Extensive Excision of Deep Infiltrative Endometriosis before In Vitro Fertilization Significantly Improves Pregnancy Rates

Original Article Extensive Excision of Deep Infiltrative Endometriosis before In Vitro Fertilization Significantly Improves Pregnancy Rates Paulo H. M. Bianchi, MD*, Ricardo M. A. Pereira, MD, Alysson Zanatta, MD, Jose Roberto Alegretti, BSc, Eduardo L. A. Motta, PhD, and Paulo C. Serafini, PhD From the Huntington Medicina Reprodutiva, São Paulo, Brazil (all authors). ABSTRACT Keywords: Study Objective: We sought to compare the outcomes of in vitro fertilization (IVF) treatments in women with infertility-associated deep infiltrative endometriosis (DIE) who underwent extensive laparoscopic excision of endometriosis before IVF with those who underwent IVF only. Design: Prospective cohort study. Setting: Infertility clinic and private hospital in São Paulo, Brazil. Patients: A total of 179 infertile patients younger than 38 years had symptoms and/or signs of endometriosis and sonographic images suggestive of DIE. Interventions: After thorough counseling, 179 women were invited to participate in a prospective cohort study with 2 treatment options: IVF without undergoing laparoscopic surgery (group A, n 5 105) and extensive laparoscopic excision of DIE before IVF (group B, n 5 64). Ten women were lost to follow-up. The IVF outcomes were compared between the 2 groups. Measurements and Main Results: In group B, patients had 5 6 2 (mean 6 SD) DIE lesions excised during laparoscopy. Patient characteristics in groups A and B, respectively, were: age (32 6 3vs326 3 years, p 5.94), infertility duration (29 6 20 vs 27 6 17 months, p 5.45), day-3 serum follicle-stimulating hormone levels (5.6 6 2.5 vs 5.9 6 2.5 IU/L, p 5.50), and previous IVF attempts (1 6 1vs26 1, p 5.01). The IVF outcomes differed between groups A and B, respectively, with regard to total dose of recombinant follicle-stimulating hormone required to accomplish ovulation induction (2380 6 911 vs 2542 6 1012 IU, p 5.01), number of oocytes retrieved (10 6 5vs96 5, p 5.04), and pregnancy rates (24% vs 41%, p 5.004), but not number of embryos transferred (3 6 1vs36 1, p 5 1). The odds ratio of achieving a pregnancy were 2.45 times greater in group B than in group A. Conclusion: Extensive laparoscopic excision of DIE significantly improved IVF pregnancy rates of women with infertilityassociated DIE. Journal of Minimally Invasive Gynecology (2009) 16, 174 180 Ó 2009 AAGL. All rights reserved. Deep infiltrative endometriosis; In vitro fertilization; Laparoscopy; Pregnancy rate As many as 60% of infertile women are afflicted with endometriosis [1 3]. The presence of endometriosis can impair fertility in several ways, from mechanical obstruction of the uterine tubes to altered ovulation and impaired implantation [4]. Surgical treatment has been an excellent option to help women with infertility-associated endometriosis who either possessed anatomically normal tubes or had their tubes functionally restored at surgery [5]. A multicentric trial [9] The authors have no commercial, proprietary, or financial interest in the products or companies described in this article. Corresponding author: Paulo H. M. Bianchi, Avenida República do Líbano 529, Ibirapuera, São Paulo SP, Brazil 04501-000. E-mail: paulobianchi35@gmail.com Submitted October 13, 2008. Accepted for publication December 12, 2008. Available at www.sciencedirect.com and www.jmig.org showed that laparoscopic excision of minimal and mild endometriosis improves fertility. Although this evidence was contested by the results of another prospective trial [10], a metaanalysis that included these 2 prospective studies concluded that laparoscopic treatment of minimal and mild endometriosis may have a beneficial effect on fertility [11]. Some authors claim in vitro fertilization (IVF) as the first treatment option to overcome infertility associated with endometriosis, despite the severity of the disease [6]. This has generated much debate and the application of IVF treatment to all women with infertility-associated endometriosis has not been universally accepted [7,8]. Data from a metaanalysis including 22 studies documented that IVF pregnancy rates sharply decreased with advanced stages of endometriosis [13]. In addition, a significant decrease occurred in the number of oocytes retrieved, peak serum estradiol 1553-4650/$ - see front matter Ó 2009 AAGL. All rights reserved. doi:10.1016/j.jmig.2008.12.009

