesion scores) as continuous data. For ordinal data (e.g. adhesion scores), if all studies report exactly the same outcomes we will calculate mean difference (MDs) between treatment groups. If similar outcomes are reported on different scoring scales we will not calculate the standardized mean difference (SMD) because the seven different adhesion score classifications have not been validated. We will reverse the direction of effect of individual studies, if required, to ensure consistency across trials. We will present 95% confidence intervals for all outcomes. We will make contact with the corresponding or first authors of all included trials that report data in a form that is not suitable for meta-analysis, for example time-to-pregnancy data (TTP). We will report the data of those reports that fail to present additional data that could be analyzed under ’other data’. We will not include TTP data in any meta-analysis. Where data to calculate ORs or MDs are not available, we will utilize the most detailed numerical data available that may facilitate similar analyses of included studies (e.g. test statistics, P-values). We will compare the magnitude and direction of effect reported by studies with how they are presented in the review, taking account of legitimate differences.

Unit of analysis issues The primary analysis will be per woman randomized; per pregnancy data will be included for some outcomes (e.g. miscarriage). If studies report only “per cycle” data, we will contact the primary study authors and request “per woman” data. If these are not available, the "per cycle" data will be briefly summarized in an additional table and will not be meta- analyzed. Multiple live births (e.g. twins or triplets) will be counted as one

159 Chapter 6

live birth event. Only first-phase data from crossover trials will be included.

Dealing with missing data We will analyze the data on an intention-to-treat basis as much as possible; if needed, attempts will be made to obtain missing data from the original researchers. Where these are unobtainable, imputation of individual values will be undertaken for the beneficial primary outcome (live birth) only: we will assume that live births would not have occurred in women without a reported outcome. For all other outcomes, we will only analyze the available data. Any imputation undertaken for missing data on the primary outcomes will be subjected to sensitivity analysis. If studies report sufficient detail to calculate mean differences but no information on associated standard deviation (SD), the outcome will be assumed to have standard deviation equal to the highest SD from other studies within the same analysis.

Assessment of heterogeneity We will consider whether the clinical and methodological characteristics of the included studies are sufficiently similar for meta-analysis to provide a clinically meaningful summary. We will carry out a formal assessment of statistical heterogeneity by using the I² statistic combined with the Q- statistic. Cochran’s Q test, a kind of Chi² statistic, is the classical measure to test significant heterogeneity. Cochran’s Q test is calculated as the weighted sum of squared differences between individual study effects and the pooled effect across studies. The Q-statistic follows Chi² distribution with k-1 degree of freedom where k is the number of studies. Q > k-1 suggests statistical heterogeneity. A low P value of Cochran’s Q test means significant heterogeneous results among different studies; usually, the P value at 0.10 is used as the cut-off. The Q-statistic has low power as a comprehensive test of heterogeneity especially when the number of studies is small. The Q-statistic informs us about the presence or absence of heterogeneity; it does not report on the extent of such heterogeneity. The I² statistic describes the percentage of variation across studies due to significant heterogeneity rather than random chance. It measures the extent of heterogeneity. An I² measurement greater than 50% will be taken to indicate substantial heterogeneity (Higgins 2003). We will explore possible explanations if substantial heterogeneity is detected.

Assessment of reporting biases Reporting biases arise when the dissemination of research findings is influenced by the nature and direction of results. Some types of reporting

160 Anti-adhesion therapy following operative hysteroscopy bias (e.g. publication bias, multiple publication bias, language bias etc) reduce the likelihood that all studies eligible for a review will be retrieved. The review may be biased if not all eligible studies are retrieved. In view of the difficulty of detecting and correcting for publication bias and other reporting biases, we will aim to minimize their potential impact by ensuring a comprehensive search for eligible studies and by being alert for duplication of data. If there are ten or more studies in an analysis, we will use a funnel plot to explore the possibility of small study effects (a tendency for estimates of the intervention effect to be more beneficial in smaller studies).

Data synthesis

One review author (JB) will enter the data and carry out the statistical analysis of the data in Review Manager 5. If the studies are sufficiently similar and substantial statistical heterogeneity can be confidently ruled out, we will combine the data from the primary studies in a meta-analysis with Review Manager 5 software using the summary Mantel-Haenszel (M-H) odds ratios (ORs) and a random-effects model (REM) for the following comparisons:

• Anti-adhesion therapy versus placebo or no active anti-adhesion therapy following operative hysteroscopy. • Anti-adhesion therapy A versus anti-adhesion therapy B following operative hysteroscopy.

The outcomes 'live birth' and 'clinical pregnancy' are considered positive outcomes of effectiveness and by consequence, higher numbers will be considered as a benefit. The outcomes 'incidence of adhesion formation', 'miscarriage', 'mean adhesion scores' and 'severity of adhesions' at second- look hysteroscopy are negative effects and higher numbers will be considered harmful. An increase in the odds of a particular outcome, which may be beneficial (e.g. live birth) or detrimental (e.g. de novo adhesions), will be displayed graphically in the meta-analyses to the right of the centre-line and a decrease in the odds of an outcome to the left of the centre-line. We will aim to define analyses that are comprehensive and mutually exclusive, so that all eligible study results can be slotted into only one stratum of the two predefined categories of randomized comparisons, and so that trials within the same stratum can be pooled. Stratification is not a requirement, but allows consideration of effects within each stratum as well as, or instead of, an overall estimate for the comparison. The use of

161 Chapter 6

the REM instead of a fixed-effect analysis model (FEM) is justified by the fact that the results of a similar surgical treatment may be different across studies; despite rigorous standardization, there might be inevitably differences in surgical skill among the different surgeons involved in the trials. If no RCTs are retrieved for some comparisons, the review will indicate their absence identifying knowledge gaps which need further research. We will undertake a narrative overview if meta-analysis is not appropriate.

162 Anti-adhesion therapy following operative hysteroscopy

Subgroup analysis and investigation of heterogeneity Where enough data are available, we will conduct subgroup analyses to determine the separate evidence within the following subgroups: • Studies that report both ‘live birth’ and ‘clinical pregnancy’ in order to assess any overestimation of the treatment effect and reporting bias. • According to the type, extent or severity of the uterine abnormality treated. We will report the interpretation of any subgroup analysis restrictively even if enough data were available; subgroup analysis is by its nature an observational study which can be helpful in generating or exploring hypotheses. Moreover the interpretation of the statistical analysis for subgroups is not without problems.

Sensitivity analysis We will conduct sensitivity analyses for the primary outcomes to determine whether the conclusions are robust to arbitrary decisions made regarding the eligibility and analysis. These analyses will include consideration of whether the review conclusions would have differed if: • Eligibility were restricted to studies without high risk of bias versus all studies • A fixed-effect rather than a random-effects model had been adopted. • Alternative imputation strategies had been implemented. • The summary effect measure was relative risk rather than odds ratio.

Grading the evidence We will generate a “Summary of findings” table for the primary outcomes 'live birth' and 'incidence of adhesion formation at second-look hysteroscopy' using GRADEPRO software. This table will evaluate the overall quality of the body of evidence for these two key outcomes, using GRADE criteria (study limitations (i.e. risk of bias), consistency of effect, imprecision, indirectness and publication bias). Judgments about evidence quality (high, moderate or low) will be justified, documented, and incorporated into reporting of results for each outcome.

163 Chapter 6

REFERENCES

1. Amer MI, Abd- El- Maeboud KH. Amnion graft following hysteroscopic lysis of intrauterine adhesions. J Obstet Gynaecol Res 2006; 32(6):559-566. 2. Arora M, Jaroudi KA, Hamilton CJ, Dayel F. Controlled comparison of Interceed and amniotic membrane graft in the prevention of postoperative adhesions in the rabbit uterine horn model. Eur J Obstet Gynecol Reprod Biol 1994; 55 (3):179-182. 3. Badawy SZ, Baggish MS, ElBakry MM, Baltoyannis P. Evaluation of tissue healing and adhesion formation after an intra-abdominal amniotic membrane graft in the rat. J Reprod Med 1989; 34(3):189- 202. 4. Belluco C, Meggiolaro F, Pressato D, Pavesio A, Bigon E, Donà M, Forlin M, Nitti D, Lise M. Prevention of Postsurgical Adhesions with an Autocross linked Hyaluronan Derivative Gel J Surg Res 2001; 100:217-221. 5. Binda MM, Molinas CR, Bastidas A, Jansen M, Koninckx PR. Efficacy of barriers and hypoxia-inducible factor inhibitors to prevent CO²pneumoperitoneum-enhanced adhesions in a laparoscopic mouse model. J Minim Invasive Gynecol 2007; 14(5):591-599. 6. Binda MM, Koninckx PR. Prevention of adhesion formation in a laparoscopic mouse model should combine local treatment with peritoneal cavity conditioning. Hum Reprod 2009; 24(6):1473-1479. 7. Binda MM, Koninckx PR. Hyperoxia and prevention of adhesion formation: a laparoscopic mouse model for open surgery. BJOG 2010; 117(3):331-339. 8. De Iaco PA, Stefanetti M, Pressato D, Piana S, Doná M, Pavesio A, Bovicelli L. A novel hyaluronan-based gel in laparoscopic adhesion prevention: preclinical evaluation in an animal model. Fertil Steril 1998; 69:318-323. 9. Deans R, Abbott J. Review of Intrauterine Adhesions. J Minim Invasive Gynecol 2010; 17: 555-569. 10. Diamond MP, DeCherney AH, Linsky CB, Cunningham T, Constantine B. Adhesion re-formation in the rabbit uterine horn model: I. Reduction with carboxymethylcellulose. Int J Fertil 1988; 33:372-375. 11. Ducarme G, Davitian C, Zarrouk S, Uzan M, Poncelet C. Interest of auto-cross linked hyaluronic acid gel in the prevention of intrauterine adhesions after hysteroscopic surgery: a case-control study. J Gynecol Obstet Biol Reprod 2006; 35(7):691-695.

164 Anti-adhesion therapy following operative hysteroscopy

12. Farhi J, Bar-Have I, Homburg R, Dicker D, Ben-Rafael Z. Induced regeneration of endometrium following curettage for abortion: a comparative study. Hum Reprod 1993; 8:1143. 13. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003; 327(7414):557-560. 14. Hori J, Wang M, Kamiya K, Takahashi H, Sakuragawa N. Immunological characteristics of amniotic epithelium. Cornea 2006; 25(10 Suppl 1):S53-58. 15. Johary J, Xue M, Zhu X, Xu D, Palani Velu P. The efficacy of estrogen therapy in patients with intrauterine adhesions: a systematic review. J Minim Invasive Gynecol 2013; Aug 8: pii: S1553- 4650(13)00426-3. 16. Kelekci S, Yilmaz B, Oguz S, Zergero÷lu S, Inan I, Tokuco÷lu S. The efficacy of a hyaluronate/carboxymethylcellulose membrane in prevention of postoperative adhesion in a rat uterine horn model. Tohoku J Exp Med 2004; 204:189-194. 17. Kodaman PH, Arici A. Intra-uterine adhesions and fertility outcome: how to optimize success? Curr Opin Obstet Gynecol 2007; 19(3):207- 214. 18. Koçak I, Unlü C, Akçan Y, Yakin K. Reduction of adhesion formation with cross-linked hyaluronic acid after peritoneal surgery in rats. Fertil Steril 1999; 72:873-878. 19. Laurent TC, Fraser JRE. Hyaluronan. FASEB J 1992; 6(7): 2397- 2404. 20. Leach RE, Burns JW, Dawe EJ, SmithBarbour MD, Diamond MP. Reduction of postsurgical adhesion formation in the rabbit uterine horn model with use of hyaluronate /carboxymethyl-cellulose gel. Fertil Steril 1998; 69(3):415-417. 21. Meller D, Tseng SC. Conjunctival epithelial cell differentiation on amniotic membrane. Invest Ophthalmol Vis Sci 1999; 40: 878-886. 22. Nimrod A, Ezra E, Ezov N, Nachum G, Parisada B. Absorption, distribution, metabolism and excretion of bacteria-derived hyaluronic acid in rats and rabbits. J Ocul Pharmacol 1992; 8:161-172. 23. Orhue AAE, Aziken ME, Igbefoh JO. A comparison of two adjunctive treatments for intrauterine adhesions following lysis. Int J Gynaecol Obstet 2003; 82: 49-56. 24. Padykula HA. Regeneration in the primate uterus: the role of stem cells. Ann N Y Acad Sci 1991; 622:47-56. 25. Rajab TK, Wallwiener M, Planck C, Brochhausen C, Kraemer B, Wallwiener CW. A Direct Comparison of Seprafilm, Adept, Intercoat and Spraygel for Adhesion Prophylaxis. J Surg Res 2010; 161: 246- 249.

165 Chapter 6

26. Schonman R, Corona R, Bastidas A, De Cicco C, Mailova K, Koninckx PR. Intercoat Gel (Oxiplex): Efficacy, Safety, and Tissue Response in a Laparoscopic Mouse Model. J Minim Invasive Gynecol 2008; 16(2):188-194. 27. Shamiyeh A, Danis J, Benkö L, Vattay P, Röth E, Tulipan L, Shebl O, Wayand W. Effect of hyaluron derivate gel in prevention of postsurgical peritoneal adhesions-an experimental study in pigs. Hepatogastroenterology 2007; 54(76):1121-1124. 28. Szabo A, Haj M, Waxsman I, Eitan A. Evaluation of seprafilm and amniotic membrane as adhesion prophylaxis in mesh repair of abdominal wall hernia in rats. Eur Surg Res 2002; 32(2):125-128. 29. Trelford Sauder M, Trelford JD, Matolo NM. Replacement of the peritoneum with amnion following pelvic exenteration. Surg Gynecol Obstet 1977; 145:699-701. 30. Wallwiener M, Brucker S, Hierlemann H, Brochhausen C, Solomayer E, Wallwiener C. Innovative barriers for peritoneal adhesion prevention: liquid or solid? A rat uterine horn model. Fertil Steril 2006; 86(4 Suppl):1266-1276. 31. Wood J, Pena G. Treatment of traumatic uterine synechias. Int J Fertil 1964; 9: 405-410. 32. Zegers-Hochschild F, Adamson GD, de Mouzon J, Ishihara O, Mansour R, Nygren K, Sullivan E, Vanderpoel S; International Committee for Monitoring Assisted Reproductive Technology; World Health Organization. International Committee for Monitoring Assisted Reproductive Technology (ICMART) and the World Health Organization (WHO) revised glossary of ART terminology 2009. Fertil Steril 2009; 92(5):1520-1524.

166

Has there been any progress in evidence- based practice in reproductive surgery?

Jan Bosteels, Steven Weyers, Chantal Mathieu, Charalampos Siristatidis, Siladitya Bhattacharya Ben W. J. Mol, Thomas M. D’Hooghe

Facts, Views and Visions in Obstetrics&Gynecology 2010; 2(4): 232-252 Facts, Views and Visions in Obstetrics&Gynecology 2011; 3(4): 238-244 http://www.fvvo.be/archive/volume-2/number-4/structured-review/the- effectiveness-of-reproductive-surgery-in-the-treatment-of-female- infertility/ http://www.fvvo.be/archive/volume-3/number-4/opinion-papers/progress- in-evidence-based-reproductive-surgery/

167

Progress in evidence-based practice in reproductive surgery

ABSTRACT

Background In 1996 the CONSORT statement (Consolidated Standards Of Reporting Trials) was published to develop a new scale for assessing the quality of published reports of randomized studies. We aimed to appraise if there has been any progress in evidence-based practice in reproductive surgery by studying the number and the quality of published reports of randomized controlled studies (RCTs) of surgical interventions in subfertile women.

Methods We searched MEDLINE, EMBASE and The Cochrane Library for RCTs evaluating laparoscopic or hysteroscopic interventions in subfertile women reporting pregnancy and/or live birth as the main outcomes. Two review authors independently assessed eligible studies for inclusion and risk of bias, and extracted data. We compared the quality of the trials based on the composite outcome measure "all six items for risk of bias" in the Cochrane ‘Risk of bias’ tool between a post-CONSORT (•1999) and a pre-CONSORT (”1995) group. Studies between 1996 and 1998 were excluded from the analysis for the assessment of study quality.

Results We retrieved 64 RCTs. Our findings demonstrated a steady increase from 1970- 2010 in the number of RCTs per decade. We excluded 10 studies between 1996 and 1998 for the assessment of study quality. About 61% of the 54 RCTs had an adequate random sequence generation and nearly 39% had adequate allocation concealment. The proportion of trials with adequate random sequence generation was higher in the post- (•1999) (33/38) versus pre- (”1995) (7/16) CONSORT group; the risk ratio (RR) was 2.0 (95% confidence interval (CI) 1.1 to 3.5, P=0.02, 54 studies). Similar trends were seen for the risk of bias items ‘adequate allocation concealment’ (RR 2.2, 95% CI 0.90 to 5.4, P=0.08), ‘incomplete outcome data’ (RR 1.4, 95% CI 0.92 to 2.3, P=0.11), ‘selective outcome reporting’ (RR 1.3, 95% CI 0.94 to 1.7, P=0.12) and ‘other potential sources of bias’ (RR 2.1, 95% CI 0.52 to 8.5, P=0.30) but nor for ‘blinding’ (RR 0.42, 95% CI 0.12 to 1.5, P=0.18). There was a higher proportion of items judged to be at low risk of bias for the composite outcome ‘all six risk of bias items’ favoring the post-CONSORT group (RR 1.5, 95% CI 1.2 to 1.9, P=0.001).

169 Chapter 7

Conclusions There has been an increase in the number of published reports of RCTs on reproductive surgical interventions in subfertile women measuring live birth or pregnancy rate; more importantly, there has been an improvement in the overall methodological quality of these publications since the publication of the CONSORT statement in 1996.

BACKGROUND

In 1996 the CONSORT statement was published (Schulz et al, 2010). CONSORT stands for CON solidated Standards Of Reporting Trials; originally started in 1993 in Ottawa, Canada as a meeting of medical journal editors, clinical trialists, epidemiologists and methodologists this initiative aimed to develop a new scale for assessing the quality of published reports of randomized studies. Two revisions of the CONSORT statement have been published in 2001 and 2010. The statement refers to the use of a checklist consisting of 25 items selected because empirical evidence indicates that not reporting the information is associated with biased estimates of treatment effect, or because the information is essential to judge the reliability or relevance of the findings. A systematic review has demonstrated that the use of the CONSORT checklist is associated with an improvement in the reporting of RCTs (Plint et al, 2006). We aimed to investigate if a similar improvement in the quality of the published reports of RCTs on reproductive surgery could be demonstrated as an indicator for progress in evidence-based practice in the field of reproductive surgery.

MATERIALS AND METHODS

Search methods for identification of studies

We aimed to identify RCTs on reproductive surgery in subfertile women with pregnancy or live birth rate as the main outcomes. We searched The Cochrane Library (1970 to June 1st 2010) for relevant trials in the CDSR, DARE, CENTRAL or HTA databases. We also searched EMBASE (1974 to June 1st 2010) and MEDLINE through Pub Med using a combination of text words and MeSH terms for “laparoscopy”, “hysteroscopy” ,“infertility”, “pregnancy rate” and “live birth rate” (1966- June 1st 2010). We also searched the Current Controlled Trials (metaRegister); the last up-dated search was done on June 1st 2010. The search was done independently by two review authors. Language restrictions were not applied.

170 Progress in evidence-based practice in reproductive surgery

Data collection and analysis

Selection of studies We included RCTs on reproductive surgical interventions in subfertile women reporting the main outcomes of pregnancy or live birth. We excluded non-randomized trials, studies not reporting pregnancy or live birth, studies in a source population with gynecological problems other than subfertility, trials on diagnostic accuracy, technical feasibility and patient compliance. Study selection was done by two authors independently based on reviewing the full-text manuscript. Any disagreement was solved by discussion or by a third review author.

Data extraction and management The characteristics of the different study populations, the control and study intervention as well as other relevant study characteristics were extracted from the full-text articles by using standardized work sheets for quality assessment of RCTs from the website of the Dutch Cochrane Centre. We contacted the corresponding authors in case of unclear study methodology or to obtain missing data. Two review authors independently extracted relevant study data. To test the study hypothesis that quality of studies changed over time after the publication of the CONSORT statement, we made two comparison groups using the year of publication of the CONSORT statement (1996) as a time marker (Begg et al, 1996); we excluded the studies between 1996 and 1998 to allow a reasonable time period for the implementation of the CONSORT statement. We compared post-CONSORT (•1999) versus pre-CONSORT (”1995).

Assessment of risk of bias in included studies We estimated the risk of bias by using the ‘Risk of bias’ tool provided by the Cochrane Collaboration based on all six items: random sequence generation, allocation concealment, blinding of participants, personnel and outcome assessors, completeness of outcome data, selective outcome reporting and other potential sources of bias. This was done independently by two review authors; any disagreement was solved by discussion or by a third review author.

Measures of treatment effect and statistical analysis The methodological quality was assessed based on the composite outcome measure "all six items for risk of bias" in the Cochrane ‘Risk of bias’ tool. For the analysis we defined the reported use of allocation concealment as ‘low risk’ of bias against the other concealment categories i.e. unclear, inadequate or concealment not used. The same principle was used for all

171 Chapter 7

other risk of bias items. The number of items judged as "at low risk of bias" was considered as a positive outcome of effectiveness and higher numbers were considered as a benefit. Statistical analysis was done using Review Manager 5. The dichotomous data (number of items judged as ‘low risk’ versus ‘other’) were extracted in 2x2 tables. The results of individual trials and of the meta-analysis were expressed as Mantel- Haenszel (M-H) risk ratios (RR) with 95% confidence limits (95% CI) using a fixed-effect model.

Subgroup analysis and investigation of heterogeneity We defined a sensible ‘a priori’ subgroup analysis based on each of the six items in the Cochrane ‘Risk of bias’ tool. We aimed to avoid over interpretation; subgroup analysis is by its nature an observational study within a study and the interpretation of the statistical significance of subgroup differences is not without problems. Statistical heterogeneity was assessed with the Chi² test and the I² test.

Sensitivity analysis We planned to conduct sensitivity analyses for the composite outcome ‘all six risk of bias items’ to investigate whether the review conclusions would have differed if the item ‘blinding’ was excluded or not. Blinding of participants and personnel is very often problematical in the design of randomized studies in surgery. Other sensitivity analyses studied the effect of the choice of a fixed-effect rather than a random-effects model and the use of an odds ratio rather than the risk ratio as the effect measure.

