Background
Many women who experience endometriosis and endometriomas also encounter problems with fertility.
Objective
The purpose of this study was to determine the impact of surgical excision of endometriosis and endometriomas compared with control subjects on ovarian reserve.
Study Design
This was a prospective cohort study of 116 women aged 18–43 years with pelvic pain and/or infertility who underwent surgical treatment of suspected endometriosis (n=58) or endometriomas (n=58). Based on surgical findings, the suspected endometriosis group was further separated into those with evidence of peritoneal disease (n=29) and those with no evidence of endometriosis (n=29). Ovarian reserve was measured by anti-Müllerian hormone and compared before surgery and at 1 month and 6 months after surgery.
Results
Baseline anti-Müllerian hormone values were significantly lower in the endometrioma vs negative laparoscopy group (1.8 ng/mL [95% confidence interval, 1.2–2.4 ng/mL] vs 3.2 ng/mL [95% confidence interval, 2.0–4.4 ng/mL]; P <.02), but the peritoneal endometriosis group was not significantly different than either of these groups. Only patients with endometriomas had a significant decline in ovarian reserve at 1 month (–48%; 95% confidence interval, –54 to –18%; P <.01; mean anti-Müllerian hormone baseline value, 1.77–1.12 ng/mL at 1 month). Six months after surgery, anti-Müllerian hormone values continued to be depressed from baseline but were no longer significantly different. The rate of anti-Müllerian hormone decline was correlated positively with baseline preoperative anti-Müllerian hormone values and the size of endometrioma that was removed. Those with bilateral endometriomas (n=19) had a significantly greater rate of decline (53.0% [95% confidence interval, 35.4–70.5%] vs 17.5% [95% confidence interval, 3.2–31.8%]; P =.002).
Conclusion
At baseline, patients with endometriomas had significantly lower anti-Müllerian hormone values compared with women without endometriosis. Surgical excision of endometriomas appears to have temporary detrimental effects on ovarian reserve.
Endometriosis is a common condition in reproductive-aged women. Although it can be asymptomatic, many women with endometriosis experience pain and/or infertility. Over one-third of women with endometriosis encounter issues with infertility, which suggests an association, although there has yet to be empiric evidence that indicates a causal relationship. Because hormonal suppression is counterproductive, surgical removal of endometriosis has been the mainstay treatment for patients with endometriosis who desire conception.
When there is ovarian involvement of endometriosis, the cysts that form are termed endometriomas . The mere presence of an endometrioma has been proposed to contribute to a decrease in ovarian reserve, with a potential negative effect on fertility. Although complete cystectomy is the procedure of choice because of decreased recurrence rates, there is concern that it may further damage healthy ovarian tissue and cause an additional decrease in ovarian reserve. As such, the removal of endometrioma for fertility purposes remains controversial.
Although there is no absolute measure of ovarian reserve, the relatively recent addition of anti-Müllerian hormone (AMH) testing as a diagnostic fertility assessment has proved to be a reliable surrogate marker of ovarian reserve with an established correlation with age, antral follicle count, and response to controlled ovarian hyperstimulation. An advantage of measuring AMH over other techniques is that it remains stable through the menstrual cycle.
The effect of pelvic endometriosis on AMH levels remains uncertain. The effect of surgical excision of ovarian endometriomas on ovarian reserve has also shown mixed results. However, most studies that have studied AMH levels after endometrioma removal have reported a decrease in ovarian reserve with a greater effect noted with bilaterality.
A recent prospective study of the removal of endometriomas on AMH levels compared with control subjects without ovarian cysts showed decreased preoperative levels in women with endometriomas and a further decline after surgical removal. However, there is a paucity of literature that compares patients with the same complaints with suspected pelvic peritoneal endometriosis with those with ovarian endometriomas. Furthermore, the published studies with control groups thus far have not recruited patients from the same subfertile population. Although many patients delay or forgo surgical management of endometriosis and/or small endometriomas (<4 cm) in favor of expediting fertility treatments, this study aims to focus on symptomatic patients who were faced with the decision to undergo surgery.
The goal of this prospective cohort trial was to compare baseline AMH levels in women with endometriomas, pelvic peritoneal endometriosis, or negative laparoscopies and to determine the impact of surgical treatment on short- and long-term ovarian reserve.
