Excisional surgery versus ablative surgery for ovarian endometriomata: a Cochrane Review

Human Reproduction Vol.20, No.11 pp. 3000 3007, 2005 doi:10.1093/humrep/dei207 Excisional surgery versus ablative surgery for ovarian endometriomata: a Cochrane Review Roger Hart 1,3, Martha Hickey 1, Panos Maouris 1, William Buckett 2 and Ray Garry 1 1 School of Women s and Infants Health, University of Western Australia, King Edward Memorial Hospital, 374 Bagot Road, Subiaco, Perth, Western Australia, WA 6008, Australia and 2 Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynaecology, McGill University, Canada 3 To whom correspondence should be addressed. E-mail: rhart@obsgyn.uwa.edu.au This paper is based on a Cochrane review published in The Cochrane Library, issue 3 (see www.cochranelibrary.net for information) with permission from The Cochrane Collaboration and John Wiley & Sons. Cochrane reviews are regularly updated as new evidence emerges and in response to comments and criticisms, and The Cochrane Library should be consulted for the most recent version of the review BACKGROUND: The objective of this review was to determine which is the most effective technique for treating an ovarian endometrioma; excision or ablation. METHODS: A systematic review employing the principles of the Cochrane Menstrual Disorders and Subfertility Group was undertaken. No randomized studies of the management of endometriomata by laparotomy were found. Two randomized studies of the laparoscopic management of ovarian endometriomata of >3 cm in size were included. RESULTS: Laparoscopic excision of the cyst wall of the endometrioma was associated with a reduced rate of recurrence of the endometrioma [odds ratio (OR) 0.41, confidence interval (CI) 0.18 0.93], reduced requirement for further surgery (OR 0.21, CI 0.05 0.79), reduced recurrence rate of the symptoms of dysmenorrhoea (OR 0.15, CI 0.06 0.38), dyspareunia (OR 0.08, CI 0.01 0.51) and non-menstrual pelvic pain (OR 0.10, CI 0.02 0.56). It was also associated with a subsequently increased rate of spontaneous pregnancy in women who had documented prior subfertility (OR 5.21, CI 2.04 13.29). CONCLUSIONS: There is some evidence that excisional surgery for endometriomata provides for a more favourable outcome than drainage and ablation, with regard to the recurrence of the endometrioma, recurrence of symptoms and subsequent spontaneous pregnancy in women who were previously subfertile. Consequently this should be the favoured surgical approach. However, we found no data to indicate the best surgical approach in women planning to undergo assisted reproductive techniques. Key words: ablation/endometriomata/endometriosis/excision/meta-analysis Introduction Endometriosis is a common condition associated with pelvic pain and infertility in women. Endometriosis is defined as the presence of ectopic deposits of endometrial tissue usually, but not exclusively, limited to the pelvis, which may lead to infertility and pelvic pain. It may be present in up to 22% of asymptomatic women and up to 45% of women with pelvic pain (Ajossa et al., 1994; Farquhar, 2000). It has been believed for almost a century by the majority of academic opinion that endometriosis is a disease caused by shedding of menstrual endometrium and its dissemination throughout the pelvis (Cullen, 1920; Sampson, 1927a,b). Endometriomata are endometriotic deposits within the ovary. The origin of ovarian endometriomata is unknown; however, most authors believe that they result initially from a deposit of endometrium passed through the Fallopian tube (the transplantation theory: Sampson, 1927a), causing adherence of the ovary to the pelvic peritoneum and progressive invagination (folding inwards) of the ovary (Hughesdon, 1957; Brosens et al., 1994, 1996; Nisolle and Donnez, 1997). If this is true, an endometrioma would be a pseudocyst (false cyst), the wall of which is the inverted ovarian cortex (centre) and hence the removal of this cyst wall might involve removal of normal ovarian tissue, with possible adverse implications for future fertility (Vercellini et al., 2003). The primary indications for treatment of ovarian endometriomata are the symptoms of pelvic pain and dyspareunia (pain during or after sexual intercourse) (Vercellini, 1997) and may impair the outcome of fertility treatment (Yanushpolsky et al., 1998). There is also a small risk of malignant (cancerous) transformation (Nishida et al., 2000). The evidence suggests that, although medical treatment will result in a reduction in size of the endometrioma of up to 57%, the most effective approach to treatment is surgical (Farquhar and Sutton, 1998). Furthermore, if they are left, as with any ovarian cyst they have a risk of rupture and torsion. In recent years, laparoscopy has become the gold standard for the treatment of ovarian endometriotic cysts (Daniell et al., 1991; Donnez et al., 1996; Sutton et al., 1997; Yuen et al., 1997). When compared to traditional surgery by laparotomy, operative 3000 The Author 2005. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: journals.permissions@oupjournals.org

