Polycystic ovarian syndrome (PCOS) is the most common endocrinopathy among reproductive-aged women [1]. In 1935, when PCOS was originally defined by Stein and Leventhal [2], seven women with enlarged ovaries, menstrual disturbances $\pm$ hirsutism, were treated with bilateral ovarian wedge resection. After treatment, all had regular cycles and five of them conceived during the follow-up period. In 1984, 62 women with PCOS were treated by laparoscopic systematic electrocautery of the ovarian capsule. Following treatment, ovulation occurred in 92% of the patients and pregnancy was achieved in 69% of the women who wished to conceive [3].

Despite a significant improvement in ovulation and pregnancy rates after LOD, the associated complications, risk of post-operative adhesions, significant decrease in ovarian reserve markers, and later the introduction of clomiphene citrate, have limited its use in recent years [4]. Currently, although clomiphene citrate (CC) or letrozole is recommended as first-line treatment for women with PCOS, laparoscopic ovarian drilling (LOD) as an alternative approach to exogenous gonadotropin, might still be preferred in the context of second-line treatment, particularly for women who fail to ovulate after taking antioestrogen pills [3, 4, 5]. The advantages of LOD include a lower risk of higher-order multiple gestations, avoidance of ovarian hyperstimulation syndrome (OHSS), no requirement to monitor follicular growth, and the persistence of ovulatory cycles for many years without any medication [5]. In addition, LOD might reduce the cost of reaching an ongoing pregnancy with a time horizon of 12 months [6].

Although several follow-up studies have provided evidence for short-term reproductive and endocrinological benefits of LOD [7, 8, 9], there are some data that indicate a decline in ovarian reserve markers [10, 11] and one report describes severe ovarian atrophy after the use of a high level of energy [12]. Therefore, the amount of energy applied during the procedure might be critical to produce a treatment effect without impairing the ovarian reserve. Nevertheless, in the recent international evidence-based guidelines for the assessment and management of PCOS, the editors highlighted the small risk of reduced ovarian reserve or loss of ovarian function [13]. In the current study, we aimed to evaluate the fertility performance and menopause status of women with PCOS who had been treated with LOD with a median (minimum to maximum) time period of 27 (25–27) years ago.

The median female age at the time of current analysis was 52 (range 46–64) years. The median age at the time of LOD was 25 (range 1935) years. When excluded cases were also analysed for their demographic features in order to confirm the representativity of our included cases, the median ages of the whole group were 26 (range 20–37) and 55 (range 45–66), respectively. These values are similar to our included cases.

The median body mass index was 29.1 (range 25.4–36.8) kg/m${}^2$ at the time of the current analysis. Of the 11 women, patients 1 and 9 had surgical menopause at the age of 35 and 47, respectively. Notably, patient number 1 had abdominal hysterectomy due to leiomyoma and prophylactic salpingo-oophorectomy. Patient number 9 had abdominal hysterectomy and bilateral salpingo-oophorectomy due to bilateral endometrioma and chronic pelvic pain (Table 1). Of the remaining nine women, four experienced natural menopause when they were between 45 to 52 years (Table 1). Although five patients were still having regular cycles at the age of 46 to 54 years at the time of the study, we do not have any biochemical data to evaluate their ovulatory status. The median menopausal age of their mothers was 48 (range 45–55) years.

Of the patients wishing to conceive, two had undergone assisted reproduction treatment due to failure to conceive after the LOD. Of these, one woman had conceived and one woman had failed to conceive. Nine patients had conceived naturally (Table 1). The corresponding number of gravidity and parity for each patient are also given in Table 1.

As already reported, patients with PCOS may reach menopausal status later than otherwise healthy women [14]. According to one study which included 27 women that had been diagnosed with PCOS 24 years earlier, whereas the mean menopausal age was 53.3 $\pm$ 2.2, it was 49.3 $\pm$ 2.3 years in the control group (p 0.01) [15]. In the current study, we saw a similar menopause age after a long-term follow-up among patients treated with LOD, compared with patients who had not undergone any ovarian surgery [15]. According to the available data, a decline in fertility begins around the age of 37–38, and menopause follows approximately 13 years later at, on average, the age of 51 years [16]. Nevertheless, in our study cohort with a median age of 52, four patients had natural menopause between the ages of 45 to 52 years, and five patients were still having regular cycles at 45 to 53 years. The favourable fecundity performance that was achieved spontaneously in this cohort was another significant finding that might encourage the use of LOD in well-selected cases.