Bianchi et al. Laparoscopic Resection Improves IVF Pregnancy Rates 175 concentration, and fertilization and implantation rates in patients with endometriosis as compared with those with tubal factor infertility only [13]. Furthermore, a reduction in pregnancy rate ranging from 21% to 54% was reported in women with endometriosis as compared with those with tubal disease. Recently, authors also revealed that surgical treatment of endometriosis boosts IVF success rates up to 70% in women with minimal and mild endometriosis [12]. Extensive laparoscopic excision of deep infiltrative endometriosis (DIE) is an effective approach to alleviate pain-associated endometriosis, and to aid women with infertility-associated endometriosis [14,15]. Thus, women with infertility-associated endometriosis may benefit from therapeutic options such as IVF, ovum donation, and IVF surrogacy [16 21]. To our knowledge, no data exist on the outcome of IVF in infertile women with DIE previously treated with extensive laparoscopy. Moreover, at the beginning of the study, no evidence existed that surgical laparoscopy, an invasive procedure with inherent risks, would increase IVF success. Yet, observational evidence on the combined approach to treat infertility-associated DIE was promising [14]. This prospective cohort study aimed to evaluate whether extensive laparoscopic excision of DIE before IVF could improve the success of IVF in an ethnic and socially homogeneous group of infertile women with clinical and sonographic evidence of DIE. Materials and Methods Participants and Eligibility Criteria Women seeking assisted reproductive care in a private clinic were invited to participate in an open prospective longitudinal cohort study conducted from January 2005 through January 2008. All patients were evaluated by obtaining a detailed history, physical examination, and laboratory work-up as recommended by the American Society for Reproductive Medicine (ASRM), including hysterosalpingogram, basal serum hormonal levels, partner s semen analysis, and a transvaginal ultrasound after bowel preparation (TVSbp) [22,23]. This last examination was performed by 2 experienced radiologists [24]. Eligibility criteria were: (1) infertility with clinical and TVS-bp evidence of DIE; (2) age between 21 and 38 years; (3) presence of a standard indication for either IVF or intracytoplasmic sperm injection (ICSI) treatment; (4) presence of at least 1 functional ovary; (5) presence of an anatomically normal uterine cavity on the basis of a recent hysterosalpingographic or hysteroscopic evaluation (%6 months); (6) early follicular phase (day 2 or 3) serum follicle-stimulating hormone (FSH) levels of 15 IU/L or less and estradiol levels of 60 pg/ml or less; (7) absence of untreated endocrinologic disorder; and (8) male partner ejaculated spermatozoa having 1% or greater strict morphology. All participants were counseled regarding the nature and purpose of the study, including 2 comprehensive consultations. During these consultations, we conducted open and detailed discussion regarding the presumptiveness of the disease and current management options, including information regarding potential risks and benefits of IVF and extensive laparoscopic excision of endometriotic lesions. Patients then provided written informed consent according to institutional review board regulations. Patient s Allocation, Laparoscopic Surgery, and IVF Protocol Within an interval of 2 to 4 weeks after comprehensive counseling regarding potential benefits and risks associated with the procedures, each patient chose between 2 intervention options. Study groups were comprised according to choice of treatment: group A underwent IVF/ICSI only and group B underwent extensive laparoscopy for removal of all identifiable DIE lesions before proceeding to IVF/ICSI. All laparoscopies were performed by an experienced surgeon (R. M. A. P.) under a general endovenous combined epidural anesthesia with morphine. All patients underwent a full bowel preparation 1 day before the procedure, wore sequential compression devices (Ethicon Endosurgery Cincinatti, OH) in the lower extremities, and were treated with low molecular weight heparin as prophylaxis against venous thromboembolism. Preoperative antibiotic prophylaxis was used in all women and extended protocols were used for patients who underwent bowel resection. Punctures for 4 ports were made in the abdomen, and the umbilical port was established with an open technique to secure a trocar. Endoscopic abdominal and pelvic inspections were followed by meticulous dissection and removal of all identifiable endometriotic lesions with a monopolar scalpel. Deep infiltrative endometriosis lesions located in the uterosacral area, retrocervical area, rectovaginal area, pouch of Douglas, vesicouterine compartment, and bladder were dissected from the surrounding normal tissue while preserving important structures such as the ureter, uterine vessels, and pelvic nerves. Endometriomas were drained and their capsules were removed, followed by oophoroplasty. Rectosigmoid lesions