Cochrane review: excision or ablation for ovarian endometrioma? laparoscopy is associated with shorter hospital stay, faster patient recovery, decreased costs (Luciano et al., 1992) and lower incidence of de novo adhesion formation (Luciano et al., 1989, 1992; Lundorff et al., 1991; Operative Laparoscopy Study Group, 1991). The pregnancy rates, monthly fecundity and cyst recurrence rates after laparoscopic surgery are comparable (Adamson et al., 1992; Bateman et al., 1994; Catalano et al.,1996; Crosignani et al., 1996; Busacca et al., 1998; Milingos et al., 1999; Sawada et al., 1999; Canis et al., 2003). Laparoscopic surgery for endometriomata does carry a risk of conversion to laparotomy, and this is associated with the experience of the surgeon, the complexity of the surgery as well as patient factors, such as body mass index (Sokol et al., 2003). A prospective randomized trial of laparotomy or laparoscopy for the management of endometriomata demonstrated benefits to the patient with regard to less analgesic requirement, earlier discharge and post-operative recovery in those women operated by laparoscopic ovarian cystectomy (Mais et al., 1996). The procedure of drainage of the endometrioma alone is not recommended due to the high rate of recurrence (Vercellini et al., 1992; Donnez et al., 1996). However, the most effective method of laparoscopic surgery (excisional or ablative) remains controversial. Several alternative laparoscopic techniques have been described for the treatment of ovarian endometriomata: cyst wall laser vaporization (destruction by burning) preceded or not by medical therapy (Brosens et al., 1996; Donnez et al., 1996; Sutton et al., 1997), drainage and coagulation, and stripping (Reich and McGlynn, 1986; Martin, 1991; Canis et al., 1992). Excision of the cyst involves the opening of the endometrioma either with or without the use of electrosurgical or laser energy. The wall of the endometrioma is then excised or stripped away from the underlying cortex using a combination of scissors (or monopolar hook) and grasping forceps. Ablation of the endometrioma also involves opening and draining of the endometrioma or fenestration (making a window in the wall of the cyst), followed by the destruction of the cyst wall using either cutting or coagulating current, or using a form of laser energy (Jones and Sutton, 2000). Whatever the surgical modality employed to treat the cyst, a sample of the endometrioma must be sent for histological assessment as there is a need to confirm the clinical diagnosis, to exclude the presence of malignancy as the risk of malignant transformation of the cyst is 0.7% (Brinton et al., 1997; Nishida et al., 2000; Del Carmen et al., 2003). The recurrence rate of ovarian endometrioma following surgery has been previously reviewed by Vercellini et al. (2003). In this study, endometrioma recurrence was observed in 39 of 212 (18.4%) women treated with coagulation or laser vaporization and in 19 of 295 (6.4%) who underwent excision. This meta-analysis also reported that the pregnancy rates following surgery were 24 60%, with conflicting data regarding the most favourable surgical approach (Vercellini et al., 2003). The effect of an ovarian endometrioma on fertility is unclear, as it is rare to have a solitary endometrioma without surrounding pelvic endometriotic disease; indeed even minimal endometriosis is associated with subfertility (Pritts and Taylor, 2003) and its treatment has been demonstrated to improve fertility outcome (Jacobson et al., 2002). The presence of an endometrioma during IVF cycles has been associated with the need for greater ovarian stimulation and the production of fewer follicles following stimulation, suggesting that the endometrioma may be compromising ovarian function (Al-Azemi et al., 2000). However, the impact of endometriomata on the outcome of fertility treatment is controversial as some studies have failed to demonstrate an adverse effect of endometriomata on IVF outcome (Garcia-Velasco et al., 2004). It has been suggested that the technique of ovarian endometrioma capsule excision may lead to removal of normal ovarian tissue (Hachisuga and Kawarabayashi, 2002) and that the procedure of capsule ablation may lead to thermal (heat) damage to the underlying ovarian cortex and a risk of incomplete destruction of the endometriotic tissue (Maouris and Brett, 2002). Thus both interventions may lead ovarian cortical damage and hence a functional loss in the ovarian reserve. Hence there is a need to review the existing literature to determine the optimum surgical management of ovarian endometriomata with regard to the recurrence rate of the endometrioma, the pain relief afforded by the surgery, the effect on ovarian function, the effect on subsequent fertility and the impact on the patient s quality of life. The aim of this systematic review of the published literature is to critically appraise the literature describing these techniques with regard to the pain relief afforded by the surgery and the effect on subsequent fertility. Recurrence rates and the impact on the patient s quality of life will also be assessed. Materials and methods Our objective was to determine whether laparoscopic surgical excision or ablation is the optimum surgical management of ovarian endometriomata with respect to pain and fertility outcomes. The patient population were women with ovarian endometriomata who were undergoing surgery for the indication of pain or infertility, and endometriomata were defined as cysts of endometriosis within the ovary. Women with gynaecological cancer were excluded. Primary outcome measures were the relief from pelvic pain as measured by visual analogue scores (VAS) or dichotomous data and subsequent fertility, including biochemical and clinical pregnancy and the live birth rate, either spontaneous or as a result of fertility treatment. Secondary outcome measures were the recurrence of endometriomata, conversion from planned laparoscopic procedure to laparotomy, ovarian function as measured by changes in FSH or onset of menopausal symptoms after surgery and quality of life as measured by patient satisfaction or objective quality of life scales. All randomized controlled trials comparing excision and ablation of ovarian endometrioma were reviewed. Non-randomized controlled trials and quasi-randomized trials were excluded. Randomized trials comparing surgical with expectant management of endometriomata were also sought. The Cochrane Menstrual Disorders & Subfertility Group s Specialised Register of controlled trials was searched for any trials that met the search criteria (searched November 15, 2004). The following electronic databases were searched using OVID Software: MEDLINE (1966 November 2004), EMBASE (1980 November 2004) and Biological Abstracts (1980 Nov 2004). The Cochrane Central Register of Controlled Trials (CENTRAL) on the current issue of the Cochrane Library was searched, November 15, 2004. Online databases of ongoing 3001

