Risk of Urogenital Infections in Patients with Type 2 Diabetes Initiating SGLT2i versus GLP-1RA: A Danish Emulated Target Trial

Background: People with type 2 diabetes have a 1.5 to 2-fold increased risk of urinary tract infections (UTI) and genital tract infections (GTI) compared to the general population. The new glucose lowering drug classes, including sodium-glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP-1RA), offer organ-protective effects against cardiovascular and kidney diseases. Unfortunately, these drugs also have side effect. There is evidence that SGLT2i increases GTI compared with GLP-1RA, whereas evidence is limited for any differential effect on UTI.

Objectives: To compare risks of UTI and GTI between new users of SGLT2i and GLP-1RA in an emulated target trial.

Methods: We conducted a cohort study that included all adult metformin users who initiated SGLT2i or GLP-1RA in Denmark between 2016 and 2021. UTI and GTI were based on hospital records and community pharmacy prescription data. To balance confounders across treatment cohorts, we computed propensity scores and used inverse-probability of treatment weight (IPTW). Both on-treatment (OT) and intention-to-treat (ITT) analyses were performed. Patients were followed from drug initiation until an outcome event, death, emigration, or the study’s end on January 31, 2024, (in OT analysis also until treatment discontinuation). Comparing SGLT2i and GLP-1RA cohorts, we computed weighted 1-year risk taking competing risk into consideration and hazard ratios (HRs) using Cox proportional hazards models.

Results: We included a total of 79,437 patients (52,414 new SGLT2i users and 27023 new GLP-1RA users). New users of SGLT2i had a slightly higher mean age (64 years vs 61 years), a greater proportion of males (64% vs 57%), lower prevalence of obesity (13% vs 22%), and used less insulin (10% vs 21%). After IPTW every variable was well-balanced.
The weighted 1-year risk of UTI was almost identical in the two groups regardless of the underlying analyses: 10.0% (95% CI: 9.7-10.3) for SGLT2i and 10.3% (95% CI: 9.9-10.7) for GLP-1RA, corresponding to a HR of 0.94 (95% CI: 0.90-0.97) in the OT analysis. In the ITT the weighted 1-year UTI risk were 10.2% (95% CI: 9.9-10.5) for SGLT2i and 10.4% (95% CI: 10.0-10.8) for GLP-1RA, corresponding to a HR of 0.96 (95 CI: 0.93-0.99).
For GTI in contrast, the weighted 1-year risks were increased for SGLT2i users: 2.1% (95% CI: 2.0-2.3) for SGLT2i and 0.5% (95% CI: 0.4-0.6) for GLP-1RA in the OT analysis (HR 2.94 95% CI: 2.60-3.32), and risks of 2.1% (95% CI: 2.0-2.2) for SGLT2i and 0.7% (95% CI: 0.6-0.8) for GLP-1RA in the ITT analysis (HR 1.81 95% CI: 1.65-1.98).

Conclusions: People with type 2 diabetes who initiate SGLT2i versus GLP-1RA as a second line therapy in Denmark had very similar risk of subsequent UTI. However, the risk of GTI was increased up to 3-fold in SGLT2i initiators.