Evaluation of the World Health Organization clinical case definition of AIDS among tuberculosis patients in Kinshasa, Zaire.

Abstract

Although the World Health Organization (WHO) clinical case definition for AIDS has been confirmed to have fair sensitivity specificity and positive prediction value in sub-Saharan Africa its application among tuberculosis patients at the Makala Sanatorium in Kinshasa Zaire were evaluated in terms of this case definition by physicians who were not aware of their human immunodeficiency virus (HIV) serostatus. Screening for HIV-1 enzyme-liked immunosorbent assay (ELISA) and Western blot indicated that 85 (36%) of these patients were HIV-positive. In this population the WHO clinical case definition had a sensitivity of 33% a specificity of 86% and a positive predictive value of 58% for HIV infection. When the case definition was modified to exclude chronic cough in tuberculosis patients as a minor criterion the sensitivity decreased to 18% and the specificity and positive predictive value increased to 97% and 77% respectively. A possible explanation for the low sensitivity of the WHO clinical case definition of HIV infection among tuberculosis patients is that tuberculosis may be an early manifestation of immunosuppression that precedes other signs and symptoms of AIDS. It is also possible that the chemotherapy administered to tuberculosis patients eliminates symptoms contained in the WHO case definition such as fever cough weight loss and lymphadenopathy. These findings suggest that periodic serosurveys of tuberculosis patients may be more effective than use of the WHO clinical case definition in detecting HIV infection.