R.Hart et al. trials, the National Research Register (NRR) and Clinical Trials register were also searched in all fields using the following words: ovarian cyst, laparotomy, laparoscopy, ovarian surgery. The citation list of relevant publications, abstracts of scientific meetings and list of included studies were all checked through hand searching and >20 experts in the field of endoscopic surgery were contacted to identify further reported trials; Professors Arici, Beretta, Chapron, Donnez and Vercellini responded to the enquiry. Planned surgical excision (stripping) of endometriomata was compared with planned ablation of endometrioma capsule. Because of the risk of conversion from laparoscopy to laparotomy (open surgery), both laparotomy and laparoscopic approaches were included. Analysis of endometrioma ablation data were to be stratified according to the modality used. No language or other limitations were imposed. The authors attempted to identify unpublished data. Randomized controlled trials addressing these secondary outcome measures were also to be included. Study selection All eligible studies were assessed for their methodological quality and relevance to the review objectives. Study selection was undertaken by three reviewers (R.H., P.M. and M.H.). Reviewers extracted data independently and assessed whether the studies met the inclusion criteria. Any discrepancies were resolved by a co-reviewer (R.G.). Further information was sought from the authors where papers contained insufficient information to make a decision about eligibility. The quality of allocation concealment was graded as adequate (A), unclear (B), or inadequate (C), following the detailed descriptions of these categories provided by the Menstrual Disorders and Subfertility Review Group. Table I. Characteristics of included studies No randomized studies of ovarian cyst excision versus ablation of the endometrioma capsule by a laparotomy approach were found. Laparoscopic ovarian cyst excision versus drainage of the endometrioma and ablation of cyst wall was compared in two studies (Beretta et al., 1998; Alborzi et al., 2004) (see Table I); excluded studies are listed in Table II. One trial was performed at two centres, although a sole operator performed in both centres (Alborzi et al., 2004). There were two prospective randomized controlled trials of the laparoscopic approach where the method of randomization was at the time of surgery by computer randomization, after surgery both patient and surgeon were aware of the procedure performed (Beretta et al., 1998; Alborzi et al., 2004). In both randomized controlled trials of the laparoscopic approach the indication for surgery was the presence of an endometrioma of >3 cm in size as assessed by ultrasound examination. Patients were excluded if they had had previous operative intervention for endometriosis, hormonal or suppressive therapy for the preceding 6 months (Beretta et al., 1998; Alborzi et al., 2004). In one study patients were aged between 20 and 40 years of age (Beretta et al., 1998) and in the other study the mean age of participants was 28 years in both groups (Alborzi et al., 2004). Description and quality assessment of included studies Of the two studies that met the inclusion criteria, both used computergenerated randomization sequences (Beretta et al., 1998; Alborzi et al., 2004). Both of the two studies that met the inclusion criteria did not conceal the allocation of the patients after randomization (Beretta et al., 1998; Alborzi et al., 2004). Both studies performed the randomization at the time of operation (Beretta et al., 1998; Alborzi et al., 2004). In both studies the patients, operators and examining doctors at Study Methods Participants Interventions Outcomes Notes Allocation concealment Alborzi (2004) Beretta (1998) Prospective randomized controlled trial Prospective randomized controlled trial Women from two tertiary centres withan endometrioma 3 cm in diameter. Women were excluded if they had had previous surgery for endometriosis or had taken hormonal or suppressive therapy in the last 6 months Women aged 20 40 years with an endometrioma 3 cm in diameter. Women were excluded if they had had previous surgery for endometriosis or had taken hormonal or suppressive therapy in the last 6 months Excision of the endometrioma versus drainage and ablation of the endometrioma Excision of the endometrioma versus drainage and bipolar ablation of the endometrioma Excisional surgery offered advantage over drainage and ablation with respect to the recurrence of the endometrioma, recurrence of symptoms of pain and dysmenorrhoea, need for re-operation and spontaneous conception. There were no intra-operative conversions from a laparoscopic approach to a laparotomy. There were no drop-outs from the study Excisional surgery offered advantage over drainage and ablation with respect to the recurrence of the symptoms of dysmenorrhoea, deep dyspareunia, non-menstrual pelvic pain and spontaneous conception. There were no intra-operative conversions from a laparoscopic approach to a laparotomy. There were no drop-outs from the study. Other outcomes: there was no difference between the two surgical modalities performed with regard to operative time, blood loss and post-operative stay Additional information provided by first author. Although this was a multicentre study, the surgery was performed by the same surgeon in two separate sites. Power calculation: not stated. Histological examination of the ovarian cyst confirmed the presence of endometriosis in 100% of cases Additional information provided by third author. Histological examination of the ovarian cyst confirmed the presence of endometriosis in 89% of cases, there being no difference between the two groups A A 3002

