Effectiveness of isoniazid treatment for latent tuberculosis infection among human immunodeficiency virus (HIV)-infected and HIV-uninfected injection drug users in methadone programs.

Injection drug users (IDUs) were heavily affected by the tuberculosis (TB) resurgence in New York City in the 1990s. We assessed the effectiveness of screening for latent TB infection in methadone users and of selective treatment with isoniazid. Risk for future TB was classified as low or high on the basis of tuberculin, anergy, and HIV test results. The cohort of 2212 IDUs was followed up for a median of 4.2 years; 25 IDUs, of whom 20 (80%) were infected with human immunodeficiency virus (HIV), developed TB. In an adjusted Cox proportional hazards model of high-risk IDUs, the risk of TB was associated with HIV infection (HR 10.3; 95% CI, 3.4-31.3); receipt of <6 months of isoniazid therapy (HR 7.6; 95% CI, 1.02-57.1); a CD4+ T lymphocyte count of <200 cells/mm3 (HR 6.6; 95% CI, 1.7-25.9); and tuberculin positivity (HR 4.0; 95% CI, 1.6-10.2). Treatment with isoniazid was beneficial in HIV-infected, tuberculin-positive IDUs.