172 Progress in evidence-based practice in reproductive surgery

RESULTS

Results of the search

The findings are presented in detail in the two primary published reports (Bosteels et al, 2010; Bosteels et al, 2011). The process of literature search and selection is described in figure 1: we included 64 RCTs on reproductive surgical techniques and pregnancy outcome; 16 RCTs in the pre-CONSORT group, 38 RCTs in the post-CONSORT group and 10 RCTs between 1996 and 1998. These latter 10 trials were taken into account for the quantitative but not the qualitative assessment of the research question.

Figure 1 PRISMA flow diagram: systematic review of benefit after reproductive surgery.

Records identified through Cochrane Library Additional records identified through Current (n=170), MEDLINE (n=106) and EMBASE Controlled Trials (n=334) (N=609) (N=37)

Records screened after duplicates removed (n=592)

Records excluded (n=495)

Full-text articles assessed for eligibility (n=97)

Excluded: n=33

Studies included in quantitative synthesis (n=64)

173 Chapter 7

An overview of the included trials and a summary of findings grouped according to the specific pathology or clinical setting are presented in table 1.

Table 1 Overview of included trials.

174 Progress in evidence-based practice in reproductive surgery

Table 1 Overview of included trials (continued).

175 Chapter 7

Table 1 Overview of included trials (continued).

176 Progress in evidence-based practice in reproductive surgery

Table 1 Overview of included trials (continued).

177 Chapter 7

Table 1 Overview of included trials (continued).

178 Progress in evidence-based practice in reproductive surgery

Effects of interventions

An overview of the RRs of all included trials is presented in figure 2. The mean value of the effect measure was 1.47 ± SD 1.51. The RR of the included studies had a range from 0.33 to 12. In the majority of studies the treatment effect was moderate: the log (RR) varies from -0.3 to 0.3, corresponding with 0.5

Figure 2 Risk ratios across all included studies

The number and the quality of RCTs post- versus pre-CONSORT

Our findings demonstrate a steady increase from 1970- 2010 in the number of published reports of randomized studies on reproductive surgical interventions per decade (Figure 3).

Figure 3 Evolution of the number of published reports of RCTs on the effectiveness of reproductive surgery.

179 Chapter 7

There is a significant increase in the proportion of trials with adequate random sequence generation favoring the post-CONSORT group (RR 2.0; 95% CI 1.1 to 3.5, P=0.02) (Figure 4).

Figure 4 Forest plot of comparison: Post-CONSORT versus pre- CONSORT. Outcome: Number of risk of bias items judged as ‘low risk’. M-H: Mantel-Haenszel, fixed-effect model, CI: confidence interval. Post-CONSORT Pre-CONSORT Risk Ratio Risk Ratio Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI 1.1.1 Random sequence generation Bosteels 2010 33 38 7 16 18.4% 1.98 [1.12, 3.51] Subtotal (95% CI) 38 16 18.4% 1.98 [1.12, 3.51] Total events 33 7 Heterogeneity: Not applicable Test for overall effect: Z = 2.36 (P = 0.02)

1.1.2 Allocation concealment Bosteels 2010 21 38 4 16 10.5% 2.21 [0.90, 5.41] Subtotal (95% CI) 38 16 10.5% 2.21 [0.90, 5.41] Total events 21 4 Heterogeneity: Not applicable Test for overall effect: Z = 1.74 (P = 0.08)

1.1.3 Blinding Bosteels 2010 4 38 4 16 10.5% 0.42 [0.12, 1.48] Subtotal (95% CI) 38 16 10.5% 0.42 [0.12, 1.48] Total events 4 4 Heterogeneity: Not applicable Test for overall effect: Z = 1.35 (P = 0.18)

1.1.4 Incomplete outcome data Bosteels 2010 31 38 9 16 23.7% 1.45 [0.92, 2.29] Subtotal (95% CI) 38 16 23.7% 1.45 [0.92, 2.29] Total events 31 9 Heterogeneity: Not applicable Test for overall effect: Z = 1.59 (P = 0.11)

1.1.5 Selective outcome reporting Bosteels 2010 36 38 12 16 31.6% 1.26 [0.94, 1.69] Subtotal (95% CI) 38 16 31.6% 1.26 [0.94, 1.69] Total events 36 12 Heterogeneity: Not applicable Test for overall effect: Z = 1.56 (P = 0.12)

1.1.6 Other potential sources of bias Bosteels 2010 10 38 2 16 5.3% 2.11 [0.52, 8.55] Subtotal (95% CI) 38 16 5.3% 2.11 [0.52, 8.55] Total events 10 2 Heterogeneity: Not applicable Test for overall effect: Z = 1.04 (P = 0.30)

Total (95% CI) 228 96 100.0% 1.50 [1.17, 1.91] Total events 135 38 Heterogeneity: Chi² = 7.12, df = 5 (P = 0.21); I² = 30% 0.01 0.1 1 10 100 Test for overall effect: Z = 3.25 (P = 0.001) Favours pre-CONSORT Favours post-CONSORT Test for subgroup differences: Chi² = 6.78, df = 5 (P = 0.24), I² = 26.2%

180 Progress in evidence-based practice in reproductive surgery

Similarly there is a consistent trend for a higher proportion of studies favoring the post-CONSORT group for the outcome of number of risk of bias items judged as ‘low risk’ for ‘adequate allocation concealment’ (RR 2.2, 95% CI 0.90 to 5.4, P=0.08), ‘completeness of outcome data’ (RR 1.4, 95% CI 0.92 to 2.3, P=0.11), ‘selective outcome reporting’ (RR 1.3, 95% CI 0.94 to 1.7, P=0.12), ‘other potential sources of bias’ (RR 2.1, 95% CI 0.52 to 8.5, P=0.30) but not for ‘blinding’ (RR 0.42, 95% CI 0.12 to 1.5, P=0.18). The tests for subgroup differences demonstrated no statistically significant differences between the different subgroups according to the risk of bias item (Chi²=6.8, df=5 (P=0.24), I²=26%). Overall, there is an effect in favor of the post-CONSORT group for the composite measure ‘all six risk of bias items’ (RR 1.5, 95% CI 1.2 to 1.9, P=0.001, 324 items from 54 studies). A sensitivity analysis did not demonstrate statistically significant differences between both comparison groups when the item ‘blinding’ was excluded from the data synthesis (RR=1.6, 95% CI 1.3 to 2.1, P=0.0001). Two other sensitivity analyses demonstrated similarly that the data are robust concerning the choice of the effect measure: odds ratio (OR) versus risk ratio (P=0.0004) and random-effects (RE) model rather than a fixed-effect model (P=0.02). Live birth was reported as the primary outcome measure in 13 out of 54 studies (24%) of the pre + post-CONSORT RCTs. In the post-CONSORT group 13 out of 38 studies (34%) either explicitly referred to the CONSORT statement or used a study flow diagram as proposed in the checklist.

181 Chapter 7

DISCUSSION

The results of the present systematic review show a steady increase in the number of published RCTs on reproductive surgical interventions measuring live birth or pregnancy as main outcome in subfertile women. Moreover we were able to demonstrate an improvement in the quality of the published reports of RCTs on reproductive surgical interventions in the post-CONSORT group. Our findings are consistent with those reported earlier by other review authors, both for non-gynecological (Plint et al, 2006) and gynecological RCTs (Selman et al, 2008); the latter authors reported a significant increase of the proportion of studies reporting adequate allocation concealment after CONSORT, as well as a trend towards improvement in precision of the estimation of the treatment effect over time (Selman et al, 2008). There were nevertheless differences in methodology. Selman and co-workers only retrieved studies from the Cochrane Database of Systematic Reviews. Moreover these authors focused only on one risk of bias item (allocation concealment) while we examined all six items of risk of bias assessment. A limitation of the present review is that we included only RCTs reporting live birth and/or pregnancy as main outcome; this means that some RCTs might not have been included, likely causing selection bias. Moreover we have only demonstrated a time-effect relationship rather than a cause- effect relationship between the publication of CONSORT and the improved methodological quality. The uptake of the study flow diagrams in the published reports since 1996 can be considered as an indirect proof that the implementation of the CONSORT might have been a causal factor for this improvement. Our findings demonstrate that gynecology has evolved to becoming a specialty in which the interventions are increasingly exposed to the gold standard of RCTs (Johnson et al, 2003). Reproductive surgery seems not to have “missed the boat”. Not all methodological problems of RCTs on reproductive surgery have been solved at the present. The methodological quality of surgical trials could be improved further by adequate training of surgeons in clinical epidemiology and evidence-based medicine or employing epidemiologists in surgical units where clinical research is being carried out (Urschel et al, 2001; Madhok et al, 2002). More research is needed on the willingness of reproductive surgeons to change practice for improving the outcomes that matter to subfertile women.

182 Progress in evidence-based practice in reproductive surgery

SUMMARY CONCLUSION

Clinical research in the field of reproductive surgery should be guided by high-quality RCTs. The limited and mediocre quality evidence provided by 64 RCTs indicates a positive role for some surgical reproductive interventions. There has been a steady increase in the number of published reports of RCTs on reproductive surgery in the last decade. Overall the methodological quality of the RCTs published after the CONSORT statement in 1996 has improved, likely because of this methodological initiative and the uptake of its implementation by the leading journals in reproductive medicine. In conclusion, we agree with others: “Evidence- based reproductive surgery is no passing fad” (Johnson et al, 2008).

183 Chapter 7

REFERENCES

1. Adhesion Study Group (Buttram V, Malinak R, Cleary R). Reduction of postoperative pelvic adhesions with intraperitoneal 32% dextran 70: a prospective, randomized clinical trial. Fertil Steril 1983; 40: 612-619. 2. Alborzi S, Momtahan M, Parsanezhad ME, Dehbashi S, Zolghadri J, Alborzi S. A prospective randomized study comparing laparoscopic ovarian cystectomy versus fenestration and coagulation in patients with endometriomas. Fertil Steril 2004; 82: 1633-1637. 3. Alborzi S, Ghotbi S, Parsanezhad ME, Dehbashi S, Alborzi S, Alborzi M. Pentoxifylline therapy after laparoscopic surgery for different stages of endometriosis: a prospective, double-blind, randomized, placebo-controlled study. J Minim Invasive Gynecol 2007; 14: 54-58. 4. Al-Mizyen E, Grudzinskas JG. Unilateral laparoscopic ovarian diathermy in infertile women with clomiphene citrate-resistant polycystic ovary syndrome. Fertil Steril 2007; 88: 1678-1680. 5. Amer SA, Li TC, Metwally M, Emarh M, Ledger WL. Randomized controlled trial comparing laparoscopic ovarian diathermy with clomiphene citrate as a first-line method of ovulation induction in women with polycystic ovary syndrome. Hum Reprod 2009; 24: 219- 225. 6. Balasch J, Creus M, Fábregues F, Carmona F, Martínez-Román S, Manau D, Vanrell JA. Pentoxifylline versus placebo in the treatment of infertility associated with minimal or mild endometriosis: a pilot randomized clinical trial. Hum Reprod 1997; 12: 2046-2050. 7. Balen A, Jacobs H. A prospective study comparing unilateral and bilateral laparoscopic ovarian diathermy in women with the polycystic ovary syndrome (PCOS). Fertil Steril 1994; 62: 921-925. 8. Batioglu S, Haberai A, Celikkanat H. Comparison of GnRH agonist administration before and after laparoscopic drainage of endometriomas. J Gynecol Surg 1997; 13:17-21. 9. Bayram N, van Wely M, Kaaijk EM, Bossuyt PMM, van der Veen F. Using an electrocautery strategy or recombinant follicle stimulating hormone to induce ovulation in polycystic ovary syndrome: randomized controlled trial. BMJ 2004; 328 (7433): 192. 10. Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, Pitkin R, Rennie D, Schulz KF, Simel D, Stroup DF. Improving the quality of reporting of randomized controlled trials: the CONSORT statement. JAMA 1996; 276: 637-639. 11. Beretta P, Franchi M, Ghezzi F, Busacca M, Zupi E, Bolis P. Randomized clinical trial of two laparoscopic treatments of

184 Progress in evidence-based practice in reproductive surgery

endometriomas: cystectomy versus drainage and coagulation. Fertil Steril 1998; 70: 1176-1180. 12. Bianchi S, Busacca M, Agnoli B, Candiani M, Calia C, Vignali M. Effects of 3 month therapy with danazol after laparoscopic surgery for stage III-IV endometriosis: a randomized study. Hum Reprod 1999; 14:1335-1337. 13. Bosteels J, Weyers S, Mathieu C, Mol BWJ, D’Hooghe TM. The effectiveness of reproductive surgery in the treatment of female infertility: facts, views and vision. Facts, Views and Visions in Obstetrics and Gynecology 2010; 2(4): 232-252. 14. Bosteels J, Weyers S, Siristatidis C, Bhattacharya S, D’Hooghe TM. Progress in evidence- based reproductive surgery. Facts, Views and Visions in Obstetrics and Gynecology 2011; 3(4): 238-244. 15. Busacca M, Somigliana E, Bianchi S, De Marinis S, Calia C, Candiani M, Vignali M. Postoperative GnRH analogue treatment after conservative surgery for symptomatic endometriosis stage III-IV: a randomized controlled trial. Hum Reprod 2001; 16: 2399-2402. 16. Campo S, Garcea N. Laparoscopic myomectomy in premenopausal women with and without preoperative treatment using gonadotrophin- releasing hormone analogues. Hum Reprod 1999; 14: 44-48. 17. Casini ML, Rossi F, Agostini R, Unfer V. Effect of the position of fibroids on fertility. Gynecol Endocrinol 2006; 22: 106-109. 18. Chong AP. Pregnancy outcome in neosalpingostomy by the cuff vs. Bruhat technique using the carbon dioxide laser. J Gynecol Surg 1991; 7: 207-210. 19. Colacurci N, De Franciscis P, Mollo A, Litta P, Perino A, Cobellis L, De Placido G. Small-diameter hysteroscopy with Versapoint versus resectoscopy with a unipolar knife for the treatment of septate uterus: a prospective randomized study. J Minim Invas Gynecol 2007; 14: 622-627. 20. Comninos AC. Salpingostomy: results of two different methods of treatment. Fertil Steril 1977; 28: 1211-1214. 21. Creus M, Fábregues F, Carmona F, del Pino M, Manau D, Balasch J. Combined laparoscopic surgery and pentoxifylline therapy for treatment of endometriosis-associated infertility: a preliminary trial. Hum Reprod 2008; 23: 1910-1916. 22. Daraï E, Dubernard G, Coutant C, Frey C, Rouzier R, Ballester M. Randomized trial of laparoscopically assisted versus open colorectal resection for endometriosis. Ann Surg 2010: 251; 1018-1023. 23. Déchaud H, Daurès JP, Arnal F, Humeau C, Hédon B. Does previous salpingectomy improve implantation and pregnancy rates in patients with severe tubal factor infertility who are undergoing in vitro

185 Chapter 7

fertilization? A pilot prospective randomized study. Fertil Steril 1998; 69: 1020-1025. 24. Demirol A. Effect of endometrioma cystectomy on IVF outcome: a prospective randomized study. Reprod Biomed Online 2006; 12: 639- 643. 25. Demirol A, Gurgan T. Effect of treatment of intrauterine pathologies with office hysteroscopy in patients with recurrent IVF failure Reprod Biomed Online 2004; 8: 590-594. 26. Farquhar CM, Williamson K, Gudex G, Johnson NP, Garland J, Sadler L. A randomized controlled trial of laparoscopic ovarian diathermy versus gonadotrophin therapy for women with clomiphene- resistant polycystic ovarian syndrome. Fertil Steril 2002; 78: 404-411. 27. Ghafarnegad M, Arjmand N, Khazaeipour Z. Pregnancy rate of gonadotrophin therapy and laparoscopic ovarian electrocautery in polycystic ovary syndrome resistant to clomiphene citrate: A comparative study. Tehran Uni Med J 2010; 67: 712-717. 28. Goldstein DB, Sasaran LH, Stadtmauer L. Selective salpingostomy- salpingectomy (SSS) and medical treatment prior to IVF in patients with hydrosalpinx. Fertil Steril 1998; 70: S320. 29. Gruppo Italiano per lo Studio dell’Endometriosi. Ablation of lesions or no treatment in minimal-mild endometriosis in infertile women: a randomized trial. Hum Reprod 1999; 14: 1332-1334. 30. Guida M, Acunzo G, Di Spiezio Sardo A, Bifulco G, Piccoli R, Pellicano M, Cerrota G, Cirillo D, Nappi C. Effectiveness of auto- cross linked hyaluronic acid gel in the prevention of intra-uterine adhesions after hysteroscopic surgery: a prospective, randomized, controlled study. Hum Reprod 2004; 19: 1461-1464. 31. Gürgan T, Urman B, Aksu T, Yarali H, Develioglu O, Kisnisci HA. The effect of short-interval laparoscopic lysis of adhesions on pregnancy rates following Nd-YAG laser photocoagulation of polycystic ovaries. Obstet Gynecol 1992; 80: 45-47. 32. Hamed HO, Hasan AF, Ahmed OG, Ahmed MA. Metformin versus laparoscopic ovarian drilling in clomiphene- and insulin-resistant women with polycystic ovary syndrome. Int J Gynecol Obstet 2010; 108: 143-147. 33. Hammadieh N, Coomarasamy A, Ola B, Papaioannou S, Afnan M, Sharif K. Ultrasound-guided hydrosalpinx aspiration during oocyte collection improves pregnancy outcome in IVF: a randomized controlled trial. Hum Reprod 2008; 23: 1113-1117. 34. Hart RJ, Hickey M, Maouris P, Buckett W. Excisional surgery versus ablative surgery for ovarian endometriomata. Cochrane Database Syst Rev. 2011; issue 3:CD004992.

186 Progress in evidence-based practice in reproductive surgery

35. Jansen RPS. Failure of intraperitoneal adjuncts to improve the outcome of pelvic operations in young women. Am J Obstet Gynecol 1985; 153: 363-371. 36. Johnson NP, Selman T, Zamora J, Khan KS. Gynaecologic surgery from uncertainty to science: evidence-based surgery is no passing fad. Hum Reprod 2008; 23: 832-839. 37. Kamel MA, Amin AF. Using thermocoagulation versus electrocoagulation in laparoscopic adhesiolysis in infertility treatment. Hum Reprod abstract book I, Abstract of 15th Annual Meeting of ESHRE, Tours, France 1999; 14: 276-277. 38. Kaya H, Sezik M, Ozkaya O. Evaluation of a new surgical approach for the treatment of clomiphene citrate-resistant infertility in polycystic ovary syndrome: Laparoscopic ovarian multi-needle intervention. J Minim Invasive Gynaecol 2005; 12: 355-358. 39. Kontoravdis A, Makrakis E, Pantos K, Botsis D, Deligeoroglou E, Creatsas G. Proximal tubal occlusion and salpingectomy result in similar improvement in in vitro fertilization outcome in patients with hydrosalpinx. Fertil Steril 2006; 86: 1642-1649. 40. Larsson B, Lalos O, Marsk L, Tronstad SE, Bygdeman M, Pehrson S, Joelsson I. Effect of intraperitoneal instillation of 32% dextran 70 on postoperative adhesion formation after tubal surgery. Acta Obstet Gynecol Scand 1985; 64: 437-441. 41. Lazovic G, Milacic D, Terzic M. Medicaments or surgical therapy of PCOS. Fertil Steril 1998; 70: 472. 42. Loverro G, Santillo V, Pansini MV. Are GnRH agonists helpful in the therapy of endometriosis after surgical treatment? Hum Reprod 2001; 16: 96. 43. Madhok R, Handoll HH. Randomized trials in surgery. Letters to the Editor of the BMJ. BMJ 2002; 325: 658-9. 44. Marcoux S, Maheux R, Bérubé S. Laparoscopic surgery in infertile women with minimal or mild endometriosis. N Engl J Med 1997; 337: 217-222. 45. Moshin V, Hotineanu A. Reproductive outcome of the proximal occlusion prior to IVF in patients with hydrosalpinx. Hum Reprod 2006; 21: i193-i194. 46. Palomba S, Zupi E, Falbo A, Russo T, Marconi D, Tolino A, Manguso F, Mattei A, Zullo F. A multicenter randomized, controlled study comparing laparoscopic versus minilaparotomic myomectomy: reproductive outcomes. Fertil Steril 2007; 88: 933-941. 47. Palomba S, Falbo A, Battista L, Russo T, Venturella R, Tolino A, Orio F, Zullo F. Laparoscopic ovarian diathermy vs clomiphene citrate plus metformin as second-line strategy for infertile anovulatory

187 Chapter 7

patients with polycystic ovary syndrome: a randomized controlled trial. Am J Obstet Gynecol 2010; 202: 577:e1-8.Epub 2010 Jan 22. 48. Pellicano M, Guida M, Bramante S, Acunzo G, Di Spiezio Sardo A, Tommaselli GA, Nappi C. Reproductive outcome after auto-cross- linked hyaluronic acid gel application in infertile patients who underwent laparoscopic myomectomy. Fertil Steril 2005; 83(2): 498- 500. 49. Parazzini F, Fedele L, Busacca M, Falsetti L, Pellegrini S, Venturini PL, Stella M. Postsurgical medical treatment of advanced endometriosis: Results of a randomized clinical trial. Am J Obstet Gynecol 1994; 171: 1205-1207. 50. Pellicano M, Guida M, Bramante S, Acunzo G, Di Spiezio Sardo A, Tommaselli GA, Nappi C. Reproductive outcome after auto-cross- linked hyaluronic acid gel application in infertile patients who underwent laparoscopic myomectomy. Fertil Steril 2005; 83: 498-500. 51. Pérez-Medina T, Bajo-Arenas J, Salazar F, Redondo T, Sanfrutos L, Alvarez P, Engels V. Endometrial polyps and their implication in the pregnancy rates of patients undergoing intrauterine insemination: a prospective randomized study Hum Reprod 2005; 20: 1632-1635. 52. Plint AC, Moher D, Morrison A, Schulz K, Altman DG, Hill C, Gaboury I. Does the CONSORT checklist improve the quality of reports of randomized controlled trials? A systematic review. Med J Aust 2006; 185:263-267. 53. Querleu D, Vankeerberghen-Deffense F, Bouteville C. The effect of noxytioline and systemic corticosteroids in infertility surgery. A prospective randomized study. J Gynecol Obstet Biol Reprod 1989; 18: 935-940. 54. Rama Raju GA, Shashi Kumari G, Krishna KM, Prakash GJ, Madan K. Assessment of uterine cavity by hysteroscopy in assisted reproduction programme and its influence on pregnancy outcome. Arch Gynecol Obstet 2006; 274: 160-164. 55. Rock JA, Siegler AM, Meisel MB, Haney AF, Rosenwaks Z, Pardo- Vargas F, Kimball AW. The efficacy of postoperative hydrotubation: a randomized prospective multicenter clinical trial. Fertil Steril 1984a; 42: 373-376. 56. Rock JA, Bergquist CA, Kimball AW Jr, Zacur HA, King TM. Comparison of the operating microscope and loupe for microsurgical tubal anastomosis: a randomized clinical trial. Fertil Steril 1984b; 41: 229-232. 57. Roy KK, Baruah J, Moda N, Kumar S. Evaluation of unilateral versus bilateral ovarian drilling in clomiphene citrate resistant cases of

