ACCELERATED COURSE OF HUMAN-IMMUNODEFICIENCY-VIRUS INFECTION AFTER TUBERCULOSIS

To determine the effect of active tuberculosis on survival and the inc idence of opportunistic infections in HIV-infected patients, we perfor med a retrospective cohort study at four U.S. medical centers to compa re the survival and incidence rate of opportunistic infections in 106 HIV-infected patients with active tuberculosis (cases) with that of 10 6 HIV-infected patients without tuberculosis (control subjects) but wi th a similar level of immunosuppression (measured by the absolute CD4 lymphocyte count) as the cases. Cases and control subjects were simil ar with regard to age, sex, race, previous opportunistic infection, an d use of antiretroviral therapy, but they were more likely than contro l subjects to have a history of intravenous drug use (49 versus 19%). The mean CD4+ counts were similar for cases and control subjects (154 versus 153 cells/mu l, respectively). The incidence rate of new AIDS-d efining opportunistic infections in cases was 4.0 infections per 100 p erson-months compared with 2.8 infections per 100 person-months in con trol subjects for an incidence rate ratio (RR) of 1.42 (95% confidence interval: 0.94-2.11). Cases also had a shorter overall survival than did controls subjects (p = 0.001). Active tuberculosis was associated with an increased risk for death (odds ratio = 2.17), even when contro lling for age, intravenous drug use, previous opportunistic infection, baseline CD4+ count, and antiretroviral therapy. Although active tube rculosis may be an independent marker of advanced immunosuppression in HIV-infected patients, it may also act as a cofactor to accelerate th e clinical course of HIV infection.