To our knowledge, this is the first prospective study to examine the effect of ovarian puncture for retrieval of immature oocytes on the endocrine profile in patients with PCOS undergoing IVM treatment. The results of our study suggest that the effect of oocyte retrieval of immature oocytes on the endocrine profile of these patients is significant but transient and that ovarian puncture for IVM does not offer long-term prospects of an alteration of the endocrine profile in patients with PCOS, as observed after LOD. Although the exact mechanism of action of LOD is unknown, it is likely that multiple ovarian punctures may destroy ovarian androgen-producing tissue and by doing so result in reduced peripheral conversion of androgen into oestrogen . This is further documented by the observation that LOD can restore ovulation, results in good pregnancy rates [14, 15] and significantly reduces AMH levels , LH/FSH ratio, serum concentration of LH, testosterone and free androgen index. The effect of LOD on these endocrine parameters often persists for many years after the procedure .
Although gonadotropin doses in IVM cycles are typically much lower than those administered in conventional ART (c-ART) cycles, clinical protocols for IVM often include mild stimulation using low-dose gonadotropins and/or oral or transdermal estrogens [18–20] to increase endometrial thickness. In our study, patients received 14.7 ± 2.5 days of transdermal oestradiol and 450 IU HP-hMG cumulatively for endometrial priming before oocyte retrieval, resulting in mean serum E2 levels (1,098 ± 1,831 ng/L) on the day of OR-IVM. The observation that serum AMH levels were significantly reduced on the day of IVM-OR (p = 0.005) is in accordance with previous studies showing reduced AMH levels after controlled ovarian stimulation for c-ART [21, 22]. Nevertheless, although in c-ART cycles the number of small antral follicles producing AMH decreases as follicles progressively grow in response to FSH stimulation, this mechanism does not occur in IVM-cycles: in the study presented here, IVM-OR was performed when follicles had a maximal mean diameter of 10 mm. Because previous studies have shown that FSH increases estradiol levels and suppresses AMH secretion [23, 24], we propose that the observed significant reduction of circulating AMH levels on the day of IVM-OR is caused by increased serum estradiol levels secondary to exogenous administration of estradiol and gonadotropins. Elevated estradiol levels also enhance hepatic SHBG synthesis and secretion, resulting in increased SHBG levels on the day of IVM-OR. TT values were also increased after mild ovarian stimulation for IVM (p = 0.002). This might be explained by the fact that the amounts of LH and hCG present in HP-hMG formulations are capable of stimulating androgen substrate production from theca cells .
Taking into account that decreased AMH biosynthesis following LOD restores ovulatory function through enhanced follicular sensitivity to circulating FSH levels, we speculated that a similar effect might occur following immature oocyte retrieval from small follicles. LOD has been proven to significantly reduce circulating AMH levels , with values remaining significantly reduced up to 6 months after the procedure . The mechanic effect of LOD on polycystic ovaries may reside in destruction of small antral and pre-antral follicles, besides stroma, resulting in a relatively long term impact on circulating levels of AMH, LH, and total and free testosterone . As a result of these changes, increased ovulation and pregnancy rates have been documented after LOD in patients with PCOS. Transvaginal aspiration of small antral follicles has been proposed as a method to induce ovulation in anovulatory patients with PCO : in a series of 18 patients, Mio et al. demonstrated a mean ovulation rate of 52.6% in subsequent cycles after multiple antral follicular punctures. Using a similar approach, Ferraretti et al. performed transvaginal ovarian puncture in PCOS patients to improve IVF results and named this procedure “transvaginal ovarian drilling” . In a case-control study in 42 patients with PCOS who had IVM treatment before conventional IVF and 48 patients with PCOS who had not had IVM before, significantly more oocytes and embryos in subsequent IVF cycles were obtained in patients who had had IVM treatment . Moreover, Frantz et al. reported three spontaneous pregnancies in women with PCOS after transvaginal ovarian punctures for IVM. Although this is small case report series the authors concluded that multiple ovarian punctures for IVM in PCOS patients may have contributed to their pregnancy in the months following the IVM procedure . However, although the above observations imply that multiple punctures for IVM may have an effect similar to that of LOD, no study has been conducted to provide a biological explanation for this effect. In the current study we examined the true impact of transvaginal puncture of minimally stimulated polycystic ovaries on the endocrine profile and conclude that the effect is only temporary: serum AMH concentrations had returned to levels before oocyte retrieval within two weeks after IVM-OR and the reduction of androgen levels had disappeared within three months after IVM-OR, at least in the subset of patients who had a follow-up at three months. This rather short-term effect compared to LOD might be explained by the relatively less invasive nature of ovarian puncture for the retrieval of immature oocytes, probably because of the absence of thermal energy. Hence, the new wave of developing small antral follicles that is observed as early as five days after IVM-OR  will result in restored steroidogenesis and AMH synthesis soon after the procedure.
In spite of its prospective design, the current study has several limitations. First, serial hormone analysis was restricted to the IVM cycle itself and only a small subset of patients were available for hormone profiling at three months’ follow-up. After one IVM cycle, the majority of patients in this study were not pregnant and proceeded with reproductive treatment shortly after this failed treatment cycle. The observation, at least in the small subset of five patients available for follow-up, that the endocrine profile was unchanged at three months’ follow-up, leads us to suggest that it is highly unlikely that the results would have been different at longer intervals. However, follow-up data were only available for five patients, which significantly limits the ability to generalize these results to a larger population.
Secondly, this study only focused on markers of the endocrine profile; the effect of ovarian puncture on ovulatory function and clinical outcomes following IVM-OR could not be investigated within the scope of this study, since the majority of patients decided to have further ART cycles shortly after the study cycle, rather than await potential effects of treatment on endocrine or ovulatory function.
Furthermore, the effect of multiple ovarian punctures for IVM on the endocrine profile described in this study has not been compared to the effect of a regular oocyte retrieval in women with PCOS undergoing a c-ART. A prospective study should be conducted in order to confirm the findings of the current study.
Our study cohort consisted of patients with PCOS, who were generally overweight or obese (mean BMI 30.3 ± 9.0 kg/m2) but who had relatively low serum levels of LH levels (7.9 ± 4.4 IU/L) and normal markers of hyperandrogenism (TT 0.36 ± 0.13 nmol/L, FTc 5.0 ± 3.6 pmol/L and SHBG 74.6 ± 59.1 nmol/L). In a study investigating parameters that are predictive of ovulation rates and pregnancy in 200 patients with PCOS who underwent LOD, it was shown that morbid obesity and high levels of androgens, among other parameters, are associated with low chances of response to LOD . According to these findings, the patients in our study would be predicted good responders to LOD in terms of ovulatory function and clinical outcomes. Again, long-term follow-up to evaluate the effect of oocyte retrieval for IVM on ovulation rates and pregnancy could not be performed in our patients.