188 Progress in evidence-based practice in reproductive surgery

polycystic ovarian syndrome. Arch Gynecol Obstet 2009; 280: 573- 578. 58. Roy KK, Baruah J, Sharma A, Sharma JB, Kumar S, Kachava G, Karmakar D. A prospective randomized trial comparing the clinical and endocrinological outcome with rosiglitazone versus laparoscopic drilling in patients with polycystic ovarian disease resistant to ovulation induction with clomiphene citrate. Arch Gynecol Obstet 2010; 281: 939-944. 59. Schulz KF, Altman DG, Moher D, for the CONSORT Group. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010; 340:c332. 60. Selman TJ, Johnson NP, Zamora J, Khan KS. Gynaecologic surgery from uncertainty to science: evolution of randomized control trials. Hum Reprod 2008; 23: 827-831. 61. Seracchioli R, Rossi S, Govoni F, Rossi E, Venturoli S, Bulletti C, Flamigni C. Fertility and obstetric outcome after laparoscopic myomectomy of large myomata: a randomized comparison with abdominal myomectomy. Hum Reprod 2000; 15: 2663-2668. 62. Sharma M, Kriplani A, Agarwal N. Laparoscopic bipolar versus unipolar ovarian drilling in infertile women with resistant polycystic ovarian syndrome. J Gynaecol Surg 2006; 22: 105-111. 63. Shokeir T, El-Shafei M, Yousef H, Allam AF, Sadek E. Submucous myomas and their implications in the pregnancy rates of patients with otherwise unexplained primary infertility undergoing hysteroscopic myomectomy: a randomized matched control study. Fertil Steril 2009; 92: S44- S45. 64. Soihet S. Three comparative techniques on tubo-plasty. Int J Fertil 1974; 19: 111-115. 65. Strandell A, Lindhard A, Waldenström U, Thorburn J, Janson PO, Hamberger L. Hydrosalpinx and IVF outcome: a prospective, randomized multicentre trial in Scandinavia on salpingectomy prior to IVF. Hum Reprod 1999; 14: 2762-2769. 66. Tanahatoe S, Lambalk CB, Hompes PGA. The role of laparoscopy in intrauterine insemination: a prospective randomized reallocation study. Hum Reprod 2005; 20: 3225-3230. 67. Telimaa S, Ronnberg L, Kauppila A. Placebo-controlled comparison of danazol and high-dose medroxyprogesterone acetate in the treatment of endometriosis after conservative surgery. Gynecol Endocrinol 1987; 1: 363-371. 68. Tulandi T, Vilos GA. A comparison between laser surgery and electrosurgery for bilateral hydrosalpinx: a 2-year follow-up. Fertil Steril 1985; 44: 846-847.

189 Chapter 7

69. Tulandi T. Salpingo-ovariolysis: a comparison between laser surgery and electrosurgery. Fertil Steril 1986; 45: 489-491. 70. Tulandi T, Falcone T, Kafka I. Second-look laparoscopy one year following reproductive surgery. Fertil Steril 1989; 52: 421-424. 71. Urschel JD, Goldsmith CH, Tandan VR, Miller JD. Evidence- based Surgery Working Group. Users’ guide to evidence- based surgery: how to use an article evaluating surgical interventions. Can J Surg 2001; 44: 95-100. 72. Vegetti W, Ragni G, Baroni E. Laparoscopic ovarian drilling versus low-dose pure FSH in anovulatory clomiphene-resistant patients with polycystic ovarian syndrome: randomized prospective study. Hum Reprod 1998; 13: 120. 73. Vercellini P, Crosignani PG, Fadini R, Radici E, Belloni C, Sismondi P. A gonadotrophin- releasing hormone agonist compared with expectant management after conservative surgery for symptomatic endometriosis. Br J Obstet Gynecol 1999; 106: 672-677. 74. Youssef H, Atallah MM. Unilateral ovarian drilling in polycystic ovarian syndrome: a prospective randomized study. Reprod Biomed Online 2007; 15: 457-462. 75. Zakherah MS, Nasr A, El Saman AM, Shaaban OM, Shahin AY. Clomiphene citrate plus tamoxifen versus laparoscopic drilling in women with clomiphene-resistant polycystic ovary syndrome. Int J Gynecol Obstet 2010; 108: 240-243.

190

Thesis summary and discussion

191

Thesis summary and discussion

SUMMARY

In 1987, only one year before his death, ‘Archie’ Cochrane referred to a systematic review of RCTs of care during pregnancy and childbirth as “a real milestone in the history of randomized trials and in the evaluation of care” (Chalmers et al, 1989). Surely, to his judgment, from that moment there was no longer any reason to leave the notorious wooden spoon described in chapter 1 in the possession of Obstetrics. But is this true for reproductive surgery as well? We summarize the findings of our thesis on evidence-based practice in reproductive surgery by answering the three predefined research questions. We also discuss the implications for clinical practice and future research. We briefly reflect on the future of evidence-based practice in reproductive surgery. Chapter 2 discusses the position of laparoscopy in current fertility practice. Although Assisted Reproductive Technology (ART) is steadily replacing tubal surgery as the first-line treatment for tubal infertility a randomized comparison between both treatment options is lacking. We retrieved limited evidence on the effectiveness of laparoscopy for treating female subfertility in some settings. The treatments for which we found evidence are summarized in the next alinea. The Canadian ENDOCAN trial has demonstrated a benefit with the laparoscopic treatment of minimal and/or mild endometriosis for the outcome ‘ongoing pregnancy at 20 weeks or live birth’ in women with otherwise unexplained subfertility: odds ratio (OR) 2.0, 95% confidence interval (CI) was 1.2 to 3.3, P=0.01, 341 women. In the smaller Gruppo Italiano trial there was no evidence for such a benefit (OR 0.85, 95% CI 0.32 to 2.3, P=0.75, 96 women). Laparoscopic ovarian drilling (LOD) in women with clomiphene-resistant polycystic ovaries is at least as effective as gonadotrophin treatment - the odds ratio (OR) was 1.0, 95% CI 0.63 to 1.6, P=0.95, 4 studies, 304 women- but LOD carries a substantially lower risk for multiple pregnancy (OR 0.13, 95% CI 0.03 to 0.59, P=0.008, 4 studies, 154 women). Doing a laparoscopy before a first or after several failed cycles of IUI does not affect the pregnancy rates (OR 0.81, 95% CI 0.43 to 1.5, P=0.52, 1 study, 154 women) nor the incidence of clinically relevant pelvic pathology (OR 0.72, 95% CI 0.28 to 1.9, P=0.51, 1 study, 89 women). The removal of hydrosalpinges visible on ultrasound may increase the delivery rate after in vitro fertilization (IVF) - the risk ratio (RR) was 2.4, 95% CI 1.1 to 5.3, P=0.038, 1 study, 75 women- especially when bilaterally detectable (RR 3.5, 95% CI 1.1 to 11, P=0.019, 1 study, 39 women). In conclusion, there is no evidence-based fundament for the shift from tubal reconstructive surgery to ART. The limited evidence on the role of laparoscopy in the treatment of female subfertility and the

193 Chapter 8

availability of reliable alternative methods for detecting tuboperitoneal subfertility are nevertheless two important reasons to question the practice of doing laparoscopy in all subfertile women as the final step in the fertility work-up. Chapter 3 aims to clarify the position of hysteroscopy in current fertility practice by a non-Cochrane systematic review of the literature. The Royal College of Obstetricians and Gynecologists (RCOG) and the European Society for Human Reproduction and Embryology (ESHRE) do not recommend hysteroscopy as a first-line procedure for screening all subfertile women. We aimed to summarize and appraise the evidence on the effectiveness of hysteroscopy in subfertile women for improving reproductive outcome. We retrieved five RCTs. The hysteroscopic removal of endometrial polyps with a mean diameter of 16mm detected by ultrasound may double the pregnancy rates in women undergoing intrauterine insemination (IUI) compared to diagnostic hysteroscopy only (RR 2.2, 95% CI 1.6 to 3.1, P<0.00001, 1 study, 215 women). There is no proof for a benefit with hysteroscopic myomectomy compared to timed intercourse in women with one submucous fibroid <4cm and otherwise unexplained subfertility, although the point estimate is almost statistically significant (RR 1.9, 95% CI 0.97 to 3.7, P=0.06, 1 study, 94 women). Hysteroscopic metroplasty for a septate uterus may be less effective in a target population of women with primary subfertility versus those with recurrent miscarriage (RR 0.7, 95% CI 0.5 to 0.9, P=0.01, 1 study, 160 women). Hysteroscopy before a next IVF attempt may nearly double the pregnancy rate in women with primary subfertility and at least two failed IVF attempts compared to starting IVF without prior hysteroscopy (RR 1.6, 95% CI 1.3 to 1.9, P<0.00001, 2 studies, 941 women) irrespective of whether pathology was present or not (RR 0.91, 95% CI 0.71 to 1.2, P=0.48, 2 studies, 465 women). In conclusion, there is limited evidence for doing hysteroscopy in some but not all subfertile women. Chapter 4 illustrates the added value of doing a Cochrane systematic review for identifying knowledge gaps to direct future primary research. We aimed to assess the effects of the hysteroscopic removal of endometrial polyps, submucous fibroids, uterine septum or intrauterine adhesions suspected on ultrasound, hysterosalpingography, diagnostic hysteroscopy or any combination of these methods in women with otherwise unexplained subfertility or prior to IUI, IVF or ICSI. We found no RCTs on the hysteroscopic removal of endometrial polyps, intrauterine adhesions or uterine septa in women with otherwise unexplained subfertility. The only included study in women with otherwise unexplained subfertility and fibroids moreover failed to report data on the predefined primary outcomes for this Cochrane review (live birth for

194 Thesis summary and discussion positive and procedure-related complications for negative outcome). We similarly found no RCTs on the effectiveness of operative hysteroscopy in subfertile women with suspected submucous fibroids, intrauterine adhesions or uterine septa prior to IUI, IVF or ICSI. The only included study for this second category of randomized comparisons (endometrial polyps before IUI) also did not present data for the primary outcomes. Moreover we found no data on the effects of the number, size/extent or the localization of the uterine cavity abnormalities on the main outcomes for both categories of randomized comparisons or on a possible time- effect relationship between the timing of the hysteroscopic intervention and subsequent IUI or ART. In conclusion, we have identified many knowledge gaps in our current understanding of the effects of hysteroscopic interventions on reproductive outcome in subfertile women. This is an opportunity for future research. Chapter 5 examines the role of anti-adhesion barrier gels following operative hysteroscopy for treating female subfertility. We aimed to assess the effects of any anti-adhesion barrier gel used after hysteroscopy for treating subfertility by a systematic review of the literature. We retrieved five studies. We assessed the risk of bias and performed a meta-analysis for some randomized comparisons. There is no evidence for a benefit with the use of any anti-adhesion barrier gel following operative hysteroscopy for the main outcomes live birth or clinical pregnancy (RR 3.0, 95% CI 0.35 to 26, P=0.32, 1 study, 30 women). The use of any gel decreases the incidence of adhesions at second-look hysteroscopy (RR 0.65, 95% CI 0.45 to 0.93, P=0.02, 5 studies, 372 women). The number needed to treat to benefit (NNTB) is 9 (95% CI 5 to 33). The use of auto-cross-linked hyaluronic acid (ACP) gel in women undergoing operative hysteroscopy for submucous fibroids, endometrial polyps or uterine septa is associated with a decrease in the mean adhesion scores at second-look hysteroscopy: the mean difference (MD) was -1.4, 95% CI -1.8 to -1.0, P<0.00001, 1 study, 24 women. The largest decrease is demonstrated in women treated by ACP gel after hysteroscopic adhesiolysis (MD -3.3, 95% CI -3.4 to - 3.2, P<0.00001, 1 study, 19 women). When adhesions occur following operative hysteroscopy, the use of any anti-adhesion barrier gel is associated with more ‘mild’ adhesions (RR 2.8, 95% CI 1.1 to 7.0, P=0.03, 4 studies, 79 women) and less ‘moderate or severe adhesions’ at second-look hysteroscopy (RR 0.25, 95% CI 0.10 to 0.67, P=0.006, 4 studies, 79 women). The NNTB is 2 with a 95% CI 1 to 2 for the former and a 95% CI 1 to 4 for the latter outcome. The quality of the evidence retrieved was graded as very low. In conclusion, gynecologists should counsel women wishing to become pregnant that intrauterine adhesion formation is the major long-term complication of operative hysteroscopy

195 Chapter 8

in women of reproductive age. The use of any anti-adhesion barrier gel might be considered after operative hysteroscopy because its use might decrease the incidence of adhesions and if these do occur, they tend to be more ‘mild’ and less ‘moderate or severe’. We retrieved only very low quality evidence from a limited number of RCTs in a general source population including- but not restricted to- subfertile women. Moreover subfertile women should be counseled that there is at the present no evidence for a benefit for the outcomes of pregnancy or live birth. Chapter 6 studies the effectiveness of different anti-adhesion strategies following operative hysteroscopy in subfertile women. Intrauterine adhesions are associated with a poor reproductive outcome due to a high incidence of subfertility or recurrent miscarriage and due to major obstetric complications as the endpoint of a successful hysteroscopic treatment of severe intrauterine adhesions. Therefore several preventive strategies were proposed in the literature, based on observational evidence. We aimed to summarize and appraise the evidence on the effectiveness of inserting an intrauterine device or a Foley catheter balloon, prescribing postoperative hormone therapy, using anti-adhesion barrier gels or human amnion membrane grafting by conducting a Cochrane systematic review. Chapter 7 aims to examine if there has been any progress in evidence- based practice in reproductive surgery. In 1996 the CONSORT statement was launched as a methodological initiative to improve the standards of reporting of clinical trials. We aimed to study the effect of the CONSORT statement on the number and the quality of published reports of RCTs on reproductive surgery in subfertile women. After a systematic review of the literature we retrieved 64 RCTs on reproductive surgical interventions in subfertile women measuring pregnancy or live birth rates: we classified 16 studies in the pre-CONSORT group (”1995) and 38 studies in the post- CONSORT group (•1999). We excluded 10 studies published between 1996 and 1998 for the assessment of the research question on the quality of RCTs to allow a reasonable time period for the implementation of the CONSORT statement by the scientific community. The number of RCTs reporting pregnancy or live birth as main outcome (N=32) in the last decade (2000-2010) equals the number of all RCTs published in the previous three decades (1970 to 1999). We compared the methodological quality post- versus pre-CONSORT by counting the number of items judged as low risk of bias by using the Cochrane ‘Risk of bias’ tool. We found a consistent trend for an improvement favoring the post-CONSORT studies for the items of random sequence generation (RR 2.0, 95% CI 1.1 to 3.5, P=0.02, 54 studies), allocation concealment (RR 2.2, 95% CI 0.90 to 5.4, P=0.08, 54 studies), completeness of outcome data (RR 1.4, 95%

196 Thesis summary and discussion

CI 0.92 to 2.3, P=0.11, 54 studies), selective outcome reporting (RR 1.3, 95% CI 0.94 to 1.7, P=0.12, 54 studies) and other potential sources of bias (RR 2.1, 95% CI 0.52 to 8.5, P=0.30, 54 studies) but not for blinding (RR 0.42, 95% CI 0.12 to 1.5, P=0.18, 54 studies). We found a significant improvement in the overall methodological quality defined by the composite outcome ‘all six risk of bias items’ favoring the post- CONSORT studies (RR 1.5, 95% CI 1.2 to 1.9, P=0.001, 54 studies). In conclusion, recent years have witnessed more and better reports on reproductive surgical interventions in subfertile women.

197 Chapter 8

DISCUSSION

Our search did not retrieve any randomized comparison of tubal surgery versus expectant management or IVF measuring live birth or pregnancy in women with tubal infertility. This is consistent with the findings of two Cochrane reviews (Pandian et al, 2008; Johnson et al, 2010). This answers the first research question of the present thesis.

Our appraisal of the evidence on the effectiveness of hysteroscopy in treating subfertile women has identified limited evidence for a benefit in women with endometrial polyps before IUI and in women with primary subfertility and at least two failed IVF attempts; moreover many knowledge gaps have been identified. The use of any barrier gel following operative hysteroscopy in subfertile women might be considered: its use may decrease the formation of adhesions as well as their extent and severity but data on the key reproductive outcomes are lacking. This answers the second research question of our thesis.

Our research has demonstrated an increase in the number and an improvement in the methodological quality of published RCTs on reproductive surgery over recent years. This answers part of the third research question of the present thesis. Our findings are consistent with those of a recently published opinion paper on ‘Twenty years of Cochrane reviews in Menstrual Disorders and Subfertility’ (Farquhar et al, 2013). The number of RCTs on the treatment of subfertility has increased from fewer than 2000 in 1997 to over 5000 in 2013. There is a move towards identifying and accumulating the highest quality evidence; inevitably, this should lead to a greater use of this evidence in the field and bridge the gap between evidence and daily clinical practice.

198 Thesis summary and discussion

1. IMPLICATIONS FOR CLINICAL PRACTICE

• Laparoscopic surgery for treating female subfertility

At the present we can no longer recommend doing a laparoscopy as the final step in the fertility work-up of all subfertile women: to attain a 100% level of certainty about the status of the tubes would increase the pool of women with minimal pelvic pathology of questionable prognostic significance leading to an inflation of health care costs (Collins and Van Steirteghem, 2004). HSG and laparoscopy are comparable for predicting natural conception in a general subfertile population (Verhoeve et al, 2011). Subfertile women with a history of PID, complicated appendicitis, pelvic surgery, ectopic pregnancy and endometriosis have a higher risk of having tuboperitoneal pathology (Luttjeboer et al, 2009) and should be offered a diagnostic laparoscopy early in the fertility work-up (NICE 2004; ASRM 2006). CAT by using the microimmunofluorescence test is the test of first choice in the initial screening for tubal pathology (Broeze et al, 2011). The combination of medical history taking and CAT testing has a higher predictive value in the selection of subfertile women for diagnostic laparoscopy than either of both alone and as a triage has the potential to decrease the number of ‘unnecessary laparoscopies’ by 82% compared to doing a laparoscopy in all cases (Coppus et al, 2007) . HSG is a useful test for screening tubal patency in all subfertile women (Broeze et al, 2011; the Practice Committee of the American Society for Reproductive Medicine, 2012). The combination of medical history taking, CAT and HSG has the best performance for the diagnosis of bilateral tubal pathology (Broeze et al, 2012). Routine diagnostic tubal patency tests in the fertility work-up are not cost-effective if considered only from the viewpoint of health policy management; when there is a need for information on the tubal status to make clinical decisions for the health of individual women, HSG followed by a tailored treatment or a diagnostic laparoscopy if HSG shows no tubal patency is more cost- effective than diagnostic laparoscopy (Verhoeve et al, 2013). LOD followed by CC and gonadotrophins if anovulation persists is an attractive alternative for the second-line treatment of CC-resistant PCOS women mainly since it lowers the risk of multiple pregnancies for at least equal effectiveness compared to starting gonadotrophin treatment immediately. Moreover a long-term follow-up study demonstrated that LOD by electrocautery significantly reduced the need for ovulation induction or ART for a first live born baby and increases the chance for a second child (Nahuis et al, 2011). Health-related quality of life was not affected for both LOD or gonadotrophin treatment except in a subanalysis

199 Chapter 8

of women without ongoing pregnancy that had more depressive symptoms in the gonadotrophin treatment arm compared to the electrocautery strategy group (van Wely et al, 2004). The total treatment costs up to an ongoing pregnancy are comparable for both treatment strategies (van Wely et al, 2004) but data from a long-term follow-up have demonstrated that the mean treatment costs per live birth within 8 to 12 years after initial treatment were significantly lower for the strategy starting with electrocautery compared to starting immediately with gonadotrophin treatment (Nahuis et al, 2012). Therefore starting with LOD first followed by clomiphene citrate and gonadotrophins when anovulation persists should be recommended as a rational second-line treatment in CC- resistant PCOS patients. Tubal reconstructive surgery was the only available treatment option for tubal infertility in the pre-ART era. It offers women the potential to achieve more than one conception by a one-time minimally invasive intervention, avoiding ART and its associated risks. The disadvantages of reconstructive surgery include the risks associated with surgery/anesthesia and recurrence of hydrosalpinx. In case of bilateral irreparable hydrosalpinges there is good-quality evidence for recommending salpingectomy or tubal occlusion to improve the pregnancy rates after IVF (Johnson et al, 2010; Johnson et al, 2011). Young women without other reasons for subfertility might be counseled to undergo laparoscopic fimbrioplasty or neosalpingostomy for restoring tubal pathology with a good fertility prognosis e.g. phimosis and adhesions (Mol et al, 1999); bilateral salpingectomy in these cases would eradicate the potential for spontaneous pregnancy (Johnson et al, 2011). Microsurgical or laparoscopic reanastomosis by a skilled reproductive surgeon might be recommended after prior tubal sterilization (the Practice Committee of the American Society for Reproductive Medicine, 2012).

• Hysteroscopic surgery for treating female subfertility

Doing a hysteroscopy prior to a subsequent IVF/ICSI treatment cycle in women with primary subfertility and at least two failed IVF/ICSI attempts may increase the likelihood of a successful reproductive outcome. A systematic review and meta-analysis of five observational studies and one randomized trial has suggested that a hysteroscopy before a first IVF attempt might improve the clinical pregnancy and live birth rates; the data should be interpreted with caution because of substantial clinical diversity and statistical heterogeneity across the included studies (Pundir et al, 2013). A hysteroscopy before a first IVF attempt might be cost-effective as shown by a cost-decision model (Kasius et al, 2013). More research is

200 Thesis summary and discussion needed before hysteroscopy prior to a first ART treatment cycle can be firmly recommended.

