Cervical cerclage outcomes by indication and cervical length across gestational-age thresholds: a historical cohort study

Abstract Background Cervical cerclage reduces the risk of preterm birth, but factors associated with its success may not be uniform. Among singleton pregnancies with ultrasound- or physical examination-indicated cerclage (no history-based indication), we examined whether factors associated with delivery differ by gestational-age threshold. Methods This single-center historical cohort included 114 women undergoing cerclage between May 2020 and December 2025: ultrasound-indicated with cervical length 10 to < 20 mm (Group 1, n = 41) or < 10 mm (Group 2, n = 32), and physical examination-indicated cervical dilatation (Group 3, n = 41). All cerclages used the McDonald technique with polypropylene monofilament suture. The primary outcome was gestational age at delivery, assessed at ≥ 34, < 32, and < 28 weeks. Firth penalized multivariable logistic regression evaluated study group, maternal age, and gestational age at cerclage as primary covariates; post-cerclage cervical length was analyzed separately as a secondary post-treatment-adjusted variable. All models were exploratory, and p values were nominal. Results In these exploratory models, factors associated with delivery differed across thresholds. For delivery ≥ 34 weeks, gestational age at cerclage was the only covariate independently associated with the outcome (aOR 1.28 per week, 95% CI 1.04 to 1.59; p = 0.021). For delivery before 32 weeks (aOR 2.96, 95% CI 1.08 to 8.66; p = 0.034) and before 28 weeks (aOR 9.36, 95% CI 1.98 to 91.11; p = 0.003), physical examination-indicated cerclage was the only covariate that reached significance; 11 of the 14 deliveries before 28 weeks occurred in this group, so this estimate is imprecise and near-separating. The numerically higher rate of delivery ≥ 34 weeks in Group 2 than Group 1 (78.1% vs. 70.7%) was not statistically significant. Apparent discrimination was moderate (AUC 0.69 to 0.77), without internal validation. Conclusions In this exploratory single-center cohort, the factors associated with delivery varied with the gestational-age threshold examined. Near term, later gestational age at cerclage was associated with delivery ≥ 34 weeks; at earlier thresholds, physical examination-indicated cerclage identified a higher-risk group for very and extreme preterm delivery, although these estimates were imprecise and partly confounded by indication severity and co-interventions. Findings should be interpreted as hypothesis-generating and require confirmation in larger prospective studies.