Materials and Methods
This was a prospective cohort study that was approved by the Institutional Review Board at the Cleveland Clinic; all patients gave written informed consent. Patients between the ages of 18-43 years who were scheduled to undergo laparoscopic surgery for a chief complaint of pelvic pain and/or infertility were recruited consecutively from the Reproductive Endocrinology and Infertility clinic between October 2013 and May 2015. Those patients in the endometrioma arm had preoperative ultrasonographic imaging that was consistent with endometrioma, and ovarian cystectomy was planned. The control arm consisted of patients with similar complaints of pelvic pain and/or infertility with suspected endometriosis, no evidence of endometrioma, and a plan for surgical management. Patients were excluded if they had had previous ovarian surgery.
The surgeries were performed laparoscopically, primarily by 3 reproductive surgeons (T.F., J.M.G., and R.L.F.) with extensive experience in laparoscopic treatment of endometriosis. All endometriomas were excised completely. In most cases, dilute vasopressin (20 units/100 mL saline solution) was injected between the ovarian cortex and cyst wall for hemostasis and to develop the surgical plane. An incision was made in the ovarian cortex over the endometrioma, and cystectomy was performed with the use of blunt and sharp dissection with traction and countertraction. The ovarian defects were made hemostatic by minimal use of bipolar electrosurgical energy, with the additional use of hemostatic agents if needed, but were not closed by suture or any other method. The endometriomas were identified by the classic intraoperative appearance and confirmed by histologic diagnosis.
The control group was separated into 2 arms based on surgical findings. The first included patients with evidence of pelvic peritoneal endometriosis. Endometriosis was staged with the use of the American Society of Reproductive Medicine staging criteria, and excised endometriosis specimens were confirmed histologically. The second control arm consisted of patients without laparoscopic or histologic evidence of endometriosis.
The primary outcomes were baseline serum AMH levels and declined over 1- and 6-month follow-up periods. Secondary outcomes included length of surgery, stage of endometriosis disease, and size and laterality of the endometriomas.
Measurement of AMH as a marker of ovarian reserve
Baseline AMH values were obtained from venous blood samples, either at the preoperative appointment or immediately before surgery. Postoperative AMH values were obtained at the postoperative visit approximately 1 month after surgery. A final blood sample was then obtained 6 months after surgery. All samples were spun, and the resulting serum was frozen at –70°C for later analysis. The samples were resulted in batches. Serum AMH was determined by a commercially available quantitative 3-step sandwich-type immunoassay kit (picoAMH ELISA; Ansh Labs, Webster, TX). The lower limit of detection was 1.2 pg/mL. The intra- and interassay coefficients of variation for the AMH assay were 3.7% and 4.5%, respectively.
Statistical methods
Quantitative values were compared for both demographic and outcome data. Parametric continuous data were compared with the use of the Student t test, and proportions were compared with the use of the Chi-square and Fisher’s Exact tests, where appropriate. Nonnormally distributed data were defined with median and interquartile range. Preoperative AMH values were compared between women with endometriomas and those without with the use of the Mann-Whitney U test and among the 3 groups with the use of analysis of variance. Pre- and postoperative AMH values were compared by Wilcoxon signed rank test. A subgroup analysis was performed that was based on laterality of endometriomas (unilateral vs bilateral).
The rate of decline in AMH was calculated based on the preoperative and 1- and 6-month postoperative AMH values; this value was expressed as a percentage. Linear regression analysis was used to create a model to control for potential confounding factors for patient and surgical outcome characteristics and to identify factors that were associated with the rate of decline of AMH 6 months after surgery. A probability value of <.05 was considered statistically significant.
Sample size justification
The sample size for this study was determined by the primary outcome of magnitude of decline in serum AMH levels over the 6-month study period. The following power analysis was based on results from Uncu et al. Using baseline AMH values of subjects with and without endometriomas, we assumed a conservative value for the standard deviation of the difference of 2.65 ng/mL, a 5% significance level, and a paired t -test. To reach an 80% power with a difference in AMH of 1 ng/mL, we estimated that 58 subjects with endometriomas should be recruited and that, if equal size groups of those with and without endometriomas were to be collected, a total sample size of 116 subjects should be included.
Results
The study population included 58 women with endometriomas and 58 control subjects. All patients in the endometrioma group had confirmation of endometrioma at the time of surgery; most of them (52/58 patients; 89.6%) also had evidence of pelvic peritoneal disease. The control group was then further segregated after surgery to those with pelvic peritoneal endometriosis (n=29) and those with no evidence of endometriosis (n=29). Of those with no evidence of endometriosis, there were 8 negative laparoscopies, 12 with peritoneal biopsy specimens consistent with fibrosis and/or chronic inflammation, 6 with adhesive disease, and 3 with evidence of pelvic inflammatory disease.