• Anti-adhesion therapy after hysteroscopic surgery

Gynecological surgeons should inform women still wishing to conceive that intrauterine adhesion formation is the major long-term complication of operative hysteroscopy in women of reproductive age. A recent survey in the Netherlands has indicated that adhesion awareness among Dutch gynecologists seems limited and that counseling is provided inadequately (Meuleman et al, 2013). Using barrier gels after operative hysteroscopy may decrease the incidence and the extent/severity of adhesions but the effects on reproductive outcome are not documented by high-quality evidence similarly to the use of barrier gels after laparoscopic myomectomy (Pellicano et al, 2005). This absence of evidence should not be neglected in the process of counseling.

201 Chapter 8

2. IMPLICATIONS FOR FUTURE RESEARCH

• Laparoscopic surgery for treating female subfertility

Additional randomized studies are needed to assess the effectiveness of HyCoSy to predict a spontaneous pregnancy; one RCT has found no evidence for a benefit with HyCoSy for improving clinical pregnancy rates in subfertile women (Lindborg et al, 2009). Several prospective comparative studies suggest a high correlation of the findings of transvaginal hydrolaparoscopy (THL) with those of conventional laparoscopy (Campo et al, 1999; Casa et al, 2002; Darai et al, 2000; Nawroth et al, 2001). In a prospective cohort study similar FRRs for one-sided tubal pathology, two sided tubal pathology and adhesions/endometriosis were found; this suggests that the capacity of THL to predict spontaneous ongoing pregnancy may be comparable to that of laparoscopy (van Tetering et al, 2007). Definitive evidence can only be derived from additional pragmatic RCTs comparing THL and laparoscopy in the same group of participants with a short time interval between both procedures. The value of clinical prediction models for tubal disease and endometriosis (all stages versus stage III or IV) in the selection of women with unexplained subfertility for THL as an outpatient screening procedure under local anesthesia should be studied. Several authors have indicated that the ongoing parallel debate on the economics of ART treatment to define the most appropriate funding framework for providing safe, equitable and cost-effective treatment is complex (Mol et al, 2000; Connolly et al, 2010; Berg Brigham et al, 2012; Chambers et al, 2013). Cost-effectiveness studies comparing restorative tubal surgery with expectant management and reproductive tubal surgery as an alternative or preliminary to ART in women with tubal infertility should be placed on top of the research agenda, especially in countries with a high ART utilization, like Belgium; the comparison between both interventions for the outcomes of adverse events and health-related quality of life should be considered as well. A major barrier for conducting this research might be that the clinical community in reproductive medicine seems reluctant to question the perceived success of ART (Kamphuis et al, 2014). We admit that there are major limitations in the practicability of designing such a study e.g. the availability of sufficient surgical expertise in tubal restorative surgery (Pandian et al, 2008) and a sufficiently high burden of tubal disease (Sabatini and Davis, 2005). A priori defined subgroup analyses should be done to study the cost-effectiveness of tubal restorative surgery compared to ART in specific subgroups (e.g. older versus younger women, poor responders, previous tubal ligation…).

202 Thesis summary and discussion

• Hysteroscopic surgery for treating female subfertility

We agree with others that there is currently no evidence from adequately powered high-quality RCTs for a benefit with hysteroscopic myomectomy for improving pregnancy or live birth rates in women with otherwise unexplained subfertility or prior to ART (Metwally et al, 2012). The findings of a systematic review of observational studies (Pritts et al, 2009) have suggested a negative impact of submucous fibroids on reproductive outcome and a benefit with their hysteroscopic removal. As reproductive surgeons we therefore face a difficult choice: either take this non-evidence based (biased) research for granted or follow the scientific path to prove the effectiveness and safety of hysteroscopic myomectomy for treating women with otherwise unexplained subfertility or before IUI/ART. Not removing the fibroids might be considered unethical by some, to proceed with ovarian stimulation, IUI or ART might be thought inappropriate by others but there is the third and overlooked possibility of asking informed consent for inclusion in a pragmatic multicentre RCT. Data from randomized studies on a plausible dose-response or location relationship and the corresponding increase in benefit or decrease in adverse events are needed as part of investigating a causal relationship. More RCTs are needed to clarify the relationship between the timing of the hysteroscopic intervention and the effects on pregnancy or live birth rates. Additional high-quality RCTs are needed to investigate whether hysteroscopy in the preceding cycle may benefit all women undergoing IVF/ICSI or should be restricted to subgroups such as women with primary subfertility and unexplained recurrent implantation failure. A recently published systematic review and meta-analysis of four RCTs and three observational studies (Potdar et al, 2012) and a Cochrane review (Nastri et al, 2012) have reported a benefit with endometrial injury (including endometrial biopsy/scratch and diagnostic hysteroscopy) during the cycle preceding ART treatment on the live birth and clinical pregnancy rates. If the endometrial injury was made on the day of oocyte retrieval, a decrease in the ongoing and clinical pregnancy rates was demonstrated (Nastri et al, 2012). The beneficial effect of repetitive endometrial sampling was first observed in a study aiming to investigate the pattern of endometrial expression of a gap junction protein. The hypothesis that local injury of the endometrium might increase the receptivity for implantation was proven by a randomized study (Barash et al, 2003). A randomized comparison between hysteroscopy in the cycle preceding ART versus on the day of oocyte retrieval could investigate a similar time-effect relationship of hysteroscopy before ART. Moreover additional research is needed to investigate whether the benefit of endometrial injury by

203 Chapter 8

scratch/biopsy or hysteroscopy in the cycle preceding ART is different according to the phase of the cycle. Since the duration of this beneficial effect is not known, more research is needed to investigate whether repeat interventions are required in cases of persistent treatment failure. This may increase costs as well as patients’ stress and discomfort; the issues of cost-effectiveness and health-related quality of life should therefore be taken into account. For the outcome of clinical pregnancy the benefit with endometrial scratch/biopsy was greater compared to hysteroscopy (Potdar et al, 2012); this result of the meta-analysis should be interpreted with great caution because all cases of operative hysteroscopy were excluded from the statistical pooling on the a priori assumption that operative hysteroscopy for unsuspected uterine anomalies is likely to improve ART success rates (Potdar et al, 2012); therefore additional RCTs are needed to measure the cost-effectiveness of endometrial biopsy/scratch versus diagnostic and/or operative hysteroscopy. Finally more basic research is needed to study the underlying biological mechanisms that may increase the endometrial receptivity during the window of implantation in the human. In 1907 a report was published on the observation of a rapid growth of endometrial stromal cells provoked by injury caused by scratching of the progestational guinea-pig uterus (Loeb, 1907). An increased influx of macrophages and dendritic cells playing a key role in decidualization of human endometrial stromal cells (HESCs) and implantation (Gnainsky et al, 2010), an increased release of pro- inflammatory cytokines and growth factors, especially leukemia- inhibitory factor (LIF), interleukin-1ȕ (IL) and heparin-binding epidermal growth factor-like growth factor (HB-EGF) (Paria et al, 2002), modulated gene expression in the endometrium with up regulation of pro- implantation proteins (Kalma et al, 2009) and synchronization of endometrial and embryo development (Zhou et al, 2008) might be plausible explanations. The basic research on the hypothetical mechanisms for the increased endometrial receptivity should assist clinicians to unravel what extent of endometrial injury is required to improve reproductive outcomes.

• Anti-adhesion therapy after hysteroscopic surgery

Our research on the effectiveness of barrier gels for preventing intrauterine adhesions is in agreement with the findings of a Cochrane systematic review (Metwally et al, 2006) but the effects of using hyaluronic acid barrier gels following operative hysteroscopy in subfertile women on the key reproductive outcomes are unknown and should be studied.

204 Thesis summary and discussion

RANDOM REFLECTIONS (*) ON THE FUTURE OF EVIDENCE- BASED PRACTICE IN REPRODUCTIVE SURGERY

There is little doubt on the validity of using a rigorously scientific approach (i.e. a RCT) to measure the efficacy of an intervention under ideal experimental conditions; the RCT offers the potential of the least biased measure of the true treatment effect. A well-designed RCT has therefore a high internal validity. This means that the estimates of the treatment effect for the conditions, intervention and the population described in the experiment are highly reliable. But is this equally true for all subfertile women? Or at least for those under the care of the reproductive surgeon who consults the findings of an “experimental” RCT to propose a treatment based on the best available evidence but in circumstances wholly different from the “experimental” condition described in the RCT? The applicability of the study findings to a general or target population constitutes the external validity of a trial. Recently there has been an emerging interest in ‘real life’ studies. In one review (Saturni et al, 2014) the trade-off between internal and external validity of a trial was erroneously simplified to the distinction between ‘experimental’ RCTs measuring efficacy versus ‘real life’ studies measuring effectiveness; this disregards the fact that explanatory (efficacy) and management trials (effectiveness) are a continuum rather than a dichotomy (Haynes et al, 2006). In criticism on the rigorous approach of the ‘experimental’ RCT has led to the introduction of ‘practice based evidence’ aiming to improve the prediction of outcomes at the individual level (Margison et al, 2000). Practice based evidence is defined as the ability to use our clinical skills and past experience to rapidly identify each patient’s unique health state and diagnosis, their individual risks and benefits of potential interventions, and their personal values and expectations. The ‘conscientious, explicit, and judicious use of the current best evidence in making decisions about the care of individual patients’ stands for the integration of individual expertise of the clinician with the best available evidence from systematic research taking into account the patient’s unique values and preferences. Practice based evidence is therefore a part of evidence base practice. The scientific approach of EBM might be less appealing for the practicing clinician who may be reluctant to follow the challenging path of finding valid proof for a given intervention rather than to continue offering non- evidence based therapy (Kamphuis et al, 2014). We agree with others that the strength of a well-conducted RCT consists of its high internal validity; this enables the clinician to distinguish between effective, undetermined, ineffective or even possibly harmful interventions (Johnson et al, 2003;

205 Chapter 8

Johnson et al, 2008; Farquhar et al, 2013). We agree that there are many barriers for designing RCTs in surgery. Many of these have been identified by others (McCulloch et al, 2002; McLeod, 1999) e.g. the problems of the surgical learning curve and differences in surgical expertise (Devereaux et al, 2005), the difficulty to standardize surgical interventions (Kapiteijn et al, 1999), the problem of blinding in surgical trials (Moseley et al, 2002), the presence of co-interventions (Haynes et al, 2006), problems of recruitment (Haynes et al, 2006) and issues of statistical power calculation (Dimick et al, 2001). Recently there has been a growing recognition of the problem of implementing EBM: the biomedical knowledge grows exponentially but the integration of research into practice is slow (Oude Rengerik et al, 2011). Clinical practice guidelines based on the best available evidence have the potential to decrease undesirable variation in care between settings and may facilitate the implementation of evidence-based practice. Attempts have been made to study the effectiveness of multi-faceted strategies to promote change and clinical daily practice implementation by using a RCT as study design (Mourad et al, 2011). It is clear that evidence-based practice is necessary but not sufficient to fulfill the needs of the caretakers and the women under their care. Recent years have witnessed the appearance of patient- centeredness in fertility care (Dancet et al, 2010). The framework for patient-centered fertility treatment places the patient/woman in the central position while taking into account effectiveness, safety, costs and burden of fertility treatment (Dancet et al, 2014). It is clear from the definitions of patient-centered care and evidence-based medicine that both are part of a continuum rather than being competitive or mutually exclusive. Clinicians might be tempted to consider the evidence provided by observational studies as an ethical basis to discard the need for additional high-quality research, as in the cases of submucous fibroids in women with otherwise unexplained subfertilitity or a uterine septum in women suffering from recurrent miscarriage (Kowalik et al, 2010). This “certainty” based on the findings of studies with an inherent risk of bias should nevertheless be put aside in deference to the reasoned uncertainty existing within the larger community of experts (Haynes et al, 2006). The true uncertainty on the part of the expert professional community about the benefit to harms balance of two or more treatments for a well- defined study population has been termed “clinical equipoise” (Freedman, 1987). The balance between the clinical equipoise and the ‘certainty’ based on studies of lesser methodological quality should be the standard to judge whether or not additional RCTs should be conducted to study the effectiveness of an intervention. To our opinion, the possibility of

206 Thesis summary and discussion participation in a RCT is therefore not only a third and often overlooked alternative but a real ‘treatment’ option if the balance is undetermined. The final question is whether there is any proof that the use of EBM indeed improves practice. Has there been any proof that evidence-based care does improve outcomes that matter for patients (Dobbie et al, 2000)? To solve this crucial question scientifically by randomly comparing evidence-based versus non-evidence-based practice would truly be a daunting challenge. An alternative but to our opinion valid answer comes from data derived from outcomes research studies. Outcomes research-the study of the end results of health services that takes patients’ experiences, preferences, and values into account-is intended to provide scientific evidence relating to decisions made by all who participate in health care (Clancy and Eisenberg, 1998). Many data from outcomes research studies have proven that patients who received evidence-based treatment often had better outcomes than those who did not in several domains of medicine (Krumholz et al, 1996; Krumholz et al, 1998) including reproductive medicine (Twisk et al, 2006; Pandian et al, 2013; Mastenbroek et al, 2011) and reproductive surgery (Johnson et al, 2008; Johnson et al, 2010). We agree with others that “demanding rigorous evidence in evaluating the effectiveness of medical interventions is a good thing” (Imrie and Ramey, 2008). This applies equally to reproductive surgery: true progress in the field of reproductive surgery is based upon surgical skills, innovation in techniques and instruments, basic research and the exposure of the clinical practice research to the validity of the RCT. If the balance of effectiveness is undetermined reproductive surgeons should follow the guiding principle of clinical equipoise rather than hide behind false certainty or pseudo-ethical objections as an excuse for not changing clinical practice or being reluctant to do additional clinical research. There couldn’t have been a more fitting end for the story of the wooden spoon in the introduction of this thesis than the powerful “Old English” words of one of Cochrane’s greatest countrymen (**): “Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning”.

207

Samenvatting proefschrift en discussie

209

Samenvatting proefschrift en discussie

SAMENVATTING

'Archie' Cochrane verwees in 1987, één jaar voor zijn dood, naar een systematisch literatuuroverzicht van gerandomiseerde studies over de zorg tijdens zwangerschap en bevalling als "een mijlpaal in de geschiedenis van gerandomiseerde studies en de evaluatie van zorg" (Chalmers et al, 1989). Van dan af was er voor hem geen reden meer om de beruchte houten lepel beschreven in hoofdstuk 1 nog verder in het bezit te laten van de discipline Obstetrie. Maar geldt dit ook voor reproductieve chirurgie? Wij vatten de bevindingen van dit proefschrift over ‘evidence- based’ praktijkvoering in de reproductieve chirurgie samen door de drie onderzoeksvragen te beantwoorden. We bespreken tevens de implicaties voor de klinische praktijk en toekomstig onderzoek. We besluiten met enkele korte beschouwingen over de toekomst voor ‘evidence-based’ praktijkvoering in het domein van de reproductieve chirurgie. Hoofdstuk 2 bespreekt de positie van de laparoscopie in de hedendaagse vruchtbaarheidskliniek. Hoewel ART geleidelijk aan tubaire chirurgie als eerstelijns behandeling van tubaire onvruchtbaarheid lijkt te vervangen, is er geen gerandomiseerde vergelijking van de beide behandelingen. We hebben beperkt bewijs gevonden voor de doeltreffendheid van de laparoscopie voor de behandeling van vrouwelijke onvruchtbaarheid in bepaalde klinische indicaties. Deze werd samengevat in de onderstaande alinea. De Canadese ENDOCAN studie heeft een voordeel aangetoond voor de laparoscopische behandeling van minimale en/of milde endometriose voor de uitkomstmaat 'doorgaande zwangerschap na 20 weken of levendgeboorte' bij vrouwen met overigens onverklaarde onvruchtbaarheid: odds ratio (OR) 2.0, 95% betrouwbaarheidsinterval (BI) was 1.2 tot 3.3, P=0.01, 341 vrouwen. De kleinere Gruppo Italiano studie kon echter geen bewijs leveren voor dergelijk voordeel (OR 0.85, 95% BI 0.32 tot 2.3, P=0.75, 96 vrouwen). Laparoscopische kapseldrilling bij vrouwen met CC-resistente polycysteuze ovaria is minstens even effectief als behandeling met gonadotrofines: de odds ratio (OR) was 1.0, 95% BI 0.63 tot 1.6, P=0.95, 4 studies, 304 vrouwen. De kans op meerlingzwangerschap was beduidend lager (OR 0.13, 95% BI 0.03 tot 0.59, P=0.008, 4 studies, 154 vrouwen). Het doen een laparoscopie voor een eerste of na verscheidene mislukte cycli van intra-uteriene inseminatie (IUI) lijkt de kans op zwangerschap (OR 0.81, 95% BI 0.43 tot 1.5, P=0.52, 1 studie, 154 vrouwen) noch het voorkomen van klinisch belangrijke pathologie van het kleinbekken niet beduidend te beïnvloeden (OR 0.72, 95% BI 0.28 tot 1.9, P=0.51, 1 studie, 89 vrouwen). De verwijdering van echoscopisch zichtbare hydrosalpinges vergroot de kans

211 Hoofdstuk 8

op bevalling na IVF –het relatief risico (RR) was 2.4, 95% BI 1.1 tot 5.3, P=0.038, 1 studie, 75 vrouwen- vooral zo bilateraal echoscopisch zichtbaar (RR 3.5, 95% BI 1.1 tot 11, P=0.019, 1 studie, 39 vrouwen). Samengevat lijkt er geen ‘evidence-based’ basis te bestaan voor de verschuiving van reconstructieve tubachirurgie naar ART. De beperkte evidentie over de rol van de laparoscopie in de behandeling van vrouwelijke onvruchtbaarheid en de beschikbaarheid van betrouwbare alternatieve methoden voor het opsporen van tuboperitoneale onvruchtbaarheid zijn niettemin twee belangrijke redenen om de praktijk van het uitvoeren van een laparoscopie bij alle gevallen van onvruchtbaarheid als de laatste stap in de exploratie in vraag te stellen. Hoofdstuk 3 tracht de plaats van de hysteroscopie in de vruchtbaarheidskliniek te verduidelijken door een non-Cochrane systematisch literatuuroverzicht. De Royal College of Obstetricians and Gynecologists en the European Society for Human Reproduction and Embryology bevelen hysteroscopie niet als een eerstelijns onderzoek aan voor het systematisch screenen van alle subfertiele vrouwen. Onze doelstelling was om de effectiviteit van de hysteroscopie bij subfertiele vrouwen voor het verbeteren van de reproductieve uitkomst samen te vatten en kritisch te beoordelen. Wij vonden vijf gerandomiseerde studies. De hysteroscopische verwijdering van endometriumpoliepen met een gemiddelde doormeter van 16mm zichtbaar via echoscopie kan de kans op zwangerschap verdubbelen in vergelijking met een louter diagnostische hysteroscopie bij vrouwen die een behandeling met IUI moeten ondergaan (RR 2.2, 95% BI 1.6 tot 3.1, P < 0.00001, 1 studie, 215 vrouwen). Er is geen bewijs voor enig voordeel voor het hysteroscopisch verwijderen van myomen versus gerichte sexuele activiteit bij vrouwen met één submuceus fibroom < 4cm en onverklaarde onvruchtbaarheid hoewel de puntschatting net niet statistisch significant is (RR 1.9, 95% BI 0.97 tot 3.7, P=0.06, 1 studie, 94 vrouwen). Hysteroscopische septumresectie voor een uterus septus lijkt minder doeltreffend bij vrouwen met primaire onvruchtbaarheid dan bij vrouwen met habitueel miskraam (RR 0.7, 95% BI 0.5 tot 0.9, P=0.01, 1 studie, 160 vrouwen). Hysteroscopie vóór een volgende IVF poging kan de kans op zwangerschap bij vrouwen met primaire subfertiliteit en minstens twee mislukte IVF pogingen bijna verdubbelen vergeleken met het opstarten van een volgende IVF poging zonder voorafgaande hysteroscopie (RR 1.6, 95% BI 1.3 tot 1.9, P< 0.00001, 2 studies, 941 vrouwen) en dit ongeacht of er pathologie was of niet (RR 0.91, 95% BI 0.71 tot 1.2, P=0.48, 2 studies, 465 vrouwen). Samengevat blijkt er bewijs om over te gaan tot een exploratie via hysteroscopie bij sommige subfertiele vrouwen voor enkele beperkte indicaties.

