RANDOMIZED TRIAL OF ENDOVAGINAL ULTRASOUND IN PRETERM PREMATURE RUPTURE OF MEMBRANES

Objective: To assess the effect of weekly endovaginal ultrasound on th e incidence of maternal infection and the time from rupture to deliver y in women with preterm premature rupture of membranes (PROM). Methods : Women with singleton pregnancies complicated by preterm PROM at 24-3 4 weeks' gestation were assigned randomly to groups having endovaginal ultrasound or no vaginal sonography. Along with the standard expectan t management, the endovaginal-ultrasound group had weekly vaginal prob e ultrasound scans. Power analysis based upon expected maternal infect ion required a sample size of 45 patients in each group. Results: Fort y-seven and 45 subjects were assigned to the no-probe and probe groups , respectively. The latency period, defined as days from rupture to de livery, was 9.8 and 11.7 days for the no-probe and probe groups, respe ctively (95% confidence interval -5.9, 2.1). There were no significant differences in the incidence of chorioamnionitis (28% and 20%), endom etritis (6% and 9%), or neonatal infection (17% and 20%). The mean lat ency period in women who went into spontaneous labor and whose initial cervical length was 3.0 cm or less was 9.4 days, compared with 11.0 d ays if the cervix exceeded 3.0 cm, a nonsignificant difference. There were three neonatal deaths, all in the probe group and none directly r elated to infection. Conclusions: Endovaginal ultrasound in patients w hose pregnancies are complicated by preterm PROM does not appear to in crease the incidence of maternal infection. Copyright (C) 1997 by The American College of Obstetricians and Gynecologists.