Cochrane review: excision or ablation for ovarian endometrioma? Table II. Characteristics of excluded studies Study Fayez and Vogel (1991) Hemmings et al. (1998) Saleh and Tulandi (1999) Reason for exclusion Inadequate randomization Non-randomized Retrospective the time of follow-up were all aware of the operative procedure performed (Beretta et al., 1998; Alborzi et al., 2004). In both studies that met the inclusion criteria there were no drop-outs and all patients were accounted for (Beretta et al., 1998; Alborzi et al., 2004). One group included a power calculation in their publication; however, the power of this study to detect a difference in the rate of disease recurrence was only 20% and the power calculation was performed after completion of the study (Beretta et al., 1998). In both studies that met the inclusion criteria both groups were comparable. Patients in the excisional surgery group and the drainage and ablation group were similar with regard to the incidence and severity of pre-operative pelvic pain, size of the endometrioma, stage of endometriosis and age at the time of surgery (Beretta et al., 1998; Alborzi et al., 2004). Neither of the studies included in this review stated a funding or sponsorship source (Beretta et al., 1998; Alborzi et al., 2004). Additional information was sought from both lead authors of the two studies that met the inclusion criteria for assessment (Beretta et al., 1998; Alborzi et al., 2004). In one group the lead author responded and provided additional information (Alborzi et al., 2004); in the other group the third author responded and provided additional information (Beretta et al., 1998). Statistical analysis Statistical analysis was performed in accordance with the guidelines for statistical analysis developed by Menstrual Disorders and Subfertility Group. All trials were initially included in one analysis of surgical laparoscopy for endometriomata. Subgroup analysis by looking at the indication for ovarian endometrioma surgery (pain or infertility) was not possible for the papers meeting the inclusion criteria. Statistical heterogeneity between the results of different studies was examined by inspecting the scatter in the data points on the graphs and the overlap in their confidence intervals (CI). The outcomes were pooled statistically when no clinical heterogeneity was apparent. Continuous data were combined for meta-analysis. Using RevMan software, mean and SD were used to derive a weighted mean difference (WMD) with 95% CI and fixed effects model. A fixed effects model was used for calculations of summary estimates and their 95% CI. For categorical outcomes the numbers reporting an outcome to each group were related. Results for each study were expressed as an odds ratio (OR) with 95% CI and combined for meta-analysis with RevMan software. Results No randomized studies of ovarian cystectomy versus ablation of the endometrioma capsule performed at the time of laparotomy were found in the literature. The procedure of laparoscopic management of ovarian endometriomata was compared in two studies (Beretta et al., 1998; Alborzi et al., 2004). In these studies ovarian cystectomy, by the stripping and total excision of the capsule of the endometrioma by pulling two atraumatic grasping forceps in opposite directions, was compared with the procedure of tissue biopsy, drainage and subsequent ablation of endometrioma by bipolar coagulation of the capsule. Neither of the two randomized studies of the laparoscopic management of endometriomata employed adjuvants to surgery to reduce the incidence of adhesions (Beretta et al., 1998; Alborzi et al., 2004). In one study follow-up was 2 years (Alborzi et al., 2004) and the other study followed patients for up to 26 months, median follow-up 20 months in the excisional group and 19.5 months in the drainage and ablation group. Of the studied outcomes the percentage of patients with significant pelvic pain remaining after surgery was derived from the publications and from contacting the authors for clarification of this information. Due to the discomfort of the operative procedures themselves a formal comparison of the two operative techniques immediately postoperatively is problematic (Beretta et al., 1998; Alborzi et al., 2004). The recurrence of pelvic pain was assessed by VAS in one publication (Beretta et al., 1998) and by dichotomous data recording in the other (Alborzi et al., 2004). One paper reported the effect on pain by dividing the general term pain into the symptoms of dysmenorrhoea, deep dyspareunia and pelvic pain and by recording on a VAS and reporting pain classed as moderate or severe (Beretta et al., 1998). The effect on fertility of laparoscopic surgery excision or ablation of ovarian endometriomata was recorded by the time taken to spontaneous conception and the percentage of patients conceiving naturally. In one paper follow-up was limited to 12 months to allow patients to seek assisted reproduction (Alborzi et al., 2004), in the other paper follow-up was extended to 24 months (Beretta et al., 1998). One study recorded the pregnancy rate (Beretta et al., 1998) without specifying the method of confirmation of pregnancy and one paper confirmed pregnancy by the