212 Samenvatting proefschrift en discussie

Hoofdstuk 4 illustreert de toegevoegde waarde van een Cochrane review voor het opsporen van kennis lacunes omdat dit toelaat om toekomstig primair onderzoek te oriënteren. Wij trachten de effecten te bestuderen van de hysteroscopische verwijdering van endometriumpoliepen, submuceuze myomen, uteriene septa of intra-uteriene adhesies opgespoord via een echoscopie, hysterosalpingografie, diagnostische hysteroscopie of een combinatie van deze onderzoeken bij vrouwen met onverklaarde subfertiliteit of voorafgaand aan behandeling met IUI, IVF of ICSI. Wij vonden geen gerandomiseerde studies over de hysteroscopische verwijdering van endometriumpoliepen, intra-uteriene adhesies of uteriene septa bij vrouwen met onverklaarde onvruchtbaarheid. De enige studie bij vrouwen met onverklaarde onvruchtbaarheid en myomen rapporteert bovendien geen gegevens over de vooraf bepaalde primaire uitkomstmaten voor deze Cochrane review (levendgeboorte als positieve en procedure gerelateerde verwikkelingen als negatieve uitkomst). We vonden evenmin gerandomiseerde studies over de doeltreffendheid van operatieve hysteroscopie bij onvruchtbare vrouwen met gediagnosticeerde submuceuze myomen, intra-uteriene adhesies of uteriene septa voorafgaand aan behandeling met IUI, IVF of ICSI. De enige studie in deze tweede categorie van gerandomiseerde vergelijkingen (behandeling van endometriumpoliepen vóór IUI) rapporteerde evenmin gegevens voor de primaire uitkomstmaten. Bovendien vonden we geen gegevens over de effecten van het aantal, de grootte/uitgebreidheid of de lokalisatie van de uteriene afwijkingen op de hoofduitkomsten voor beide categorieën van gerandomiseerde vergelijkingen of over een verband tussen het tijdstip van de hysteroscopische interventie en effecten hiervan op de uitkomsten van een daaropvolgende behandeling met IUI of ART. Wij concluderen dat er nog vele lacunes zijn in onze huidige kennis over de effecten van hysteroscopische interventies op de reproductieve uitkomst van subfertiele vrouwen. Dit is een opportuniteit voor verder onderzoek. Hoofdstuk 5 onderzoekt de rol van het gebruik van anti-adhesieve gels na operatieve hysteroscopie voor de behandeling van vrouwelijke onvruchtbaarheid. De bedoeling van dit systematisch literatuuroverzicht was om de doeltreffendheid te onderzoeken van het gebruik van anti- adhesieve barrière gels na een hysteroscopische interventie voor onvruchtbaarheid. Wij vonden vijf studies. Na beoordeling van het risico op systematische vertekening besloten wij om een meta-analyse te doen voor sommige gerandomiseerde vergelijkingen. Er lijkt geen bewijs te zijn voor enig voordeel voor het gebruik van gelijk welke anti-adhesieve gel na operatieve hysteroscopie voor de hoofduitkomsten levendgeboorte of klinische zwangerschap (RR 3.0, 95% BI 0.35 tot 26, P=0.32, 1 studie, 30

213 Hoofdstuk 8

vrouwen). Het gebruik van een anti-adhesieve gel vermindert wel de incidentie van intra-uteriene adhesies bij een second-look hysteroscopie (RR 0.65, 95% BI 0.45 tot 0.93, P=0.02, 5 studies, 372 vrouwen). De NNTB is 9 (95% BI 5 tot 33). Het gebruik van ACP gel bij vrouwen die een operatieve hysteroscopie ondergaan voor submuceuze myomen, endometriumpoliepen of uteriene septa is geassocieerd met lagere gemiddelde adhesiescores bij second-look hysteroscopie (MD -1.4, 95% BI -1.8 tot -1.0, P<0.00001, 1 studie, 24 vrouwen); de grootste daling kan worden aangetoond bij vrouwen behandeld met ACP gel na hysteroscopische adhesiolyse (MD -3.3, 95% BI -3.4 tot -3.2, P<0.00001, 1 studie, 19 vrouwen). Als er toch intra-uteriene adhesies ontstaan na operatieve hysteroscopie, dan blijkt het gebruik van een anti-adhesieve gel meer met ”milde” adhesies (RR 2.8, 95% BI 1.1 tot 7.0, P = 0.03, 4 studies, 79 vrouwen) en minder met ”matige of ernstige adhesies” bij second-look hysteroscopie geassocieerd te zijn (RR 0.25, 95% BI 0.10 aan 0.67, P = 0.006, 4 studies, 79 vrouwen). De NNTB is 2 met een 95% BI 1 tot 2 voor de eerstgenoemde en een 95% BI 1 tot 4 voor de laatstgenoemde uitkomst. De kwaliteit van de gevonden evidentie is zeer laag. Samengevat zouden gynaecologen vrouwen met kinderwens moeten informeren dat intra-uteriene adhesies de belangrijkste langetermijn verwikkeling is van operatieve hysteroscopie bij vrouwen in de reproductieve periode. Men kan het gebruik van anti-adhesieve gel na operatieve hysteroscopie steeds overwegen omdat dit de incidentie van intra-uteriene adhesies vermindert en als er toch vergroeiingen ontstaan dan blijken deze meer ‘mild' en minder 'matig of ernstig'. Wij vonden enkel bewijs van zeer lage methodologische kwaliteit uit een beperkt aantal gerandomiseerde studies in een algemene populatie die niet beperkt was tot vrouwen met subfertiliteit. Bovendien moeten onvruchtbare vrouwen worden geïnformeerd dat er momenteel geen bewijs is voor een voordeel voor het gebruik van anti-adhesieve gels voor de uitkomstmaten zwangerschap of levendgeboorte. Hoofdstuk 6 bestudeert de doeltreffendheid van verschillende anti- adhesie strategieën na operatieve hysteroscopie bij subfertiele vrouwen. Intra-uteriene adhesies hebben een slechte reproductieve uitkomst door een hoge incidentie van onvruchtbaarheid of habitueel miskraam en door ernstige obstetrische verwikkelingen als eindpunt van een succesvolle hysteroscopische behandeling van uitgebreide intra-uteriene adhesies. Daarom stelt de literatuur verschillende preventieve strategieën voor maar vaak zijn deze enkel gebaseerd op observationeel onderzoek. Onze Cochrane systematische literatuurstudie tracht om de doeltreffendheid van verschillende preventieve maatregelen zoals het achterlaten van een IUD of een Foley katheter ballon, het voorschrijven van postoperatieve

214 Samenvatting proefschrift en discussie hormoontherapie, het aanbrengen van anti-adhesieve gels of enten van menselijk amnion membraan samen te vatten en kritisch te beoordelen. Hoofdstuk 7 heeft als doel na te gaan of er enige vooruitgang is geboekt in de ‘evidence-based’ praktijkvoering in reproductieve chirurgie. In 1996 werd de CONSORT verklaring gepubliceerd als een methodologisch initiatief om de normen van de rapportering van klinische studies te verhogen. Het doel van ons onderzoek was om de effecten van CONSORT op het aantal en de kwaliteit van de wetenschappelijke publicaties van gerandomiseerde studies in het domein van de reproductieve chirurgie bij subfertiele vrouwen te onderzoeken. Via een systematische literatuurstudie vonden wij 64 gerandomiseerde studies over chirurgische ingrepen bij onvruchtbare vrouwen met zwangerschap of levendgeboorte als uitkomstmaat: 16 studies behoren tot de pre CONSORT groep (”1995) en 38 studies tot de post CONSORT groep (•1999). Voor de beoordeling van het kwalitatieve luik van de onderzoeksvraag werden 10 studies gepubliceerd tussen 1996 en 1998 uitgesloten om een redelijke termijn toe te laten voor het implementeren van de CONSORT verklaring. Het aantal gerandomiseerde studies die zwangerschap of levendgeboorte als hoofd uitkomstmaat hebben gerapporteerd tijdens 2000-2010 (N=32) is hetzelfde als dat van alle gerandomiseerde studies die tijdens de periode 1970-1999, zijnde drie decennia werden gepubliceerd. Wij hebben de methodologische kwaliteit post versus pre CONSORT met elkaar vergeleken door het aantal punten beoordeeld als laag risico op bias te tellen via de Cochrane 'Risk of bias tool’. We hebben een consistente trend gevonden voor een verbetering in het voordeel van de post CONSORT studies voor de punten ‘genereren van willekeurige volgorde’ (RR 2.0, 95% BI 1.1 tot 3.5, P=0.02, 54 studies), ‘blinderen van toewijzing’ (RR 2.2, 95% BI 0.90 tot 5.4, P = 0.08, 54 studies), ‘volledigheid van uitkomst gegevens’ (RR 1.4, 95% BI 0.92 tot 2.3, P= 0.11, 54 studies), ‘selectieve uitkomst rapportering’ (RR 1.3, 95% BI 0.94 tot 1.7, P=0.12, 54 studies) en ‘andere mogelijke oorzaken van systematische vertekening’ (RR 2.1, 95% BI 0.52 tot 8.5, P=0.30, 54 studies), maar niet voor het punt ‘blindering’ (RR 0.42, 95% BI 0.12 tot 1.5, P=0.18, 54 studies). Er blijkt een belangrijke verbetering in de globale methodologische kwaliteit omschreven door de composiet uitkomstmaat 'alle zes punten voor risico voor systematische vertekening' in het voordeel van de post CONSORT studies (RR 1.5, 95% BI 1.2 tot 1.9, P=0.001, 54 studies). Wij besluiten dat er de laatste jaren een duidelijke toename is geweest in zowel het aantal als de kwaliteit van de wetenschappelijke publicaties over reproductieve chirurgie bij vrouwen met vruchtbaarheidsproblematiek.

215 Hoofdstuk 8

DISCUSSIE

Onze systematische literatuurstudie kon geen gerandomiseerde vergelijking van reconstructieve tuba chirurgie met expectatief beleid of ART vinden in een doelpopulatie van vrouwen met tubaire onvruchtbaarheid waarbij levendgeboorte of zwangerschap werd bestudeerd als hoofduitkomstmaat. Dit stemt overeen met de bevindingen van twee Cochrane reviews (Pandian et al, 2008; Johnson et al, 2010). Dit beantwoordt de eerste onderzoeksvraag van dit proefschrift.

Onze beoordeling van de beschikbare evidentie over de doeltreffendheid van hysteroscopie voor het behandelen van onvruchtbare vrouwen kon een beperkt positief effect aantonen bij vrouwen met endometriumpoliepen vóór IUI en bij vrouwen met primaire subfertiliteit en minstens twee gefaalde IVF pogingen vóór een volgende poging IVF. Bovendien hebben wij tal van lacunes in de kennis opgemerkt. Het gebruik van anti-adhesie barrière gels kan worden overwogen na operatieve hysteroscopie bij subfertiele vrouwen omdat dit de incidentie, de ernst en de uitgebreidheid van adhesies kan verminderen; gegevens over de effecten op de belangrijkste reproductieve uitkomsten ontbreken echter. Dit beantwoordt de tweede onderzoeksvraag van ons proefschrift

Ons onderzoek heeft een toename in aantal en methodologische kwaliteit van publicaties over gerandomiseerd onderzoek over de doeltreffendheid van de reproductieve chirurgie in de afgelopen jaren aangetoond. Dit beantwoordt althans voor een gedeelte de derde onderzoeksvraag van ons proefschrift. Onze bevindingen stemmen overeen met deze van een recent verschenen opinie artikel over ‘Twenty years of Cochrane reviews in Menstrual Disorders and Subfertility' (Farquhar et al, 2013). Het aantal RCTs over de behandeling van subfertiliteit is volgens de auteurs toegenomen van bijna 2000 in 1997 naar meer dan 5000 in 2013. Er is een duidelijke verschuiving naar het opzoeken en verzamelen van bewijs van effectiviteit van de hoogste kwaliteit; dit moet ongetwijfeld kunnen leiden naar een groter gebruik van deze bewijsvoering op de werkvloer met de mogelijkheid om de kloof tussen de theorie en klinische praktijkvoering te overbruggen.

216 Samenvatting proefschrift en discussie

1. KLINISCHE IMPLIKATIES

• Laparoscopische chirurgie voor de behandeling van onvruchtbaarheid

Men kan niet langer verdedigen om routinematig een laparoscopie te doen bij alle subfertiele vrouwen als laatste stap in de fertiliteit exploratie: om 100% zekerheid te hebben over de staat van de eileiders zou men enkel de groep van onvruchtbare vrouwen met minimale of prognostisch dubieuze pathologie groter maken wat leidt tot een inflatie van de kosten voor het gezondheidszorgsysteem (Collins en Van Steirteghem, 2004). HSG en laparoscopie zijn even doeltreffend voor het voorspellen van de kansen op een spontane conceptie in een algemene subfertiele populatie (Verhoeve et al, 2011). Onvruchtbare vrouwen met een voorgeschiedenis van PID, verwikkelde appendicitis, kleinbekken chirurgie, ectopische graviditeit en endometriose hebben een hogere kans op tuboperitoneale pathologie (Luttjeboer et al, 2009) en moeten worden gecounseld om vroeg in de exploratie een diagnostische laparoscopie te ondergaan (NICE 2004; ASRM 2006). CAT via de microimmuno-fluorescentie test lijkt de eerste keuze voor de initiële screening op tuba pathologie (Broeze et al, 2011). Het combineren van de anamnese met CAT heeft een grotere voorspellende waarde voor de selectie van subfertiele vrouwen die in aanmerking komen voor een diagnostische laparoscopie vergeleken met ieder van beide apart en biedt als triage de mogelijkheid om het aantal 'onnodige laparoscopies' te verminderen met 82% vergeleken met het routinematig uitvoeren van een laparoscopie bij alle gevallen (Coppus et al, 2007). Het HSG is een nuttige screeningstest voor het opsporen van tuba afwijkingen bij alle subfertiele vrouwen (Broeze et al, 2011; the Practice Committee of the American Society for Reproductive Medicine, 2012). De combinatie van anamnese, CAT en HSG heeft de hoogste performantie voor de diagnostiek van bilaterale tuba pathologie (Broeze et al, 2012). Het routinematig uitvoeren van testen voor het opsporen van eileiderdoorgankelijkheid bij de vruchtbaarheidsexploratie is niet kosteneffectief bekeken vanuit het standpunt van de zorgverzekeraar of gezondheidsautoriteit: als het nodig is om informatie te verkrijgen over de status van de eileiders om klinische beslissingen te nemen voor de gezondheid van individuele vrouwen dan is het HSG gevolgd door geïndividualiseerde behandeling of een diagnostische laparoscopie in gevallen van niet doorgankelijke eileiders op het HSG meer rendabel dan de routinematig uitgevoerde diagnostische laparoscopie (Verhoeve et al, 2013).

217 Hoofdstuk 8

De ovariële kapseldrilling gevolgd door CC en gonadotrofines bij persisterende anovulatie is een aantrekkelijk alternatief voor de tweede lijn behandeling van vrouwen met CC-resistente PCOS vooral omwille van het lager risico op meerlingzwangerschap voor minstens dezelfde doeltreffendheid vergeleken met het onmiddellijk opstarten van gonadotrofine behandeling. Bovendien heeft een studie met langere termijn opvolging aangetoond dat ovariële kapseldrilling met elektrocauterisatie aanzienlijk de nood tot ovulatie-inductie of ART voor een eerste levendgeborene vermindert en de kans op een tweede kind verhoogt (Nahuis et al, 2011). Er bleken geen beduidende verschillen te bestaan in levenskwaliteit tussen ovariële kapseldrilling en gonadotrofine behandeling, behalve voor een subgroep van vrouwen zonder doorgaande zwangerschap waarbij meer depressieve symptomen in de studie arm van de gonadotrofine behandeling werden opgemerkt vergeleken met de groep van de kapseldrilling (van Wely et al, 2004). De totale kosten voor de behandeling voor een doorgaande zwangerschap zijn gelijk voor beide behandelstrategieën (van Wely et al, 2004) maar de gegevens van een langere termijn follow-up hebben aangetoond dat de gemiddelde behandelingskosten per levendgeboorte binnen 8 tot 12 jaar na de eerste behandeling aanzienlijk lager zijn voor de strategie waarbij eerst gestart wordt met ovariële kapseldrilling vergeleken met gonadotrofines (Nahuis et al, 2012). Daarom dient het starten met ovariële kapseldrilling gevolgd door clomiphene citraat en gonadotrofines in geval van persisterende anovulatie te worden aanbevolen als een rationele tweede lijn behandeling bij subfertiele vrouwen met CC-resistente PCOS. Reconstructieve chirurgie was de enige beschikbare behandelingsoptie voor tubaire onvruchtbaarheid in het tijdperk vóór de IVF. Tubaire chirurgie biedt vrouwen de mogelijkheid om meer dan één maal zwanger te kunnen worden door een eenmalig uitgevoerde minimaal invasieve ingreep waardoor ART en de hiermee verbonden risico’s kunnen worden vermeden. De nadelen van reconstructieve chirurgie zijn de risico’s gepaard met de chirurgische ingreep/narcose en de kans op recidief hydrosalpinx. In geval van bilaterale onherstelbare hydrosalpinges is er kwaliteitsvol bewijs dat salpingectomie of tubaire occlusie ondersteunt om de kansen op een zwangerschap via IVF te verhogen (Johnson et al, 2010; Johnson et al, 2011). Aan jonge vrouwen zonder verdere problemen kan een laparoscopische fimbrioplastie of neosalpingostomie worden aanbevolen om tuba pathologie met een goede prognose zoals phimosis of adhesies te herstellen (Mol et al, 1999); het uitvoeren van een bilaterale salpingectomie zou alle mogelijkheden voor een spontane zwangerschap in deze gevallen onherroepelijk wegnemen (Johnson et al, 2011). Microchirurgische of laparoscopische reanastomose door een deskundige

218 Samenvatting proefschrift en discussie reproductieve chirurg verdient mogelijks de voorkeur na een voorafgaande eileidersterilisatie (the Practice Committee of the American Society for Reproductive Medicine, 2012).

• Hysteroscopische chirurgie voor de behandeling van onvruchtbaarheid

Het uitvoeren van een hysteroscopie voorafgaand aan een daaropvolgende IVF/ICSI behandeling kan de kans op een succesvolle reproductieve uitkomst vergroten bij vrouwen met primaire subfertiliteit en minstens twee mislukte IVF/ICSI pogingen. Een systematisch literatuuronderzoek met meta-analyse van vijf observationele studies en één gerandomiseerde studie heeft aangetoond dat een hysteroscopie vóór een eerste IVF-poging de kansen op een klinische zwangerschap en levendgeboorte kan vergroten; deze gegevens moeten omzichtig worden geïnterpreteerd wegens de aanzienlijke klinische verscheidenheid en de statistische heterogeniteit van de geïncludeerde studies (Pundir et al, 2013). Een hysteroscopie vóór een eerste IVF poging is mogelijk kosteneffectief zoals blijkt uit een studie van een kostenbeslissing model (Kasius et al, 2013). Bijkomend onderzoek is nodig alvorens men met zekerheid kan aanbevelen om routinematig een hysteroscopie uit te voeren voorafgaand aan iedere eerste ART behandeling.

• Anti-adhesieve behandeling na hysteroscopische chirurgie

Gynaecologische chirurgen moeten vrouwen met zwangerschapswens informeren dat de vorming van intra-uteriene adhesies de belangrijkste lange termijn verwikkeling is na operatieve hysteroscopie bij vrouwen in de reproductieve leeftijd. Een recent Nederlands onderzoek heeft aangetoond dat het bedacht zijn op de problematiek van adhesies bij Nederlandse gynaecologen beperkt is en dat de counseling over dit probleem niet adequaat verloopt (Meuleman et al, 2013). Het gebruik van anti-adhesieve gels na operatieve hysteroscopie kan de incidentie van adhesies verminderen net als de mate/ernst van de intra-uteriene adhesies; de effecten op de reproductieve uitkomst zijn niet echter niet onderbouwd door bewijs van hoge kwaliteit vergelijkbaar met het gebruik van anti- adhesieve gel na laparoscopische myomectomie (Pellicano et al, 2005). Dit gebrek aan voldoende bewijs mag men uiteraard niet over het hoofd zien tijdens de counseling.

219 Hoofdstuk 8

2. TOEKOMSTIG ONDERZOEK

• Laparoscopische chirurgie voor de behandeling van onvruchtbaarheid

Bijkomend onderzoek is nodig om de doeltreffendheid van HyCoSy te onderzoeken voor het voorspellen van spontane zwangerschap; één gerandomiseerd onderzoek kon geen voordeel aantonen voor HyCoSy voor het vergroten van de kansen op een klinische zwangerschap bij subfertiele vrouwen (Lindborg et al, 2009). Verschillende vergelijkende studies suggereren een sterke overeenkomst tussen de bevindingen van transvaginale hydrolaparoscopie (THL) en conventionele laparoscopie (Campo et al, 1999; Casa et al, 2002; Darai et al, 2000; Nawroth et al, 2001). In een prospectieve cohort studie werden concordante FRRs gevonden voor unilaterale tuba pathologie, bilaterale tuba pathologie en adhesies/endometriose; dit suggereert dat de capaciteit van THL om de kans op een spontane doorgaande zwangerschap te voorspellen dezelfde is als deze van de gewone laparoscopie (van Tetering et al, 2007). Definitief bewijs kan slechts komen van bijkomend pragmatisch gerandomiseerd onderzoek dat THL vergelijkt met de standaard laparoscopie in een vergelijkbare doelgroep met een kort tijdsinterval tussen beide procedures. Bijkomend onderzoek is ook nodig om de waarde van klinische predictiemodellen voor tubaire pathologie en endometriose (alle stadia versus fase III of IV) te bestuderen voor het selecteren van vrouwen met onverklaarde onvruchtbaarheid die bij voorkeur vroeg in de onderzoeksfase THL als een poliklinische procedure onder lokale verdoving zouden moeten ondergaan. Verschillende auteurs melden dat het parallel debat over de gezondheidseconomische aspecten van ART om het meest geschikte financieringskader voor het verstrekken van veilige, billijke en kosteneffectieve behandeling te omschrijven uiterst ingewikkeld is (Mol et al, 2000; Connolly et al, 2010; Berg Brigham et al, 2012; Chambers et al, 2013). Kosteneffectiviteit studies die reconstructieve tuba chirurgie vergelijken met expectatief beleid als een alternatief of voorafgaande aan ART bij vrouwen met tubaire infertiliteit dienen hoog bovenaan op de onderzoeksagenda te worden geplaatst, zeker in landen met een hoog verbruik van ART, zoals België; de vergelijking tussen beide interventies moet ook de uitkomsten van ongewenste neveneffecten en gezondheidsgerelateerde levenskwaliteit in acht nemen. De terughoudendheid vanwege de artsen in de reproductieve geneeskunde om het als hoog gepercipieerde succes van ART openlijk in vraag te durven stellen lijkt een grote belemmering te zijn voor het uitvoeren van dergelijk

220 Samenvatting proefschrift en discussie onderzoek (Kamphuis et al, 2014). We moeten anderzijds ook toegeven dat er belangrijke beperkingen zijn voor de uitvoerbaarheid van dergelijke studies bijvoorbeeld voor wat betreft de beschikbaarheid van voldoende chirurgische expertise voor reconstructieve tuba chirurgie (Pandian et al, 2008) en een voldoende hoge prevalentie van tubaire pathologie (Sabatini en Davis, 2005). A priori gedefinieerde subgroep analyses zijn nodig om de kosteneffectiviteit van tubaire reconstructieve chirurgie te vergelijken met ART in specifieke subgroepen (bvb. oudere versus jongere vrouwen, poor responders, eerdere tubasterilisatie...).