were removed by either segmental or discoid resection depending on the number of lesions, extension, and distance from anal border [25]. Primary bowel anastomosis was performed in all cases of bowel resection. All patients had temporary suspension of ovaries and uterus with nonabsorpable suture 7 days after the surgery to prevent adhesion formation. The IVF treatments were not commenced until at least 3 months after laparoscopy. The IVF/ICSI protocol was performed as described previously [26]. Briefly, a gonadotropin-releasing hormone analog (leuprolide acetate, Abbott Laboratories, Abbott Park, IL) was started in the midluteal phase of the menstrual cycle preceding the treatment cycle (long protocol) to prevent premature luteinizing hormone surge and ovulation. After pituitry

176 Journal of Minimally Invasive Gynecology, Vol 16, No 2, March/April 2009 down-regulation, the women were started on recombinant FSH (r-fsh) (follitropin alpha, Merck-Serono Laboratories, Geneva, Switzerland) administered daily in the early morning, at a dosage ranging from 150 to 300 IU based on age as follows: 30 years or younger, 150 IU; 31 to 35 years, 225 IU; and 36 years or older, 300 IU. Ovarian response was monitored by transvaginal ultrasound and r-fsh doses were adjusted to achieve satisfactory follicular development. When at least 2 follicles reached measurements of 18 mm or greater mean diameter, 250 mg of recombinant human chorionic gonadotropin (r-hcg) (choriogonadotropin alpha, Merck- Serono Laboratories) was administered subcutaneously. Oocyte retrieval was carried out in an outpatient facility under mild sedation and analgesia 35 to 36 hours after r-hcg administration. None of the women were prescribed gonadotropinreleasing hormone analogs or hormonal treatments for endometriosis before IVF. Oocyte identification, gamete preparation and handling, conventional insemination or ICSI, IVF culture environment, and preembryo handling (including transfer) were all described previously [26]. Embryo quality evaluation was based on developmental stage and morphologic criteria including cell number, regularity of blastomeres, fragmentation, and other morphologic aspects such as granulation. When less than 10% fragmentation was evident, developmental stage was appropriate for their age (8 blastomeres), and no multinucleated blastomeres existed, the embryos were described as top quality [26]. Embryo transfer was performed on the third day of extracorporeal development by loading up to 4 embryos using a Sydney IVF embryo transfer catheter, K-jets 7019-SIVF (Cook OB/GYN, Indianapolis, IN). The choice of the number of embryos transferred was based on patient s age, and embryos number and quality. Luteal phase hormonal support consisted of natural micronized progesterone administered vaginally (400 mg every 8 hours) and oral intake of estradiol valerate (2 mg twice daily) [26]. Luteal phase was assessed and monitored by measuring serum estradiol, progesterone, and b-hcg 12 days after oocyte retrieval. A clinical pregnancy was considered when a pelvic ultrasound showed a gestational sac with embryonic heart activity. Sample Size, Measures, and Statistical Analyses A previous study showed a reduction in oocytes retrieved (2 3) from women with advanced endometriosis after undergoing surgery [27]. Therefore, the number of mature oocytes retrieved was the primary outcome measure. Sample size calculation was performed assuming that if less than 3 oocytes were retrieved from women treated with extensive laparoscopic excision of endometriosis before IVF treatment, an 80% likelihood existed of detecting a clinically significant difference (at p 5.05). To detect this level of difference, 45 patients were required in each arm. However, to minimize biases we selected patients in an approximate ratio of 1:2 (case:control). Secondary outcome measures were average total dose of r-fsh required to accomplish follicle development, serum estradiol levels on the day of ovulatory r-hcg, fertilization rates, number of top-quality embryos, number of embryos transferred to the uterine cavity, implantation rate, and pregnancy rate. The c 2 and binary logistic regression tests were performed to test associations between pregnancy and operative status, along with odds ratio (OR) calculation. The proportion of women in each group with an additional infertility factor was compared using a z test for 2 proportions. Women s age, basal serum FSH and estradiol levels, infertility duration in months, number of previous IVF procedures, average total dose of r-fsh to accomplish follicle maturation, number of oocytes retrieved, number of embryos transferred, number of top-quality embryos produced, fertilization rate, implantation rate, and pregnancy rate were assessed using the