Abstract
Background: The use of different definitions and diagnostic approaches of polycystic ovary syndrome (PCOS) and recurrent miscarriage (RM) has led to a wide range of prevalence rates in the literature. Despite the persistent controversy about the factual prevalence of PCOS in RM, a vast number of studies have revealed evidence about their association with each other. The goals of this study were to evaluate the prevalence of polycystic ovarian morphology and PCOS within the RM population, performing meta-analyses with the obtained data from this study, together with previous reports on this topic and evaluating reproductive outcome in women with RM and PCOS. Methods: A retrospective cohort study with 452 women with RM and a meta-analysis were conducted. The main outcome parameter was the prevalence of PCOS in RM patients. Results: In the retrospective study, the prevalence of PCOS in RM was 9.5%. Negative results for the selected risk factors for RM were present in 283 patients (62.6%). From all evaluated possible underlying causes for RM, only the presence of thrombophilic disorders was significantly associated with PCOS (PCOS: 20.9% versus no PCOS: 7.8%, p = 0.010). In the meta-analysis of three studies on PCOS in RM patients, which used the revised Rotterdam criteria for defining PCOS, an estimated pooled prevalence of 14.3% (95% CI: 6.2-24.9) was found. In the retrospective data set, women in the PCOS group revealed significantly higher luteinizing hormone (LH), testosterone, and Anti-Mullerian hormone (AMH) levels than age- and body mass index (BMI)-matched controls with RM negative for the selected risk facotrs (p < 0.05). The rate of further miscarriages was significantly higher in PCOS women than in controls (71.4% versus 53.6%, respectively; p = 0.031). Conclusions: The prevalence of PCOS seems slightly increased in women with RM. Women with PCOS suffering from RM showed a significantly higher risk for further miscarriage and decreased chances of having a life birth of about 18% which did not reach statistical significance. Therefore, we assume that PCOS plays a moderate role in RM.