• Hysteroscopische chirurgie voor de behandeling van onvruchtbaarheid

Wij zijn het eens met andere auteurs dat er op dit ogenblik onvoldoende hoog kwalitatief bewijs is gebaseerd op gerandomiseerd onderzoek met voldoende statistisch onderscheidingsvermogen dat enig voordeel kan aantonen voor de hysteroscopische myoomverwijdering voor het vergroten van de kans op zwangerschap of levendgeboorte bij vrouwen met onverklaarde onvruchtbaarheid of voorafgaand aan ART (Metwally et al, 2012). De bevindingen van een systematisch literatuuroverzicht van observationele studies (Pritts et al, 2009) suggereren een negatief impact van submuceuze myomen op de reproductieve uitkomst en een voordeel voor de hysteroscopische verwijdering van de fibromen. Wij staan als reproductieve chirurgen bijgevolg voor een moeilijke keuze: ofwel nemen wij dit niet op bewijs gebaseerd onderzoek voor (vertekende) waarheid ofwel kiezen wij het pad van de wetenschap om de effectiviteit en veiligheid van hysteroscopische myomectomie voor behandeling van vrouwen met onverklaarde onvruchtbaarheid of voorafgaande aan IUI/ART aan te tonen. Het niet verwijderen van de fibromen wordt wellicht door sommigen als onethisch aanzien, een aanvullende behandeling opstarten met ovariumstimulatie, IUI of ART lijkt dan voor anderen weer niet aangwezen; er is echter een derde en vaak over het hoofd geziene mogelijkheid om geïnformeerde toestemming te vragen aan de patiënte voor haar deelname aan een pragmatische multicentrische gerandomiseerde studie. Bijkomende gegevens uit gerandomiseerd onderzoek over verbanden tussen dosisrespons of lokalisatie van de fibromen en de overeenkomstige toename in voordeel of afname in nadeel zijn nodig om de causaliteit tussen submuceuze fibromen en onvruchtbaarheid te bestuderen. Er is verder onderzoek nodig naar het verband tussen de timing van de hysteroscopische interventie en de effecten op de kansen voor zwangerschap of levendgeboorte. Bijkomend onderzoek dient aan te tonen of hysteroscopie in de voorgaande cyclus

221 Hoofdstuk 8

voordeel biedt aan alle vrouwen die behandeld moeten worden met IVF/ICSI of beperkt is tot de subgroep van vrouwen met primaire subfertiliteit en onverklaard herhaald implantatiefalen. Een recent gepubliceerd systematisch literatuuroverzicht met een meta-analyse van vier gerandomiseerde en drie observationele studies (Potdar et al, 2012) en één Cochrane review (Nastri et al, 2012) hebben een positief effect aangetoond van het aanbrengen van endometriaal trauma (endometriumbiopsie/endometrial scratch en diagnostische hysteroscopie) tijdens de cyclus die voorafgaat aan de ART behandeling op de kansen op levendgeboorte en klinische zwangerschap. Wanneer het endometriaal trauma werd aangebracht op de dag van de follikelaspiratie werd een beduidende daling van de kansen op een doorgaande en klinische zwangerschap aangetoond (Nastri et al, 2012). Het positief effect van het herhaald afnemen van een endometriumbiopsie werd voor het eerst geobserveerd in een onderzoek naar het patroon van expressie van een gap junction eiwit in het endometrium. Een gerandomiseerde studie kon vervolgens de hypothese bewijzen dat lokale schade van het endometrium de receptiviteit voor implantatie kan verhogen (Barash et al, 2003). Een gerandomiseerde vergelijking tussen hysteroscopie in de cyclus voorafgaand aan ART vergeleken met de dag van de follikelaspiratie biedt de mogelijkheid om eenzelfde verband tussen tijdstip en effect van de hysteroscopie vóór ART te bestuderen. Daarnaast moet ook worden onderzocht of het voordeel van het aanbrengen van endometriaal trauma via biopsie/scratching of hysteroscopie vóór ART verschilt volgens het tijdstip in de menstruele cyclus. De duur van dit positief effect is ongekend maar klinisch relevant; aanvullend onderzoek moet een antwoord kunnen geven op de vraag of het zinvol is om de interventie te herhalen bij persisterend implantatiefalen. Deze praktijk kan leiden tot een aanzienlijke meerkost naast toegenomen stress en ongemak bij de patiënten; daarom moeten aspecten van kosteneffectiviteit en gezondheidsgerelateerde levenskwaliteit eveneens in rekening worden gebracht. Het nuteffect van de endometrium-biopsie/scratch lijkt groter dan dat van de hysteroscopie voor de uitkomstmaat klinische zwangerschap (Potdar et al, 2012); dit resultaat moet met de nodige voorzichtigheid worden geïnterpreteerd, omdat alle gevallen van operatieve hysteroscopie uit de statistische pooling in deze review werden uitgesloten vanuit de voorafgaande assumptie dat operatieve hysteroscopie voor alle niet vermoede intracavitaire afwijkingen de kansen op een succesvolle uitkomst van ART vergroot (Potdar et al, 2012). Verder onderzoek is nodig naar vergelijkende kosteneffectiviteit van endometriaal trauma via biopsie/scratch vergeleken met hysteroscopie. Bovendien is basis wetenschappelijk onderzoek nodig om

222 Samenvatting proefschrift en discussie de onderliggende biologische werkingsmechanismen voor het verhogen van de receptiviteit van het endometrium gedurende het implantatievenster te verklaren. In 1907 werd een artikel gepubliceerd over de observatie van een toegenomen groei van endometriale stromale cellen uitgelokt door letsels aan het endometrium door schrapen van progestationele baarmoeders bij de cavia (Loeb, 1907). Een grotere influx van macrofagen en dendritische cellen belangrijk voor decidualizatie van humane endometriale stroma cellen (HESCs) en implantatie (Gnainsky et al, 2010), een verhoogde vrijzetting van pro-inflammatoire cytokines en groeifactoren, vooral leukemia-inhibiting factor (LIF), interleukin-1ȕ (IL) en heparin-binding epidermal growth factor-like growth factor (HBEGF) (Paria et al, 2002), een gewijzigde genexpressie in het baarmoederslijmvlies met opregulatie van pro-implantatie eiwitten (Kalma et al, 2009) en synchronisatie van endometrium en embryonale ontwikkeling (Zhou et al, 2008) zijn plausiebele verklaringen. Dit basisonderzoek naar onderliggende hypothetische mechanismen voor het verhogen van de receptiviteit van het endometrium moet voor de klinische praktijk kunnen beantwoorden hoeveel endometriaal trauma nodig is om betere reproductieve uitkomsten met ART te kunnen behalen.

• Anti-adhesieve behandeling na hysteroscopische chirurgie

Het onderzoek naar de effectiviteit van anti-adhesieve barrière gel voor het voorkomen van intra-uteriene adhesies stemt overeen met de bevindingen van een Cochrane review (Metwally et al, 2006). Aanvullend onderzoek is nodig omdat de effecten van het gebruik van anti-adhesieve barrière gel na operatieve hysteroscopie bij de behandeling van subfertiele vrouwen op de reproductieve uitkomst onbekend zijn.

223 Hoofdstuk 8

3. WILLEKEURIGE BESCHOUWINGEN(*) OVER DE TOEKOMST VAN OP BEWIJS GEFUNDEERDE PRAKTIJK- VOERING IN DE REPRODUCTIEVE CHIRURGIE

Er is weinig twijfel over de betrouwbaarheid van het toepassen van een streng wetenschappelijke benadering voor het meten van de doeltreffendheid van een interventie onder ideale experimentele omstandigheden via een gerandomiseerde studie; de RCT biedt de mogelijkheid om het echte behandelingseffect te meten met de kleinste kans op systematische vertekening. Een lege artis opgezette RCT heeft net daarom een grote interne validiteit. Dit betekent dat de schatting van het behandelingseffect voor de voorwaarden, de interventie en de populatie beschreven in het experiment uiterst betrouwbaar zijn. Maar geldt dit evenzeer voor alle onvruchtbare vrouwen? Of minstens voor die onvruchtbare vrouwen onder de behandeling van een reproductieve chirurg die de bevindingen van een "experimentele" gerandomiseerde studie raadpleegt om een behandeling gebaseerd op de beste beschikbare evidentie voor te stellen maar werkt in omstandigheden die mogelijk sterk verschillen van deze beschreven in het experiment? De toepasbaarheid van de bevindingen van een studie op een algemene of doelpopulatie noemt men de externe validiteit van een studie. Sinds kort lijkt er een toenemende interesse in zogenaamde 'real life' studies. In een review (Saturni et al, 2014) werd de afweging tussen de interne en externe validiteit van een studie onterecht herleid tot het onderscheid tussen 'experimentele' RCTs die de werkzaamheid (efficacy) bestuderen versus 'real life' studies die beogen om de doeltreffendheid (effectiveness) te meten; deze vereenvoudiging gaat voorbij aan het gegeven dat men studies die de werkzaamheid of doeltreffendheid onderzoeken eerder dient te plaatsen in een continuüm dan een dichotomie (Haynes et al, 2006). Kritiek op de streng wetenschappelijke benadering van de 'experimentele' RCT heeft binnen het vakgebied van de psychiatrie geleid tot wat omschreven wordt als door ‘praktijk effectief verondersteld bewijs’ of ‘practice-based evidence’ om beter de uitkomsten op het niveau van de individuele patiënt te kunnen voorspellen (Margison et al, 2000). ‘Practice-based evidence’ wordt gedefinieerd als de mogelijkheid om klinische vaardigheden en ervaring te gebruiken voor het snel herkennen van de gezondheidstoestand en diagnose van elke individuele patiënt, de afzonderlijke voor- en nadelen van de alternatieve behandelingen en de persoonlijke waarden en verwachtingen. Het 'gewetensvol, expliciet en verstandig gebruiken van de beste evidentie bij het nemen van beslissingen over de zorg van individuele patiënten' staat voor de integratie van individuele expertise van de zorgverstrekker met de beste

224 Samenvatting proefschrift en discussie beschikbare gegevens van systematisch literatuuronderzoek rekening houdend met de unieke waarden en voorkeur van de individuele patiënt. ‘Practice-based evidence’ is daarom een deel van ‘evidence-based practice’. De wetenschappelijke benadering van EBM lijkt minder aantrekkelijk voor de praktiserende arts die mogelijk opziet tegen de uitdaging om betrouwbaar bewijs te zoeken voor een bepaalde interventie in plaats van het verder aanbieden van behandelingen die niet op bewijs gebaseerd zijn (Kamphuis et al, 2014). Wij zijn het eens met andere auteurs dat de sterkte van een goed uitgevoerde RCT berust op de grote interne validiteit; dit laat de clinicus toe om effectieve interventies te onderscheiden van ineffectieve, onzekere of mogelijks schadelijke interventies (Johnson et al, 2003; Johnson et al, 2008; Farquhar et al, 2013). Wij geven toe dat er veel belemmeringen kunnen zijn voor het opzetten van een degelijke chirurgische RCT. Een aantal barrières werd al door andere auteurs beschreven (McCulloch et al, 2002; McLeod, 1999) bijvoorbeeld de problematiek van de chirurgische leercurve en de verschillen in chirurgisch-technische vaardigheid (Devereaux et al, 2005), de moeilijkheid om chirurgische ingrepen te standaardiseren (Kapiteijn et al, 1999), het probleem van blindering in chirurgische klinische trials (Moseley et al, 2002), het bestaan van tal van co-interventies (Haynes et al, 2006), problemen van rekrutering (Haynes et al, 2006) en problemen met het berekenen van het statistisch onderscheidingsvermogen en de grootte van de steekproef (Dimick et al, 2001). Er is de laatste tijd ook een toenemende erkenning van het probleem over het toepassen van EBM in de praktijk: de biomedische kennis blijft exponentieel toenemen terwijl de integratie van de bevindingen en de conclusies van het klinisch wetenschappelijk onderzoek in de dagelijkse praktijkvoering te traag verloopt (Oude Rengerik et al, 2011). Praktijkrichtlijnen gebaseerd op de beste beschikbare evidentie kunnen niet wenselijke variabiliteit in zorg tussen instellingen verminderen en bijgevolg de toepassing van op bewijs gebaseerde medische praktijkvoering faciliteren. Men heeft al geprobeerd om via een gerandomiseerde studie de doeltreffendheid te meten van veelzijdige strategieën die beogen om gedragsverandering en toepassing in de dagelijkse klinische praktijk te bevorderen (Mourad et al, 2011). Het is duidelijk dat op bewijs gefundeerde medische praktijkvoering een nodige maar op zich niet voldoende voorwaarde is voor de behoeften van zowel de zorgverstrekkers als de patiënten. De afgelopen jaren kwam ook de ‘patiënt centeredness’ in de kijker binnen de vruchtbaarheidskliniek (Dancet et al, 2010). Het kader voor ‘patiënt-centered care’ binnen de fertiliteitskliniek plaatst de vrouw centraal maar houdt wel rekening met bijkomende aspecten zoals doeltreffendheid, veiligheid, kosten en

225 Hoofdstuk 8

belasting door IVF-behandeling (Dancet et al, 2014). Uit de definities van ‘patient centered’ en ‘evidence-based practice’ wordt het in één oogopslag duidelijk dat beide behoren tot eenzelfde continuüm eerder dan competitief of wederzijds uitsluitend te zijn. Voor de clinici lijkt de verleiding soms groot om evidentie gebaseerd op observationeel onderzoek als een ethisch imperatief te gebruiken om geen bijkomend klinisch wetenschappelijk onderzoek van hogere methodologische kwaliteit te moeten opzetten, zoals in de gevallen van vrouwen met onverklaarde onvruchtbaarheid en submuceuze myomen of vrouwen met habitueel miskraam en uterien septum (Kowalik et al, 2010). Deze "zekerheid" gebaseerd op resultaten van onderzoek met een inherent risico op systematische vertekening moet niettemin opzij worden geschoven uit respect voor de beredeneerde onzekerheid binnen de grotere gemeenschap van deskundigen (Haynes et al, 2006). De echte onzekerheid van de deskundige professionele gemeenschap over de balans tussen winst en verlies voor twee of meer behandelingen voor een wel omschreven populatie wordt in de Angelsaksische literatuur omschreven als "clinical equipoise"(Freedman, 1987). Het tegen elkaar afwegen van de “clinical equipoise”en de 'zekerheid' op basis van studies van lagere methodologische kwaliteit zou de norm moeten zijn om te oordelen of bijkomend onderzoek voor het toetsen van de effectiviteit van een interventie al dan niet nodig is. Wij zijn daarom van oordeel dat het aanbieden van deelname aan een gerandomiseerde studie niet alleen een derde en vaak over het hoofd gezien alternatief is maar bovendien moet worden gezien als een echte behandelingsoptie in alle gevallen van “clinical equipoise”. Een laatste vraag betreft het bewijs voor de doeltreffendheid van EBM in de praktijkvoering. Is er enig bewijs dat op bewijs gebaseerde praktijkvoering doeltreffend is om positieve uitkomsten die belangrijk zijn voor patiënten te verbeteren (Dobbie et al, 2000)? Het lijkt een haast onoverkomelijke uitdaging om deze kapitale vraag wetenschappelijk te benaderen via een random vergelijking van ‘evidence-based’ met non- ‘evidence-based’ praktijkvoering. Gegevens van uitkomsten onderzoek kunnen naar onze mening een alternatief maar betrouwbaar antwoord geven op deze vraag. Uitkomsten onderzoek kan worden omschreven als de studie van de eindresultaten van gezondheidszorg die rekening houdt met patiënten ervaringen, voorkeuren en waarden; deze onderzoeksvorm heeft als doel om wetenschappelijk bewijs te leveren voor beslissingen genomen door allen die deelnemen aan gezondheidszorg (Clancy en Eisenberg, 1998). Tal van gegevens uit het uitkomsten onderzoek tonen aan dat patiënten die werden behandeld met ‘evidence-based practice’ vaak betere uitkomsten hadden dan hen die geen ‘evidence-based’

226 Samenvatting proefschrift en discussie behandelingen kregen in verschillende domeinen van de geneeskunde (Krumholz et al, 1996; Krumholz et al, 1998) waaronder ook de reproductieve geneeskunde (Twisk et al, 2006; Pandian et al, 2013; Mastenbroek et al, 2011) en de reproductieve chirurgie (Johnson et al, 2008; Johnson et al, 2010). Wij zijn het eens met andere auteurs dat "de vraag naar streng bewijs voor de beoordeling van de doeltreffendheid van medische interventies een goede zaak is" (Imrie en Ramey, 2008). Dit geldt volgens ons evenzeer voor de reproductieve chirurgie: echte vooruitgang op het gebied van reproductieve chirurgie moet steunen op chirurgisch-technische vaardigheden, innovatie in technieken en instrumenten, basis wetenschappelijk onderzoek en het toetsen van het klinisch wetenschappelijk onderzoek aan de gouden standaard van de gerandomiseerde studie. Wanneer de balans van de doeltreffendheid in evenwicht, dus onbepaald is, zouden reproductieve chirurgen de leidraad moeten volgen van ‘clinical equipoise’ eerder dan zich te verschuilen achter valse zekerheid of pseudo-ethische bezwaren als excuus om hun klinische praktijkvoering niet te veranderen of terughoudend te zijn om aanvullend hoog kwalitatief klinisch onderzoek op te zetten. Er lijkt geen passender einde voor het verhaal van de houten lepel bij het begin van dit proefschrift dan de krachtige Oud-Engelse woorden uit de mond van één van Cochrane’s grootste landgenoten (*): "Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning".

227 Hoofdstuk 8

REFERENCES/REFERENTIES

1. Barash A, Dekel N, Fieldust S, Segal I, Schechtman E, Granot I. Local injury to the endometrium doubles the incidence of successful pregnancies in patients undergoing in vitro fertilization. Fertil Steril 2003; 79(6): 1317-1322. 2. Berg Brigham K, Cadier B, Chevreul K. The diversity of regulation and public financing of IVF in Europe and its impact on utilization. Hum Reprod 2013; 28(3): 666-675. 3. Broeze KA, Opmeer BC, Van Geloven N, Coppus SFPJ, Collins JA, Den Hartog JE, Van der Linden PJQ, Marianowski P, Ng EHY, Van der Steeg JW, Steures P, Strandell A, Van der Veen F, Mol BWJ. Are patient characteristics associated with the accuracy of hysterosalpingography in diagnosing tubal pathology? An individual patient data meta-analysis. Hum Reprod Update 2011; 17(3): 293- 300. 4. Broeze KA, Opmeer BC, Coppus SFPJ, Van Geloven N, Alves MFC, Anestad G, Bhattacharya S, Allan J, Guerra-Infante MF, Den Hartog JE, Land JA, Idahl A, Van der Linden PJQ, Mouton JW, Ng EHY, Van der Steeg JW, Steures P, Svenstrup HF, Tiitinen A, Toye B, Van der Veen F, Mol BW. Chlamydia antibody testing and diagnosing tubal pathology in subfertile women: an individual patient data meta- analysis. Hum Reprod Update 2011; 17(3): 301-310. 5. Broeze KA, Opmeer BC, Coppus SF, Van Geloven N, Den Hartog JE, Land JA, Van der Linden PJQ, Ng EHY, Van der Steeg JW, Steures P, van der Veen F, Mol BW. Integration of patient characteristics and the results of Chlamydia antibody testing and hysterosalpingography in the diagnosis of tubal pathology: an individual patient data meta- analysis. Hum Reprod 2012; 27(10): 2979-2990. 6. Campo R, Gordts S, Rombauts L, Brosens I. Diagnostic accuracy of transvaginal hydro laparoscopy in infertility. Fertil Steril 1999; 71(6): 1157-1160. 7. Casa A, Sesti F, Marziali M, Piccione E. Transvaginal hydro laparoscopy versus conventional laparoscopy for evaluating unexplained primary infertility in women. J Reprod Med 2002; 47(8): 617-620. 8. Chambers GM, Adamson GD, Eijkemans MJC. Acceptable cost for the patient and society. Fertil Steril 2013; 100(2): 319-327. 9. Clancy CM, Eisenberg JM. Outcomes research: measuring the end results of health care. Science 1998; 282: 245–246.

228 Samenvatting proefschrift en discussie

10. Cochrane AL. Foreword. In: Chalmers I, Enkin M, Keirse MJNC, eds. Effective care in pregnancy and childbirth. Oxford: Oxford University Press, 1989. 11. Collins JA, Van Steirteghem A. Overall prognosis with current treatment of infertility. Hum Reprod Update 2004; 10: 309-316. 12. Connolly MP, Hoorens S, Chambers GM on behalf of the ESHRE Reproduction and Society task Force. The costs and consequences of assisted reproductive technology: an economic perspective. Hum Reprod Update 2010; 16(6): 603-613. 13. Coppus SFPJ, Opmeer BC, Logan S, Van der Veen F, Bhattacharya S, Mol BWJ. The predictive value of medical history taking and Chlamydia IgG ELISA antibody testing (CAT) in the selection of subfertile women for diagnostic laparoscopy: a clinical prediction model approach. Hum Reprod 2007; 22(5): 1353-1358. 14. Dancet EAF, Nelen WLDM, Sermeus W, De Leeuw L, Kremer JAM, TM D’Hooghe. The patients’ perspective on fertility care: a systematic review. Hum Reprod Update 2010; 16(5): 467-487. 15. Dancet EAF, D’Hooghe TM, van der Veen F, Bossuyt PMM, Sermeus W, Mol BWJ, Repping S. Patient-centered fertility treatment: what is required? Fertil Steril 2014; 101(4): 924-926. 16. Darai E, Dessolle L, Lecuru F, Soriano D. Transvaginal hydro laparoscopy compared with laparoscopy for the evaluation of infertile women: a prospective comparative blind study. Hum Reprod 2000; 15(11): 2379-2382. 17. Devereaux PJ, Bhandari M, Clarke M, Montori VM, Cook DJ, Yusuf S, Sackett DL, Cinà CS, Walter SD, Haynes B, Schünemann HJ, Norman GR, Guyatt GH. Need for expertise based randomized controlled trials. BMJ 2005; 330: 88-91. 18. Dimick JB, Diener- West M, Lipsett PA. Negative results of randomized clinical trials published in the surgical literature: equivalence or error? Arch Surg 2001; 136: 796-800. 19. Dobbie AE, Schneider FD, Anderson AD, Littlefield J. What evidence supports teaching evidence-based medicine? Academic Medicine 2000; 75(12): 1184-1185. 20. Farquhar C, Moore V, Bhattacharya S, Blake D, Vail A, Thomas J, Cheong Y, Showell M, Nagels H, Marjoribanks J, on behalf of the Cochrane Editorial Board of the Menstrual Disorders and Subfertility Group. Twenty years of Cochrane reviews in menstrual disorders and subfertility. Hum Reprod 2013; 28(11): 2883-2892. 21. Farquhar C, Rishworth JR, Brown J, Nelen WLDM, Marjoribanks J. Assisted reproductive technology: an overview of Cochrane reviews. Cochrane Database Syst Rev 2013, Issue 8. Art. No.: CD010537.

