Background:  There are few descriptions of HIV-infected and -exposed infants in the first 10 weeks of life. Such data may guide diagnostic algorithms. We report on the clinical and immunological parameters of HIV-infected infants and clinical features of exposed, uninfected (EU), and unexposed, uninfected (UU) infants aged 4 to 10 weeks, screened for 2 interlinked trials in the Comprehensive International Program for Research in AIDS program in South Africa.

Methods:  Infants were recruited through the prevention of mother-to-child transmission (PMTCT)programs in Cape Town and Soweto. HIV-infected (n = 540), EU (n = 125), and UU (n = 125) infants were identified. Clinical data was collected in a standardized format. Characteristics of infants in the three groups were compared using Kruskal-Wallis and Fisher’s exact tests. 

Results:  The median age of all infants was 44 days (range 28 to 78). HIV-infected infants had significantly lower weight and were more likely to have generalized lymphadenopathy (OR 6.4, 95%CI 2.8 to 14.9), oral Candida (OR 3.7, 95%CI 2.0 to 6.8), and hepatomegaly (18.5, 95%CI 2.5 to 134.3) than EU infants. Compared to UU infants, both HIV-infected (OR 26.6, 95%CI 6.5 to 108.8) and EU (OR 7.1, 95%CI 1.6 to 32.3) infants were more likely to have oral Candida. Clinical gastroesophageal reflux disease and splenomegaly were found in 3% and 6% of the HIV-infected infants, respectively, and none of the uninfected infants. All associations persisted when adjusted for infant age. The sensitivity of each of these symptoms for HIV infection was low (<20%), but specificity was high (98 to 100%) in predicting absence of HIV infection. Infection/exposure status was not associated with a Bacille-Calmette-Guerin (BCG) local adverse event, sepsis, or encephalopathy. Oral thrush and upper respiratory tract infections were the only findings having strong associations with CD4% <25% (both p = 0.006). These characteristics, although highly specific, were also poorly sensitive for immunosuppression (specificity 79 to 98%; sensitivity 2 to 21%).

Conclusions:  This is the one of the first clinical descriptions of a large cohort of very young HIV-infected and exposed infants from South Africa. When evaluating children at risk for HIV whose status is unknown, the lack of certain findings may be useful for identifying those who are less likely to be HIV infected when HIV polymerase chain reaction (PCR) is not possible, and in identifying those without severe immunosuppression when CD4 percentage is not available.