Abstract
RATIONALE: HIV infection increases the risk of reactivation of latent tuberculosis (TB). The present study evaluates how latent TB is detected and treated to determine the effectiveness of screening in HIV-infected patients with diverse risk profiles.METHOD: A retrospective medical record database review (1988 to 2007) was conducted at a tertiary care HIV clinic. The proportion of patients receiving tuberculin skin tests (TSTs) and the rate of active TB at each stage of screening and prevention were estimated. Predictors of receiving a TST at baseline, testing positive by TST and developing active TB were evaluated.RESULTS: In the present study, 2123 patients were observed for a total of 9412 person-years. Four hundred seventy-six (22.4%) patients were tested by TST within 90 days of first clinic visit. Having a first clinic visit during the highly active antiretroviral therapy era (OR 3.64; 95% CI 2.66 to 4.99), country of birth (ORs: Africa 3.11, Asia 2.79, Haiti 3.14, and Latin America and the Caribbean 2.38), time between HIV diagnosis and first visit (OR per one-year change 0.97; 95% CI 0.94 to 0.99) and previous antiretroviral exposure (OR 0.61; 95% CI 0.45 to 0.81) were independent predictors of receiving a TST at baseline. Of the 17 patients who developed active TB during follow-up, nine (53%) had no documented TSTs at baseline or during follow-up. Forty-one per cent of all TB patients and 56% of TB patients who were not screened were born in Canada.CONCLUSION: The administration of TSTs to newly diagnosed HIV patients was inconsistent and differential according to country of birth, among other factors, resulting in missed opportunities for TB prevention.