Sodium-Glucose Cotransporter-2 Inhibitors and Urinary Tract Infections: A Propensity Score-matched Population-based Cohort Study

Abstract

Objectives: Sodium-glucose cotransporter-2 (SGLT2) inhibitor-induced glycosuria is hypothesized to increase the risk of urinary tract infections (UTIs). We assessed the risk of UTIs associated with SGLT2 inhibitor initiation in type 2 diabetes.

Methods: We conducted a population-based cohort study using primary care data from the United Kingdom's Clinical Practice Research Datalink (CPRD) and administrative health-care data from Alberta, Canada. From a base cohort of new metformin users, we constructed 5 comparative cohorts, wherein the exposure contrast was defined as new use of SGLT2 inhibitors or 1 of 5 active comparators: dipeptidylpeptidase-4 (DPP-4) inhibitors, sulfonylureas (SU), glucagon-like peptide-1 receptor agonists (GLP-1 RA), thiazolidinediones (TZD) and insulin. We defined a composite UTI outcome based on hospitalizations or physician visit records. For each comparative cohort, we used high-dimensional propensity score matching to adjust for confounding and Cox proportional hazards regression to estimate the hazard ratios (HRs) in each database. We meta-analyzed estimates using a random-effects model.

Results: SGLT2 inhibitor use was not associated with a higher risk of UTI compared with DPP-4 inhibitors (pooled HR, 1.08; 95% confidence interval [CI], 0.89 to 1.30), SU (pooled HR, 1.08; 95% CI, 0.90 to 1.30), GLP-1 RA (pooled HR, 0.81; 95% CI, 0.61 to 1.09) or TZD (pooled HR, 0.81; 95% CI, 0.55 to 1.19). The risk of UTI was lower compared with insulin (pooled HR, 0.74; 95% CI, 0.63 to 0.87). The risk of UTI did not differ based on the SGLT2 inhibitor agent or dose. Last, SGLT2 inhibitor initiation was not associated with an increased risk of UTI recurrence.

Conclusion: SGLT2 inhibitor use is not associated with an increased risk of UTIs, compared with other antidiabetic agents.

Keywords: IVU; SGLT2 inhibitors; UTI; cohort study; diabète de type 2; drug safety; inhibiteurs du SGLT2; innocuité des médicaments; pharmaco-epidemiology; pharmacoépidémiologie; type 2 diabetes; étude de cohorte.