Kolmogorov-Smirnov test and F test for normality and homoscedasticity, respectively. Differences between groups A and B were compared by Student t test or a z test for 2 proportions when appropriate. Data were expressed as mean 6 SD and 95% confidence intervals where appropriated. Levels below 0.05 (a 5 5%) were considered significant. Results As shown in Fig. 1, we enrolled 179 consecutive infertile, Caucasian, middle- to upper-class Brazilian women with symptoms and/or signs of endometriosis (dysmenorrhea, dyspareunia, acyclic and chronic pelvic pain, tenesmus, dysuria, painful nodule on palpation of the anterior or posterior vaginal fornices, or tenderness on palpation of uterosacral ligaments). Their ages ranged from 24 to 38 years with a mean of 32 6 3 years. After counseling, 105 women chose to undergo IVF treatment (group A) and 66 chose to undergo extensive laparoscopy for removal of endometriosis before IVF. Eight women were lost to follow-up after providing written informed consent and were considered early dropouts. Surgical procedures were carried out without major complications such as hemorrhage, surgical infection, anastomosis dehiscence, or bowel fistulas; laparotomy and blood transfusion were not required for any patient. One patient had a pudendal nerve lesion caused by extensive dissection requiring postoperative physical therapy for 6 months. Two women who underwent laparoscopy declined IVF treatment and were considered late dropouts. Baseline characteristics of patients comprising the 2 study groups are shown in Table 1. The ages of the women enrolled in groups A and B were comparable, as were proportion of women with associated tubal factor infertility (blockage, severe distortion, or surgical absence), proportion of patients with associated ovulatory disorders (anovulation or polycystic ovary syndrome), proportion of patients with associated male factor infertility according to ASRM criteria [28], and basal serum levels of FSH and estradiol. Women in

Bianchi et al. Laparoscopic Resection Improves IVF Pregnancy Rates 177 Fig. 1. Study flowchart and patient allocations. LSC 5 Laparoscopy with complete excision of DIE lesions. group B had undergone significantly more IVF procedures than those in group A (2 6 1 vs 16 1, respectively, p 5.01) and had a longer duration of infertility (Table 1). Multiple DIE lesions were observed in all women in group B with an average of 5 6 2 lesions/woman, identified by laparoscopic inspection before removal. Nineteen (29.7%) women had DIE lesions in the retrocervical space, 19 (29.7%) had retrovaginal endometriosis, 4 (6%) had lesions infiltrating the vaginal wall, and 23 (35.9%) had DIE within the bowel (19 rectosigmoid lesions, 3 appendiceal lesions, and 1 ileal node). Examples of these are shown in Figs. 2 and 3. Three (4.7%) patients had bladder lesions, 8 (12.5%) had endometriotic nodes in the pouch of Douglas, and 29 (45.3%) had uterosacral ligament involvement. In addition, endometriomas were removed from 29 (45.3%) women. Of those, 20 (31.2%) endometriomas were larger than 2 cm whereas 9 (14%) were smaller than 2 cm. The duration of surgery varied from 3 to 9 hours. The time interval between surgery and IVF was 14.6 6 12.5 months, ranging from 3 to 18 months. Patients that spontaneously conceived were excluded from the study, as our objective was to evaluate surgery on IVF outcomes. Table 1 Baseline characteristics of patients who did not (group A) and did (group B) undergo laparoscopy Group A (n 5 105) Group B (n 5 64) p Age (yrs) 32 6 3 326 3.94 Body mass index (kg/m 2 ) 21.4 6 2.3 22.4 6 4.1.26 Basal FSH (IU/L) 5.6 6 2.5 5.9 6 2.5.50 Basal estradiol (pg/ml) 38.6 6 30.1 38.7 6 30.4.60 No. of previous IVF attempts 1 6 1 26 1.01 Infertility duration (mo) 29 6 20 35 6 18.01 Associated tubal factor (%) 11.4 15.6.44 Associated ovarian factor (%) 8.6 3.1.21 Associated male factor (%) 31.4 18.8.06 FSH 5 Follicle-stimulating hormone; IVF 5 in vitro fertilization. Values are expressed as mean 6 SD. In all, 153 IVF treatments were performed with an average of 1.6 6 0.8 attempts in group A and 86 IVF cycles were performed with an average of 1.9 6 1.3 attempts in group B. The IVF outcomes are summarized in Table 2. A significantly higher r-fsh was required to achieve adequate follicular development in group B compared with group A (2542 6 1012 vs 2380 6 911 IU, respectively; p 5.01). The number of oocytes retrieved from group B was significantly lower than from group A (9 6 5vs106 5, respectively; p 5.04). However, fertilization rates, number of top-quality embryos, and number of embryos transferred did not differ between the groups. Significantly higher implantation (32.1% vs 19%, p 5.03) and pregnancy rates (41% vs 24%, p 5.004) were identified in group B. The