presence of a gestational sac on ultrasound examination (Alborzi et al., 2004). In both studies of either laparoscopic excision or ablation of ovarian endometriomata the recurrence rate of an ovarian endometrioma was recorded by ultrasound examination every 3 months for the first year and then either 6 monthly (Alborzi et al., 2004) or annually (Beretta et al., 1998). Both studies of the laparoscopic approach to the management of endometriomata reported the conversion rate as an outcome variable although in neither group was a conversion performed (Beretta et al., 1998; Alborzi et al., 2004). Both studies reported the requirement for further surgery after the initial laparoscopic procedure as an outcome variable (Beretta et al., 1998; Alborzi et al., 2004). Primary outcome measures Relief from pelvic pain The two studies that met the inclusion criteria for the laparoscopic approach to the management of endometriomas reported results that suggest that both excision, and drainage and coagulation, of endometriomata treated the symptoms related to the endometrioma and endometriosis 100% effectively in all cases. Both lead authors were contacted for further clarification (Beretta et al., 1998; Alborzi et al., 2004). The authors do not report immediate post-operative relief from pain but recurrence of pain over time. 3003

R.Hart et al. The recurrence of dysmenorrhoea, dyspareunia and nonmenstrual pelvic pain was significantly greater in the drainage and ablation group and the time to the recurrence of the pain was significantly shorter in the Beretta study (Figure 1). The recurrence of pelvic pain and dysmenorrhoea was also significantly greater in the drainage and ablation group in the Alborzi (2004) study. After contacting the authors of this study directly they were unable to differentiate between those patients with recurrence of pain and recurrence of dysmenorrhoea alone. Consequently these patients are analysed as recurrence of dysmenorrhoea only in the analysis and are excluded from the analysis of non-menstrual pain and dyspareunia. The laparoscopic excision of endometriomata was significantly associated with a benefit from the recurrence of dysmenorrhoea (OR 0.15, CI 0.06 0.38), recurrence of dyspareunia (OR 0.14, CI 0.05 0.44) and recurrence of non-menstrual pelvic pain (OR 0.10, CI 0.02 0.56). Subsequent spontaneous conception In one of the studies patients were followed up for 1 year and at this time the spontaneous pregnancy was significantly greater in the excision group (Alborzi et al., 2004) and the other study follow-up was performed for 24 months and the spontaneous conception rate observed (Beretta et al., 1998) (Figure 2). The Berreta paper was analysed and the 12 month spontaneous fecundity derived and plotted. After combining with the data from Alborzi et al.(2004) the chance of a spontaneous pregnancy at up to 12 months significantly favoured laparoscopic excision of endometriomata (OR 5.24, CI 1.92 14.27) (Figure 3). The overall subsequent spontaneous conception rate also favoured laparoscopic excision (OR 5.21, CI 2.04 13.29; Figure 2). Secondary outcome measures Recurrence of endometriomata The two studies that met the inclusion criteria for the laparoscopic approach to the management of endometriomas reported that excision of endometriomata, as compared to drainage and coagulation, leads to a significant reduction in the rate of recurrence of an endometrioma (OR 0.41, CI 0.18 0.93) (Beretta et al., 1998; Alborzi et al., 2004) (Figure 4). Conversion from a planned laparoscopic procedure to laparotomy The two studies that met the inclusion criteria for the laparoscopic approach to the management of endometriomas reported that both excision, and drainage and coagulation, of endometriomata have a low chance of intra-operative conversion to a laparotomy and that there is no difference according to the operative technique used (Beretta et al., 1998; Alborzi et al., 2004). Study Excisional surgery Ablative surgery OR (fixed) Weight OR (fixed) or sub-category n/n n/n 95% CI % 95% CI Quality 01 Recurrence of dysmenorrhea Alborzi 2004 6/38 17/30 66.67 0.14 [0.05, 0.44] A Beretta 1998 3/19 9/17 33.33 0.17 [0.04, 0.79] A Subtotal (95% CI) 57 47 100.00 0.15 [0.06, 0.38] Total events: 9 (Excisional surgery), 26 (Ablative surgery) Test for heterogeneity: Chi² = 0.02, df = 1 (P = 0.88), I² = 0% Test for overall effect: Z = 4.04 (P < 0.0001) 02 Recurrence of dyspareunia Beretta 1998 3/15 9/12 100.00 0.08 [0.01, 0.51] A Subtotal (95% CI) 15 12 100.00 0.08 [0.01, 0.51] Total events: 3 (Excisional surgery), 9 (Ablative surgery) Test for heterogeneity: not applicable Test for overall effect: Z = 2.68 (P = 0.007) 03 Recurrence of non-menstrual pelvic pain Beretta 1998 2/20 9/17 100.00 0.10 [0.02, 0.56] A Subtotal (95% CI) 20 17 100.00 0.10 [0.02, 0.56] Total events: 2 (Excisional surgery), 9 (Ablative surgery) Test for heterogeneity: not applicable Test for overall effect: Z = 2.60 (P = 0.009) 0.01 0.1 1 10 100 Favours excision Favours ablation Figure 1. Recurrence of pelvic pain. Favours excision for recurrence of dysmenorrhea, dyspareunia and non-menstrual pain. Study Exisional surgery Ablative surgery OR (fixed) Weight OR (fixed) or sub-category n/n n/n 95% CI % 95% CI Quality Alborzi 2004 19/32 7/30 76.08 4.80 [1.60, 14.45] A Beretta 1998 6/9 4/17 23.92 6.50 [1.09, 38.63] A Total (95% CI) 41 47 100.00 5.21 [2.04, 13.29] Total events: 25 (Exisional surgery), 11 (Ablative surgery) Test for heterogeneity: Chi² = 0.08, df = 1 (P = 0.78), I² = 0% Test for overall effect: Z = 3.45 (P = 0.0006) Figure 2. Subsequent spontaneous conception. Favours excision OR 5.21 (CI2.04 13.29). 3004 0.01 0.1 1 10 100 Favours ablation Favours excision

