OBJECTIVE: To evaluate the value of serial post-cerclage cervical length measurements in predicting preterm birth. METHODS: This was a retrospective study of women with a singleton pregnancy who underwent prophylactic transvaginal cerclage from 2017 to 2023. Prophylactic cerclage was performed in women with an obstetric history of cervical insufficiency. After cerclage, all participants underwent routine transvaginal cervical measurement every 2-4 weeks until 32-34 weeks of gestation. Trajectory patterns of changes in cervical length were identified through latent class trajectory modeling. The association of patterns with spontaneous preterm birth and pregnancy interval was analyzed by multivariable logistic regression analysis and Cox proportional hazards regression models. The specific characteristics of changes in cervical length associated with preterm birth were identified. RESULTS: We involved 225 women and acquired a total of 1248 measurements of cervical length. Two distinct trajectories of cervical length change along gestation were identified: Class 1 (persistent shortening of cervix, n = 52 [23.11%]) and class 2 (stable cervix, n = 173 [76.89%]). Women in class 1 were highly associated with spontaneous preterm birth, with a hazard ratio (95% confidence interval [CI]) of 2.00 (1.36, 2.96). The average shortening rate ≥0.24 cm per week and cervical length ≤1.1 cm was significantly associated with spontaneous preterm birth at different gestational age. CONCLUSION: Serial sonographic cervical length measurements after prophylactic cerclage are valuable in predicting preterm birth.