In vitro maturation in subfertile women with polycystic ovarian syndrome undergoing assisted reproduction.

Background: Polycystic ovarian syndrome (PCOS) occurs in 8% to 13% of all women of reproductive age and 50% of women presenting with infertility (i.e. inability to reach a pregnancy after 12 months or more of regular unprotected sexual intercourse). A proportion of these women ultimately need assisted reproductive technology. In vitro fertilisation (IVF)/intracytoplasmic sperm injection (ICSI) are assisted reproduction techniques used to raise the chances of a pregnancy. In women with PCOS, the supra-physiological doses of gonadotrophins used for controlled ovarian hyperstimulation (COH) often result in an exaggerated ovarian response characterised by the development of a large cohort of follicles of uneven quality, retrieval of immature oocytes, and increased risk of ovarian hyperstimulation syndrome (OHSS). A potentially effective intervention for women with PCOS-related infertility involves earlier retrieval of immature oocytes at the germinal-vesicle stage followed by in vitro maturation (IVM). This is the third update of this Cochrane review on the subject (after the last update on 27 June 2018).

Objectives: To assess the benefits and harms of IVM followed by IVF or ICSI versus conventional IVF or ICSI among women with PCOS.

Search methods: On 27 February 2023, we searched the Cochrane Gynaecology and Fertility Group Specialised Register of Controlled Trials, CENTRAL, MEDLINE, Embase, and the Open Grey database. We further searched the National Institute for Health and Care Excellence (NICE) fertility assessment and treatment guidelines. We also searched reference lists of relevant papers and Google Scholar for any additional trials.

Selection criteria: We included randomised controlled trials (RCTs) comparing IVM before IVF or ICSI with conventional IVF or ICSI for infertile women with PCOS, irrespective of language and country of origin.

Data collection and analysis: Two review authors independently selected studies, assessed the risk of bias, extracted data from studies, and, where needed, attempted to contact the authors for missing data. Our primary outcomes were live birth per woman randomised and miscarriage. We performed statistical analysis using Review Manager. We assessed the certainty of the evidence using GRADE and the risk of bias using the Cochrane RoB 2 tool.

Main results: We found four published trials suitable for inclusion in this update. The studies involved 810 subfertile women undergoing assisted reproductive technology. Two of four were already included in the previous version of the review, were published as abstracts in international conferences, and were at high risk of bias. The two new studies were at low risk of bias in all domains and in terms of all outcomes. We implemented the random-effects model for the quantitative analyses and restricted the primary analysis to studies at low risk of bias in all domains. We are very uncertain about the effect of IVM or capacitation IVM (a new biphasic IVM system improving the developmental competence of oocytes) on live birth when compared to IVF when a GnRH antagonist protocol was applied (odds ratio (OR) 0.47, 95% confidence interval (CI) 0.17 to 1.32; I² = 91%; 2 studies, 739 participants; very low-certainty evidence). This suggests that if the chance of live birth following standard IVF is assumed to be 45.7%, then the chance of IVM would be 12.5% to 52.6%. In contrast, IVM or capacitation IVM increases miscarriage per clinical pregnancy (where clinical pregnancy was defined as evidence of a fetal heart beat on ultrasound at seven gestational weeks) in women with PCOS when compared to IVF (OR 1.66, 95% CI 1.02 to 2.70; I² = 0%; 2 studies, 378 clinical pregnancies; high-certainty evidence). This suggests that if the chance of miscarriage following standard IVF is assumed to be 20.1%, then the chance using IVM would be 20.4% to 40.4%. Results remained similar when using the risk ratio (RR) as the measure of effect. We are uncertain about the effect of IVM or capacitation IVM on clinical pregnancy when compared to IVF when a GnRH antagonist protocol was applied (OR 0.49, 95% CI 0.14 to 1.70; I² = 94%; 2 studies, 739 participants; very low-certainty evidence). The results were similar after pooling the RRs. IVM or capacitation IVM results in a large reduction in the incidence of moderate or severe OHSS as compared to IVF when a GnRH antagonist protocol was applied (OR 0.08, 95% CI 0.01 to 0.67; I² = 0%; 2 studies, 739 participants; high-certainty evidence). This suggests that if the incidence of OHSS following IVF is assumed to be 3.5%, then the incidence with IVM would be 0% to 2.4%. Also, there is probably little to no difference in preterm birth between IVM or capacitation IVM and IVF after the application of a GnRH antagonist protocol (OR 0.69, 95% CI 0.31 to 1.52; I² = 45%; 2 studies, 739 participants; moderate-certainty evidence). As for congenital anomalies, one study reported no events, while another showed an uncertain effect of IVM (OR 0.33, 95% CI 0.01 to 8.24; 1 study, 351 participants; low-certainty evidence). Results remained similar when using the RR as the measure of effect. There were no data from any of the studies for cycle cancellation, oocyte fertilisation, or subgroup analyses.

Authors' conclusions: There is continuous scientific interest in IVM, and promising data have been published. Concerning live birth and clinical pregnancy, we are very uncertain about the effect of the technique when compared to IVF after using a GnRH antagonist protocol. In contrast, high-certainty evidence shows that IVM increases miscarriage per clinical pregnancy and reduces the incidence of moderate or severe OHSS in women with PCOS compared to IVF after a GnRH antagonist protocol. Regarding the rest of the outcomes, low- to moderate-certainty evidence showed little to no difference in preterm birth and risk of congenital anomalies between the two modalities. We eagerly anticipate further evidence from high-quality trials in the field (we found five ongoing trials).