The effect of pregnancy on disease progression and survival in HIV-infected women was investigated through a systematic literature review and meta-analysis of the results of eligible studies. The literature search identified seven studies--all prospective cohort analyses--on this topic published in 1983-96. The summary odds ratios (ORs) for adverse maternal outcomes related to HIV infection were as follows: maternal death, 1.8 (95% confidence interval (CI), 0.99-3.3); HIV disease progression, 1.41 (95% CI, 0.85-2.33); progression to an AIDS-defining illness, 1.63 (95% CI, 1.00-2.67); and fall of CD4 cell count to below 200, 0.73 (95% CI, 0.17-3.06). Sensitivity analyses indicated HIV progression during pregnancy was significantly more common in developing countries (OR, 3.71; 95% CI, 1.82-7.75) than developed countries (OR, 0.55; 95% CI, 0.27-1.11) as well as in studies of high methodological quality (OR, 3.71; 95% CI, 1.82-7.57) compared with low-quality studies (OR, 0.55; 95% CI, 0.27-1.11). When studies attempted to control for confounding factors by matching or restriction techniques, there was less progression of HIV disease, but this was not statistically significant. Although this meta-analysis suggests that pregnancy in HIV-positive women may be associated with at least small increases in risk of adverse maternal outcomes, further large-scale observational studies of pregnant and nonpregnant HIV-infected women with long-term follow-up are required. Data on immune function should be available before pregnancy to facilitate determination of the effect of pregnancy on immune status.