BACKGROUND: The natural history of Kaposi's sarcoma (KS) is poorly documented. We attempted to identify factors predictive of progression and survival in HIV-infected patients with KS and CD4+ cell counts greater than 100/microL. PATIENTS AND METHODS: We studied retrospectively 78 HIV-infected patients diagnosed as having KS between 1989 and 1995. The following variables were assessed as potential predictors of progression and death, in a Cox proportional hazards model: age, sex, ethnic group, transmission group, site of the first KS lesions, duration of KS, concomitant opportunistic infections or malignancies, antiretroviral drug therapy (excluding protease inhibitors), antiherpes treatments, neutrophil counts, CD4+ and CD8+ cell counts, plasma HIV load, p24 antigenaemia, beta2-microglobulinaemia and immunoglobin A and G serum levels. RESULTS: During a median follow-up of 22 months (3-81 months), KS progressed in 66 of the 78 patients. The median survival time after progression was 68 months (9-126 months). Multivariate analysis identified only visceral KS, a high neutrophil count and a high serum immunoglobulin (Ig) level as independent predictors of progression (P < 0.05). Previous and concomitant opportunistic diseases (P = 0.003) and low CD4+ cell counts (P = 0.013) were independently associated with shorter survival; in contrast KS therapy did not independently influence survival. CONCLUSION: Progression of KS is predicted by markers of KS severity, while overall survival is best predicted by markers of immunodeficiency (opportunistic diseases and the CD4+ cell count).
BACKGROUND: The natural history of Kaposi's sarcoma (KS) is poorly documented. We attempted to identify factors predictive of progression and survival in HIV-infectedpatients with KS and CD4+ cell counts greater than 100/microL. PATIENTS AND METHODS: We studied retrospectively 78 HIV-infectedpatients diagnosed as having KS between 1989 and 1995. The following variables were assessed as potential predictors of progression and death, in a Cox proportional hazards model: age, sex, ethnic group, transmission group, site of the first KS lesions, duration of KS, concomitant opportunistic infections or malignancies, antiretroviral drug therapy (excluding protease inhibitors), antiherpes treatments, neutrophil counts, CD4+ and CD8+ cell counts, plasma HIV load, p24 antigenaemia, beta2-microglobulinaemia and immunoglobin A and G serum levels. RESULTS: During a median follow-up of 22 months (3-81 months), KS progressed in 66 of the 78 patients. The median survival time after progression was 68 months (9-126 months). Multivariate analysis identified only visceral KS, a high neutrophil count and a high serum immunoglobulin (Ig) level as independent predictors of progression (P < 0.05). Previous and concomitant opportunistic diseases (P = 0.003) and low CD4+ cell counts (P = 0.013) were independently associated with shorter survival; in contrast KS therapy did not independently influence survival. CONCLUSION: Progression of KS is predicted by markers of KS severity, while overall survival is best predicted by markers of immunodeficiency (opportunistic diseases and the CD4+ cell count).