Abstract
PURPOSE: Recent large-scale studies have investigated the risk of type 2 diabetes mellitus (T2DM) associated with the use of 5-alpha reductase inhibitors (5-ARIs) in benign prostatic hyperplasia (BPH) patients, but heterogeneous findings have been reported. This study aimed to examine the risk of T2DM associated with 5-ARIs (finasteride and dutasteride) compared to tamsulosin. MATERIALS AND METHODS: We identified nationwide cohorts of new users of finasteride, dutasteride, and tamsulosin without a history of prescriptions for 5-ARIs or tamsulosin within the previous 3 years among patients covered by national health insurance between 2010 to 2020 in South Korea. Patients were free of T2DM and had received at least 90 days of dispensation of the medications. We balanced cohorts on risk factors for T2DM based on age, health insurance status, index year, comorbidities, and co-medications using stabilized inverse probability of treatment weighting. Adjusted hazard ratios (HRs) were estimated using Cox proportional hazards models, with censoring for treatment discontinuation, switching, or augmentation, end of enrollment, and death. RESULTS: A total of 34874 tamsulosin users, 16953 finasteride users, and 19480 dutasteride users were included. The adjusted HR for finasteride was 1.06 [95% confidence interval (CI), 1.01-1.11] and that for dutasteride was 0.97 (95% CI, 0.92-1.02), when compared with tamsulosin users. These estimates were consistent in lag time analyses and attenuated among patients with good compliance. CONCLUSION: This study suggests minimal effects of finasteride and dutasteride on the risk of T2DM when compared with tamsulosin among patients with BPH.