OBJECTIVE: To compare the tocolytic efficacy and side effects of parenteral and oral magnesium and terbutaline. METHODS: Ninety-eight patients in labor between 23-35 weeks were prospectively entered into a controlled trial of intravenous and oral magnesium versus subcutaneous and oral terbutaline. Tocolytic effectiveness was judged by delay of delivery for 48 hours or 1 week, and to 37 weeks or more. The need to change therapy to the alternate drug was identified, as were side effects. Entrance characteristics of the population, initial pelvic examination, and concomitant infection or cervicovaginal isolates were noted. Outcomes included gestational age at delivery, birth weights, and Apgar scores. Outcome analysis was based on initial tocolytic therapy. RESULTS: Significantly more patients in the magnesium group delivered at 37 weeks or more: 34 of 46 versus 27 of 52 (P < .05). No significant differences were found for delivery by 48 hours or 1 week. The interval between treatment and delivery was greater for magnesium: 7.1 +/- 3.9 versus 5.0 +/- 3.2 weeks (P