Effect of pretreatment with oral contraceptives and progestins on IVF outcomes in women with polycystic ovary syndrome.

STUDY QUESTION: Do oral contraceptives (OCs) and progestins impact live birth rate of IVF when used for cycle scheduling in women with polycystic ovary syndrome (PCOS)? SUMMARY ANSWER: OCs used for scheduling IVF cycle were associated with lowered rates of pregnancy and live birth after fresh embryo transfer, whereas progestins used for this purpose yield higher rates of pregnancy and live birth than OCs. WHAT IS KNOWN ALREADY: Due to oligo-menorrhea in PCOS, OCs and progestin are extensively used to schedule the start of an IVF cycle in women with PCOS. Little is known about the effect of such pretreatments on outcomes, especially, the rate of live birth. STUDY DESIGN, SIZE, DURATION: This was a nested cohort study and secondary analysis of a multicenter randomized trial, which was designed to compare live birth rate after fresh embryo transfer vs frozen embryo transfer (FET) in women with PCOS (Frefro-PCOS). A total of 1508 women were enrolled from 14 centers between June 2013 and May 2014. PARTICIPANTS/MATERIALS, SETTING, METHODS: At the discretion of local investigators, subjects were instructed to wait for spontaneous menses (Control group, n = 323), or were prescribed progestins (P group, n = 283) or OCs (OCs group, n = 902) to induce menstruation prior to the start of ovarian stimulation. GnRH antagonist protocol was initiated at Day 2 or 3 of induced or spontaneous menses cycle. The rates of pregnancy, pregnancy loss and live birth after either fresh embryo transfer or FET were compared among these three groups. MAIN RESULTS AND THE ROLE OF CHANCE: With fresh embryo transfer, women with OC-induced menses had lower rates of clinical pregnancy (48.8% vs 63.6%, relative rate (RR): 0.77, 95% CI: 0.66-0.89) and live birth (36.1% vs 48.1%, RR: 0.75, 95% CI: 0.61-0.92) than women with spontaneous menses. With freeze-all and deferred FET, women with OC-induced menses had a similar pregnancy rate but a higher pregnancy loss rate (27.7% vs 13.0%, RR: 2.13, 95% CI: 1.28-3.52) after FET than women with spontaneous menses. The live birth rate after FET in women with OC-induced menses, progestin-induced menses and spontaneous menses was 49.4%, 50.7% and 60.2%, respectively (P = 0.06). Progestin-induced menses was associated with similar rates of pregnancy, pregnancy loss and live birth after transfer of either fresh or frozen embryos compared with spontaneous menses. Multivariate logistic regression analysis showed that OCs used for menses induction was associated with lower rate of live birth. LIMITATIONS, REASONS FOR CAUTION: The methods for menses induction were not assigned randomly, thus selection bias was highly likely because of the study design and significant differences that were observed in the baseline characteristics of the women in the different groups. The mean BMI in this study population was relatively normal; the applicability of this result to obese PCOS women needs to be evaluated in further study. WIDER IMPLICATIONS OF THE FINDINGS: Our results suggest that either waiting for a spontaneous menses or using progestin is a better option than using OCs to induce menses in women with PCOS prior to ovarian stimulation using GnRH antagonist protocol for IVF. Further randomized controlled studies are needed to confirm our findings.