Double-blind, randomized, controlled trial of atosiban and ritodrine in the treatment of preterm labor: A multicenter effectiveness and safety study

Objective: This study was undertaken to compare the efficacy and safety of intravenous administration of atosiban versus ritodrine for the treatment of preterm labor. Study Design: Women with preterm labor and intact membranes diagnosed at 23 to 33 gestational weeks (n = 247) were randomly assigned to treatment arms and received atosiban (6.75 mg intravenous bolus, 300 μg/min for 3 hours, then 100 μg/min intravenously) or ritodrine (0.10-0.35 mg/min intravenously) for as long as 18 hours. Tocolytic effectiveness was assessed in terms of the numbers of women who had not been delivered after 48 hours and after 7 days. Safety was assessed in terms of maternal side effects and neonatal morbidity. Secondary outcomes included mean gestational age at delivery and mean birth weight. An intent-to-treat analysis was performed with the Cochran-Mantel-Haenszel test. Results: The proportion of women who had not been delivered at 48 hours was 84.9% (n = 107) in the atosiban group and 86.8% (n = 105) in the ritodrine group. At 7 days 92 women had still not been delivered in both the atosiban (73.0%) and ritodrine (76.0%) groups. Neither of these differences was statistically significant. The incidence of maternal cardiovascular side effects was substantially lower in the atosiban group (4.0% vs 84.3%, P < .001). In addition, intravenous therapy was terminated more frequently as a result of maternal adverse events in the ritodrine group (29.8%) than in the atosiban group (0.8%). The overall occurrences of fetal adverse events in the two treatment groups were comparable. Neonatal morbidity was similar between the treatment groups after adjustment for unbalanced enrollment of women with multiple pregnancies and for gestational ages within treatment groups. Conclusion: Atosiban was comparable in clinical effectiveness to conventional ritodrine therapy but was better tolerated than ritodrine, with no evidence of significant maternal or fetal adverse events. Neonatal morbidity, which was similar between the two treatment arms, was apparently related to the gestational age of the infant rather than to the exposure to either tocolytic agent. (Am J Obstet Gynecol 2000;182:1191-9.)