Different pregnancy outcomes according to the polycystic ovary syndrome diagnostic criteria: a systematic review and meta-analysis of 79 studies.

ABSTRACT

OBJECTIVE: To correlate the distinct diagnostic criteria of polycystic ovary syndrome (PCOS) with the development of maternal and neonatal complications. DESIGN: Systematic review and meta-analysis. SETTING: Not applicable. PATIENT(S) Pregnant women with PCOS. INTERVENTION(S) Maternal and neonatal complications were compared among women with PCOS diagnosed with different criteria. MAIN OUTCOME MEASURE(S) The primary outcomes of gestational diabetes mellitus and preeclampsia (PE) were assessed for every diagnostic criterion. RESULT(S) Seventy-nine studies were included. Regarding gestational diabetes, the overall pooled prevalence was 14% (95% confidence interval [CI], 11%-18%; I2, 97%), reaching the highest level when polycystic ovarian morphology on ultrasound and 1 of the remaining 2 Rotterdam criteria (1/2 Rotterdam criteria) were used (18%; 95% CI, 13%-24%; I2, 20%) and the lowest when polycystic morphology on ultrasound and hyperandrogenism were used (3%; 95% CI, 0%-19%; I2, not applicable). Regarding PE, the overall pooled prevalence was 5% (95% CI, 4%-7%; I2, 82%). The highest PE prevalence was reported when the National Institutes of Health criteria were used (14%; 95% CI, 5%-33%; I2, 90%) and the lowest when menstrual irregularities and 1 of the 2 Rotterdam criteria were used (2%; 95% CI, 1%-3%; I2, not applicable). CONCLUSION(S) The prevalence of gestational diabetes mellitus in pregnant women with PCOS does not differ according to the criteria used; however, women diagnosed with PCOS per the National Institutes of Health criteria are at higher risk of PE.