Cochrane review: excision or ablation for ovarian endometrioma? Study Excisional surgery Ablative surgery OR (fixed) Weight OR (fixed) or sub-category n/n n/n 95% CI % 95% CI Quality Alborzi 2004 19/32 7/30 86.41 4.80 [1.60, 14.45] A Beretta 1998 3/9 1/17 13.59 8.00 [0.69, 92.70] A Total (95% CI) 41 47 100.00 5.24 [1.92, 14.27] Total events: 22 (Excisional surgery), 8 (Ablative surgery) Test for heterogeneity: Chi² = 0.14, df = 1 (P = 0.71), I² = 0% Test for overall effect: Z = 3.24 (P = 0.001) 0.01 0.1 1 10 100 Favours ablation Favours excision Figure 3. 12 month spontaneous conception. Favours excision OR 5.24 (CI1.92 14.27). Study Excisional surgery Ablative surgery OR (fixed) Weight OR (fixed) or sub-category n/n n/n 95% CI % 95% CI Quality Alborzi 2004 9/52 15/48 69.64 0.46 [0.18, 1.18] A Beretta 1998 2/32 6/32 30.36 0.29 [0.05, 1.56] A Total (95% CI) 84 80 100.00 0.41 [0.18, 0.93] Total events: 11 (Excisional surgery), 21 (Ablative surgery) Test for heterogeneity: Chi² = 0.22, df = 1 (P = 0.64), I² = 0% Test for overall effect: Z = 2.14 (P = 0.03) Figure 4. Recurrence of endometrioma. Favours excision OR 0.41 (CI0.18 0.93). Effect on ovarian function No studies were found which addressed the effect of laparoscopic excision versus drainage and coagulation of endometriomata on ovarian function or the quality of life of patients after surgery. Other outcome measures The rate of re-operation One study analysed the effect of the laparoscopic excision versus drainage and coagulation of endometriomata on the need for a further operation within 2 years; this demonstrated a significantly reduced requirement for a further operation in those women undergoing excision of an endometrioma (OR 0.21, CI 0.05 0.79) (Alborzi et al., 2004). However, there may be an element of bias introduced due to the lack of operator blinding as to the initial operation performed. Operative time and post-operative stay One study addressed the effect of laparoscopic excision versus drainage and coagulation of endometriomata on operating time and post-operative stay: there were no reported differences between the two groups (Beretta et al., 1998). Adverse effects There were no adverse effects of either laparoscopic excision or drainage and coagulation of endometriomata reported in either study that met the inclusion criteria for analysis (Beretta et al., 1998; Alborzi et al., 2004). Sensitivity analysis Heterogeneity Both the included papers report very similar outcome data and consequently there was no significant heterogeneity detected between these studies. 0. 1 0.2 0.5 1 2 5 10 Favours excision Favours ablation Both the studies report that the patient and surgeon were unblinded as to the procedure performed (Beretta et al., 1998; Alborzi et al., 2004). It is impossible to blind the surgeon; however, the patient and medical staff performing the followup assessments should be blinded in further studies. Discussion This review addresses the controversial issue of the most appropriate surgical approach to the management of endometriomata either excision or drainage and ablation of the cyst. There were no randomized studies identified where the approach to the management of the endometrioma was by laparotomy. A prospective randomized trial of laparotomy or laparoscopic ovarian cystectomy for the management of endometriomata has demonstrated the benefits to the patient of the laparoscopic approach as manifest by less analgesic requirement, earlier discharge and a shorter post-operative recovery (Mais et al., 1996). Consequently this discussion will focus on the laparoscopic approach to the management of ovarian endometriomata. Both studies meeting the criteria for inclusion in the metaanalysis have the potential for bias derived from the fact that both patient and surgeon were not blind as to the procedure performed (Beretta et al., 1998; Alborzi et al., 2004). However, both studies had full retention of patients for the 2 years of follow-up, with regular interim review. The studies had very similar designs and inclusion and exclusion criteria. In the analysis there was no statistical heterogeneity detected. Neither study addressed ovarian function other than by the surrogate of spontaneous pregnancy, nor was a study found that addressed the issue of patient satisfaction or quality of life. These should be priorities for further research. An assessment of the effect of excision of an endometrioma versus drainage and ablation on the outcome variable relief 3005

R.Hart et al. from pelvic pain (immediately post operation) was impossible to derive as both studies suggest complete relief of pelvic pain by both approaches, and this highlights the difficulty in determining when the initial assessment should be made. However, the outcome variable recurrence of pelvic pain was significantly worse in the drainage and ablation treatment arms. The concern that this may be not related to the management of the endometrioma but to the background disease of endometriosis can be dismissed, as in both studies there was no difference in the disease severity between the two treatment modalities. In the analysis of the recurrence of pelvic pain, one of the studies only analysed the recurrence of dysmenorrhoea (Alborzi et al., 2004). The authors were approached but were unable to provide further information as to the recurrence of other symptoms. This meta-analysis demonstrates that excisional surgery for the management of endometriomata is associated with a reduced recurrence rate of dysmenorrhoea, dyspareunia and non-menstrual pelvic pain when compared to the drainage and ablation of the cysts. The recurrence rate of the endometriomata as assessed by pelvic ultrasonography was analysed in two studies that met the inclusion criteria (Beretta et al., 1998; Alborzi et al., 2004). These studies both followed the patients