455 © Rapid Science Publishers ISSN 0269-9370 Site of disease and opportunistic infection predict survival in HIV-associated tuberculosis Christopher Whalen, C. Robert Horsburgh Jr*, David Hom † , Christopher Lahart ‡ , Michael Simberkoff § and Jerrold Ellner † Objective: Infection with HIV adversely affects survival in patients with tuberculosis (TB), even when TB is effectively treated. The aim of this study was to identify the determinants of survival in HIV-associated TB. Design: Retrospective cohort study. Setting: Four US academic medical centers. Patients: An inception cohort of 112 HIV-infected patients (mean age 41 years, 96% men, 46% African American) with their first episode of culture-proven TB. Outcome measures: Observed survival from the date of diagnosis of TB to the date of death or censoring. Independent variables included demographics, HIV-related conditions, risk behavior for HIV, absolute CD4+ counts, and site of disease with Mycobacterium tuberculosis. Results: Of the 112 patients, 54 (48%) had pulmonary TB alone, 36 (32%) had both pulmonary and extra-pulmonary TB and 22 (20%) had extrapulmonary TB alone. Median CD4+ count was 95 × 10 6 /l (range, 2–767 × 10 6 /l). During follow-up, 45 patients (40%) died. Median survival was shortest in patients with both pulmonary and extrapulmonary disease (8.4 months), followed by extrapulmonary disease alone (15.6 months), then pulmonary disease (30.4 months; P < 0.001, log- rank test). Median survival was also reduced in patients with previous opportunistic infection and in those with CD4+ < 200 × 10 6 /l. In a proportional hazards regression analysis, which adjusted for CD4+ count, extrapulmonary disease and previous opportunistic infection were the only factors independently associated with shorter survival. Of the extrapulmonary sites of disease, TB meningitis was associated with the greatest risk of death. Conclusion: The site of culture-proven TB at presentation and the history of previous opportunistic infection are important predictors of survival in HIV-infected patients with TB. AIDS 1997, 11:455–460 Keywords: HIV infection, tuberculosis, survival, prognosis From the Department of Epidemiology and Biostatistics and the Division of General Internal Medicine, Departments of Medicine, University Hospitals of Cleveland and Case Western Reserve University School of Medicine, Cleveland, Ohio, the *Division of Infectious Diseases, Department of Medicine, Emory University School of Medicine and Grady Memorial Hospital, Atlanta, Georgia, the † Division of Infectious Diseases, Departments of Medicine, University Hospitals of Cleveland, and Case Western Reserve University School of Medicine, Cleveland, Ohio, the ‡ AIDS Unit, Medical Service, Department of Medicine, Houston Veterans Affairs Medical Center, Baylor College of Medicine, Houston, Texas and the § Division of Infectious Diseases, Manhattan Veterans Affairs Medical Center, Department of Medicine, New York University School of Medicine, New York, New York, USA. Sponsorship: Supported in part by the faculty development fund in the Department of Medicine, Cleveland Veterans Affairs Medical Center and Case Western Reserve University. Note: Information about the survival of this cohort has been presented, in part, in another publication. This paper presents a different view point with new results. Requests for reprints to: Dr Christopher Whalen, Department of Epidemiology and Biostatistics, WG-49, Case Western Reserve University, 10900 Euclid Boulevard, Cleveland OH 44106-4945, USA. Date of receipt: 12 October 1995; revised: 6 November 1996; accepted: 9 December 1996.