OR for women in group B to achieve pregnancy was 2.45 (95% CI 5 1.34 4.51). Therefore, for every 6 women who underwent surgery, an additional IVF pregnancy was achieved compared with group A (CI 95% 5 3.2 34.1). In groups A and B, respectively, pregnancy outcomes as live birth rates (87.5% vs 94.4%, p 5.41), birth weight in singleton pregnancies (3.216 6 0.416 vs 2.836 6 0.753 kg, p 5.42), and mean birth weight in twin pregnancies (2.542 6 0.227 vs 2.573 6 0.671 kg, p 5.93) were similar. To evaluate whether lower ovarian response to stimulation in group B was related to surgery itself or to the removal of endometriomas, and associated loss of normal ovarian cortex, we compared the subgroup of women who underwent laparoscopic removal of endometriomas (n 5 29) to those who underwent laparoscopy without undergoing ovarian procedures (n 5 35). As shown in Table 3, these 2 subgroups of women had comparable intervals of infertility, number of previous IVF attempts, and basal serum levels of FSH and estradiol. The average total dose of r-fsh required to achieve follicular development was comparable between these 2 groups; yet, patients who underwent removal of endometriomas yielded significantly fewer oocytes (7 6 4vs106 4, p 5.01) (Table 3). The number of top-quality embryos and fertilization rates were similar among the 2 groups.

178 Journal of Minimally Invasive Gynecology, Vol 16, No 2, March/April 2009 Fig. 2. Sonographic aspects of bowel endometriosis. Longitudinal transvaginal ultrasonography showing hypoechoic nodule (arrow) infiltrating anterior rectal wall up to muscularis propria layer (M)(A). Same lesion in transverse perspective (B). In both images, curved arrow shows hypoechoic nodule of retrocervical region contiguous to intestinal nodule. To further investigate whether laparoscopic excision of DIE had a negative impact on ovarian function, we compared the data obtained from the subgroup of women who underwent laparoscopy without ovarian surgery (n 5 35) with women in group A, who did not have sonographic signs of endometriomas (n 5 56). These 2 subgroups had comparable age, infertility duration, number of previous IVF attempts, and basal serum levels of FSH and estradiol (Table 4). The average total amount of r-fsh required to achieve follicular development, total number of oocytes retrieved, and number of top-quality embryos were statistically similar between these subgroups, but fertilization rates were significantly higher in the group of patients who underwent laparoscopy (Table 4). Discussion The findings of this study showed that laparoscopy for extensive removal of DIE before IVF treatment significantly enhanced the likelihood of conception in women with DIEassociated infertility. Several questions regarding infertilityassociated DIE were answered using a study model that included consecutive women with long-standing DIE-associated infertility. The addition of TVS-bp to the routine workup of infertility, signaled by patients symptoms and physical findings, was of great value. By permitting an accurate and realistic estimation of the presence, depth of penetration, and extensiveness of DIE, interpretation of examinations provided grounds for judicious counseling. Moreover, every woman in group B had surgical removal of multiple DIE lesions (average 5 6 2), a finding that correlated with the observations obtained by TVS-bp. Allowing patient self-allocation is not without biases. Based on the lack of practice standards for combining laparoscopy with extensive excision of DIE and IVF, and the significant and potentially additive risks to the patient, selfallocation appeared to be ethically necessary to the investigators. To prevent investigators biases, patients received full access to information published in the scientific literature and popular press [29] and were provided 2 comprehensive visits where all questions were answered. In addition, the lack of significant differences between groups regarding age; incidence of tubal, ovarian, and male factor infertility; and baseline serum levels of FSH and estradiol support that patients were similar in both groups. The diminished number of oocytes retrieved from patients who underwent laparoscopic excision of DIE along with the Table 2 In vitro fertilization outcomes in groups A and B Group A Group B p Fig 3. Resected segment of rectosigmoid colon showing 3 infiltrative endometriotic nodules. Total dose 2380 6 911 2542 6 1012.01 of FSH (IU) No. of oocytes retrieved 10 6 5 96 5.04 Fertilization rate (%) 77.9 78.76 No. of top-quality embryos/patient.59 6 1.57 6 1.48 No. of embryos transferred 3 6 1 36 1 1 Implantation rate (%) 19 6 25.1 32.1 6 30.6.03 Pregnancy rate (%) 24 41.004 FSH 5 Follicle-stimulating hormone. Absolute values are expressed as mean 6 SD.

