Prediction of uterine dehiscence using ultrasonographic parameters of cesarean section scar in the nonpregnant uterus: a prospective observational study

BACKGROUND: Every year 1.5 million cesarean section procedures are performed worldwide. As many women decide to get pregnant again, the population of pregnant women with a history of cesarean section is growing rapidly. For these women prediction of cesarean section scar performance is still a serious clinical problem. METHODS: Starting in 2005, the study included 308 nonpregnant women with a history of low transverse cesarean section. The following ultrasonographic parameters of the cesarean section scar in the nonpregnant uterus were assessed: the residual myometrial thickness (RMT) and the width (W) and the depth (D) of the triangular hypoechoic scar niche. During 8 years of follow-up, 41 of these women were referred to our department for delivery. In all cases, a repeat cesarean section was performed and the lower uterine segment was assessed. Two independent statistical methods namely the logit model and Decision Tree analysis were used to determine the relation between the appearance of the cesarean section scar in the nonpregnat state and the performance of the scar in the next pregnancy. RESULTS: The logit model revealed that the D/RMT ratio showed significant correlation with cesarean section scar dehiscence (P-value of 0.007). Specifically, a D/RMT ratio value greater than 1.3035 indicated that the likelihood of dehiscence was greater than 50%. The Decision Tree analysis revealed that a diagnosis of dehiscence versus non-dehiscence could be based solely on one criterion, a D/RMT ratio of at least 0.785. The sensitivity of this method was 71%, and the specificity was 94%. CONCLUSIONS: Assessment of the cesarean section scar in the nonpregant uterus can be used to predict the occurrence of scar dehiscence in the next pregnancy.