As already established, PCOS is the most frequent endocrinopathy among reproductive-aged women [1]. In 2010, a total of 116 million women worldwide were diagnosed with PCOS, according to the World Health Organization [17]. For anovulatory women with PCOS, antioestrogen pills have been recommended as first-line treatment for infertility in addition to lifestyle management [18]. In recent years, this first-line treatment has tended to be letrozole due to the associated live birth rate as compared with clomiphene citrate (OR 1.68, 95% CI: 1.42 to 1.99) [19]. However, in patients who fail to ovulate or fail to conceive with antioestrogen pills, either LOD or exogenous gonadotropin treatment might be employed as second-line treatment [18]. When compared with exogenous gonadotropin treatment, the main advantages of LOD include lower multiple pregnancy rates [20], reduced risk of cycle cancellation due to hyperstimulation, and no requirement to monitor follicular growth [21]. Furthermore, according to an economic analysis of a randomized sample of 168 women, LOD might significantly yield a lower cost per first live birth when compared with exogenous FSH treatment (mean difference €3247; 95% CI: €650–€5814, p 0.05) [9].

Despite the well-documented advantages of LOD, a possible decrease in ovarian reserve markers is a disadvantage of the intervention that might obscure the future fertility potential. A recent meta-analysis of seven studies consistently showed a statistically significant fall in serum AMH concentration after LOD (WMD -2.13 ng/mL; 95% confidence interval (CI) -2.97 to -1.30) [10]. This was irrespective of duration of follow-up, type of AMH assay, laterality of surgery, and amount of energy applied during LOD. There are also reports under systematic review that depict a statistically significant difference between Day 3 FSH, inhibin B levels, ovarian volume and antral follicle count before and after LOD [11]. However, whilst LOD seems to significantly influence ovarian reserve markers, it remains uncertain whether this reflects real damage to ovarian reserves or normalization of high pre-operative serum AMH levels that might inherently inhibit initial follicle recruitment, FSH-dependent growth and selection of preantral follicles. In this context, the amount of energy applied during the procedure might be critical in order to observe the effect of treatment without impairing the ovarian reserve. The amount of thermal energy (J) used in LOD can be calculated by multiplying power (W) by duration (s) for each puncture [11]. When this formula is applied to the available data, it can be seen that the energy level used varies between surgeons. However, severe ovarian atrophy has been reported after 16,000 J (8 holes x 400 W x 5 s) [12].

Other than a post-interventional drop in ovarian reserve markers, postoperative adhesion formation has also been suggested as a condition that might negatively affect the likelihood of conception. In an earlier study of 40 anovulatory women with CC resistance, patients were randomly assigned to have either (i) second-look laparoscopy with lysis of adhesions within three to four weeks of the initial laparoscopy or (ii) expectant management [22]. In the second-look laparoscopy arm, 14 patients had minimal to moderate adhesion formation and three patients had none. Notably, there were no cases with severe adhesion. Of interest, when laparoscopic adhesiolysis was performed with second-look laparoscopy, the subsequent conception rates within six months were comparable with the expectant management arm (47% vs 55%, p $>$ 0.05) [22]. These limited data suggest that although there might be some degree of adhesion formation after LOD, it does not appear to be severe and does not have a detrimental effect on the overall conception rate.

The obvious limitation of the current study is the small sample size. Since we aimed to evaluate the time of menopause in patients who had undergone LOD, we enrolled patients who were the first to receive surgery at our institute. Therefore, we were only able to reach a limited number of patients via patient files and contact numbers. Although the median (minimum–maximum) age of the whole group (n = 83) at the time of surgical procedure were similar among patients who were analysed (n = 11), the representativity of the study group should be questioned and expressed as another limitation in addition to the retrospective design, possibility of recall bias and self-reported data on menopausal status without blood work-up. However, since there have been no published prospective long-term LOD follow-up studies, we believe that our retrospective data is valuable with respect to menopausal age and conception rate. With regard to the absence of a control group, since LOD is usually performed on clomiphene-failed patients, we could not identify a control group of women in which exogenous gonadotropin was preferred as an alternative strategy in the second step. Since the computer-based file system and ICD coding were not available 29 years ago, it was not possible to properly screen and identify those patients. Whilst we could have used a control group comprising otherwise healthy women who had been admitted to the outpatient clinic during the same period, the measurement of androgens and examination with ultrasonography was missing in those patients and therefore it was not possible to exclude the risk of PCOS.

In conclusion, our results suggest that patients undergoing LOD appear to have favourable conception rates and the procedure itself does not have any negative effects on menopausal age when compared with the data obtained from large-scale cohort studies.

LOD, Laparoscopic ovarian drilling; PCOS, polycystic ovarian syndrome; AMH, anti-Müllerian hormone; CC, clomiphene citrate; POF, premature ovarian failing.

GÖ, GB and SM designed the research study. FA and PC provided help in the data collection. GB, GÖ and HY wrote the manuscript and acted in editing. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

All patients gave written informed consent to participate in the study and protocol was approved by Hacettepe University Non-Interventional Clinical Research Ethics Board with the registration file number of 2019/07 (GO–19/253).

We would like to express our gratitude to all the peer reviewers for their opinions and suggestions.

This research received no external funding.

The authors declare no conflict of interest.