Objective: To assess the risk of subsequent miscarriage in pregnant women with a prior diagnosis of polycystic ovarian syndrome (PCOS). Methods: Using a nationwide, population-based database (Taiwan National Health Insurance Research Database) during 1998-2012, the study retrieved 1,000,000 randomly-sampled insured citizens as research subjects. The women with a diagnosis of pre-pregnancy PCOS (n = 13,562) who had chromosomal anomalies, artificial abortion, inconsistent diagnoses, and who were initially diagnosed with PCOS at >45 or <15 year-old were excluded, respectively. The records of gynecologic ultrasonography and/or blood tests were checked to verify the accuracy of the diagnoses of both PCOS and miscarriage (ICD-9 CM codes). After pregnancy, every woman with prior PCOS was age-matched to four women without prior PCOS. Results: Pregnant women with prior PCOS (the case group; n = 1926) and those without prior PCOS (the control group; n = 7704) were compared. The incidence of subsequent miscarriage was much higher in the case group compared with the control group (33.80% vs. 4.09%, p < 0.0001). Logistic regression analysis revealed that the risk of subsequent miscarriage was significantly higher in the case group than the control group (odds ratio [OR] 11.98; 95% CI 10.34-13.87, p < 0.0001), and the result remained similar while adjusted with covariates (adjusted OR 11.97; 95% CI 10.27-13.95, p < 0.0001). In the case group, the patient who used metformin had a lower risk of subsequent miscarriage (adjusted OR 9.53; 95% CI 6.69-13.57) when compared with those who did not receive metformin treatment (adjusted OR 12.13; 95% CI 10.38-14.18). Conclusion: For pregnant women, a pre-pregnancy diagnosis of PCOS is an independent and significant risk factor for subsequent miscarriage. The risk of subsequent miscarriage is reduced by about 1/4 for the PCOS patients who undergo metformin treatment compared with those who do not.
Keywords: abortion; metformin; miscarriage; polycystic ovarian syndrome (PCOS).