Predictors of incident tuberculosis among HIV-1-infected women in Tanzania.

SETTING: The development of tuberculosis (TB) in HIV-1-infected individuals is associated with accelerated HIV-1 disease progression. OBJECTIVE: To examine the predictors of incident TB in HIV-1-infected Tanzanian women. DESIGN: A prospective cohort of 1078 HIV-1-infected pregnant women was enrolled in a randomized clinical trial to examine the role of vitamin supplements in HIV-1 disease progression and fetal outcomes. RESULTS: Of 1008 women evaluated for TB, 88 (8.7%) developed TB. After controlling for age, education and hemoglobin concentration, in multivariate analysis, low CD4 cell count, elevated erythrocyte sedimentation rate (ESR), decreased mid-upper arm circumference, and high viremia were associated with an increased risk of TB. CD4 <200 vs. > or = 500 cells/mm3 was associated with a 4.44-fold increase in risk of TB (95%CI 2.10-9.40). Individuals with high viremia (> or = 50,000 copies/ml) had a 2.43-fold increase in risk of TB (95%CI 1.24-4.76). Elevated malarial parasite density was slightly associated with a 65% (95%CI 19-85) decreased risk of TB. CONCLUSIONS: The risk of developing TB was elevated among women with low CD4 cell counts, elevated ESR, coinfections with other pathogens, poor nutrition and high viremia. There is a slight inverse association between malarial infection and TB, possibly because treating malaria may reduce the risk of TB.

The use of total lymphocyte count as a surrogate for low CD4+ T lymphocyte cell counts among HIV-1-infected women in Tanzania.

BACKGROUND: Human Immunodeficiency Virus type 1 (HIV-1) infection leads to a progressive decline in CD4+ T-lymphocyte (CD4) cells. Initiation of prophylaxis against Opportunistic infections in adults (CD4% used for children) and antiretroviral therapy is usually based on CD4 cell counts, but CD4 cell counts measurement is not affordable in most African countries. OBJECTIVE: To examine whether total lymphocyte counts (TLC) may be used as proxies for low CD4 cell counts. DESIGN: Cross-sectional at baseline when women were pregnant and at least six months postpartum. METHODS: 1,078 HIV-1-infected pregnant women from Dar es Salaam, Tanzania were enrolled in a randomized clinical trial. A series of receiver operator characteristic (ROC) curves were created at baseline and at least 6 months postpartum and among women in WHO Stage 3 and above. The sensitivity and specificity of TLC and hemoglobin in predicting an absolute CD4 count < 200 cells/mm3 were determined for various clinically relevant cut points. RESULTS: TLC was not a good predictor of low CD4 cell counts during pregnancy or at least six months postpartum as exhibited by low ROC Area Under the Curve (AUCs) of .57 and .62 respectively. No other variable had the ability to predict CD4 < 200 cells/mm3. CONCLUSIONS: The use of TLC as a proxy for the estimation of low CD4 cell counts in a population of HIV-1-infected adults from sub-Saharan Africa was not substantiated. Inexpensive methods to quantify CD4 cell counts in Africa are needed.