探讨食管肿瘤内镜黏膜下剥离术（ESD）后电凝综合征（PEECS）的影响因素，构建列线图预测模型并验证。

回顾性选择2019年1月至2024年1月江苏省人民医院收治的食管肿瘤患者1336例为研究对象，收集患者性别、肿瘤位置、病理分型等临床资料，以7∶3比例将患者随机分为建模组和验证组，比较两组患者临床资料；采用建模组患者临床资料，以是否发生PEECS为因变量，进行单因素和多因素分析，基于多因素分析结果，构建列线图模型，并对模型的区分度、预测准确性、拟合度及临床效用进行验证。

1336例患者中，135例发生PEECS，发生率为10.10%。基于建模组患者临床资料，单因素分析结果显示，非PEECS和PEECS患者的性别、病理分型、肿瘤性质、术中出血、糖尿病、年龄、肿瘤直径、手术时间比较，差异均有统计学意义（P<0.05）；多因素分析结果显示，性别（OR=3.228，95% CI 1.521~6.851）、糖尿病（OR=7.666，95% CI 1.614~36.400）、肿瘤直径（OR=1.240，95% CI 1.167~1.317）、手术时间（OR=1.185，95% CI 1.142~1.230）均为食管肿瘤患者PEECS的独立危险因素（P<0.05）。基于建模组多因素分析结果构建列线图模型，ROC曲线分析显示，AUC=0.954（95% CI=0.930~0.979）；Hosmer-Lemeshow检验χ²=3.678（P=0.885）。DCA曲线和CIC分析显示，风险阈值范围在0.0~1.0时，使用该模型预测食管肿瘤PEECS，可获得较高的正向净收益，且使用该模型可较准确地预测食管肿瘤是否发生PEECS。

本研究构建列线图模型对食管肿瘤PEECS的区分度良好，预测准确性较高，具有较好的临床效用，可为食管肿瘤患者PEECS的早期干预提供参考依据。

To explore the influencing factors of post endoscopic submucosal dissection (ESD) electrocoagulation syndrome (PEECS) in esophageal tumors, construct a nomogram prediction model, and validate it.

A total of 1 336 patients with esophageal tumor admitted to Jiangsu Provincial People's Hospital from January 2019 to January 2024 were selected as the study objects. Clinical data such as gender, tumor location and pathological classification of patients were collected. Patients were randomly divided into modeling group and validation group at a ratio of 7∶3, and clinical data of patients in the two groups were compared. The clinical data of patients in the modeling group were used, and the occurrence of PEECS was taken as the dependent variable to conduct univariate and multi-factor analysis. Based on the results of multi-factor analysis, the nomogram model was constructed, and the differentiation, prediction accuracy, fit degree and clinical effectiveness of the model were verified.

Among 1 336 patients, 135 developed PEECS, with an incidence rate of 10.10%. Based on the clinical data of patients in the modeling group, the results of single factor analysis showed that there were statistically significant differences between non PEECS and PEECS patients in gender, pathological classification, tumor nature, intraoperative bleeding, diabetes, age, tumor diameter, and operation time (P<0.05); Multivariate analysis showed that gender (OR=3.228, 95% CI: 1.521-6.851), diabetes (OR=7.666, 95% CI: 1.614-36.400), tumor diameter (OR=1.240, 95% CI: 1.167-1.317), and operation time (OR=1.185, 95% CI: 1.142-1.230) were all independent risk factors for PEECS in patients with esophageal tumors (P<0.05). A nomogram graphical model was constructed based on the results of the multifactorial analysis of the modeling group, and the results of the ROC curve analysis showed that the AUC=0.954 (95% CI: 0.930-0.979); the Hosmer-Lemeshow test showed that χ²=3.678 (P=0.885). The results of the DCA curves and the CIC analyses showed that, with the risk thresholds in the range of 0.0 to 1.0, the use of the model to predict PEECS in esophageal tumors resulted in a high positive net benefit, and the use of the model was more accurate in predicting whether PEECS occurred in esophageal tumors.

This study constructs a nomogram model with good discrimination and high prediction accuracy for PEECS in esophageal tumors, which has good clinical efficacy and can provide reference for early intervention of PEECS in esophageal tumor patients.