Bianchi et al. Laparoscopic Resection Improves IVF Pregnancy Rates 179 Table 3 Baseline characteristics and in vitro fertilization outcomes of patients who underwent laparoscopy with and without removal of endometriomas No endometriomas (n 5 35) Endometriomas (n 5 29) p Age (yrs) 32.6 6 3 32.8 6 3.84 Infertility duration (mo) 35 6 16 36 6 20.88 No. of previous IVF attempts 2 6 2 36 2.89 Basal FSH (IU/L) 5.7 6 1.9 6.6 6 2.5.23 Basal estradiol (pg/ml) 36.3 6 21.1 32.8 6 20.60 Total dose 2611 6 963 2450 6 1087.50 of FSH (IU) No. of oocytes retrieved 10 6 4 76 4.01 Fertilization rate (%) 86.5 6 21 83.9 6 20.7.55 No. of top-quality embryos/patient.67 6 1.63 6 1.87 FSH 5 Follicle-stimulating hormone; IVF 5 in vitro fertilization. Values are expressed as mean 6 SD. increased r-fsh requirement to accomplish ovulation induction most likely result from the surgical removal of endometriomas. These observations are supported by the fewer number of oocytes obtained from women who underwent extensive laparoscopy with removal of endometriomas than those who did not undergo ovarian surgery (Table 3). Furthermore, comparing these women with women from group A who did not have sonographic signs of endometriomas provided evidence that extensive laparoscopic DIE excision did not result in poorer ovarian response when no ovarian disease existed (Table 4). Several other authors have studied the effects of conservative surgery for endometriomas on the ovarian response to gonadotropins. Although still controversial, a large number agree that removal of endometriomas is associated with poorer ovarian response to hyperstimulation [8,30 33]. Authors [7] suggested that laparoscopic removal of endometriosis should be an option for women who have failed Table 4 Baseline characteristics and in vitro fertilization outcomes of patients who underwent laparoscopy without removal of endometriomas and patients who did not undergo laparoscopy and showed no sonographic signs of endometriomas Laparoscopy (n 5 35) No laparoscopy (n 5 56) p Age (yrs) 32.6 6 3 32.9 6 3.63 Infertility duration (mo) 35 6 16 28 6 18.06 No. of previous IVF attempts 2 6 2 26 1.87 Basal FSH (IU/L) 5.7 6 1.9 5.6 6 2.5.85 Basal estradiol (pg/ml) 36.3 6 21.1 40.7 6 34.49 Total dose 2611 6 963 2471 6 939.39 of FSH (IU) No. of oocytes retrieved 10 6 5 106 5.74 Fertilization rate (%) 86.5 6 21 78.5 6 17.8.01 No. of top-quality.67 6 1.53 6 1.46 embryos/patient Implantation rate (%) 21.7 6 30 11.7 6 23.7.051 FSH 5 Follicle-stimulating hormone; IVF 5 in vitro fertilization. Values are expressed as mean 6 SD. IVF. The effects of surgical removal of DIE lesions on fertility were previously addressed by 2 groups of investigators [14,15]. The excision of infiltrative uterosacral [14] and colorectal [15] lesions improved women s fertility. In contrast, our study aimed to evaluate the potential benefits of extensive laparoscopic excision of DIE before IVF. Although the data from this study showed that fewer oocytes were retrieved from women in group B than group A, fertilization rates and the number of top-quality embryos transferred were similar. Notwithstanding, a comparable number of embryos transferred resulted in greater implantation and pregnancy rates in the group that underwent surgery. This evidence adds to the benefit of surgery before IVF for women with DIE-associated infertility. Because the exact mechanisms by which endometriosis impairs fertility are still largely unknown, an explanation for the beneficial effects of excision of DIE on IVF outcomes as seen in our study would be at least speculative. Probably disease removal is associated with improvements in oocyte functional quality and embryo implantation. The observation that pregnancies after IVF occurred within an average of 14.6 6 12.5 months (3 18 months) of the extensive laparoscopic excision of DIE is of relevance. It might indicate that beneficial effects of surgery begins shortly after surgery and persists for a long