Lower uterine segment thickness assessed by transvaginal ultrasound before labor induction: reproducibility analysis and relationship with delivery outcome

ABSTRACT Objectives To evaluate the reproducibility of lower uterine segment (LUS) thickness measurement before induction of labor (IOL), and to assess the relationship between LUS thickness and IOL outcomes. Methods This was a prospective cohort study of pregnant women undergoing IOL at term, conducted in a single tertiary hospital between July 2014 and February 2017. Women with a singleton pregnancy at ≥ 37 weeks' gestation, with a live fetus in cephalic presentation and a Bishop score of ≤ 6, were eligible for inclusion. Both nulliparous and parous women, and those with a previous Cesarean section (CS), were eligible. All women underwent transvaginal ultrasound assessment before IOL admission, and cervical length and LUS thickness were measured offline after delivery. Maternal and obstetric characteristics and Bishop score were recorded. The main outcome was the overall rate of CS after IOL, and secondary outcomes were CS for either failure to progress in the active phase of labor or failed IOL, and CS for failed IOL only. Interobserver agreement for measurement of LUS thickness between two operators was assessed using the intraclass correlation coefficient (ICC) and Bland–Altman analysis with the ANOVA test to evaluate systematic bias. Univariable and multivariable analysis were employed to evaluate the relationship between clinical and sonographic characteristics and IOL outcomes. Results Of 265 women included in the analysis, 195 (73.6%) had a vaginal delivery and 70 (26.4%) required a CS after IOL. Reproducibility analysis showed excellent interobserver agreement for the measurement of LUS thickness (ICC, 0.96 (95% CI, 0.93–0.98)). On Bland–Altman analysis, the mean difference in LUS thickness between the two operators was 0.15 mm (95% limits of agreement, –1.84 to 2.14 mm), and there was no evidence of systematic bias (ANOVA test, P = 0.46). Univariable analysis showed that LUS thickness was associated significantly with overall CS ( P = 0.002), CS for failure to progress in the active phase of labor or failed IOL ( P = 0.03) and CS for failed IOL ( P = 0.037). On multivariable logistic regression analysis, LUS thickness was an independent predictive factor for overall CS (odds ratio (OR), 1.149 (95% CI, 1.031–1.281)) and CS for failure to progress in the active phase of labor or failed IOL (OR, 1.226 (95% CI, 1.039–1.445)). Conclusions In women undergoing IOL at term, measurement of LUS thickness is feasible and reproducible, and is associated significantly with IOL outcome. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.