A randomized clinical trial was designed to determine whether there are clinically demonstrable advantages of phenytoin over magnesium sulfate in preeclamptic patients because of the latter drug's uterine relaxant properties. An intravenous infusion, immediately after randomization, of either phenytoin or magnesium sulfate, with subsequent measurement of serum concentrations and maintenance of therapeutic levels was given to 103 preeclamptic and two eclamptic women. Observed were the rate of cervical dilation during active labor and change in hematocrit between predelivery and 24-hour postdelivery values and the incidence of side effects ascertained by interview. Compared with those receiving magnesium sulfate, patients receiving phenytoin had more rapid cervical dilation (3.3 cm/hr versus 1.5 cm/hr, p = 0.016) and a smaller fall in hematocrit after delivery (-4.7% versus -7.6%, p = 0.034). A significantly lower incidence of hot flushes (15% versus 46%, p < 0.005) and a trend toward less dyspnea and weakness were reported by phenytoin-treated patients. Our phenytoin regimen produced acceptable serum phenytoin levels (10 to 25 micrograms/ml) in 96% of patients.