# Sodium-glucose transport protein 2 inhibitor use in diabetes and its association with tuberculosis incidence

## TL;DR

This study finds that diabetes patients using SGLT2 inhibitors have a lower risk of developing tuberculosis compared to those who do not use these drugs.

## Contribution

The study provides novel evidence linking SGLT2 inhibitor use to reduced tuberculosis risk in diabetes patients.

## Key findings

- SGLT2i users had a 0.43 adjusted hazard ratio for TB compared to non-users.
- A dose-response relationship was observed, with higher drug doses associated with lower TB risk.
- Sensitivity analyses confirmed the robustness of the protective effect of SGLT2i against TB.

## Abstract

Sodium-glucose cotransporter 2 inhibitors (SGLT2i) are increasingly prescribed in diabetes mellitus (DM) patients, yet their impact on tuberculosis (TB) incidence remains unclear. This study analyzed TB risk in DM patients with and without SGLT2i use.

We conducted a retrospective cohort study using data from the Taiwan National Health Insurance Research Database between 2017 to 2020. DM patients treated with or without SGLT2i were identified, and propensity score (PS) matching was performed to balance baseline characteristics, including age, sex, comorbidities, and Diabetes Complications Severity Index (DCSI) between these two groups. TB incidence was assessed until December 31, 2021, and compared between groups. To consider death as a competing risk, a Cox regression model with the Fine and Gray approach was used to estimate the incidence of TB. Additionally, sensitivity analyses were conducted by excluding patients with concomitant antidiabetic medications and by redefining the exposure period based on the timing of SGLT2i initiation.

After PS matching, 76,159 SGLT2i users and 152,318 non-users were included. The majority of enrolled patients were aged over 50 years (63.96%), with males predominant (61.92%). Over an average follow-up of 3.08 years, TB incidence was lower in SGLT2i users than in non-users (0.10% vs. 0.17%, SMD = 0.0192). The time to TB diagnosis was longer in SGLT2i users compared to non-users (2.56 ± 1.13 vs. 2.27 ± 0.02, SMD = -0.2853). Cox regression showed a reduced TB risk in SGLT2i users, with an adjusted hazard ratio of 0.43 (95% CI: 0.33–0.56). The protective effect persisted across age, gender and concomitant use of other antidiabetic drugs. A dose–response relationship was observed, where patients receiving higher cumulated defined daily doses of SGLT2i exhibited progressively lower TB risk. Finally, sensitivity analyses reinforced the robustness of the findings and suggested a temporal association between SGLT2i exposure and a reduced risk of TB occurrence.

This study provides novel evidence that SGLT2i use is associated with a lower TB risk in DM patients, suggesting a potential host-directed therapeutic role warranting further investigation.

The online version contains supplementary material available at 10.1186/s12916-025-04460-w.