G A Macones1, C A Robinson. 1. Department of Obstetrics and Gynecology, University of Pennsylvania Health System, Philadelphia, USA.
Abstract
OBJECTIVE: Recent reports have suggested that the use of indomethacin for tocolysis may independently increase the risk for major adverse neonatal events such as intraventricular hemorrhage and necrotizing enterocolitis. The objective of this study was to determine whether this potential risk of indomethacin is outweighed by the benefit of delivery delay at gestational ages < 32 weeks. STUDY DESIGN: We constructed separate decision trees to compare strategies of tocolysis with indomethacin versus no tocolysis for hypothetic cohorts of patients with idiopathic preterm labor at 24, 26, 28, 30, and 32 weeks' gestation. Probabilities for these decision models, including estimates of indomethacin efficacy and the potential for increase in adverse neonatal events with indomethacin, were obtained from the medical literature. The primary outcome was the number of expected adverse neonatal events per 1000 women for each strategy at each gestational age. RESULTS: In the base case analysis tocolysis with indomethacin was a more favorable strategy than no tocolysis across all gestational ages that we studied. As expected, the difference in the number of events between the two strategies declined with advancing gestational age because of a decreasing baseline risk for adverse neonatal events as gestational age increased. The models at 26, 28, 30, or 32 weeks were not sensitive to our estimates of indomethacin efficacy, nor to our estimates of baseline neonatal morbidity or steroid efficacy, or to the relative increase in some neonatal morbidities with indomethacin use. CONCLUSIONS: On the basis of current estimates, the benefits of indomethacin outweigh the potential risks to the neonate at gestational ages < or = 32 weeks. Thus the use of indomethacin for tocolysis at these ages is a reasonable strategy.
OBJECTIVE: Recent reports have suggested that the use of indomethacin for tocolysis may independently increase the risk for major adverse neonatal events such as intraventricular hemorrhage and necrotizing enterocolitis. The objective of this study was to determine whether this potential risk of indomethacin is outweighed by the benefit of delivery delay at gestational ages < 32 weeks. STUDY DESIGN: We constructed separate decision trees to compare strategies of tocolysis with indomethacin versus no tocolysis for hypothetic cohorts of patients with idiopathic preterm labor at 24, 26, 28, 30, and 32 weeks' gestation. Probabilities for these decision models, including estimates of indomethacin efficacy and the potential for increase in adverse neonatal events with indomethacin, were obtained from the medical literature. The primary outcome was the number of expected adverse neonatal events per 1000 women for each strategy at each gestational age. RESULTS: In the base case analysis tocolysis with indomethacin was a more favorable strategy than no tocolysis across all gestational ages that we studied. As expected, the difference in the number of events between the two strategies declined with advancing gestational age because of a decreasing baseline risk for adverse neonatal events as gestational age increased. The models at 26, 28, 30, or 32 weeks were not sensitive to our estimates of indomethacin efficacy, nor to our estimates of baseline neonatal morbidity or steroid efficacy, or to the relative increase in some neonatal morbidities with indomethacin use. CONCLUSIONS: On the basis of current estimates, the benefits of indomethacin outweigh the potential risks to the neonate at gestational ages < or = 32 weeks. Thus the use of indomethacin for tocolysis at these ages is a reasonable strategy.