229 Hoofdstuk 8

22. Freedman B. Equipoise and the ethics of clinical research. N Engl J Med 1987; 317: 141-145. 23. Gnainsky Y, Granot I, Aldo P, Barash A, Or Y, Schechtman E, Mor G, Dekel N. Local injury of the endometrium induces an inflammatory response that promotes successful implantation. Fertil Steril 2010; 94: 2030-2036. 24. Haynes BR, Sackett DL, Guyatt GH, Tugwell P. Clinical epidemiology- how to do clinical practice research. Third edition, Lipincott, Williams and Wilkins, 2006. 25. Imrie RH, Ramey DW. The evidence for evidence-based medicine. Complementary therapies in medicine 2008; 8: 123-6. 26. Johnson N, Proctor M, Farquhar C. Gaps in the evidence for fertility treatment-an analysis of the Cochrane Menstrual Disorders and Subfertility Group database. Hum Reprod 2003; 18: 947-954. 27. Johnson NP, Selman T, Zamora J, Khan KS. Gynecologic surgery from uncertainty to science: evidence-based surgery is no passing fad. Hum Reprod 2008; 23: 832-839. 28. Johnson N, van Voorst S, Sowter MC, Strandell A, Mol BWJ. Surgical treatment for tubal disease in women due to undergo in vitro fertilisation. Cochrane Database Syst Rev 2010, Issue 1. Art. No.: CD002125. 29. Johnson N, van Voorst S, Sowter MC, Strandell A, Mol BWJ. Tubal surgery before IVF. Hum Reprod Update 2011; 17(1): 3. 30. Kalma Y, Granot I, Gnainsky Y, Or Y, Czernobilsky B, Dekel N, Barash A. Endometrial biopsy-induced gene modulation: first evidence for the expression of bladder-transmembranal uroplakin Ib in human endometrium. Fertil Steril 2009; 91: 1042-1049. 31. Kamphuis EI, Bhattacharya S, van der Veen F, Mol BWJ, Templeton A, for the Evidence Based IVF Group. Are we overusing IVF? BMJ 2014; 348: g252. 32. Kapiteijn E, Kranenbarg EK, Steup WH, Taat CW, Rutten HJ, Wiggers T, van Krieken JH, Hermans J, Leer JW, van de Velde CJ. Total mesorectal excision (TME) with or without preoperative radiotherapy in the treatment of primary rectal cancer. Prospective randomized trial with standard operative and histopathological techniques. Eur J Surg 1999; 165: 410-420. 33. Kasius JC, Eijkemans RJC, Mol BWJ, Fauser BCJM, Fatemi HM, Broekmans FJM. Cost-effectiveness of hysteroscopy screening for infertile women. RBM Online 2013; 26: 619-626. 34. Kowalik CR, Mol BWJ, Veersema S, Goddijn M. Critical appraisal regarding the effect on reproductive outcome of hysteroscopic

230 Samenvatting proefschrift en discussie

metroplasty in patients with recurrent miscarriage. Arch Gynecol Obstet 2010; 282: 465. 35. Krumholz HM, Radford MJ, Ellerbeck EF, Hennen J, Meehan TP, Petrillo M, Wang Y, Jencks SF. Aspirin for secondary prevention after acute myocardial infarction in the elderly; prescribed use and outcomes. Ann Intern Med 1996; 124; 292–298. 36. Krumholz HM, Radford MJ, Wang Y, Chen J, Heiat A, Marciniak TA. National use and effectiveness of beta-blockers for the treatment of elderly patients after acute myocardial infarction. National Cooperative Cardiovascular Project. JAMA 1998; 280:623–629. 37. Lindborg L, Thorburn J, Bergh C, Strandell A. Influence of HyCoSy on spontaneous pregnancy: a randomized controlled trial. Hum Reprod 2009; 24(5): 1075-1079. 38. Loeb L. Uber die experimentelle erzeugung von knoten von deciduagewebe in dem uterus des meerschweinchens nach stattgefundener copulation. Zbl Allg Path Anat 1907; 18: 563-565. 39. Luttjeboer FY, Verhoeve HR, van Dessel HJ, van der Veen F, Mol BWJ, Coppus SFPJ. The value of medical history taking as risk indicator for tuboperitoneal pathology: a systematic review. BJOG 2009; 116: 612-625. 40. Margison FR, Barkham M, Evans C, Mc Grath G, Mellor Clark J, Audin K, Connell J. Measurement and psychotherapy. Evidence- based practice and practice-based evidence. Br J Psych 2000; 177: 123-130. 41. Mastenbroek S, Twisk M, van der Veen F, Repping S. Preimplantation genetic screening: a systematic review and meta- analysis of RCTs. Hum Reprod Update 2011; 17(4): 454-466. 42. McCulloch P, Taylor I, Sasako M, Lovett B, Griffin D. Randomised trials in surgery: problems and possible solutions. BMJ 2002; 324: 1448-1451. 43. McLeod RS. Issues in surgical randomized controlled trials. World J Surg 1999; 23: 1210-1214. 44. Metwally M, Watson A, Lilford R, Vanderkerchove P. Fluid and pharmacological agents for adhesion prevention after gynaecological surgery. Cochrane Database Syst Rev 2006, Issue 2. Art. No.: CD001298. 45. Metwally M, Cheong YC, Horne AW. Surgical treatment of fibroids for subfertility. Cochrane Database Syst Rev 2012, Issue 11. Art. No.: CD003857. 46. Meuleman T, Schreinemacher MH, van Goor H, Bakkum EA, Dörr PJ. Adhesion awareness: a nationwide survey of gynecologists. Eur J Obstet Gynecol Reprod Biol 2013; 169(2): 353-359.

231 Hoofdstuk 8

47. Mol BWJ, Collins JA, Burrows EA, van der Veen F, Bossuyt PMM. Comparison of hysterosalpingography and laparoscopy in predicting fertility outcome. Hum Reprod 1999; 44(5): 1237-1242. 48. Mol BWJ, Bonsel GJ, Collins JA, Wiegerinck MAHM, van der Veen F, Bossuyt PMM. Cost-effectiveness of in vitro fertilization and embryo transfer. Fertil Steril 2000; 73(4): 748-754. 49. Moseley JB, O'Malley K, Petersen NJ, Menke TJ, Brody BA, Kuykendall DH, Hollingsworth JC, Ashton CM, Wray NP. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med 2002; 347: 81-88. 50. Mourad S, Hermens RPMG, Liefers J, Akkermans RP, Zielhuis GA, Adang E, Grol RPTM, Nelen WLDM, Kremer JAM. A multi-faceted strategy to improve the use of national fertility guidelines; a cluster- randomized controlled trial. Hum Reprod 2011; 26(4): 817-826. 51. Nahuis MJ, Kose N, Bayram N, van Dessel HJHM, Braat DDM, Hamilton CJCM, Hompes PGA, Bossuyt PM, Mol BWJ, van der Veen F, van Wely M. Long-term outcomes in women with polycystic ovary syndrome initially randomized to receive laparoscopic electrocautery of the ovaries or ovulation induction with gonadotrophins. Hum Reprod 2011; 26(7): 1899-1904. 52. Nahuis MJ, Oude Lohuis E, Kose N, Bayram N, Hompes P, Oosterhuis GJE, Kaaijk EM, Cohlen BJ, Bossuyt PMM, van der Veen F, Mol BWJ, van Wely M. Long-term follow-up of laparoscopic electrocautery of the ovaries versus ovulation induction with recombinant FSH in clomiphene citrate-resistant women with polycystic ovary syndrome: an economic evaluation. Hum Reprod 2012; 27(12): 3577-3582. 53. Nastri CO, Gibreel A, Raine-Fenning N, Maheshwari A, Ferriani RA, Bhattacharya S, Martins WP. Endometrial injury in women undergoing assisted reproductive techniques. Cochrane Database Syst Rev 2012, Issue 7. Art. No.: CD009517. 54. National Institute for Clinical Excellence (NICE). Fertility: assessment and treatment for people with fertility problems. 2004. 55. Nawroth F, Foth D, Schmidt T, Römer T. Results of a prospective comparative study of transvaginal hydro laparoscopy and chromo laparoscopy in the diagnostics of infertility. Gynecol Obstet Invest 2001; 52: 184-188. 56. Oude Rengerik K, Thangaratinam S, Barnfield G, Suter K, Horvath AR, Walczak J, Welminska A, Weinbrenner S, Meyerrose B, Arvanitis TN, Ónody R, Zanrei G, Kunz R,Arditi C, Burnand B, Gee H, Khan KS, Mol BWJ. How can we teach EBM in clinical practice?

232 Samenvatting proefschrift en discussie

An analysis of barriers to implementation of on-the-job EBM teaching and learning. Medical Teacher 2011; 33: e125-e130. 57. Pandian Z, Akande VA, Harrild K, Bhattacharya S. Surgery for tubal infertility. Cochrane Database Syst Rev 2008, Issue 3. Art. No.: CD006415. 58. Pandian Z, Marjoribanks J, Ozturk O, Serour G, Bhattacharya S. Number of embryos for transfer following in-vitro fertilization or intra-cytoplasmic sperm injection. Cochrane Database Syst Rev 2013, Issue 7. Art. No.: CD003416. 59. Paria BC, Reese J, Sanjoy KD, Dey SK. Deciphering the cross-talk of implantation: advances and challenges. Science 2002; 296: 2185- 2188. 60. Potdar N, Gelbaya T, Nardo LG. Endometrial injury to overcome recurrent embryo implantation failure: a systematic review and meta- analysis. Reprod Biomed Online 2012; 25: 561-571. 61. The Practice Committee of the American Society for Reproductive Medicine. Optimal evaluation of the infertile female. Fertil Steril 2006; 86(Suppl4):S264-267. 62. The Practice Committee of the American Society for Reproductive Medicine. Committee opinion: role of tubal surgery in the era of assisted reproductive technology. Fertil Steril 2012; 97: 539-545. 63. Pritts EA, Parker WH, Olive DL. Fibroids and infertility: an updated systematic review of the evidence. Fertil Steril 2009; 91(4): 1215-23. 64. Pundir J, Pundir V, Omanwa K, Khalaf Y, El-Toukhy T. Hysteroscopy prior to the first IVF cycle: A systematic review and meta-analysis. RBM Online 2013; 28(2): 151-161. 65. Sabatini L, Davis C. The management of hydrosalpinges: tubal surgery or salpingectomy? Curr Opin Obstet Gynecol 2005; 17: 323– 328. 66. Saturni S, Bellini F, Braido F, Paggiaro P, Sanduzzi A, Scichilone N, Santus PA, Morandi L, Papi A. Randomized controlled trials and real life studies. Approaches and methodologies: a clinical point of view. Pulm Pharmacol Ther 2014; 27(2):129-138. 67. Twisk M, Mastenbroek S, van Wely M, Heineman MJ, Van der Veen F, Repping S. Preimplantation genetic screening for abnormal number of chromosomes (aneuploidies) in vitro fertilization or intracytoplasmic sperm injection. Cochrane Database Syst Rev 2006, Issue 1. Art. No.: CD005291. 68. Van Tetering EAA, Bongers MY, Wiegerinck MAHM, Mol BWJ, Koks CAM. Prognostic capacity of transvaginal hydro laparoscopy to predict spontaneous pregnancy. Hum Reprod 2007; 22(4): 1091-1094.

233 Hoofdstuk 8

69. van Wely M, Bayram N, Bossuyt PMM, van der Veen F. Laparoscopic electrocautery of the ovaries versus recombinant FSH in clomiphene citrate-resistant polycystic ovary syndrome. Impact on women’s health-related quality of life. Hum Reprod 2004; 19(10): 2244-2250. 70. van Wely M, Bayram N, van der Veen F, Bossuyt PMM. An economic comparison of a laparoscopic electrocautery strategy and ovulation induction with recombinant FSH in women with clomiphene-resistant polycystic ovary syndrome. Hum Reprod 2004; 19(8): 1741-1745. 71. Verhoeve HR, Coppus SFPJ, van der Steeg JW, Steures P, Hompes PGA, Bourdrez P, Bossuyt PMM, van der Veen F, Mol BWJ, for the Collaborative Effort on the Clinical Evaluation in Reproductive Medicine. Hum Reprod 2011; 26(1): 134-142. 72. Verhoeve HR, Moolenaar LM, Hompes P, van der Veen F, Mol BWJ. Cost-effectiveness of tubal patency tests. BJOG 2013; 120: 583-593. 73. Zhou L, Li R, Wang R, Huang H, Zhong K. Local injury to the endometrium in controlled ovarian hyperstimulation cycles improves implantation rates. Fertil Steril 2008; 89: 1166-1176.

(*) Reference to Cochrane AL. Effectiveness and Efficiency. Random Reflections on Health Services. London: Nuffield Provincial Hospitals Trust, 1972. (Reprinted in 1999 for Nuffield Trust by the Royal Society of Medicine Press, London).

(**) Reference to Winston Spencer Churchill, 1942, 10 November. Mansion House, London.

234

Appendices

Abbreviations Co-authors and affiliations List of publications PhD Portfolio Dankwoord Curriculum vitae

235

Abbreviations

ABBREVIATIONS

ACP auto-cross-linked polysaccharide AFS American Fertility Society ASRM American Society for Reproductive Medicine ART assisted reproductive techniques/technology CCT clinical controlled trial CDSR the Cochrane Database of Systematic Reviews CENTRAL the Cochrane Central Register of Controlled Trials CI confidence interval CINAHL the Cumulative Index to Nursing & Allied Health CMC carboxymethylcellulose COH controlled ovarian hyperstimulation CONSORT Consolidated standards of reporting trials CPR cumulative pregnancy rate DARE the Database of Abstracts of Reviews of Effects EBM Evidence-based medicine EMBASE Excerpta Medica dataBASE ESGE European Society for Gynecological Endoscopy ESHRE European Society for Human Reproduction and Embryology FRR fecundity rate ratio FSH follicle stimulating hormone GIS gel instillation sonography GRADE Grading of Recommendations Assessment, Development and Evaluation HA hyaluronic acid HA-CMC hyaluronic acid-carboxymethylcellulose HFEA Human Fertilisation & Embryology Authority HSG hysterosalpingography HyCoSy hysterosalpingo contrast sonography ICMART International Committee for Monitoring Assisted Reproductive Technology ICSI intracytoplasmic sperm injection IPD individual patient data ISRCTN International Standard Randomized Controlled Trial Number ITT intention-to-treat IUA intrauterine adhesion IUD intrauterine device IUI intrauterine insemination IV inverse variance IVF in vitro fertilization LOD laparoscopic ovarian diathermy

237 Abbreviations

MAR medically assisted reproduction MD mean difference MDSG Menstrual Disorders and Subfertility Group MEDLINE Medical Literature Analysis and Retrieval System Online MeSH Medical subject headings MH Mantel Haenszel MICD minimally important clinical difference MOOSE Meta-analysis of Observational Studies in Epidemiology NICE National Institute for Clinical Excellence NNTB number needed to treat to benefit OR odds ratio PCOS polycystic ovary syndrome PICO(T) population-intervention-comparison-outcome-(time) PRISMA Preferred Reporting Items for Systematic Reviews and Meta- Analyses PUBMED Public/Publisher MEDLINE QUOROM Quality of Reporting of Meta-analyses RCOG Royal College of Obstetricians and Gynaecologists RCT randomized controlled trial RR risk ratio SART Society for Assisted Reproductive Technologies SD standard deviation SIS saline infusion sonography SR systematic review THL transvaginal hydro laparoscopy TTP time-to-pregnancy TVUS transvaginal ultrasonography US ultrasonography WHO World Health Organization

238 List of co-authors

LIST OF CO-AUTHORS AND THEIR AFFILIATIONS

Prof. Dr. S. Bhattacharya, Department of Obstetrics and Gynecology, Aberdeen Maternity Hospital, Aberdeen, United Kingdom.

Prof. Dr. F. J.M. Broekmans, Division of Reproductive Medicine, Department of Obstetrics and Gynecology, UMC, Utrecht, the Netherlands.

Prof. Dr. T. M. D’Hooghe, Leuven University Fertility Centre, Department of Obstetrics and Gynecology, University Hospital Gasthuisberg, Leuven, Belgium.

Prof. Dr. V. Gomel, Department of Obstetrics and Gynecology, University of British Columbia, Vancouver BC, Canada.

Dr. J. C. Kasius, Division of Reproductive Medicine, Department of Obstetrics and Gynecology, UMC, Utrecht, the Netherlands.

Prof. Dr. C. Mathieu, Department of Endocrinology, University Hospital Gasthuisberg, Leuven, Belgium.

Prof. Dr. B.W.J. Mol, The Robinson Institute, School of and Reproductive Health University of Adelaide, 5000 SA Australia.

Dr. C. Panayotidis, Department of Obstetrics and Gynecology, Royal Oldham Hospital, University of Manchester, United Kingdom.

Dr. P. Puttemans, the Leuven Institute for Fertility and Embryology, Leuven, Belgium.

Prof. Dr. C. Siristatidis, Assisted Reproduction Unit, 3rd Department of Obstetrics and , “Attikon” University Hospital, Athens, Greece.

Prof. Dr. B. van Herendael, Endoscopic Training Centre Antwerp, Antwerpen, Belgium.

Prof. Dr. S. Weyers, Universitaire Vrouwenkliniek, Ghent University Hospital, Ghent, Belgium.

239

List of publications

LIST OF PUBLICATIONS

Prostacyclin and thromboxane in pregnancy. F.A. Van Assche, B. Spitz, M. Hanssens, R. Pynenborg, J. Bosteels. Current Obstetrics and Gynaecology 1992; 2: 27-30.

Laparoscopic management of small bowel strangulation in a first trimester pregnant patient. J. Bosteels, Ph. Delattin, J.P. Van Boxelaer, M. De Bruyn, P. Vermylen. Gynecological Endoscopy 1996; 5(6): 355 – 356.

Measurement of ȕ - hCG in cul-de-sac fluid versus serum in the rapid detection of ectopic pregnancy. J. Bosteels, Ph. Delattin, J.P. Van Boxelaer, M. De Bruyn, J. Moerman, P. De Vos, P. Vermylen. Gynecological Endoscopy 1998; 7(1): 9 -12.

Sexual dysfunction in Women with type 1 diabetes. P. Enzlin, C. Mathieu, A. Van den Bruel, J. Bosteels, D. Vanderschueren, K. Demyttenaere. Diabetes Care 2002; 25(4): 672 – 677.

Diaphragmatic Endometriosis: Diagnosis and Surgical Management – A case report. A.M. Wolthuis, C. Aelvoet, J. Bosteels, J.P. Van Ryckel. Acta chirurgica Belgica 2003; 103: 519 – 520.

Seksuele problemen bij vrouwen met diabetes mellitus type 1 en bij een vergelijkbare controlegroep. P. Enzlin, C. Mathieu, A. Van den Bruel, J. Bosteels, D. Vanderschueren, K. Demyttenaere. Tijdschrift voor Geneeskunde 2002; 58(23): 1589 – 1597.

Optimisation of Prenatal Group B Streptococcal Screening H. Blanckaert, J. Frans, J. Bosteels, M. Hanssens, J. Verhaegen. European Journal of Clinical Microbiology and Infectious Diseases 2003; 22: 619 – 621.

Peripartumcardiomyopathie: Een case report en literatuuroverzicht. T. Mesens, Ph. Delattin, J. Bosteels. Tijdschrift voor Geneeskunde 2007; 63(1): 535-540.

241 List of publications

Spontaneous midgestation abortion associated with Bacteroides fragilis: a case report. J. Baekelandt, M. Crombach, G. Donders, J. Bosteels. Infectious Diseases in Obstetrics and Gynecology 2005; 13 (4):241-243.

The position of diagnostic laparoscopy in current fertility practice. Bosteels J., Van Herendael B., Weyers S., D’Hooghe T. Human Reproduction Update 2007; 13(5):477-485*.

Intrauterine adhesions: Has there been progress in understanding and treatment over the last twenty years? Panayotidis C., Weyers S., Bosteels J., Van Herendael B. Gynecological Endoscopy 2009; 6 (3): 197-211.

A case of didelphic uterus and blind hemivagina with renal dysplasia and ectopic ureter presenting with vulvodynia and recurrent fever. E. Stevens, J. Baekelandt, L. Lemmens, E. Dufraimont, A. Valcke, M. De Bruyn, J. Bosteels. Gynecological Surgery 2010; 7: 279-283.

The effectiveness of hysteroscopy in improving pregnancy rates in subfertile women without other gynaecological symptoms: a systematic review. Bosteels J. , Weyers S. , Puttemans P. , Panayotidis C. , Van Herendael B., Gomel V. , Mol B.W., Mathieu C., D’Hooghe T. Human Reproduction Update 2010; 16(1):1-11*.

Fibromen en infertiliteit. T. Marynissen, D.Timmerman, J.Bosteels, J. Baekelandt. Tijdschrift voor geneeskunde 2010; 66(12):571-581.

The effectiveness of reproductive surgery in the treatment of female infertility: facts, views and vision. J. Bosteels, S. Weyers, C. Mathieu, B.W. Mol, T. D’Hooghe. Facts, views and vision in Obstetrics and Gynaecology 2010; 2(4): 232- 252*.

Is reproductieve chirurgie effectief in de behandeling van subfertiliteit? J.Bosteels, S.Weyers, T. Dhooghe. Tijdschrift voor Geneeskunde 2011; 67(23): 1132-1138.

242 List of publications

Progress in evidence- based reproductive surgery. J. Bosteels, S. Weyers, C. Siristatidis, S. Batthacharya, T. D’Hooghe. Facts, views and visions in Obstetrics and Gynaecology 2011; 3(4): 238- 44*.

Clinical Practice Guideline- Laparoscopic Entry. ESGE- European Society of Gynecological Endoscopy Bakkum E.A. , Bongers M., Bosteels J., Gorostidi Pulgar M., Jansen F.W., Pérez Medina T., Thurkow A., Varma R., van Herendael B., Weyers S. http://www.esge.org/media/files/esge-clinical-practice-guideline.pdf.

Hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities. Bosteels J., Kasius J.C., Weyers S., Broekmans F.J., Mol B.W.J., D’Hooghe T.M. Cochrane Database of Systematic Reviews 2013, Issue 1. Art. No.:CD009461*.

Handboek Voortplantingsgeneeskunde. Hoofdstuk 17. Reproductieve heelkunde. J. Bosteels, S. Weyers, T. D’Hooghe. Redactie: dr Annemiek Nap, prof. dr Thomas D’Hooghe en prof.dr Jan Kremer. 2013, De Tijdstroom, Utrecht.

