Objective: The age-related changes in the proportion of CD4 and CD8 lymphocytes in human immunodeficiency virus (HIV)-seronegative children born to HIV-infected mothers (seroreverters) were compared with the changes in these lymphocyte subsets in children born to seronegative women to assess a possible effect of exposure to HIV without infection.
Design: There were 146 seroreverter and 72 seronegative children. The median CD4 and CD8 percentages for each of these two groups of children were compared retrospectively at 3-month intervals from birth through 27 months and at a tenth interval for the time beyond 27 months. The weighted average of the within-subject rate of change of CD4 and CD8 percentages were also compared between the two groups. Finally, for each subject, the proportion of the subject's CD4 percentage assays which were <10th percentile of the entire study population (30%) was calculated, and the distributions of the subject-specific proportions were then compared between the seronegative and seroreverter groups using the Wilcoxon rank sum test. The proportion of CD8 assays <10th percentile (12%) or > 90th percentile (26%) were also computed for each subject, and the distributions of the proportions were compared similarly. .
Results: The median CD4 percentage of seroreverter children was lower than that for the seronegative children at every interval from birth through 27 months and for the last interval for values obtained at greater than 27 months, although the comparison was statistically significant only at the 4- to 6-month period. The weighted average of the within-subject rate of change of CD4 percentage was -0.09 and -3.0 per year (P = .04), and of CD8 percentage was 1.3 and 1.0 (P = .67), for the seroreverter and seronegative children, respectively. There were significantly more children in the seroreverter group than in the seronegative group who had repeated assays in which the CD4 percentage was < 10th percentile for age (P < .00005). In addition, there was a subset of 10 seroreverter children (6.8%) who had CD4 percentages < 30% on > 50% of their assays, as compared with only one (1.4%) seronegative child. The proportion of CD8 assays < 10th percentile or > 90th percentile were not significantly different between the two groups of children.
Conclusions: The CD4 proportions were persistently lower in the seroreverter than in the seronegative population, although only reaching statistical significance in 1 of 10 3-month intervals. This finding may be due to a subgroup of seroreverter children who have persistently low CD4 lymphocyte percentages.