One hundred twenty infants were randomly assigned to receive either 15 mg vitamin A or placebo with each of three DPT/OPV (diphtheria, pertussis, tetanus/oral polio vaccine) immunizations at monthly intervals. Sixty-two received vitamin A and 58 received placebo. One month after the third supplementation dose, the response to the delayed cutaneous hypersensitivity test [multitest cell-mediated immunity (CMI) skin evaluation] for tetanus, diphtheria, and tuberculin (purified protein derivative, PPD) was the same in the vitamin A and placebo infants. The number of anergic infants was 17 (27%) and 19 (33%) in the vitamin A and placebo groups, respectively. The number of positive tests among well-nourished infants was significantly higher than that in malnourished infants irrespective of supplementation (P < 0.001). Among the infants with adequate serum retinol concentrations (> 0.7 mumol/L) after supplementation, the vitamin A-supplemented infants had a significantly higher proportion of positive CMI tests than the placebo infants (chi-square test: 8.99, P = 0.008). Among the infants with low serum retinol concentrations (< 0.7 mumol/L) after supplementation, vitamin A supplementation had no effect on CMI response. These results indicate that CMI in young infants was positively affected by vitamin A supplementation only in those infants whose vitamin A status was adequate (ie, serum retinol > 0.7 mumol/L) at the time of the CMI test. CMI was consistently better in well-nourished infants irrespective of supplementation.