for up to 2 years. There was a significantly reduced rate of recurrence in the patients who underwent excisional surgery and one study demonstrated a significantly reduced requirement for further surgery in the excisional group (Alborzi et al., 2004). Failure to complete the laparoscopic operation requiring an intra-operative conversion from a laparoscopic approach to a laparotomy was assessed in both studies. In neither study did a patient require an intra-operative conversion to be performed. The spontaneous fertility rate after surgery was assessed in both studies that met the inclusion criteria (Beretta et al., 1998; Alborzi et al., 2004). One study followed-up the patients for 2 years (Beretta et al., 1998), and one study followed-up the patients for 1 year prior to them being referred for assisted reproduction (Alborzi et al., 2004). Both studies were rigorous in their inclusion in the subfertile group of only patients without another cause for their subfertility other than their pelvic pathology. The meta-analysis of the rate of spontaneous conception demonstrated a significantly greater rate of conception in the excisional group. From the information supplied in the publication data (Beretta et al., 1998) the spontaneous conception rate at 1 year was derived, as this is a more commonly used follow-up time, to allow for spontaneous conception before referral for assisted reproduction. This also confirmed the favourable outcome with excisional surgery. As there are only two studies that address the issue of the most effective surgical approach to the management of endometriomata, there exists the possibility of publication bias. In summary there is some evidence from two non-blinded randomized controlled trials that excisional surgery for endometriomata provides for a more favourable outcome than drainage and ablation using bipolar coagulation, with regard to the recurrence of the endometrioma, recurrence of symptoms and subsequent spontaneous pregnancy in women who were previously infertile. Consequently this surgical approach should be the favoured approach. However, in women who may subsequently 3006 undergo assisted reproductive techniques no data exist as to the favoured surgical approach with regard to fertility outcome after the surgical treatment of the endometrioma. Future studies comparing excisional surgery with drainage and ablation in the management of ovarian endometriomata should address: (i) quality of life after surgery; (ii) ovarian function after surgery; (iii) assisted reproductive performance post surgery. Acknowledgements The following members of the Cochrane Menstrual Disorders and Subfertility Group have helped with trial searches: Michelle Proctor and Ruth Buist. References Included trials Alborzi S, Momtahan M, Parsanezhad ME, Dehbashi S and Zolghadri J (2004) A prospective, randomized study comparing laparoscopic ovarian cystectomy versus fenestration and coagulation in patients with endometriomas. Fertil Steril 82,1633 1637. Beretta P, Franchi M, Ghezzi F, Busacca M, Zupi E and Bolis P (1998) Randomized clinical trial of two laparoscopic treatments of endometriomas: cystectomy versus drainage and coagulation. Fertil Steril 70,1176 1180. Excluded trials Fayez JA and Vogel MF (1991) Comparison of different treatment methods of endometriomas by laparoscopy. Obstet Gynecol 78,660 665. Hemmings R, Bissonnette F and Bouzayen R (1998) Results of laparoscopic treatments of ovarian endometriomas: laparoscopic ovarian fenestration and coagulation. Fertil Steril 70,527 529. Saleh A and Tulandi T (1999) Reoperation after laparoscopic treatment of ovarian endometriomas by excision and by fenestration. Fertil Steril 72,322 324. Additional references Adamson GD, Subak LL, Pasta DJ, Hurd SJ, von Franque O and Rodriguez BD (1992) Comparison of CO 2 laser laparoscopy with laparotomy for treatment of endometriomata. Fertil Steril 57,965 973. Ajossa S, Mais V, Guerriero S, Paoletti AM, Caffiero A, Murgia C and Melis GB (1994) The prevalence of endometriosis in premenopausal women undergoing gynecological surgery. Clin Exp Obstet Gynecol 21,195 197. Al-Azemi M, Bernal AL, Steele J, Gramsbergen I, Barlow D and Kennedy S (2000) Ovarian response to repeated controlled stimulation in in-vitro fertilization cycles in patients with ovarian endometriosis. Hum Reprod 15,72 75. Bateman BG, Kolp LA and Mills S (1994) Endoscopic versus laparotomy management of endometriomas. Fertil Steril 62,690 695. Brinton LA, Gridley G, Persson I, Baron J and Bergqvist A (1997) Cancer risk after a hospital discharge diagnosis of endometriosis. Am J Obstet Gynecol 176,572 579. Brosens IA, Puttemans PJ and Deprest J (1994) The endoscopic localization of endometrial implants in the ovarian chocolate cyst. Fertil Steril 61,1034 1038. Brosens IA, Van Ballaer P, Puttemans P and Deprest J (1996) Reconstruction of the ovary containing large endometriomas by an extraovarian endosurgical technique. Fertil Steril 66,517 521. Busacca M, Fedele L, Bianchi S, Candiani M, Agnoli B, Raffaelli R and Vignali M (1998) Surgical treatment of recurrent endometriosis: laparotomy versus laparoscopy. Hum Reprod 13,2271 2274. Canis M, Mage G, Wattiez A, Chapron C, Pouly JL and Bassil S (1992) Secondlook laparoscopy after laparoscopic cystectomy of large ovarian endometriomas. Fertil Steril 58,617 619. Canis M, Mage G, Wattiez A, Pouly JL and Bruhat MA (2003) The ovarian endometrioma: why is it so poorly managed? Laparoscopic treatment of large ovarian endometrioma: why such a long learning curve? Hum Reprod 18,5 7. Catalano GF, Marana R, Caruana P, Muzii L and Mancuso S (1996) Laparoscopy versus microsurgery by laparotomy for excision of ovarian cysts in patients with moderate or severe endometriosis. J Am Assoc Gynecol Laparosc 3,267 270.

