Emergency cerclage: Neonatal outcomes and evaluation of prognostic scores – A decade of experience at a tertiary center
Objectives: Emergency cervical cerclage is a high-risk surgical procedure associated with maternal and fetal risks, including preterm birth. Authors have developed scores to try to predict the probability of success following emergency cerclage, but these tools have yet to be externally validated. Our main objective was to assess the preterm birth rate before 32 weeks following emergency cerclage. Secondary objectives included assessing perinatal outcomes and the predictive validity of two published scores in our setting. We hypothesized that predictive scores, such as Fuchs and Kokia scores, may not perform optimally in our population due to variability in clinical and biological characteristics that influence eligibility for cerclage across different settings.
Study design: We conducted a retrospective, monocentric study in a tertiary maternity hospital. All patients who underwent emergency cerclage between January 1st, 2010, and December 31st, 2021, were included. The main outcome was preterm birth rate before 32 weeks. Secondary outcomes were need for maternal hospitalization after initial hospitalization for cerclage, preterm premature rupture of membrane, and perinatal death. The rate of preterm birth before 32 weeks and adverse perinatal outcomes were estimated. To assess the predictive validity of Fuchs’ score in our population, we performed a multivariable logistic regression model including the score’s variables and constructed receiver operating characteristic (ROC) curve. To assess the validity of Kokia’s score in our population, we studied the correlation between the initial score and gestational age at delivery using Pearson’s correlation coefficient.
Results: Our population was composed of 64 women who underwent an emergency cerclage. The median gestational age at delivery was 30.6 weeks (IQR (24.2; 37.6)). The preterm birth rate before 32 weeks was 54.7 %. The rate of perinatal death was 30.7 %. There were no statistical differences between the patients delivered before and after 32 weeks. The multivariable logistic regression model showed that none of the variables of Fuchs’s score were significantly associated with preterm birth risk in our population. The AUC of the ROC curve using Fuchs’ score was 0.63. There was a minor yet significant correlation between Kokia’s score and gestational age at delivery (r = 0.26 – p = 0.04).
Conclusions: Our study underscores the persistent risks associated with cervical insufficiency, including very preterm birth. It highlights the variability in performance of predictive scores across different populations, underscoring the need for external validation in diverse clinical settings.
https://doi.org/10.1016/j.ejogrb.2024.10.019
By Camille Le Ray
