基于列线图构建食管癌术后早期胃排空障碍的预测模型

The prediction model of early gastric emptying disorder after esophageal cancer surgery was constructed based on the Nomogram

目的:探讨影响食管癌术后早期发生胃排空障碍的危险因素,描绘列线图风险预测模型并进行验证。方法:选取2020年1月-2023年1月医院220例食管癌病人作为研究对象,依据病人术后早期是否发生胃排空障碍分为发生组和未发生组,对相关因素进行Logistic回归分析,构建列线图风险预测模型并进行验证。结果:220例食管癌病人术后早期胃排空障碍发生率为16.8%。Logistic回归分析结果显示,是否合并糖尿病、不良心理反应、术后补液量、手术时间、术后血清清蛋白是食管癌病人术后早期发生胃排空障碍的独立危险因素(P<0.05),以此构建的风险预测模型的Hosmer-Lemeshow拟合优度检验结果显示,χ^(2)=15.027,P=0.059;ROC曲线下面积为0.888[95%CI(0.820,0.956),P<0.001],灵敏度为89.6%,特异性为75.7%,最大约登指数为-0.971,模型预测食管癌术后早期发生胃排空障碍的校正曲线总体趋势与理想曲线基本吻合,表明该模型拟合优度良好,具有较好的预测效能。模型回归方程Logit(P)=1.240×合并糖尿病+1.098×不良心理反应-0.101×术后血清清蛋白+0.017×术后补液量+0.077×手术时间。结论:基于危险因素构建的预测模型具有良好的预测效能,可为临床工作者预防及干预提供参考依据,以减少食管癌术后胃排空障碍发生。

Objective:To explore the risk factors affecting the early gastric emptying disorder after esophageal cancer surgery,and to describe and verify the Nomogram prediction model.Methods:A total of 220 patients with esophageal cancer in the hospital from January 2020 to January 2023 were selected as research objects.According to whether the patients had gastric empting disorder in the early postoperative period,the patients were divided into occurrence group and non-occurrence group.Logistic regression analysis was performed on relevant factors,and the Nomogram prediction model was constructed and verified.Results:The incidence of gastric emptying disorder was 16.8%in 220 patients with esophageal cancer.Logistic regression analysis showed that diabetes mellitus,adverse psychological reaction,postoperative fluid rehydration,operation time and postoperative serum albumin were independent risk factors for early gastric emptying disorder in patients with esophageal cancer(P<0.05).The Hosmer-Lemeshow goodness of fit test results of the risk prediction model constructed by this method show thatχ^(2)=15.027,P=0.059.The area under ROC curve was 0.888[95%CI(0.820,0.956),P<0.001],the sensitivity was 89.6%,the specificity was 75.7%,and the maximum approximate entry index was-0.971.The overall trend of the correction curve of the model for predicting the early gastric emptying disorder after esophageal cancer surgery is basically consistent with the ideal curve,indicating that the model has a good degree of fit and a good prediction efficiency.Model regression equation Logit(P)=1.240×diabetes mellitus+1.098×adverse psychological reaction-0.101×postoperative serum albumin+0.017×postoperative fluid intake+0.077×operative time.Conclusions:The prediction model based on risk factors has good predictive efficacy,and can provide reference for clinical workers to prevent and intervene,so as to reduce the occurrence of gastric emptying disorder after esophageal cancer surgery.