Effect of transabdominal versus transvaginal cerclage on preterm birth and neonatal outcomes among patients with a history of cervical insufficiency.

Abstract

The 2020 Multicentre Abdominal vs Vaginal Randomised Intervention of Cerclage trial found that transabdominal cerclage placed via open laparotomy reduces the risk of spontaneous preterm birth before 32 weeks relative to transvaginal cerclage in a very high-risk obstetric population. It is not known whether the results of the trial generalize to obstetric populations with fewer risk factors, on average, for recurrent spontaneous preterm birth, or to patients without history of failed transvaginal cerclage (an inclusion criterium in the trial). 1) To estimate the effect of transabdominal cerclage vs transvaginal cerclage on early preterm delivery (<34 weeks) among patients with history of cervical insufficiency at a quaternary care center; 2) to estimate the effect within a subgroup of patients without history of failed transvaginal cerclage; and 3) to describe associated complications of placement and delivery and neonatal outcomes. We identified a cohort of adult singleton pregnancies who received history-indicated cerclage at a quaternary care center in the United States. The cohort consisted of patients with a history of ≥1 spontaneous deliveries (birth or fetal loss) <28 weeks and patients with a history of failed transvaginal cerclage, defined as ≥1 spontaneous preterm deliveries <34 weeks with cervical cerclage in situ. All eligible patients who received transabdominal cerclage were compared to a sample of eligible patients who received transvaginal cerclage. We performed survival analysis with inverse probability weights to adjust for potential sources of bias. Effects were estimated as risk difference, risk ratio, and 95% confidence intervals. Subgroup analyses were performed among patients without a history of failed transvaginal cerclage. The risks of surgical, delivery, and neonatal outcomes were described. One hundred eighty-eight patients were included, of whom 87 received transabdominal cerclage (99% laparoscopic) and 101 received transvaginal cerclage. Twenty-six patients (30%) with transabdominal cerclage had no history of failed transvaginal cerclage, the majority of whom had additional clinical reasons why transabdominal cerclage was offered. After adjusting for confounding, the adjusted risk of early preterm delivery was 5.5% in the transabdominal group (95% confidence interval, 2.0%, 9.4%) and 18.7% in the transvaginal group (95% confidence interval, 6.8%, 31.4%), risk difference = -13.1% (95% confidence interval, -26.6%, -0.5%), and risk ratio = 0.30 (95% confidence interval, 0.10, 0.94). Among patients without history of failed transvaginal cerclage (N = 104 total, N = 26 in transabdominal group), the risk difference was -12.6% (-21.6%, -4.1%). Risks of placement and delivery complications were comparable across groups but included a few serious uterine complications in the transabdominal cerclage group. Neonatal intensive care unit admission occurred in 35% and 23% of neonates in the transabdominal and transvaginal groups, respectively. Among women with history of ≥1 preterm deliveries <28 weeks or failed transvaginal cerclage <34 weeks, transabdominal cerclage reduced the risk of preterm delivery <34 weeks compared to transvaginal cerclage, consistent with the Multicentre Abdominal vs Vaginal Randomised Intervention of Cerclage trial findings. Conclusions about efficacy among patients without history of failed transvaginal cerclage are limited by small sample size. Further research is necessary to understand whether the benefits of transabdominal cerclage in subgroups of patients outweigh risks of abdominal surgery and cesarean delivery.