Ultrasound surveillance of the cervix in twin gestations: management of cervical incompetency.

Fifty-one consecutive twin pregnancies studied prospectively were compared with 153 consecutive control twin gestations that were concurrently delivered, but chosen retrospectively. Study patients were managed by a protocol that used weekly ultrasound surveillance combined with clinical assessment. Control subjects were not managed by protocol or ultrasound surveillance. Seven of 51 study patients (13.7%) were diagnosed as having cervical incompetency before 30 weeks' gestation (range 18-28 weeks, mean 24). In comparison, 14 of 153 controls (9.2%) delivered before 30 weeks' gestation (range 18-29 weeks, mean 25). The mean cervical length, dilatation, and membrane herniation (funneling) at diagnosis were 1.9, 2.2, and 3.6 cm, respectively, for patients with cervical incompetency. There was a significant difference in perinatal mortality. All study twins survived, whereas nine control subjects delivered at a mean gestational age of 22.7 weeks, with a loss of 17 infants. Life table analysis demonstrated that a significantly greater proportion of controls delivered very low birth weight, premature twins. The use of ultrasound and clinical criteria to select patients for cerclage placement helped prevent birth of the youngest and smallest twins and significantly decreased perinatal mortality in the study group. Routine cerclage placement is not recommended for twin gestations, but multifetal gestations may benefit from ultrasound surveillance for cervical incompetency.