Treating suspected uterine cavity abnormalities by hysteroscopy to improve reproductive outcome in women with unexplained infertility or prior to IUI, IVF or ICSI. Bosteels J. , Kasius J.C. , Weyers S. , Broekmans F. J., Mol B.W.J., D’Hooghe T.M. Gynecological Surgery 2013; 10: 165-167.

Diagnostic endoscopy and infertility. C. Tomassetti, C. Meuleman, J. Bosteels, T. M. D’Hooghe. Médecine de la reproduction, gynécologie et endocrinologie 2013; 15 (3): 239-243.

Anti-adhesion barrier gels following operative hysteroscopy for treating female subfertility: a systematic review and meta-analysis. J. Bosteels, S. Weyers, Ben W.J. Mol, T.M. D’Hooghe. Gynecological Surgery 2014; 11: 113-127*.

243 List of publications

Anti-adhesion therapy following operative hysteroscopy for treating female subfertility. Bosteels J. , Kasius J.C. , Weyers S. , Broekmans F. J., Mol B.W.J., D’Hooghe T.M. The Cochrane Library, Cochrane Database of Systematic Reviews 2014, Issue 5. Art. No.:CD011110*.

Complications of hysteroscopic surgery-how to avoid and manage. J.Bosteels, B. van Herendael. Submitted in April 2014 as chapter 9 in: Minimally invasive gynecologic surgery: a practical and evidence-based guide. Editors: Einarsson J, Wattiez A.

The risk of uterine rupture after myomectomy: a systematic review of the literature and meta-analysis. J. Claeys, I. Hellendoorn, T. Hamerlynck, J. Bosteels, S. Weyers. Gynecological Surgery 2014; URL: http://dx.doi.org/10.1007/s10397- 014-0842-8.

Quality assessment in surgery: where do we stand now and where should we be heading? S. Weyers, S. Van Calenbergh, Y. Van Nieuwenhove, G. Mestdagh, M. Coppens, J. Bosteels. Gynecological Surgery 2014; 11: 89-95.

Prevention of multiple pregnancies after IUI. D. de Neubourg, J. Bosteels, T. D’Hooghe. Chapter 27 in: Intra-Uterine Insemination: Evidence-Based Guidelines for Daily Practice. Edited by B. Cohlen and W. Ombelet. CRC Press, Taylor&Francis Group, 2014.

* presented in this thesis.

244 PhD portfolio

PHD PORTFOLIO

1. 1.PhD PhD Training Trainin g

General courses(*) year ECTS

Advanced topics in clinical epidemiology 2013-14 0.98 Clinical epidemiology 2010 0.6 2009 0.7 Systematic reviews 2008 0.2 (*) These courses were not followed at the AMC Graduate School

Seminars, workshops and master classes

Pre-congress course Reproductive Surgery 2014 0.2 Workshop meta-analysis 19th Cochrane Coll. 2011 0.2 Pre-congress course Reproductive Surgery 2011 0.2 Transvaginal Endoscopy Course, Leuven 2009 0.5 Royal Academy of Medicine 2007 0.07 Workshop EBM Francqui Chair 2007 0.2

Oral presentations

• Surgery in reproductive medicine: a meta-analysis 2014 31 st BSRM meeting, Mechelen 0.02 • De endoscopische chirurgie in de fertiliteitskliniek 2014 AZ Sint Jan, Brugge 0.03 • Should all intrauterine pathology be removed in women of child bearing age? 2013 ESGE, Berlin 0.7 • Hoe moeten we omgaan met complicaties in de laparoscopische chirurgie? 2013 19de Doelencongres, Rotterdam 0.7

245 PhD portfolio

• SWOT analyse van de ambulante hysteroscopische chirurgie in België. 2013 19de Doelencongres, Rotterdam 0.7 • Verbetert stoppen met roken de kans op succesvolle zwangerschap? 2013 19de Doelencongres, Rotterdam 0.7 • Meta-analyse fertiliteitschirurgie 2012 Dienst Gynaecologie/Verloskunde UZ Brussel 0.03 • Is reproductieve chirurgie effectief in behandeling van infertiliteit? 2012 Dienst reproductieve geneeskunde, UZ Gent 0.03 • The impact of food supplements and antioxidants- impact on ART results 2012 BSRM, Spa 0.2 • The effectiveness of hysteroscopic surgery in treating female subfertility 2012 Indian Fertillity Society, Jodhpur, India 0.5 • Diagnostic laparoscopy in the fertility exploration 2012 Indian Fertillity Society, Jodhpur, India 0.5 • Cochrane in gynaecologie en verloskunde 2011 VVOG vergadering, Genk 0.1 • Wat is de waarde van diagnostische en operatieve hysteroscopie/ laparoscopie in de reproductieve geneeskunde? 2011 18de Doelencongres, Rotterdam 0.7 • The role of hysteroscopy in the diagnosis and treatment of infertility 2011 ESGE, London, United Kingdom 0.7 • The effectiveness of reproductive surgery in the treatment of female infertility 2011 ESGE, London, United Kingdom 0.7 • The evidence-based use of hysteroscopy and laparoscopy in the fertility setting 2010 Hôpital Universitaire Saint Pierre, Brussel 0.03

246 Phd portfolio

• Nut van diagnostische en operatieve hysteroscopie en laparoscopie bij infertiliteit 2010 Pentalfa KU Leuven 0.1 • The documentation and registration of complications in endoscopy 2009 ESGE, Florence, Italy 0.7 • A European Consensus Practice Guideline on techniques of laparoscopic entry 2009 ESGE, Florence, Italy 0.7 • The role of hysteroscopy in current reproductive medicine 2009 ESGE, Florence, Italy 0.7 • The role of laparoscopy in the evaluation of the infertile couple 2008 1st World Congress on Office Hysteroscopy, Bari, Italy 0.5 • An unusual case of infertility and its hysteroscopic treatment 2008 1st World Congress on Office Hysteroscopy, Bari, Italy 0.5 • Registration of complications in minimally invasive procedures 2008 ESGE, Amsterdam 0.7 • A European Consensus Guideline on Laparoscopic Entry 2008 ESGE, Amsterdam 0.7 • The place of surgery in reproductive medicine 2007 Leuven University Fertility Centre, KU Leuven 0.1 • Plaats van de diagnostische laparoscopie 2006 VVOG vergadering, Brussel 0.1 • The position of diagnostic laparoscopy in current fertility practice 2006 ESGE, Strasbourg, France 0.7

247 PhD portfolio

Posters

• The effectiveness of anti-adhesion therapy following operative hysteroscopy for treating female subfertility: a systematic review and meta-analysis 2013 ESHRE, London, United Kingdom 1.0 • Is reproductive surgery effective in the treatment of female subfertility: an evidence-based approach? 2011 ESHRE, Stockholm, Sweden 1.0 • The position of hysteroscopy in fertility practice 2008 ESGE congress, Amsterdam 0.7 (Inter) national conferences

ESGE 2006-2013 3.5 ESHRE 2011-2014 3.0 VVOG 2006-2014 3.5 BSRM 2008-2013 1.0 ISGE 2008 0.5 De Doelen 2007-2013 2.8 Gynaecongres 2008 0.5 Cochrane Colloquium 2011 0.75

2. Teaching

EBM terugkomdagen 2010-2014 1.7 EBM driedaagse cursus 2010-2014 1.5 Cursus Systematic review 2014 0.5 Postgraduate Course ESGE 2012-2013 0.3 EBM in 5 stappen 2010-2013 0.7 Short Course Clinical Research/EBM 2014 0.14 ETCA Training courses for ASO’s 2007-2013 2.3 VVOG Endoscopy Training courses 2008-2013 2.8

248 Dankwoord

DANKWOORD Ik wil vele mensen van harte danken voor hun bijdrage aan mijn groeiproces. Dit proefschrift is geen einde of doel op zich, maar het begin van een volgend avontuur met nieuwe uitdagingen. Rust roest… Ik dank de Rector Magnificus Professor dr. D.C. Van den Boom, de decaan, Professor dr. M.M. Levi en Professor dr. W.M. Wiersinga om mij toe te laten tot de promotie aan de Faculteit der Geneeskunde van de Universiteit van Amsterdam. Welgemeende dank aan de leden van de promotiecommissie Professor dr. Fulco van der Veen, Professor dr. Patrick M.M. Bossuyt, Professor dr. Petra de Sutter, dr. W.M. Ankum en dr. Anne Timmermans. Oprechte dank aan Professor dr. Frans André Van Assche, al mijn Professoren-opleiders en collega’s assistenten tijdens mijn opleiding tot specialist aan de KU Leuven. Bijzondere dank aan Professor dr. Myriam Hanssens: je zette niet alleen onze zoon gezond en wel op de wereld maar je gaf mij een gouden raad mee op de voorpagina van een boek over prostaglandines. Sarwana Nuradi, mijn eerste collega in ‘donkere’ tijden, jij wordt speciaal vernoemd. Professor Kamiel van den Berghe (†), mijn eerste mentor, een groot man met een warm hart en een grote integriteit: je afscheid was veel te vroeg. Jos Caudron (†) en Lieven Londers (†): jullie waren inspirerende voorbeelden. I thank Professor Emeritus dr. John A. Collins from the Mc Master University, Hamilton, Ontario, Canada and Professor Emeritus dr. Victor Gomel, the University of British Columbia, Vancouver, British Columbia, Canada for their advice on their areas of expertise and their wisdom: EBM and methodology (John) and reproductive surgery (Victor). I was blessed that our ways have crossed. Professor dr. Marc Keirse, presently School of Medicine at Flinders University, Adelaide, Southern Australia. Thank you for the inspiring Francqui Chair lectures on EBM and for being honest. I tore up the draft of my manuscript as suggested, rewrote it and it got accepted. Your advice has proven to be effective beyond chance. I thank Professor dr. Jan J. Brosens, Biomedical Research Unit in Reproductive Health, University of Warwick, Coventry, United Kingdom for assistance concerning the ‘black box’ of implantation of the human embryo. Dank aan Bert Aertgeerts, Filip Cools, Dirk Ramaekers, Trudy Bekkering, Stijn Van de Velde, Patrick Haentjens, Manu Simons, Hans van Brabandt, Patrick van Krunkelsven, Bart Geurden, Jef Adriaensens, Martine Goossens, Elizabeth Bosselaers en mijn overige collega’s en vrienden van CEBAM aan het Academisch Centrum voor Huisartsgeneeskunde te Leuven. Dank aan Professor dr. Walter Sermeus, Centrum voor Ziekenhuis- en Verplegingswetenschap te Leuven.

249 Dankwoord

Many thanks to Professor dr. Cindy Farquhar, Auckland, New Zealand, coordinating editor of the Cochrane Menstrual Disorders and Subfertility Group: writing a Cochrane systematic review is indeed not for the faint hearted! Many thanks to Ms. Helen Nagels, managing editor and Ms. Marion Showell, trials search coordinator at Auckland, New Zealand. James Duffy, Khalid Khan, Jane Daniels, Harry Siristatidis, Artin Ternamian, Costas Panayotidis, Siladitya Bhattacharya, Stephan Gordts, Rudi Campo, Patrick Puttemans, Siegfried Heylen, Ivo Brosens, Tarek El- Toukhy, Mark Possover, Arnaud Wattiez, Philippe Koninckx,.... Your impact upon my career as friends, researchers and colleagues is ever lasting. Paul Vermylen (†), Marcel De Bruyn, Jean Pierre Van Boxelaer, Philippe Delattin, Carine De Rop, Els Dufraimont, Els Lannoey, Jan Baekelandt, Sofie Pelckmans, Lore Lannoo en Nele Roggen, de collega’s van mijn maatschap. Dit was nooit mogelijk geweest zonder jullie begrip. Dank aan al onze assistenten in opleiding voor de jarenlange intellectuele kruisbestuiving. Dank aan Renilde Bastaens en alle Imelda vroedvrouwen, aan Monique Van Hiel en het OK team, aan Marleen van Dessel en de verpleging van Heelkunde 5. Dank aan de directie en de hele medische staf van Imeldaziekenhuis. Dank aan alle collega’s klinische biologie in het bijzonder voor ons bondgenootschap in het Zorgprogramma A Reproductieve Geneeskunde. Dank aan Sofie en Christel De Wit, Sarah Collin, Petra Praet, Marleen Parisis, Krista Grootaert, Katja Reyniers en Nicole Saenen voor secretariële ondersteuning in Vlaanderen. Cootje Kusters, Helene Boeren, Shirley Abels, Marjan du Prie, Ellis Hanko en Bettina Batista voor de Nederlandse ondersteuning. Dank aan alle bestuursleden en collega’s van de Vlaamse Vereniging voor Obstetrie en Gynaecologie: Prof. dr. Ignace Vergote, dr. Johan van Wiemeersch, dr. Geert Debruyne, Prof. dr. Willem Ombelet, dr. Piet Hinoul, Prof. dr. Jan Gerris, Prof. dr. Steven Weyers, Prof. dr. Jean Jacques Amy, Prof. dr. Marc Dhont, Prof. Dr. Marleen Temmerman, dr. Bart Bollen, dr. Koen Traen, dr. Guy Orye, dr. Dirk Smet, Prof. dr. Michel Degueldre, Prof. dr. Herman Tournaye, dr. Guy Verhulst, dr. Luc Debaene, dr Benoit Rombaut, dr. Geert Page, dr. Herwig Trappeniers, dr. Geert Vlaemynck, Prof. dr. Eric De Jonghe, Prof. dr. Yves Jacquemyn, Prof. dr. Paul Defoort, Prof. dr. Paul Devroey, Prof. dr. Patrick Neven, dr. Patrick Berteloot, Prof. dr. Jan Deprest, Prof. dr. Dirk Timmerman, Prof. dr. Herman Depypere, dr. Greet Mestdagh, dr. Geertrui Meganck, dr. Willem Verpoest… en zovele anderen. Ik heb enige schroom om de lange lijst verder af te gaan uit vrees om iemand te vergeten. Bijzondere dank

250 Dankwoord aan de sterke VVOG vrouwen: Mireille Merckx, Carine De Rop, Diane De Neubourg, Carla Tommassetti, Elke Sleurs, Anke Thaens en Anne Loccufier. Dank aan ‘the Dutch connection’: Hans Brölmann, Erika Bakkum, Frank Willem Jansen, Andreas Thurkow, Marlies Bongers, Matthé Burger, Ben Cohlen, Bas Veersema, Judith Huirne, Peggy Geomini, Petra Hajenius, Evert Slager, Hans Emmanuel, Minouche van Rumste… Niets is toeval, het is ook door en voor jullie dat ik in Nederland kom promoveren. Dank aan Professor dr. Jan Deprest, Professor dr. Stephan Gordts, Professor dr. Steven Weyers, Professor dr. Frederic Amant en dr. Thierry Van den Bosch: het is fijn met en onder jullie te mogen werken als associate editor van Gynaecological Surgery. Professor dr. Bruno Van Herendael en Professor dr. Thomas D’Hooghe: jullie plaats in dit project is heel bijzonder. Bruno, zonder jouw aanbod om mee cursus te komen geven bij het ETCA had ik wellicht nooit de reviews over hysteroscopie geschreven. Je cruciale rol in dit proefschrift kan dan ook nooit genoeg worden benadrukt. Beste Tom, dank om mijn hand vast te willen houden bij mijn eerste stapjes in een ondertussen uit de hand gelopen hobby, het schrijven van wetenschappelijke publicaties. Jij staat steeds op de laatste plaats, althans van de auteurslijst van mijn belangrijkste publicaties: ingewijden weten echter heel goed wat dit inhoudt… Professor dr. Frank Broekmans en dr. Jenneke Cornelia ‘Cochrane’ Kasius. Beste Frank en Jenneke, het was een eer om onder jullie vleugels te mogen promoveren. Jullie zijn nu ‘gepromoveerd’ van mijn vaste Cochrane coauteurs naar mijn (co)-promotors. Ik wens jullie hiervoor te feliciteren. Een consistent robuuste vriendschap is echt wel het minste wat ik jullie terug kan geven. Beste Ben Willem, jij doet je familienaam alle eer aan. Ik kan mij namelijk niet meer herinneren waar en wanneer onze wegen elkaar hebben gekruist. Je invloed blijft echter duidelijk tastbaar. Dank om mij om te toveren van een grijze muis via een kritisch denkende promovendus naar een zelfzekere Cochrane auteur. Dank om mij, de Nestor van al je promovendi, steeds door dik en dun te willen verdedigen. Je bent een intellectuele reus met een onwrikbaar geloof in de waarde van een wetenschappelijk onderbouwde praktijkvoering omdat dit in het belang is van onze patiënten. Dank voor de geboden kansen. Ik besluit met de essentie van mijn zijn. Ik bedank mijn vader Herman Bosteels, mijn moeder Maria De Mulder (†) die ons vorig jaar te vroeg heeft verlaten en mijn zus Kathleen. Jullie waren samen dat warme nest waar ik alle kansen kreeg om mij te ontplooien. Bijzondere dank aan mijn grootmoeder Johanna Vinck (†), een intelligente vrouw die niet de kans

251 Dankwoord

heeft gekregen om verder te studeren. Jij hebt je kleine jongen steeds het juiste advies willen geven. Wim en Patricia, Kristien en Benoni, onze kleine ‘circle of trust’: jullie lief en leed is ook het onze en dat blijft zo. Dank aan de CEO van de familie, Trinette Pots, mijn schoonmoeder, een heel bijzondere vrouw die het werk thuis heel wat lichter maakt. Resten nog mijn twee oogappels. Chantal, je bent een ‘bangelijk’ echte geestesgenoot naast een uitzonderlijke arts en een gedreven topwetenschapper. Ik hoop nog lang samen met jou onderweg te zijn om creatief te blijven zowel professioneel als persoonlijk. Jeff, mijn zoon, mijn grootste en meest waarachtige ‘werkstuk’, van alles en iedereen behoort aan jou het laatste woord van dank. Ik draag daarom heel bijzonder de woorden van Winston Churchill aan het begin van dit proefschrift op aan jou: maak er je credo van in alles wat je doet en bent…

252 Curriculum vitae

CURRICULUM VITAE

Jan Bosteels werd op 2 april 1962 geboren in Dendermonde, België waar hij in 1980 zijn diploma middelbaar onderwijs aan het Heilige Maagdcollege behaalde in de afdeling Latijn-Wiskunde. Op 30 juni 1987 behaalde hij zijn artsendiploma met de grootste onderscheiding aan de Faculteit Geneeskunde van de KU Leuven. Van 1987 tot 1992 genoot hij opleiding tot specialist in de verloskunde-gynecologie aan de Universitaire Ziekenhuizen van de KU Leuven onder supervisie van Prof. Dr. Frans André Van Assche. Vanaf juli 1992 is hij verbonden als staflid van de dienst verloskunde-gynecologie van het Imeldaziekenhuis te Bonheiden met bijzondere interesse in de reproductieve geneeskunde en de endoscopische heelkunde. In 1994 startte hij samen met Dr. Jan Moerman het Zorgprogramma reproductieve geneeskunde A op. Van 1995 tot 2009 was hij lid van de Commissie Medische Ethiek van het Imeldaziekenhuis waarvan zetelend voorzitter van 2003 tot 2009. Hij is vanaf 2007 verbonden als tutor aan het ETCA, Endoscopic Training Centre Antwerp, geleid door Prof. Dr. Bruno van Herendael. Vanaf 2001 is hij actief binnen de Vlaamse vereniging voor Verloskunde en Gynaecologie (VVOG), eerst als bestuurslid werkgroep Reproductieve Geneeskunde en medeoprichter van de werkgroep endoscopie, vervolgens als lid van de raad van bestuur als secretaris (2007-2010), voorzitter van de werkgroep Algemene Gynaecologie (2010-2013) en ten slotte als voorzitter van de raad van bestuur vanaf maart 2013. Vanaf januari 2010 is hij deeltijds stafmedewerker van CEBAM, the Belgian Branch of the Dutch Cochrane Centre, waar hij mee verantwoordelijk is voor het organiseren van opleiding in evidence-based medicine (EBM) en als ‘Ad hoc’ externe validator voor CEBAM bij het Federaal Kenniscentrum (KCE), het Belgische analoog van NICE. Hij is vanaf 2012 Associate- Editor van Gynaecological Surgery, het ‘peer-reviewed’ vaktijdschrift van de European Society of Gynaecological Endoscopy (ESGE). Jan Bosteels is gehuwd met Chantal Mathieu. Zij zijn de trotse ouders van Jeff, die net als zijn beide ouders zijn weg wil vinden binnen de gezondheidszorg.

253 Stellingen behorende bij het proefschrift van Jan Bosteels

1. Het routinematig uitvoeren van een laparoscopie bij alle gevallen van onvruchtbaarheid is niet zinvol (dit proefschrift). 2. Op grond van de huidige kennis is er onvoldoende grond om een hysteroscopie uit te voeren bij alle vrouwen met subfertiliteit (dit proefschrift). 3. Omdat bij niet-effectieve behandelingen de veiligheid er niet toe doet, dient bij evaluatie van medische interventies eerst de effectiviteit en pas dan de veiligheid geëvalueerd te worden. 4. Trauma aan het endometrium zou misschien de kans kunnen verhogen op succesvolle implantatie van het embryo bij vrouwen die behandeld worden met Medisch Begeleide Voortplanting (Nastri et al, Cochrane Database Syst Rev 2012, Issue 7. Art. No.: CD009517). 5. Het hoge verbruik van IVF in België ( bijna 3000 cycli per miljoen inwoners) en de daarmee gepaard gaande hoge kosten (bijna 100 miljoen Euro geraamd in 2008) vereisen een maatschappelijke doorlichting (Berg Brigham et al, Hum Reprod 2013, 28 (3): 666-675). 6. Om de wetenschappelijke onderbouwing van de reproductieve chirurgie te versterken moet men ofwel chirurgen klinische epidemiologie aanleren ofwel klinische epidemiologen inschakelen op diensten waar klinisch-wetenschappelijk onderzoek wordt uitgevoerd (Madhok R and Handoll HHG; Bridgman S and Maffulli N. Randomized trials in surgery. Letters to the Editor of the BMJ. BMJ 2002; 325: 658-9). 7. Een ‘fee for service’ systeem kan leiden tot zowel een slechte coördinatie als een overconsumptie van zorg. 8. Een optimale zorgfinanciering hoort een effectiviteitgestuurde betalingscomponent (P4P) te bevatten. 9. “Efficiency is doing things right; effectiveness is doing the right things” (Peter F.Drucker). 10. “Absence of evidence is not evidence of absence” (Carl Sagan).