Incidence of tuberculosis after HAART initiation in a cohort of HIV-positive patients in Burkina Faso

Abstract

Objective: To determine the incidence rates of tuberculosis (TB) after the initiation of highly active antiretroviral treatment (HAART).

Methods: We conducted a retrospective cohort study on four human immunodeficiency virus (HIV) treatment centres in Ouagadougou, Burkina Faso. TB incidence was measured at different intervals after HAART initiation. Cox regression models were used to identify factors associated with TB incidence.

Results: We analysed a cohort of 2383 subjects with a mean follow-up period of 836 days (standard deviation +/- 443.4). TB incidence rate was highest during the first trimester of HAART; after 3 months of treatment, the total TB case incidence dropped by 60% from 5.77/100 person-years (py) to 2.23/100 py. World Health Organization clinical Stage III or IV, CD4+ T-cell count < 50 cells/microl and body mass index (BMI) < 18.5 were associated with increased risk of TB on univariate analysis. In the Cox regression, BMI < 18.5 and CD4+ T-cell count < 50 cells/microl at HAART initiation were independently associated with a two-fold higher risk of TB.

Conclusions: Delaying HAART initiation until the CD4+ T-cell count drops to <50 cells/microl significantly increases TB incidence in the first 3 months after HAART initiation. Active case finding for TB is an essential element of standard clinical care in HIV-positive patients during the initial period of HAART.