The values of quantitative serum HIV-1 RNA levels and CD4 cell counts for predicting survival time among HIV-positive individuals with CD4 counts of <= 50 x 10(6) cells/l

Objective: To evaluate the HIV-1 RNA level as a predictor of survival time among individuals with advanced AIDS. Methods: The serum HIV-1 RNA level, the CD4 cell count, and other clinical variables were evaluated at baseline, as predictors of survival time, among 56 retrospectively identified HIV-l positive individuals with less than or equal to 50 X 10(6) CD4 cells/l who attended the Beth Israel Deaconess Med ical Center, Division of Infectious Diseases, between 1 July 1989 and 30 Se ptember 1993. Results: During follow-up, 55 of these 56 patients died. The median surviva l time was 20.5 months. In univariate Cox proportional hazard modeling neit her the baseline HIV-1 RNA level nor the CD4 cell count were predictive of survival time. However, in multivariate models longer survival time was ass ociated with the use of trimethoprim-sulphamethoxazole at entry [hazard rat io (HR), 0.42; P = 0.007], whereas shorter survival time was associated wit h a history of an AIDS-defining illness other than Pneumocystis carinii pne umonia (HR, 2.87; P = 0.007). Correlative analysis revealed a modest correl ation of the baseline CD4 cell count with survival time (Spearman rho = 0.4 1; P = 0.002). However, no correlation was found between HIV RNA levels and survival time(P= 0.5). Conclusions: In this population with very advanced disease, the HIV-1 RNA l evel was a poor discriminator of survival time, being inferior to the CD4 c ell count and to specific clinical variables such as the nature of the prio r AIDS-defining illness and the type of Pneumocystis carinii pneumonia prop hylaxis employed. Among individuals with advanced AIDS, these data emphasiz e the relative importance of the CD4 cell count and of specific clinical fa ctors, over the HIV-1 RNA level in predicting survival (C) 2000 Lippincott Williams & Wilkins.