SAN DIEGO — Women with prior early spontaneous preterm birth and a midtrimester sonographic cervical length of less than 25 mm may benefit from cerclage, but the benefit is most pronounced when the cervical length is less than 15 mm, results from a large, multicenter, randomized study showed.
“Although clinicians have recommended cerclage for shortened cervical lengths, previous randomized trials have not supported this practice,” Dr. John Owen said at the annual meeting of the Society for Maternal-Fetal Medicine.
A recent meta-analysis of four randomized trials of cerclage for shortened cervical length uncovered a relationship between pregnancy history and cerclage effectiveness. Cerclage was helpful only in singletons—it was harmful in multiples—but it was especially helpful in women who'd had a prior preterm birth (Obstet. Gynecol. 2005;106:181–9).
“Our hypothesis was that in women with a prior early spontaneous preterm birth [gestational age less than 34 weeks] and cervical length less than 25 mm, cerclage would reduce the rate of preterm birth before 35 weeks' gestation,” said Dr. Owen of the department of obstetrics and gynecology at the University of Alabama at Birmingham.
To test the hypothesis, he and his colleagues at 15 centers in the United States, known as the Vaginal Ultrasound Trial Consortium, studied 1,014 women with a prior spontaneous birth at less than 34 weeks and a current singleton pregnancy who underwent serial ultrasound evaluation in the period beginning at 16 weeks and ending before 23 weeks (that is, no later than 22 weeks and 6 days). Of these, 301 women with a cervical length of less than 25 mm were randomized to either cerclage or no cerclage.
Vaginal ultrasound exams lasted a minimum of 5 minutes to allow the clinician to observe any spontaneous shortening, and included fundal pressure as a provocative measure to induce cervical shortening. The scans were scheduled every 2 weeks as long as the cervical length remained at least 30 mm. They were performed weekly if the cervical length shortened to 25–29 mm. The last scan was scheduled to occur just before the 23 weeks' gestational point.
Dr. Owen reported that the cerclage and no-cerclage groups were similar in terms of race/ethnicity, mean cervical length (18.7 vs. 19.5 mm, respectively), mean gestational age at randomization (19.4 vs. 19.5 weeks), and mean gestational age of earliest prior preterm birth (24.4 vs. 24.9 weeks).
Preterm birth before 35 weeks occurred in 42% of the no-cerclage group, compared with 32% of the cerclage group, a difference that revealed a statistical trend (P = .09).
However, further analysis revealed that women in the cerclage group maintained their pregnancies significantly better if their cervical length was less than 15 mm (odds ratio, 0.23), but there was no significantly positive effect if their cervical length was 15–24 mm (OR, 0.84).
Dr. Owen had no conflicts to disclose.