胃癌前病变内镜下黏膜剥离术术后迟发性出血的风险评分模型预测  

A risk-prediction model for endoscopic submucosal dissection delayed bleeding of gastric pre-cancerous lesions

作　　者：朱怡颖 季梦遥[1] 沈磊[1] ZHU Yiying;JI Mengyao;SHEN Lei(Department of Gastroenterology,Renmin Hospital of Wuhan University,Wuhan 430060,China)

机构地区：[1]武汉大学人民医院消化内科,湖北武汉430060

出　　处：《胃肠病学和肝病学杂志》2025年第2期244-249,共6页Chinese Journal of Gastroenterology and Hepatology

摘　　要：目的探讨胃癌前病变内镜下黏膜剥离术(endoscopic submucosal dissection,ESD)术后发生迟发性出血的危险因素,并构建风险评分预测模型。方法回顾性分析2016年11月至2022年6月期间在武汉大学人民医院接受ESD治疗的胃癌前病变患者的临床资料,使用XGBoost模型筛选与术后迟发性出血相关的危险因素并构建风险预测模型,采用ROC曲线对模型进行评价,SHAP分析对模型进行解释。结果XGBoost模型筛选出了年龄、病理类型为低级别上皮内瘤变、高血压、病灶大小≥40 mm、手术时间≥120 min、女性、未使用止血夹等7个因素为与胃癌前病变ESD术后迟发性出血相关的危险因素并建立了风险预测模型,在训练组中,该模型的AUC值为0.98(0.97~0.99),灵敏度为0.91,特异性为0.95,F1值为0.92;在验证组中,该模型的AUC值为0.95(0.91~0.99),灵敏度为0.85,特异性为0.89,F1值为0.85;在测试组中,该模型的AUC值为0.95(0.90~0.99),灵敏度为0.80,特异性为0.93,F1值为0.84,显示出较强的预测能力。根据整个队列中的尤登指数(临界值=0.56),可将患者分为高危和低危两组。结论本研究建立和验证了能够预测胃癌前病变ESD术后迟发性出血风险的XGBoost预测模型,该模型可应用于临床工作中,有效预测个体的ESD术后出血风险。Objective To investigate the risk factors for postoperative delayed bleeding after endoscopic submucosal dissection(ESD)in gastric pre-cancerous lesions and construct a risk scoring prediction model.Methods A retrospective analysis of clinical data from patients with gastric pre-cancerous lesions who underwent ESD treatment at Renmin Hospital of Wuhan University between Nov.2016 and Jun.2022 was conducted.An XGBoost model was employed to identify risk factors associated with postoperative delayed bleeding and to build a risk prediction model.The model was evaluated using ROC curves,and SHAP analysis was used to interpret the model.Results The XGBoost model identified 7 factors associated with postoperative delayed bleeding in gastric precancerous lesions after ESD:age,low-grade intraepithelial neoplasia,hypertension,lesion size≥40 mm,operative time≥120 minutes,female gender,and non-use of hemoclips.A risk prediction model was established.In the training cohort,the model achieved an AUC of 0.98(0.97-0.99),a sensitivity of 0.91,a specificity of 0.95,and an F1 score of 0.92.In the validation cohort,the AUC was 0.95(0.91-0.99),with a sensitivity of 0.85,a specificity of 0.89,and an F1 score of 0.85.In the test cohort,the AUC was 0.95(0.90-0.99),the sensitivity was 0.80,the specificity was 0.93,and the F1 score was 0.84,indicating strong predictive capability.Patients were divided into high-risk and low-risk groups based on the Youden index(cut-off=0.56)in the whole cohort.Conclusion This study developed and validated an XGBoost prediction model for assessing the risk of delayed postoperative bleeding after ESD in gastric pre-cancerous lesions.This model can be applied in clinical practice to effectively predict the risk of post-ESD bleeding for individual patients.

关 键 词：胃癌前病变 内镜黏膜下剥离术 术后迟发性出血 预测模型 

分 类 号：R735.2[医药卫生—肿瘤]