period. Authors [34] showed that no difference exists in the outcomes of IVF treatments performed before or after 6 months after laparoscopy for endometriosis resection. Although these authors concluded that no effect of surgery existed on IVF outcomes, comparisons were made between 2 groups of patients that underwent surgery; moreover, implantation rates and continuing pregnancy rates were higher than those expected for patients with endometriosis [13]. Further, our work showed a positive benefit of surgery before IVF providing additional evidence to the recent review [35]. Patients who chose laparoscopy followed by IVF had previously undergone more IVF attempts than those in group A (2 vs 1) and had a longer duration of infertility. This could partially explain why some patients chose laparoscopy, because they had experienced more failures and desired to change therapies. However, if women with multiple failures and longer infertility duration had worse prognosis as a result of more severe disease, we possibly underestimated the positive value of extensive laparoscopy for removal of DIE. In conclusion, the results of this prospective cohort study showed that symptomatic infertile women with sonographic signs of DIE might benefit from extensive laparoscopic excision of endometriosis before proceeding to IVF. Ovarian responses are expected to be poorer with regard to the number of oocytes retrieved after endometriomas are removed, but overall pregnancy rates are presumed to be higher among those undergoing surgery. The authors acknowledge the valuable work of the embryologists, nurses, surgical technician (Mr. Jamil Fonseca), surgical team (Dr. Paula Beatriz Fettback, Dr. Claudia Messias Gomes, Dr. Thais Domingues), and Dr. Milton Reitzfeld

180 Journal of Minimally Invasive Gynecology, Vol 16, No 2, March/April 2009 for anesthesia expertise. The authors are also very grateful for the outstanding work of radiologists Dr. Luciana Pardini Chamie and Dr. Manuel Orlando Gonçalves. References 1. Eskenazi B, Warner ML. Epidemiology of endometriosis. Obstet Gynecol Clin North Am. 1997;24:235 258. 2. Strathy JH, Molgaard CA, Coulam CB, et al. Endometriosis and infertility: a laparoscopic study of endometriosis among fertile and infertile women. Fertil Steril. 1982;38:667 672. 3. D Hooghe TM, Debrock S, Hill JA, et al. Endometriosis and subfertility: is the relationship resolved? Semin Reprod Med. 2003;21:243 254. 4. De Hondt A, Meuleman C, Tomassetti C, et al. Endometriosis and assisted reproduction: the role for reproductive surgery? Curr Opin Obstet Gynecol. 2006;18:374 379. 5. Kennedy S, Bergqvit A, Chapron C, et al. ESHRE guideline for the diagnosis and treatment of endometriosis. Hum Reprod. 2005;20: 2698 2704. 6. Marcoux S, Maheux R, Berube S. Laparoscopic surgery in infertile women with minimal or mild endometriosis. N Engl J Med. 1997;337: 217 222. 7. Gruppo Italiano. Ablation of lesions or no treatment in minimal-mild endometriosis in infertile women: a randomized trial. Hum Reprod. 1999;14:1332 1334. 8. Jacobson TZ, Barlow DH, Koninckx PR, Olive D, Farquhar C. Laparoscopic surgery for subfertility associated with endometriosis. Cochrane Dat Syst Revs. 2002, Issue 4. Art. No.: CD001398. 9. Pal L, Shifren JL, Isaacson JB, et al. Impact of varying stages of endometriosis on the outcome of in vitro fertilization-embryo transfer. J Assist Reprod Genet. 1998;15:27 31. 10. Littman E, Giudice L, Lathi R, et al. Role of laparoscopic treatment of endometriosis in patients with failed in vitro fertilization cycles. Fertil Steril. 2005;84:1574 1578. 11. Somigliana E, Arnoldi M, Benaglia L, et al. IVF-ICSI outcome in women operated on for bilateral endometriomas. Hum Reprod. 2008; 23:1526 1530. 12. Barnhart K, Dunsmoor-Su R, Coutifaris C. Effects of endometriosis on in vitro fertilization. Fertil Steril. 2002;77:1148 1155. 13. Coccia ME, Rizzello F, Cammilli F, et al. Endometriosis and infertility surgery and ART: an integrated approach for successful management. Eur J Obstet Gynecol Reprod Biol. 2008;138:54 59. 