Association Between Thiazolidinedione Therapy and Tuberculosis-Related Events in Type 2 Diabetes: A Population-Based Cohort Study.

Tuberculosis (TB) remains a major global health burden, and diabetes mellitus increases the risk of active TB. Thiazolidinediones (TZDs) have immunomodulatory effects, but their association with TB remains unclear. This study investigated the associations between TZD use and TB-related events in patients with type 2 diabetes (T2D), including TB requiring anti-TB treatment, pulmonary TB, and TB-related hospitalization. Using Taiwan's National Health Insurance Research Database, adults with newly diagnosed T2D between 2000 and 2017 were identified. Propensity score matching and Cox proportional hazards models were used to estimate adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) for TB requiring anti-TB treatment, pulmonary TB, and TB-related hospitalization. After 1:1 matching, 44,095 TZD user-nonuser pairs and 9225 rosiglitazone user-nonuser pairs were identified. Overall, TZD use was not significantly associated with TB-related outcomes. However, rosiglitazone use was associated with lower risks of TB requiring anti-TB treatment (aHR, 0.56; 95% CI, 0.36-0.86), pulmonary TB (aHR, 0.79; 95% CI, 0.68-0.92), and TB-related hospitalization (aHR, 0.59; 95% CI, 0.43-0.81). Rosiglitazone use was associated with lower risks of TB-related outcomes in the primary Cox analyses; however, these associations were attenuated in time-dependent analyses and were no longer statistically significant after accounting for death as a competing event. These observational findings should therefore be considered hypothesis-generating rather than evidence of a causal or protective effect and require confirmation using more robust causal-inference approaches.