Home Search Abstracts Browse Sessions Program Committee E-mail Abstract Author View Session


Session 82 Poster Abstracts
HIV in Resource Limited Settings: Clinical Manifestations and Testing Strategies
Wednesday, 1:30 - 3:30 pm
Poster Hall


572    
Low Baseline CD4 Counts Do Not Lead to Faster HIV Disease Progression in Ethiopians
Y Mekonnen1,2, R B Geskus3, J C M Hendriks4, T Messele1,2, F Miedema5, R A Coutinho*1,3,6, and N H T M Dukers1,3
1Ethio-Netherlands AIDS Res. Project (ENARP); 2Ethiopian Hlth. and Nutrition Res. Inst., Addis Ababa, Ethiopia; 3Municipal Hlth. Svc. Amsterdam, The Netherlands; 4Univ. of Nijmegen, The Netherlands; 5Sanquin Res. and Landsteiner Lab., Amsterdam, The Netherlands; and 6Academic Med. Ctr., Univ. of Amsterdam, The Netherlands

Background:  HIV-negative Ethiopians have lower CD4+ T-cell (CD4) counts compared with any other population in Africa as well as in industrialized countries. Furthermore, survival time following HIV infection is not known in Ethiopians. Therefore we studied whether the unique immunological profile, i.e. low baseline CD4 counts, would result in shorter survival time in HIV-infected Ethiopians.

Methods:  We used data from an open cohort among factory workers in Ethiopia, February 1997 to June 2002. Each 6 months, 149 HIV-infected participants, including 17 with known date of seroconversion, visited the study. To estimate survival time a continuous-time Markov model was designed. To define stages of disease progression three different data sets were used in the model:  absolute CD4 cell counts; smoothed CD4 cell counts, to account for measurement error and short time-scale variability; and WHO clinical staging for HIV infection and disease. Using a random effect model, the rate of CD4 decline was compared between HIV-infected Ethiopians and Dutch individuals from the Amsterdam Cohort Studies on HIV and AIDS. Since none of the HIV (subtype C) infected Ethiopians have the syncytium-inducing (SI)/X4 phenotype, we repeated analyses excluding Dutch individuals with this type.

Results:  The 149 Ethiopians contributed 791 visits. Their mean age at intake was 34 years and 69% were males. At the first HIV-positive visit, the median CD4 count was 327 x 106/mL. During active follow-up 29 individuals died, giving an incidence of death of 6.7 per 100 person-years. We estimated median survival time from HIV seroconversion to death in the range of 9.1 to 13.0 years using three different approaches. Regardless baseline CD4 cell counts, the annual cell loss was lower for Ethiopians compared to Dutch individuals (p <0.01) (see the figure). For example, at 327 CD4 cell counts, the annual loss for Ethiopians was 32 cells, while this was 68 and 79 cells for Dutch drug users and homosexual men, respectively. Although Dutch individuals who lack SI strains have a lower rate of CD4 decline than Dutch who have such strains, they still have a faster CD4 decline than Ethiopians (p <0.05).

 

INSERT GRAPH

 

Conclusions:  Survival time in HIV-infected Ethiopians is similar to that of populations in industrialized countries before the advent of antiretroviral therapy. The generally low baseline CD4 counts among Ethiopians do not imply shorter HIV survival time than in other populations.

Keywords: HIV disease progression; CD4; Ethiopia