14. Chapron C, Fritel X, Dubuisson JB. Fertility after laparoscopic management of deep endometriosis infiltrating the uterosacral ligaments. Hum Reprod. 1999;14:329 332. 15. Darai E, Marpeou O, Thomassin I, et al. Fertility after laparoscopic colorectal resection for endometriosis: preliminary results. Fertil Steril. 2005;84:945 950. 16. Remohi J, Vidal A, Pellicer A. Oocyte donation in low responders to conventional ovarian stimulation for in vitro fertilization. Fertil Steril. 1993;59:1208 1215. 17. Díaz I, Navarro J, Blasco L, et al. Impact of stage III-IV endometriosis on recipients of sibling oocytes: matched case-control study. Fertil Steril. 2000;74:31 34. 18. Serafini P. Outcome and follow-up of children born after IVF-surrogacy. Hum Reprod Update. 2001;7:23 27. 19. Suzuki T, Izumi S, Matsubayashi H, et al. Impact of ovarian endometrioma on oocytes and pregnancy outcome in in vitro fertilization. Fertil Steril. 2005;83:908 913. 20. De Hondt A, Peeraer K, Meuleman C, et al. Endometriosis and subfertility treatment: a review. Minerva Ginecol. 2005;57:257 267. 21. The Practice Committee of the American Society for Reproductive Medicine. Intracytoplasmic sperm injection (ICSI). Fertil Steril. 2006; 86(Suppl):S122. 22. Bazot M, Darai E. Sonography and MR imaging for the assessment of deep pelvic endometriosis. J Minim Invasive Gynecol. 2005;12: 178 185. 23. Bazot M, Malzy P, Cortez A, et al. Accuracy of transvaginal sonography and rectal endoscopic sonography in the diagnosis of deep infiltrating endometriosis. Ultrasound Obstet Gynecol. 2007;30:994 1001. 24. Abrao MS, Gonçalves MO, Dias JA Jr, et al. Comparison between clinical examination, transvaginal sonography and magnetic resonance imaging for the diagnosis of deep endometriosis. Hum Reprod. 2007; 22:3092 3097. 25. Pereira RMA, Zanatta A, Reich H, et al. Use of circular stapler for laparoscopic excision of rectosigmoid anterior wall endometriosis. Surg Tech Int. 2008;17:181 186. 26. Serafini P, Yadid I, Motta EL, et al. Ovarian stimulation with daily late follicular phase administration of low-dose human chorionic gonadotropin for in vitro fertilization: a prospective, randomized trial. Fertil Steril. 2006;86:830 838. 27. Aboulghar MA, Mansour RT, Serour GI, et al. The outcome of in vitro fertilization in advanced endometriosis with previous surgery: a casecontrolled study. Am J Obstet Gynecol. 2003;188:371 375. 28. Male Infertility Best Practice Policy Committee of the American Urological Association; Practice Committee of the American Society for Reproductive Medicine. Report on optimal evaluation of the infertile male. Fertil Steril. 2004;82(Suppl1):S123 S130. 29. Wieand S, Murphy K. A commentary on treatment at random: the ultimate science or the betrayal of Hippocrates? J Clin Oncol. 2004; 22:5009 5011. 30. Gerber S, Ferreira GP, Spyer Prates LF, et al. Effects of previous ovarian surgery for endometriosis on the outcome of assisted reproduction treatment. Reprod Biomed Online. 2002;5:162 166. 31. Garcia-Velasco JA, Mahutte NG, Corona J, et al. Removal of endometriomas before in vitro fertilization does not improve fertility outcomes: a matched, case-control study. Fertil Steril. 2004;81:1194 1197. 32. Esinler I, Bozdag G, Aybar F, et al. Outcome of in vitro fertilization/intracytoplasmic sperm injection after laparoscopic cystectomy for endometriomas. Fertil Steril. 2006;85:1730 1735. 33. Demirol A, Guven S, Bayak C, et al. Effect of endometrioma cystectomy on IVF outcome: a prospective randomized study. Reprod Biomed Online. 2006;12:639 643. 34. Surrey ES, Schoolcraft WB. Does surgical management of endometriosis within 6 months of an in vitro fertilization-embryo transfer cycle improve outcome? J Assist Reprod Genet. 2003;20:365 370. 35. Catenacci M, Falcone T. The effect of endometriosis on in vitro fertilization outcome. Minerva Ginecol. 2008;60:209 221.