# Safety of Initiating Sodium-Glucose Cotransporter-2 Inhibitors in Patients with Heart Failure or Type 2 Diabetes and a History of Urinary Tract Infections

## TL;DR

This study finds that a history of urinary tract infections does not significantly increase the risk of new infections after starting SGLT2 inhibitors for heart failure or diabetes.

## Contribution

The study provides evidence that prior UTI history should not prevent SGLT2i initiation in eligible patients.

## Key findings

- 20.4% of UTI-naive patients developed a UTI after SGLT2i initiation.
- 30.2% of patients with a prior UTI developed a new UTI after SGLT2i initiation.
- No significant difference in UTI incidence was found between the two groups over 90 days.

## Abstract

Background: Despite being a pillar of heart failure (HF) management, the guideline-directed initiation of sodium-glucose cotransporter-2 inhibitors (SGLT2is) may be challenging due to the barrier of associated urinary tract infections (UTIs). Although there is a known risk, it remains unclear whether UTI incidence differs between patients with and without a prior history of UTIs. Methods: This study aimed to evaluate the risk–benefit profile of initiating an SGLT2i in patients with a history of UTIs. This retrospective, single-center healthcare system cohort analysis included adult patients hospitalized and taking an SGLT2i between 1 January 2020, and 31 August 2024. The included patients were divided into two cohorts: patients with and without a history of UTI pre-SGLT2i (described in this study as UTI-naive). Patients with urogenital structural abnormalities, indwelling catheters, or high-risk profiles were excluded. The primary outcome was the incidence of UTIs post-SGLT2i initiation. Secondary outcomes included the number of UTIs within 30, 60, and 90 days after starting an SGLT2i. Results: A total of 280 patients were evaluated for this study, of which 250 were included for analysis. Of those, 197 were UTI-naive, and 53 had a history of UTI pre-SGLT2i use. The most utilized SGLT2i was empagliflozin (75.6%). Amongst the cohorts, 20.4% of the UTI-naive patients developed a UTI post-SGLT2i versus 30.2% in patients with a historical UTI (p = 0.13). Conclusions: There was no significant difference in UTIs developed up to 90 days post-SGLT2i initiation, regardless of previous infections, suggesting that a history of UTI should not be a barrier to differing first-line therapy.