Vertical transmission of HIV-1: maternal immune status and obstetric factors. The European Collaborative Study.

OBJECTIVE

To estimate the effect of maternal factors and events around the time of delivery on HIV-1 vertical transmission risk.

DESIGN

Prospective study.

SETTING

Twenty-two obstetric and paediatric clinics in seven European countries.

PATIENTS OR OTHER PARTICIPANTS

Mothers identified as HIV-infected before or at delivery and their children.

MAIN OUTCOME MEASURE

Paediatric HIV infection.

RESULTS

By November 1995, 1846 mothers with 1945 children had been enrolled. The vertical transmission rate was 16.4% (95% confidence interval, 14.5-18.3). Parity, maternal age, race, mode of HIV acquisition, injecting drug use and sex of infant were not statistically significantly associated with risk of transmission. Children delivered vaginally were more likely to be infected than those delivered by Caesarean section. However, in vaginal deliveries the procedures used, duration of ruptured membranes or length of second-stage labour were not related to transmission. Transmission increased almost linearly with decreasing CD4 cell count, but there was no such trend for CD8 cell count. Women with CD4 cell counts below 200 x 10(6)/l were significantly more likely to deliver early (chi 2 for trend, 14.02; P < 0.001). Very premature infants were at increased risk of infection, but after about 35 weeks gestation the transmission rate remained stable, with no increase in late pregnancy. This trend was confirmed after allowing for maternal CD4 cell count.

CONCLUSIONS

The rate of vertical transmission increases linearly with decreasing maternal CD4 cell count. Women with fewer than 200 x 10(6) CD4 cells/l have an increased risk of premature delivery, which would affect timing of interventions. The stable transmission rate after 35 weeks gestation suggests little acquisition of infection during late pregnancy.