There was no difference in age, body mass index, indication for surgery, parity, smoking status, a history of surgery or oral contraceptive use between endometrioma and the overall control group, between each of the control groups (pelvic peritoneal endometriosis and no evidence of endometriosis) and among the 3 groups. There were significantly more white women in the control group than the endometrioma group ( Table 1 ). Overall, age was the only demographic factor that correlated with baseline preoperative AMH values ( r =-.33; P <.01).
| Variable | Endometrioma (n=58) | Control (n=58) | P value |
|---|---|---|---|
| Age, y a | 32.1±5.6 | 30.5±5.3 | .10 |
| Race, n (%) | .01 | ||
| White | 37 (64) | 48 (83) | |
| African American | 12 (21) | 7 (12) | |
| Asian | 6 (10) | 0 | |
| Other | 3 (5) | 3 (5) | |
| Body mass index, kg/m 2 a | 26.7±5.3 | 26.7±7.0 | .94 |
| Indication for surgery, n (%) | .57 | ||
| Pelvic pain | 32 (55) | 28 (48) | |
| Infertility | 4 (7) | 5 (9) | |
| Both | 22 (38) | 25 (43) | |
| Nulliparous, n (%) | 48 (83) | 39 (67) | .09 |
| Smoker (current), n (%) | 2 (3) | 6 (10) | .27 |
| Previous laparoscopy, n (%) | 19 (33) | 28 (48) | .09 |
| Current oral contraceptive pills use, n (%) | 4 (7) | 7 (12) | .34 |
| Recent gonadotropin-releasing hormone agonist use, n (%) | 3 (5) | 6 (11) | .49 |
Comparison of ovarian reserve between women with and without endometrioma
Women with endometriomas had significantly lower baseline AMH values compared with all control subjects (1.77 ng/mL [95% confidence interval (CI), 1.18–2.37] vs 2.75 ng/mL [95% CI 1.98 – 3.51]; P =.05). For the control group, surgery did not have an effect on the 1- and 6-month postoperative AMH levels. However, after endometrioma cystectomy, there was a significant drop in ovarian reserve at 1 month. The levels rose by the sixth month but remained nonsignificantly lower than at baseline ( Table 2 ).
| Variable | Endometrioma (n=58), mg/mL a | Control (n=58), mg/mL a | P value |
|---|---|---|---|
| Baseline | 1.77 (1.18–2.37) | 2.75 (1.98–3.51) | .05 |
| 1 Month | 1.12 (0.81–1.45) <.01 | 2.86 (2.04–3.67) .49 | <.01 |
| 6 Months | 1.41 (0.97–1.85) .22 | 2.93 (2.09–3.77) .21 | <.01 |
Comparison of ovarian reserve among women with endometriomas, pelvic peritoneal endometriosis, and negative control subjects
There was no difference in the baseline AMH levels between the endometrioma and the endometriosis groups ( P =.38) or between the endometriosis and the negative control groups ( P =.19). However, those patients with endometriomas had significantly lower baseline AMH values than did negative control subjects ( P =.02; Table 3 ; Figure ). This difference persisted at 1 and 6 months. In addition, although there was no difference at baseline, women with endometriomas had significantly lower AMH values than those with endometriosis at the 1- and 6-month follow up ( P =.03 and .03, respectively). At 1 and 6 months, there remained no difference between the AMH values of subjects with pelvic peritoneal endometriosis compared with those without endometriosis ( P =.16 and .59, respectively). Length of surgery was significantly longer in the endometrioma group, followed by the pelvic peritoneal endometriosis group, and was shortest in the negative group (123.2±45.4 min vs 88.8±48.4 min vs 71.1±31.2 min; P <.01). The use of hemostatic agents did not differ among groups ( P =.09).
| Variable | Endometrioma (n = 58), mg/mL a | Pelvic peritoneal endometriosis (n = 29), mg/mL a | No endometriosis (n = 29), mg/mL a | P value |
|---|---|---|---|---|
| Baseline | 1.77 (1.18–2.37) | 2.29 (1.34–3.25) | 3.20 (1.96–4.43) | .06 |
| 1 Month | 1.12 (0.81–1.45) <.01 | 2.38 (1.26–3.50) .41 | 3.22 (2.04–4.49) .78 | <.01 |
| 6 Months | 1.41 (0.97–1.85) .22 | 2.76 (1.58–3.95) .12 | 3.14 (1.83–4.43) .67 | .01 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
