剖宫产术后再次妊娠阴道分娩预测模型的建立与验证

Establishment and validation of predictive model for trial of labor after cesarean section

目的建立适合我国剖宫产术后晚期妊娠阴道试产风险的预测模型。方法对住院分娩的剖宫产术后阴道试产(TOLAC)晚期妊娠孕妇进行前瞻性队列研究,结合前期研究足月妊娠TOLAC的预测模型以及利用Logistic回归分析方法建立并验证晚期妊娠TOLAC预测模型,绘制受试者工作特征曲线(ROC),并进行模型拟合优度检验与准确度验证。结果共纳入研究对象532例,其中阴道分娩组499例,试产失败转剖宫产组33例。试产成功率为93.8%,子宫破裂3例(0.6%)。多因素Logistic回归分析显示,新生儿出生体质量、入院时宫颈Bishop评分以及胎膜破裂是TOLAC结局的影响因素,其OR值分别为1.001、0.374和0.114(P均<0.05)。Logistic回归预测方程表达式为:P=1/[1+exp(0.001×新生儿出生体质量-0.98×宫颈评分-2.17×胎膜破裂-3.22)](新生儿出生体质量单位为kg;胎膜破裂取1,胎膜完整取0)。通过Hosmer-Lemeshow检验,该模型的拟合程度好(χ~2=526.190,P=0.416),模型总的预测准确率为95.30%。ROC曲线下面积为0.926(P<0.001)。结论新生儿出生体质量越小、宫颈Bishop评分越高以及入院时胎膜破裂是TOLAC成功的有利因素。由这3个因素建立的剖宫产术后再次妊娠阴道试产风险预测模型预测效能好,准确率高。

Objective To establish a suitable model for predicting the success of trial of labor after cesarean section（ TOLAC） during the third trimester for the Chinese pregnant women. Methods Pregnant women receiving TOLAC during the third trimester were recruited in this prospective cohort study. The predictive models for TOLAC during the third trimster were established and validated by the predictive models for fullterm TOLAC and logistic regression analysis. The receiver operating characteristic curve（ ROC） was delineated and the goodness-of-fit test and accuracy verification of the predictive models were performed. Results A total of 532 pregnant women were enrolled,including the vaginal delivery group（ n = 499） and the cesarean section group（ n = 33）. The success rate of trial of labor was 93. 8% and 3（ 0. 6%） had urerine rupture. Multivariable logistic regression analysis showed that neonatal birth weight（ OR = 1. 001,P〈 0. 05）,cervical Bishop score（ OR = 0. 374,P〈 0. 05） and rupture of membranes（ OR = 0. 114,P〈 0. 05） were independent factors affecting TOLAC outcomes. Logistic regression model was expressed as follows： P = 1/[1 ＋ exp（ 0. 001 ×neonatal birth weight（ kg）-0. 98 × Bishop score-2. 17 × rupture of membranes（ Yes = 1,No = 0）-3. 22） ]. The Hosmer-Lemeshow test demonstrated that the model fitted well（ χ2= 526. 190,P = 0. 416）,and the overall accuracy of the predictive model was 95. 30%. The model performed well with an AUC of 0. 926（ P 〈0. 001）. Couclusions The lower neonatal birth weight,higher cervical Bishop score,and rupture of membranes contribute to the success of TOLAC. A predictive model consisting of these three variables yields high prediction efficiency and accuracy for predicting the success of TOLAC during the third trimester after cesarean section.