凶险性前置胎盘剖宫产术中大量出血的危险因素分析及风险预测模型的建立

Analysis of risk factors for massive hemorrhage during cesarean section of pernicious placenta previa and establishment of risk prediction model

目的探讨凶险性前置胎盘剖宫产术中大出血的影响因素,并建立风险预测模型。方法收集2017年1月至2021年12月该院妇产科收治的凶险性前置胎盘并行剖宫产终止妊娠的340例孕妇临床资料,分为普通出血组(术中出血量<2000 mL,n=200)和大量出血组(术中出血量≥2000 mL,n=140)。比较两组孕妇临床特征、本次妊娠临床资料及胎儿情况、影像学信息等,综合单因素分析中P<0.05的因素及临床上凶险性前置胎盘孕妇剖宫产术中大出血的可能影响因素,行二分类多因素logistic回归分析并建立风险预测模型,采用Hosmer-Lemeshow拟合优度检验和受试者工作特征(ROC)曲线评估模型的拟合效果及区分度。结果多因素logistic回归分析结果显示,流产次数、胎盘厚度、合并胎盘植入、既往剖宫产次数及胎儿性别是凶险性前置胎盘孕妇剖宫产术中发生大量出血(≥2000 mL)的独立影响因素(P<0.05)。预测模型公式:P=Log(Y/1-Y),Y=0.396+1.371×(流产次数=3次)+1.248×(流产次数≥4次)-0.351×(胎盘厚度)+0.624×(合并胎盘植入)+0.974×(既往剖宫产次数≥2次)+0.638×(女=0,男=1)。Hosmer-Lemeshow拟合优度检验结果显示该预测模型具有较好的校准能力(χ~2=77.825,P<0.001),ROC曲线下面积为0.768(95%CI:0.717~0.820),特异度为83.0%,阳性预测值为70.2%,阴性预测值为73.5%。结论构建的凶险性前置胎盘孕妇剖宫产术中大量出血风险预测模型具有良好的效能,用于产前评估有助于识别高危产妇,为临床制订输血方案和防治不良妊娠结局提供依据。

Objective To explore the influencing factors of massive hemorrhage during cesarean section of pernicious placenta previa,and establish a risk prediction model.Methods The clinical data of 340 pregnant women with pernicious placenta previa who underwent cesarean section for termination of pregnancy in this hospital from January 2017 to December 2021 were collected.They were divided into the common hemorrhage group(the amount of intraoperative blood loss<2000 mL,n=200)and massive hemorrhage group(the amount of intraoperative blood loss≥2000 mL,n=140).The clinical characteristics of pregnant women,clinical data of this pregnancy,situation of the fetus,and imaging information were compared between the two groups.Combining the variables with a P value<0.05 in the univariate analysis and the possible influencing factors of massive hemorrhage during cesarean section in pregnant women with pernicious placenta previa in clinical practice,the binary multivariate logistic regression analysis was conducted,and a risk prediction model was established.Hosmer-Lemeshow goodness of fit test and receiver operating characteristic(ROC)curve were used to evaluate the fitting effect and discrimination of the model.Results The results of multivariate logistic regression analysis showed that the number of abortions,placental thickness,combining with placental implantation,number of previous cesarean sections and fetal gender were the independent influencing factors for massive hemorrhage(≥2000 mL)during cesarean section in pregnant women with pernicious placenta previa(P<0.05).The prediction model formula:P=Log(Y/1-Y),Y=0.396+1.371×(number of abortions=three times)+1.248×(number of abortions≥four times)-0.351×(placental thickness)+0.624×(combining with placental implantation)+0.974×(two or more previous cesarean sections)+0.638×(female=0,male=1).The results of Hosmer-Lemeshow goodness of fit test showed that the prediction model had good calibration ability(χ2=77.825,P<0.001).The area under the ROC curve was 0.768(95%CI:0.717-0.820),the specificity was 83.0%,the positive predictive value was 70.2%,and the negative predictive value was 73.5%.Conclusion The risk prediction model of massive hemorrhage during cesarean section in pregnant women with pernicious placenta previa has good performance.It is helpful to identify high-risk pregnant women in the prenatal evaluation,and provide a basis for formulating the blood transfusion plan in clinic,and prevention and treatment of adverse pregnancy outcomes.