Impact of latent tuberculosis infection on the incidence of type 2 diabetes mellitus in HIV-infected participants in the Swiss HIV Cohort Study. Journal of Infectious Diseases
Tepekule et al. assessed whether latent tuberculosis infection (LTBI) is a risk factor for diabetes among people living with HIV in the Swiss HIV Cohort Study.
The authors compared the risk of developing diabetes in patients with LTBI (defined as having a positive tuberculin skin test or interferon-γ release assay) and in those without LTBI. They used time-dependent Cox proportional hazards regression models adjusted for sex, ethnicity, and time-varying calendar year, age, BMI, ART treatment status, CD4 cell counts, and smoking status. The adjustment was performed using multivariable regression and inverse probability weighting, the latter allowing an estimation of the average causal effect
The study included 10’841 individuals with an LTBI test available (70% male, 14.2% of African origin, 37% were overweight or obese). Overall, 974 individuals had LTBI (69% by means of tuberculin skin testing, and 25% by interferon-γ release assay). Within the first 10 years after LTBI testing, 5.5% of patients with a positive test developed diabetes (60 cases), compared to 3.8% of individuals without LTBI (433 cases, log-rank test p = 0.006). The hazard ratios were 1.44 (95% CI 1.09-1.90) using multivariable Cox regression, and 1.47 (95% CI 1.06-2.03) after adjustment with inverse probability weighting, indicating a 47% risk increase for diabetes with LTBI. These findings were robust across several sensitivity analyses.
In summary, the study shows that LTBI may be a risk factor for diabetes mellitus. In people living with HIV, LTBI has previously been associated with lower HIV viral loads and a lower risk for opportunistic infections. However, the present findings indicate that this state of chronic low-grade inflammation may lead to increases in individuals’ risk for metabolic diseases.