OBJECTIVE: To evaluate the risk of gestational diabetes mellitus (GDM) in women with polycystic ovary syndrome (PCOS).DESIGN: Systematic review and meta-analysis of observational studies.SETTING: Tertiary Department of Reproductive Endocrinology.PATIENT(S): Five thousand two hundred ninety-three pregnant women (721 with PCOS and 4,572 controls without PCOS).INTERVENTION(S): Literature search in the electronic databases MEDLINE, EMBASE, and CENTRAL, study of the references of all relevant trials or reviews, and manual search of the abstracts from the major meetings in the field of human reproduction.MAIN OUTCOME MEASURE(S): Gestational diabetes mellitus odds ratio.RESULT(S): Women with PCOS demonstrated a significantly higher risk for the development of GDM as compared with women without PCOS (odds ratio 2.89, 95% confidence interval [CI] 1.68-4.98), yet with significant statistical heterogeneity (I(2) = 59.3%), durable to sensitivity analysis. In the subgroup of cohort studies, this finding remained robust (7.11, 95% CI 2.95-17.12), whereas in the subgroup of case-control studies, it did not (0.89, 95% CI 0.38-2.06). Metaregression modeling revealed a linear dependence of the outcome on study type and baseline risk (post hoc).CONCLUSION(S): Significant heterogeneity among studies and dependence of the outcome on study type make the higher risk of GDM in women with PCOS a questionable finding. The conduction of properly designed studies should precede any recommendation to pregnant women with PCOS in regard to the risk of GDM.