Two hundred primigravid women with mild preeclampsia at 26-35 weeks' gestation were randomly allocated to treatment with hospitalization alone or combined with labetalol. There were no differences between the two groups in mean systolic or diastolic pressures, mean gestational age, or initial laboratory findings at time of entry. Patients in the labetalol group demonstrated a statistically significant (P less than .005) decrease in blood pressure during treatment. No such decrease occurred in the hospitalization-alone group. Both groups exhibited significant deterioration in proteinuria, creatinine, and uric acid. The average days of pregnancy prolongation (mean +/- SD) were 21.3 +/- 13 and 20.1 +/- 14 days in the hospitalization and labetalol groups, respectively. There were no differences between the groups regarding any of the following: gestational age at delivery, birth weight, number of infants admitted to the special care unit, or cord blood gas measurements. However, the incidence of small-for-gestational-age infants was significantly higher in the labetalol group (19 versus 9%, P less than .05). There were no stillbirths, but one neonatal death in the labetalol group. Treatment of maternal blood pressure in preeclamptic pregnancies with labetalol did not improve perinatal outcome, and was associated with a higher frequency of fetal growth retardation.