External cephalic version beginning at 34 weeks' gestation versus 37 weeks' gestation: a randomized multicenter trial.
OBJECTIVE: In about 3% to 4% of all pregnancies at term, the fetal presentation will be noncephalic. External cephalic version (ECV) at term has been shown to decrease the rate of noncephalic presentation at birth and to decrease the rate of cesarean section associated with breech presentation. However, success rates for ECV are low. We did a randomized trial to compare a policy of beginning ECV early, at between 34 and 36 weeks' gestation, and beginning ECV at 37 to 38 weeks' gestation. STUDY DESIGN: At 25 centers in seven countries, 233 women with a singleton breech fetus were randomly assigned to having an ECV procedure done early (at between 34 weeks 0 days and 36 weeks 0 days), or delayed (at between 37 weeks 0 days and 38 weeks 0 days). An experienced practitioner undertook the ECV procedure, and repeat ECV procedures were allowed. Tocolytics and use of epidural analgesia were included as part of the protocol. The primary outcome was the rate of noncephalic presentation at birth. An intention-to-treat analysis was used. RESULTS: Data were received for 232 women, with 116 women in each of the early and delayed ECV groups. Of these, 86.2% in the early ECV group and 67.2% in the delayed ECV group had at least one ECV performed. The rate of noncephalic presentation at birth in the early ECV group was 66 of 116 (56.9%) and 77 of 116 (66.4%) in the delayed ECV group (relative risk [RR] [95% CI] 0.86 [0.70-1.05], P =.09). The rate of serious fetal complications and the rate of preterm birth at <37 weeks were not significantly increased in the early ECV group compared with the delayed ECV group (6.9% vs 7.8%, RR [95% CI] 0.89 [0.36-2.22], P =.69 and 8.6% vs 6.1%, RR [95% CI] 1.42 [0.56-3.59], P =.31, respectively). The rate of cesarean section in the early ECV group was 75 of 116 (64.7%) and 83 of 116 (71.6%) in the delayed ECV group (RR [95% CI] 0.90 [0.76-1.08], P =.32). Neonatal outcomes were comparable in the two groups. The rate of reversion to noncephalic was low in both groups. The majority of women in both groups indicated that they would consider having an ECV in another pregnancy. CONCLUSION: Early ECV performed at 34 to 36 weeks compared with 37 to 38 weeks may reduce the risk of noncephalic presentation at delivery. A large pragmatic trial of early ECV is now required to assess this approach further in terms of cesarean section rates and neonatal outcomes before changes in clinical practice.