Cochrane review: excision or ablation for ovarian endometrioma? Crosignani PG, Vercellini P, Biffignandi F, Costantini W, Cortesi I and Imparato E (1996) Laparoscopy versus laparotomy in conservative surgical treatment for severe endometriosis. Fertil Steril 66,706 711. Cullen T (1920) The distribution of adenomyoma containing uterine mucosa. Archs Surg 1,215 283. Daniell JF, Kurtz BR and Gurley LD (1991) Laser laparoscopic management of large endometriomas. Fertil Steril 55,692 695. Del Carmen MG, Smith Sehdev AE, Fader AN, Zahurak ML, Richardson M, Fruehauf JP, Montz FJ and Bristow RE (2003) Endometriosis-associated ovarian carcinoma: differential expression of vascular endothelial growth factor and estrogen/progesterone receptors. Cancer 98,1658 1663. Donnez J, Nisolle M, Gillet N, Smets M, Bassil S and Casanas-Roux F (1996) Large ovarian endometriomas. Hum Reprod 11,641 646. Farquhar C and Sutton C (1998) The evidence for the management of endometriosis. Curr Opin Obstet Gynecol 10,321 332. Farquhar CM (2000) Extracts from the clinical evidence. Endometriosis. Br Med J 320,1449 1452. Garcia-Velasco JA, Mahutte NG, Corona J, Zuniga V, Giles J, Arici A and Pellicer A (2004) Removal of endometriomas before in vitro fertilization does not improve fertility outcomes: a matched, case control study. Fertil Steril 81,1194 1197. Hachisuga T and Kawarabayashi T (2002) Histopathological analysis of laparoscopically treated ovarian endometriotic cysts with special reference to loss of follicles. Hum Reprod 17,432 435. Hemmings R, Bissonnette F and Bouzayen R (1998) Results of laparoscopic treatments of ovarian endometriomas: laparoscopic ovarian fenestration and coagulation. Fertil Steril 70,527 529. Hughesdon P (1957) The structure of endometrial cysts of the ovary. J Obstet Gynaecol Br Emp 44,481 487. Jacobson TZ, Barlow DH, Koninckx PR, Olive D and Farquhar C (2002) Laparoscopic surgery for subfertility associated with endometriosis. Cochrane Database Syst Rev, CD001398. Jones KD and Sutton CJ (2000) Laparoscopic management of ovarian endometriomas: a critical review of current practice. Curr Opin Obstet Gynecol 12,309 315. Luciano AA, Maier DB, Koch EI, Nulsen JC and Whitman GF (1989) A comparative study of postoperative adhesions following laser surgery by laparoscopy versus laparotomy in the rabbit model. Obstet Gynecol 74,220 224. Luciano AA, Lowney J and Jacobs SL (1992) Endoscopic treatment of endometriosis-associated infertility. Therapeutic, economic and social benefits. J Reprod Med 37,573 576. Lundorff P, Hahlin M, Kallfelt B, Thorburn J and Lindblom B (1991) Adhesion formation after laparoscopic surgery in tubal pregnancy: a randomized trial versus laparotomy. Fertil Steril 55,911 915. Mais V, Ajossa S, Guerriero S, Piras B, Palomba M and Melis GB (1996) Laparoscopic management of endometriomas: a randomized trial versus laparotomy. J Gynecol Surg 12,41 46. Maouris P and Brett L (2002) Endometrioitic ovarian cysts: the case for excisional laparoscopic surgery. Gynaecol Endosc 11,231 234. Martin DC (1991) Laparoscopic treatment of ovarian endometriomas. Clin Obstet Gynecol 34,452 459. Milingos S, Loutradis D and Kallipolitis G (1999) Comparison of laparoscopy with laparotomy for the treatment of extensive endometriosis with large endometriomata. J Gynecol Surg 15,131 136. Nishida M, Watanabe K, Sato N and Ichikawa Y (2000) Malignant transformation of ovarian endometriosis. Gynecol Obstet Invest 50(Suppl 1),18 25. Nisolle M and Donnez J (1997) Peritoneal endometriosis, ovarian endometriosis and adenomyotic nodules of the rectovaginal septum are three different entities. Fertil Steril 68,585 596. Operative Laparoscopy Study Group (1991) Postoperative adhesion development after operative laparoscopy: evaluation at early second-look procedures. Operative Laparoscopy Study Group. Fertil Steril 55,700 704. Pritts EA and Taylor RN (2003) An evidence-based evaluation of endometriosisassociated infertility. Endocrinol Metab Clin North Am 32,653 667. Reich H and McGlynn F (1986) Treatment of ovarian endometriomas using laparoscopic surgical techniques. J Reprod Med 31,577 584. Sampson J (1927a) Peritoneal endometriosis due to menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 14,422 469. Sampson J (1927b) Metastatic or embolic endometriosis, due to menstrual dissemination of endometrial tissue into venous circulation. Am J Pathol 3,93. Sawada T, Satoshi O and Kawakami S (1999) Laparoscopic surgery versus laparotomy management for infertile patients with ovarian endometriomata. Gynaecol Endosc 8,17 19. Sokol AI, Chuang K and Milad MP (2003) Risk factors for conversion to laparotomy during gynecologic laparoscopy. J Am Assoc Gynecol Laparosc 10,469 473. Sutton CJ, Ewen SP, Jacobs SA and Whitelaw NL (1997) Laser laparoscopic surgery in the treatment of ovarian endometriomas. J Am Assoc Gynecol Laparosc 4,319 323. Vercellini P (1997) Endometriosis: what a pain it is. Semin Reprod Endocrinol 15, 251 261. Vercellini P, Vendola N, Bocciolone L, Colombo A, Rognoni MT and Bolis G (1992) Laparoscopic aspiration of ovarian endometriomas. Effect with postoperative gonadotropin releasing hormone agonist treatment. J Reprod Med 37,577 580. Vercellini P, Chapron C, De Giorgi O, Consonni D, Frontino G and Crosignani PG (2003) Coagulation or excision of ovarian endometriomas? Am J Obstet Gynecol 188,606 610. Yanushpolsky EH, Best CL, Jackson KV, Clarke RN, Barbieri RL and Hornstein MD (1998) Effects of endometriomas on oocyte quality, embryo quality and pregnancy rates in in vitro fertilization cycles: a prospective, case-controlled study. J Assist Reprod Genet 15,193 197. Yuen PM, Yu KM, Yip SK, Lau WC, Rogers MS and Chang A (1997) A randomized prospective study of laparoscopy and laparotomy in the management of benign ovarian masses. Am J Obstet Gynecol 177,109 114. Submitted on June 9, 2005; accepted on June 16, 2005 3007