Chemoprophylaxis for tuberculosis and survival of HIV infected patients in Brazil

Objective: To study the impact of chemoprophylaxis for tuberculosis on the survival of HIV-infected patients with a positive tuberculin skin test. Design: Prospective observational cohort study. Setting: Outpatient clinic of a university hospital, in Rio de Janeiro, Brazil. Patients: Two-hundred and ninety-seven patients with a positive tuberculin skin test (reaction greater than or equal to 5mm) who were admitted to the cohort between January 1991 and December 1994. Follow-up ended on September 30, 1998. Intervention: The use of chemoprophylaxis for tuberculosis. Main outcome measures: Death was the primary outcome variable. The occurrence of tuberculosis was studied as a secondary outcome. Cox regression models were used in these analyses. Results: The median follow-up time was 43.6 months. Chemoprophylaxis was used by 128 (43%) of the patients. The use of chemoprophylaxis was associated with a reduction in risk for tuberculosis (hazard ratio, 0.38; 95% confidence interval, 0.14-1.04; P=0.05). In a regression model adjusted for baseline CD4 cell count, chemoprophylaxis was associated with longer survival (hazard ratio, 0.24; 95% confidence interval, 0.09-0.65; P=0.002). Conclusions: Anti-tuberculosis chemoprophylaxis was associated with a substantially prolonged survival among purified protein derivative-positive HIV-infected patients in Brazil. These data have important implications for the clinical care of patients with HIV infection in areas of the world with a high prevalence of Mycobacterium tuberculosis infection. (C) 2001 Lippincott Williams & Wilkins.