超声内镜参数联合临床资料构建的列线图预测食管胃底静脉曲张治疗后再出血的价值

Value of nomogram constructed with endoscopic ultrasonography parameters and clinical data in predicting rebleeding after treatment of esophageal and gastric varices

目的探讨超声内镜(EUS)参数联合临床资料构建的列线图模型预测食管胃底静脉曲张(EGV)治疗后再出血的应用价值。方法选取我院收治的186例EGV患者,根据治疗后是否出血分为出血组102例和未出血组84例,比较两组EUS参数及临床资料的差异。应用多因素Logistic回归分析筛选EGV治疗后再出血的独立影响因素,并构建列线图模型;绘制受试者工作特征(ROC)曲线分析模型预测EGV治疗后再出血的诊断效能;采用校准曲线评价模型的拟合优度;临床决策曲线分析模型的临床适用性。结果两组EGV程度、红色征、腹水、肝性脑病、终末期肝病模型评分、Child-Pugh分级评分、食管旁侧支静脉直径、门静脉直径、奇静脉直径、脾静脉直径比较差异均有统计学意义(均P<0.05)。多因素Logistic回归分析显示,红色征、腹水、Child-Pugh分级评分、食管旁侧支静脉直径、奇静脉直径均为EGV治疗后再出血的独立影响因素(均P<0.05)。ROC曲线分析显示,构建的列线图模型预测EGV治疗后再出血的曲线下面积为0.967;校准曲线显示模型拟合度较好(χ^(2)=7.303,P=0.504);临床决策曲线显示模型具有较好的临床适用性。结论EUS参数联合临床资料构建的列线图模型在预测EGV治疗后再出血中具有较好的临床应用价值。

Objective To explore the application value of the nomogram model constructed with endoscopic ultrasonography(EUS)and clinical data in predicting rebleeding after treatment of esophagogastric varices(EGV).Methods A total of 186 patients with EGV treated in our hospital were selected and divided into bleeding group(n=102)and non-bleeding group(n=84)according to whether bleeding after treatment.EUS parameters and clinical data of the two groups were compared.Multivariate Logistic regression was used to screen the independent influencing factors of rebleeding after EGV treatment,and a nomogram model was constructed.Receiver operating characteristic(ROC)curve was drawn to analyze the diagnostic efficacy of the model in predicting rebleeding after EGV treatment.Calibration curve was used to evaluate the goodness of fit of the model,and the clinical applicability of the model was analyzed by clinical decision curve.Results There were significant differences in EGV degree,red sign,ascites,hepatic encephalopathy,end-stage liver disease model score,Child-Pugh grading score,para-esophageal collateral vein diameter,portal vein diameter,azygo vein diameter and splenic vein diameter between the two groups(all P<0.05).Multiple Logistic regression analysis showed that red sign,ascites,Child-Pugh grading score,para-esophageal collateral vein diameter and azygos vein diameter were independent influencing factors for rebleeding after EGV treatment(all P<0.05).ROC curve analysis showed that the area under the curve of constructed nomogram model in predicting rebleeding after EGV treatment was 0.967.Calibration curve showed a good fitting of the model(χ^(2)=7.303,P=0.504).The clinical decision curve showed that the model had good clinical applicability.Conclusion Nomogram constructed with EUS parameters combined with clinical data have good clinical application value in predicting rebleeding after EGV treatment.