The association of maternal-to-infant transmission of human immunodeficiency virus type 1 (HIV-1) with maternal p24 antigenemia was assessed in 86 HIV-1-infected mothers. We retrospectively examined serum or plasma samples collected in the peripartum period (delivery +/- 11 days; sd 16.89 days; range, delivery +/- 2 months). Immune complexes of p24 antigen and anti-p24 antibody were dissociated using acid hydrolysis (Method A, glycine-HCl buffer; Method B, HCl) in an attempt to increase the sensitivity of the test. The detection of HIV-1 p24 antigenemia in serum was increased from 23 of 86 (26.7%) to 37 of 82 (45.1%) following acid hydrolysis with Method A (chi square = 5.4, P = 0.02) and to 36 of 78 (46.1%) with Method B (chi square = 5.874, P = 0.015). Mothers of HIV-1-infected children were no more likely to have p24 antigenemia than mothers of seroreverted infants when untreated samples were assayed (7 of 23 vs. 10 of 48; chi square = 0.348, P = 0.55). Although acid hydrolysis increased the ability to detect p24 antigen, it did not enhance any association between p24 antigenemia and maternal-to-infant transmission of HIV infection: Method A, 9 of 23 in mothers of infected children vs. 21 of 45 in mothers of seroreverted children (chi square = 0.112, P = 0.738); and Method B, 9 of 22 in mothers of infected children vs. 18 of 42 in mothers of seroreverted children (chi square = 0.014; P = 0.907), respectively.(ABSTRACT TRUNCATED AT 250 WORDS)