Tuberculosis Incidence and All-Cause Mortality among Human Immunodeficiency Virus-Infected Patients on Isoniazid Preventive Therapy in Dar Es Salaam, Tanzania

Bstract Objective: We determined rates of tuberculosis (TB) incidence and allcause mortality among patients who received Isoniazid preventive therapy (IPT) for 6 months and among a control group that did not receive IPT. Methods: A prospective cohort study was conducted among 2564 HIVinfected patients in Dar-es-Salaam, Tanzania between February 2012 and March 2014. The Tanzania National Tuberculosis and Leprosy Program clinical screening tool was used to exclude active TB before enrollment into the study. Patients were followed up for a total of 24 months. Multivariate Cox proportional hazards were used for analysis and results are presented as adjusted relative hazard (aHR) and 95% confidence intervals (CI). Results: TB incidence was 91 cases/100,000 person-years (PY) (95% CI 11-328) in patients who received IPT and 511 cases/100,000 PY (95% CI 255-915) in control group. There was 79% reduction in TB risk among patients receiving IPT (aHR=0.21, 95% CI 0.25-1.77, p=0.15) after adjusting for use and duration of antiretroviral (ARV) and current CD4 T cell counts. All-cause mortality rate was 136 deaths/100,000 PY (95% CI 28-398) and 1115 deaths/100,000 PY (95% CI 715-1659) in IPT and the control groups, respectively. There was an 83% reduction in the risk of death among patients receiving IPT (aHR=0.17, 95% CI 0.64-0.74, p=0.02) after adjusting for use and duration of ARV and current CD4 T cell counts. Conclusion: IPT does not significantly reduce the risk of TB but that of allcause mortality in HIV-infected patients whose majority were already on ARV medication.