| Home | Search Abstracts | Browse Sessions | Program Committee | E-mail Abstract Author | View Session |
|
|
|
Session 83
Poster Abstracts Antiretroviral Agents in Resource Limited Settings Wednesday, 1:30 - 3:30 pm Poster Hall |
Background: Though prices of highly active antiretroviral therapy (HAART) continue to fall, standard lab tests are prohibitively expensive for monitoring therapy in resource-limited countries. Total lymphocyte count, hemoglobin (Hg), and weight changes may be able to predict CD4 count change in patients on HAART. In addition, heat denatured p24 antigen immunocomplexes prior to ELAST may be as sensitive as RT-PCR in detecting viral load in India.
Methods: CD4 count, total lymphocyte count, Hg, and weight change were analyzed in 131 patients on generic HAART attending an HIV-care clinic in Chennai, India. Sensitivity, specificity, positive predictive value, and negative predictive value of total lymphocyte count as a predictor of CD4 count response and multivariate regression as performed. An additional 39 subjects had HIV viral load assessed by RT-PCR (Amplicor v1.5, Roche Molecular) and signal-boosted, heat-denatured p24 antigen assays (Perkin Elmer p24 ELISA and ELAST Amplification System). Association between HIV-1 RNA and p24 antigen levels was compared using Spearman’s rank correlation.
Results: In the first 12 months of HAART, the sensitivity and PPV of a total lymphocyte count increase as a marker for CD4 count increase was 72% and 96%, respectively. The specificity and negative predictive value of a total lymphocyte count decrease as a marker for CD4 count decrease was 75% and 24%, respectively. Mean changes in total lymphocyte count with a CD4 count response of < or >100 cells/mm3 were 168 cells/mm3, and 707 cells/mm3, respectively (p <0.01). The slope of dCD4 cell count was [145+(0.08)(d total lymphocyte count)]. In multivariate regression, change in total lymphocyte count (p <0.001), and change in weight (p <0.03) predicted change in CD4 count. Of 39 subjects with viral load analysis, 17 had less than 400 copies/mL by RT-PCR; all of these subjects had a p24 antigen assay of 1500 fg/mL, the lowest level of detection; 19 subjects had a p24 antigen assay of 1500 fg/mL, of which 17 had an HIV RNA <400 copies/mL. The sensitivity of p24 antigen to predict an undetectable viral load by HIV RNA was 100%, and the specificity was 91%. Correlation of both tests by Spearman rank test and analysis of variance was R = 0.958 (n = 22, R2 = 0.8635, p <0.001) when values below cut-off were excluded and R = 0.908 (n = 39, R2 = 0.8375, p <0.001) when values below cutoff were included in the analysis.
Conclusions: Total lymphocyte count and weight could serve as components of a low cost, effective CD4 monitoring system and heat-denatured p24 antigen assays may serve as a surrogate marker for HIV RNA viral load in resource-limited countries.
Keywords: Low Cost Monitoring; total lymphocyte count; p24
