Sodium-Glucose Co-Transporter 2 Inhibitor Use and Risk of Major Adverse Cardiovascular Events in Patients With Rheumatoid Arthritis and Type 2 Diabetes: A Retrospective Observational Cohort Study Using Real-World Data

Abstract

Background: Patients with rheumatoid arthritis (RA) and type 2 diabetes mellitus (T2DM) face elevated cardiovascular risk. Sodium-glucose co-transporter 2 (SGLT2) inhibitors have demonstrated cardiovascular benefits in diabetic populations, but their effectiveness in RA patients with T2DM remains inadequately explored. The study aimed to compare the risk of major adverse cardiovascular events (MACEs) between SGLT2 inhibitor users and dipeptidyl peptidase-4 (DPP-4) inhibitor users among patients with concurrent RA and T2DM.

Methods: This retrospective cohort study utilized the TriNetX US Collaborative Network database, encompassing 67 healthcare organizations. We identified adult patients with documented RA and T2DM prescribed either SGLT2 inhibitors (n = 823) or DPP-4 inhibitors without SGLT2 inhibitor exposure (n = 284) between 2006 and 2026. Propensity score matching (1:1) balanced baseline demographics and comorbidities, yielding 277 matched pairs. The primary outcome was a composite MACE endpoint comprising myocardial infarction, cerebral infarction, heart failure, cardiovascular death, and cardiac arrest. Kaplan-Meier survival analysis and Cox proportional hazards regression were employed to assess time-to-event outcomes.

Results: After propensity score matching, baseline characteristics were well-balanced between cohorts (standardized differences < 0.2). The SGLT2 inhibitor cohort showed a numerically lower, but statistically non-significant, MACE risk (43.0% vs. 48.7%; risk difference −5.8%, 95% confidence interval (CI): −14.1% to 2.5%; P = 0.172; odds ratio 0.792, 95% CI: 0.567– 1.107). Median survival time was 2,227 days in the SGLT2 inhibitor group versus 2,750 days in the DPP-4 inhibitor group. Kaplan-Meier analysis showed a borderline difference in event-free survival (log-rank P = 0.051), though interpretation is limited by substantially different follow-up durations between groups (median 3.1 vs. 6.4 years). However, the Schoenfeld residual test indicated significant violation of the proportional hazards assumption (P = 0.026), rendering the Cox hazard ratio of 1.284 (95% CI: 0.998–1.652) non-interpretable as a summary measure of treatment effect.

Conclusions: Among patients with RA and T2DM, SGLT2 inhibitor use was associated with a numerically lower but statistically non-significant incidence of MACE compared to DPP-4 inhibitors. These findings are hypothesis-generating and do not provide statistically significant evidence of cardiovascular benefit. The lack of significance may reflect limited statistical power (277 matched pairs) and substantial differential follow-up duration between groups. Adequately powered prospective studies with standardized follow-up periods and comprehensive adjustment for RA-specific confounders are needed to determine whether SGLT2 inhibitors confer cardiovascular benefit in this high-risk population.