Differential predictive utility of preoperative cervical length for preterm birth among various indications for cervical cerclage: a retrospective cohort study

To investigate the predictive utility of preoperative cervical length (CL) for preterm birth (PTB) in cervical cerclage with different surgical indications. This study included data from 289 pregnant women who underwent transvaginal cervical cerclage at the Wuhan Tongji Hospital (Hubei, China) between 2015 and 2024. Participants were divided into three groups according to surgical indication(s) for cerclage: history indicated (<i>n</i> = 173); ultrasound indicated (<i>n</i> = 89); and examination indicated (<i>n</i> = 27). The area under the receiver operating characteristic curve (AUC) revealed that preoperative CL for PTB was greater for ultrasound-indicated cerclage (AUC 0.6613, cutoff 13.5 mm) than that for history-indicated cerclage (AUC 0.5675, optimal cutoff 30.5 mm). In history-indicated cerclage, there were no difference in the rate of PTBs between preoperative CL &gt; 30.5 mm and ≤ 30.5 mm, nor in the rate of postoperative CL ≤ 15 mm during the fourth week. In the ultrasound-indicated group, compared with preoperative CL &gt; 13.5 mm, women with a CL ≤ 13.5 mm exhibited a higher rate of PTB (82.1% versus 48.0%; <i>p</i> &lt; .001), and a higher rate of postoperative CL ≤ 15 mm in the fourth week (71.4% versus 37.5%; <i>p</i> = 0.009). The correlation between preoperative CL and PTB varied depending on the indication(s) for cerclage, and changes in postoperative CL contributed to varying predictive utility.