Thirteen trials (11,843 women) were included in the review.
Three studies were unblended; all others were double-blinded. The quality scores ranged from 3 to 5.
Zidovudine versus placebo (5 studies): zidovudine significantly reduced the risk of mother-to-child transmission (pooled RR 0.57, 95% CI: 0.45, 0.71) and low birth weight (pooled RR 0.75, 95% CI: 0.57, 0.99). There were no significant effects of treatment on stillbirth, infant death or premature delivery.
Zidovudine plus lamivudine versus placebo (1 study): the antiretroviral treatment reduced the risk of infant HIV infection by 37% (95% CI: 10, 55). There were no significant effects of treatment on stillbirth or infant death.
Nevirapine plus zidovudine versus placebo plus zidovudine (1 study): the additional dose of nevirapine reduced the risk of mother-to-child HIV transmission by 68% (95% CI: 39, 83) and reduced the risk of death by 80% (95% CI: 10, 95). There was no effect on stillbirth, premature delivery or low birth weight.
Zidovudine versus nevirapine (1 study): nevirapine reduced the risk of mother-to-child transmission of HIV by 40% (95% CI: 13, 59). There were no differences between the treatments in terms of infant death, stillbirth and low birth weight.
Zidovudine for all women during the third trimester of pregnancy followed by a single dose of nevirapine versus placebo in mothers and infants (1 study): in the group receiving the single dose of nevirapine, the risk of mother-to-child transmission of HIV was reduced by 55% (95% CI: 19, 75).
The following treatment comparisons showed no differences in outcomes between intervention and comparison groups: zidovudine plus nevirapine versus nevirapine (2 studies); various regimens of zidovudine (1 study); zidovudine plus lamivudine versus nevirapine (1 study); nevirapine plus standard antiretroviral therapy versus placebo plus standard antiretroviral therapy (1 study).