前置胎盘剖宫产术中MT的预测模型

【目的】探讨前置胎盘(placenta previa,PP)剖宫产术中大量输血(massive transfusion,MT)的预测模型。【方法】选择2016年1月至2020年6月在日喀则市人民医院接受剖宫产手术的PP产妇临床资料,根据术中是否MT(massive transfusion,MT)分为MT组[输注≥5单位浓缩红细胞(packed red blood cells,PRBC)]和NMT组(输注<4单位PRBC)。采用单因素和多因素Logistic分析术中MT的危险因素,采用多因素Logistic分析结果拟定术中MT危险系数(MT risk factor,MTR),采用受试者工作曲线(receiver operating curve,ROC)分析MTR与单个危险因素预测MT的曲线下面积(area under curve,AUC)。【结果】MT组和NMT组前置状态、流产史、胎盘陷窝、低回声层缺失、前壁胎盘等比较差异有统计学意义(P<0.05)。二分类多因素Logistic分析显示:完全前置胎盘、低回声层缺失和前壁胎盘是NMT的独立风险因素(P<0.05)。MTR公式为MTR=2.437×(完全前置胎盘)+4.382×(低回声层缺失)+1.167×(前壁胎盘)。ROC曲线分析结果显示:MTR评分预测术中MT的AUC为0.918,高于完全前置胎盘(0.787)、低回声层缺失(0.718)和前壁胎盘(0.717)的AUC(Z=3.785、5.266、4.298,P均<0.001)。【结论】本研究通过多因素Logistic分析建立了PP患者剖宫产术中MT预测模型,临床医师可根据MTR预测术中MT发生的风险,提前做好MT的准备工作。

Predictive Model for Massive Blood Transfusion during Cesarean Section of Placenta Previa

【Objective】To establish a predictive model for MT(massive transfusion)during cesarean section of placenta previa(PP).【Methods】The medical records of PP patients who underwent cesarean section in our hospital from January 2016 to June 2020 were analyzed.According to the occurrence of massive intraoperative blood transfusion(MT),patients were divided into the MT group(infusion≥5 units of packed red blood cells)and the non MT group(NMT,infusion<4 units of PRBC).Relationship between clinical factors and intraoperative massive blood transfusion was examined by univariate and multivariate logistic analysis.Single factor and multivariate logistic analysis were used to analyze the risk factors of intraoperative MT,the results of multivariate logistic analysis were used to draw up the intraoperative MT risk factor(MTR),and the receiver operating curve(ROC)was used to analyze the area under curve(AUC)that predicted MT with MTR and single risk factor.【Results】There were significant differences between the MT group and the NMT group in preposition,history of miscarriage,placental lacuna,hypoechoic layer loss,and anterior placenta(P<0.05).Binary multivariate logistic analysis showed that complete placenta previa,hypoechoic layer loss and anterior placenta were independent risk factors of NMT(P<0.05).The MTR formula was MTR=2.437×(complete placenta previa)+4.382×(hypoechoic layer loss)+1.167×(anterior placenta).ROC curve showed that the AUC of MTR score predicting intraoperative MT was 0.918,which was higher than that of complete placenta previa(0.787),hypoechoic layer loss(0.718)and anterior placenta(0.717)(Z=3.785,5.266,4.298,P<0.001).【Conclusion】This study established the prediction model of MT during cesarean section in PP patients by multivariate logistic analysis.Clinicians can predict the risk of MT during cesarean section according to MT risk score,and prepare for massive blood transfusion in the early stage.