Gout incidence in metformin versus sodium-glucose co-transporter-2 inhibitor users: a retrospective cohort study.

Abstract

To compare the incidence of gout in individuals with diabetes receiving metformin vs sodium-glucose co-transporter-2 inhibitors (SGLT-2is). This new-user comparative effectiveness study included adults from the JMDC claims database with health check-up and administrative claims data from 2014 to 2022. Individuals initiated on metformin were compared with those initiated on SGLT-2is. The primary outcome was the incidence of new gout diagnoses. After propensity-score inverse probability treatment weighting (IPTW), Cox proportional hazards models were fitted to estimate the hazard ratios (HRs) and 95% CIs. A linear mixed model was employed to assess the association between the two groups and changes in serum uric acid levels. A total of 21 561 individuals with diabetes were identified, including 17 636 males with a mean age of 53 years. The metformin and SGLT-2i groups included 13 535 and 8026 individuals, respectively. In the metformin and SGLT-2i groups, the gout incidence rates were 2.40 and 3.15 events per 1000 person-years, respectively. After IPTW, metformin was not associated with a decreased risk of gout compared with SGLT-2i (HR 0.90, 95% CI 0.63-1.28; rate difference -0.28, 95% CI -1.24 to 0.68 events per 1000 person-years). The mean difference in serum uric acid level change 1 year after the index date was 0.48 mg/dl (95% CI 0.43-0.52) for the metformin group relative to the SGLT-2i group. Gout risk may be comparable between metformin and SGLT-2is in individuals with diabetes, with metformin showing a lesser reduction in serum uric acid levels than SGLT-2is.