794    

Background:  It has been hypothesized that maternal-to-fetal microtransfusions mediate HIV mother-to-child-transmission (MTCT). To test this hypothesis, we measured placental alkaline phosphatase (PLAP) activity in umbilical cord sera. PLAP is a large (135 kD), placental specific enzyme that under normal conditions does not cross the placental barrier; therefore, PLAP activity in fetal circulation is thought to be a marker of maternal-fetal microtransfusions.

Methods:  We conducted a case-cohort study on samples selected from a prospective cohort study in Malawi that was designed to determine the effect of malaria on MTCT. For the case-cohort study, 150 HIV⁺ pregnant women were randomly selected, regardless of HIV transmission status, and compared with 36 cases of in utero MTCT and 48 cases of intrapartum MTCT. The in utero and intrapartum cases were similar to the sub-cohort in terms of CD4 counts, age, and parity. Data were analyzed with logistic regression using generalized estimating equations to account for the inclusion of cases in the sub-cohort. Using this study design, calculated odds ratios estimate risk ratios for the whole cohort.

Results:  Similar to previous reports, median cord PLAP levels in elective caesarean sections were lower than median cord PLAP levels in spontaneous vaginal deliveries. For women who delivered via spontaneous vaginal deliveries, cord PLAP activity strongly associated with intrapartum MTCT (RR: 3.00, 95% CI 1.31 to 6.84) while no association was seen with in utero MTCT (RR: 0.74, 95% CI 0.33 to 1.65). In a multivariate analysis of women who delivered via spontaneous vaginal deliveries, cord PLAP activity remained strongly associated with intrapartum MTCT (RR: 2.76, 95% CI 1.20 to 6.31), independent of HIV RNA load.

Conclusions:  These data support the hypothesis that increased cord PLAP activity is associated with an increased risk of intrapartum HIV-1 MTCT.

Keywords: HIV Mother-to-Child Transmission; placental microtransfusions; case-cohort study