Abstract
PURPOSE: This population-based, retrospective cohort study was to investigate whether metformin is associated with a lower risk of subsequent age-related macular degeneration (AMD) in patients with type 2 diabetes. METHODS: Using the Taiwan National Health Insurance Research Database from 2001 to 2013, 68205 subjects with type 2 diabetes were enrolled in the study cohort. Among them, 45524 were metformin users and 22681 were nonusers. The metformin and nonmetformin groups were followed until the end of 2013. Cox regression analyses were used to estimate hazard ratios (HRs) for AMD development associated with metformin use. Confounders included for adjustment were age, sex, and comorbidities (hypertension, hyperlipidemia, coronary artery disease, obesity, diabetic retinopathy, chronic kidney disease, and insulin treatment). Furthermore, propensity score (PS) matching method was used to choose the matched sample, and PS-adjusted Cox regression was performed. Finally, how HRs changed according to metformin treatment duration and dose was also evaluated in the metformin group. RESULTS: After adjusting for confounders, the metformin group had a significantly lower risk of AMD (adjusted HR = 0.54; 95% confidence interval [CI], 0.50-0.58). In the PS-matched sample, the significance remained (adjusted HR = 0.57; 95% CI, 0.52-0.63). In the metformin group, the adjusted HRs for the second (1.5-4 years) and third (≥4 years) tertiles of metformin treatment duration were 0.52 and 0.14, respectively, compared with the first tertile (<1.5 years). We also found significant trends of lower HRs (all p-value for trend <0.05) with increasing total and average doses. CONCLUSIONS: Among patients with type 2 diabetes, those who use metformin are at a significantly lower risk of developing AMD relative to individuals who do not use metformin. Also, the trend of a significantly lower AMD risk was found with a higher dose of metformin.