Total lymphocyte count, hemoglobin, and delayed-type hypersensitivity as predictors of death and AIDS illness in HIV-1-infected women receiving highly active antiretroviral therapy

Background: Total lymphocyte count (TLC) and hemoglobin level have been suggested as useful and inexpensive parameters to indicate need for HAART in settings in which CD4(+) cell counts are unavailable. If delayed-type hypersensitivity (DTH) response predicts clinical response in persons using highly active antiretroviral therapy (HAART), it may also prove useful in resource-poor settings. Objective: To examine whether TLC, hemoglobin, and DTH response observed prior to initiation of HAART predict post-HAART clinical response. Design: Prospective cohort study. Participants: 873 women in the Women's Interagency HIV Study. Measurements: TLC, hemoglobin, CD4(+) cell counts, and DTH testing using mumps, candida, and tetanus toxoid antigens, performed within 1 year prior to HAART initiation; death; self-report of initiation of HAART use and AIDS-defining illness (ADI). Results: Three different multivariate analyses were performed: 2 models that excluded CD4(+) cell count and assessed TLC at either <850 or <1250 cells/muL, and 1 model that excluded TLC and included CD4(+) <200 cells/muL. TLC <850, TLC <1250, CD4(+) <200 cells/muL, anergy to DTH testing, hemoglobin <10.6 g/dL, and a pre-HAART report of ADI were each consistently independently associated both with death and with incident ADI. Log likelihood chi(2) values suggested similar power among the 3 models in predicting both death and incident ADI. Conclusions: Pre-HAART TLC, hemoglobin level, anergy to DTH testing, and clinical disease each independently predicted morbidity and death after HAART initiation. These findings support the use of TLC to guide decision-making for HAART initiation and suggest that further study of TLC, hemoglobin level, and DTH responses as an indication to provide HAART may be useful in resource-limited settings.