Objective: To determine the prevalence and predictors of reactivity to tuberculin purified protein derivative (PPD) and skin test anergy in patients with human immunodeficiency virus (HIV) infection and in HIV-seronegative controls.
Design: Cross-sectional analysis of baseline data from a prospective, multicenter study of pulmonary complications of HIV infection.
Setting: Community-based cohort of persons with and without HIV infection.
Patients: A total of 1171 HIV-seropositive patients without AIDS (841 homosexual men, 274 intravenous drug users, and 56 women with heterosexually acquired infection); 182 HIV-seronegative persons (125 homosexual men and 57 intravenous drug users).
Measurements: Delayed-type hypersensitivity response to tuberculin PPD, trichophytin, mumps, and Candida antigens; T-lymphocyte subsets.
Results: The prevalence of tuberculin PPD reactivity was higher among intravenous drug users than among homosexual men, in both HIV-seronegative (19.1% compared with 6.8%, P = 0.03) and HIV-seropositive persons (15.1% compared with 2.5%, P < 0.001). Among HIV-infected patients, the prevalence of tuberculin reactivity varied directly and that of anergy inversely with the absolute CD4 lymphocyte count. Prevalences were 1% and 72%, respectively, in patients with fewer than 200 CD4 cells/mm3, and 8.4% and 25.5%, respectively, in those with 600 CD4 cells/mm3 (P < 0.001 for both comparisons). Patients with HIV infection and fewer than 400 CD4 lymphocytes/mm3 had a lower prevalence of PPD reactivity than HIV-seronegative controls (2.7% compared with 10.0%, P < 0.001). The strongest predictors of tuberculin reactivity were intravenous drug use, black race, a previous positive PPD test result, and a history of Calmette-Guérin bacillus vaccination. The strongest predictor of anergy was HIV seropositivity.
Conclusions: The response to delayed-type hypersensitivity antigens depends on immune status. The value of PPD and anergy testing in HIV-seropositive patients depends on the ability of such testing to predict subsequent tuberculosis, which is imprecisely known. Until more data or better methods are available, these tests should be done as early as possible in the course of HIV infection.