High risk of AIDS or death in HIV-positive individuals in South Africa ineligible for antiretrovirals

A longitudinal study of 1,399 untreated patients in the Cape Town urban area has identified a greater short-term risk of progression to AIDS or death in South African patients with CD4 cell counts of 200-350 cells/mm³ as compared to European and Australian cohorts of patients with similar CD4 values and at the same clinical stage of disease, as reported in the October 7th edition of The Lancet. The researchers hope that findings from this study will help to revise present criteria for the start of antiretroviral therapy towards an earlier start of treatment in resource-limited settings.

World Health Organization (WHO) guidelines for the start of antiretroviral therapy have in the past been formulated based on epidemiological information from cohort studies carried out in high-income countries and assessing mostly the long-term risk of disease progression. Recommendations for low-income countries have focused on WHO staging of disease progression and on CD4 cell counts, with treatment starting at CD4 values of 3.

Data from 1,399 patients in the Cape Town AIDS Cohort (CTAC) were analysed to separately assess the short-term (six months) risk of AIDS, death, and combined risk of AIDS and death according to CD4 cell counts and WHO stage of disease. All study participants were HIV-1 infected individuals referred by health-care facilities in the Cape Town urban area between 1992 and 2005 and who had either no access to antiretroviral therapy or were receiving zidovudine monotherapy alone. Death was identified from patients’ records or hospital/municipal registry.

Data for the 852 male and 547 female young adults (median age 32) in the study cohort were stratified in several categories based on CD4 cell counts (350 cells (405-29%)) and WHO clinical stages (stage 1 and 2 combined (n=757-54%), stage 3 (461-33%) and stage 4 (181-33%)). The occurrence of AIDS, death, AIDS and death, progression to another stratum, or loss to follow-up during the six month period after entry into the respective category were registered and the values obtained were summed into patient-years of follow-up.

Univariate and multivariate methods were used for the statistical analysis of findings and each of the potential confounding effects of age, sex, socioeconomic status and use of antiretroviral monotherapy, was separately explored with Poisson regression models. Median follow-up was 16.6 months (IQR, 8.7-29.3 months) and a median of five measurements of CD4 cell count (IQR, 2-11) were available per patient.

The incidence rate of death was 0.1 (95% CI 0.1-0.2) per patient-year for the study group as a whole, with 14% (n=24) of deaths occurring in patients in WHO stages 1 and 2, 38% (66) in stage 3, and 48% (82) in stage 4, revealing that 52% of patients had not progressed to AIDS prior to their death and that, contrary to the findings of studies carried out in high-income countries, in this setting WHO stage 3 disease is associated with a high risk of short-term mortality (3 – 10.08% (n=47), 200-350 cells/mm³ – 4.3% (10), >350 cells/mm³ – 4.9% (9)), with risk doubling for patients in WHO group 3 as compared to patients in groups 1-2, and again becoming twice as high in group 4 as compared to group 3.

The incidence rate of AIDS was 0.11 (95% CI 0.09-0.13) per patient-year, with greater risk of progression to AIDS for patients with lower CD4 cell counts and in WHO stage 3 of the disease. At WHO stage 3 and CD4 cell count of 200-350cells/mm3 the risk estimate was 1.9 higher (3 – 17.4% (n=79), 200-350 cells/mm³ – 7.0% (16), >350 cells/mm3 – 2.2%(4)) than in high-income countries cohorts.

The incidence rate of AIDS or death combined was 0.19 (95% CI 0.16-0.22) per patient-year, with short-term risk ranging from 1.2% (WHO stage 1-2 – CD4>350cells/mm³), to 23.6% (WHO stage 3 – CD43).

The researchers believe that the results from this study may be reflective of the situation of other similar settings, although the effects of some other variables may also play a role, as co-morbid diseases such as malaria are not present in the South African setting.

They conclude that, due to the scarcity of laboratory and clinical tools to assess disease progression in sub-Saharan Africa, both the short-term risk of AIDS and the short-term risk of death should be considered for the start of antiretroviral therapy, and hope this study will inform the revision and refinement of WHO guidelines for scaling up antiretroviral treatment in favour of the start of treatment at levels above 200 cells/mm³, leading to the prevention of two-thirds of early deaths in patients accessing antiretroviral therapy in resource-limited settings.