Suppr超能文献

用于预测首次食管静脉曲张结扎术后再出血事件的视觉预测模型。

Visual prediction models for predicting rebleeding events after the first esophageal variceal ligation.

作者信息

Li Bingbing, Li Yong, Peng Yu, Yi Jun, Gu Lei, Li Shizhe, Liu Xiaowei, Li Fujun, Wu Yu

机构信息

Department of Gastroenterology, Xiangya Hospital, Central South University, Changsha, 410008, Hunan, China.

Hunan International Scientific and Technological Cooperation Base of Artificial Intelligence Computer Aided Diagnosis and Treatment for Digestive Disease, Xiangya Hospital, Central South University, Changsha, 410008, Hunan, China.

出版信息

Surg Endosc. 2025 Apr 21. doi: 10.1007/s00464-025-11714-4.

Abstract

BACKGROUND

Sequential endoscopic variceal ligation (EVL) is a standard therapy for esophageal variceal bleeding, yet poor compliance often compromises its efficacy. This study aimed to identify high-risk factors for rebleeding after initial EVL and establish a visual predictive model to improve patient follow-up and education.

METHODS

In this retrospective study, we enrolled patients who underwent first-time EVL for esophageal variceal bleeding at Xiangya Hospital, Central South University, between January 2017 and January 2023. Patients were allocated to training and validation cohorts. Clinical data were systematically collected, and rebleeding incidence rates within 6 weeks and 6 months post-EVL were analyzed. Logistic regression was used to explore high-risk factors influencing postoperative rebleeding and to establish visual models, evaluated with a validation set.

RESULTS

The study involved 629 patients in the training set and 312 in the validation set. Rebleeding occurred in 34.8% (219/629) and 80.1% (504/629) of patients within 6 weeks and 6 months after initial EVL, respectively. Esophageal variceal severity and platelet count emerged as shared independent risk factors for both 6-week and 6-month rebleeding. The nomogram models demonstrated robust predictive accuracy in the validation cohort, with areas under the curve (AUC) of 0.942 (95% CI 0.901-0.983) for 6-week rebleeding risk and 0.852 (95% CI 0.759-0.945) for 6-month rebleeding risk.

CONCLUSION

This study proposes a visual prediction model designed to assess the risk of rebleeding following initial EVL in patients with esophageal variceal hemorrhage to optimize clinical decision-making and improve patient management strategies.

摘要

背景

序贯内镜下静脉曲张结扎术(EVL)是治疗食管静脉曲张出血的标准疗法,但依从性差常常会影响其疗效。本研究旨在确定初次EVL后再出血的高危因素,并建立一个可视化预测模型,以改善患者随访及健康教育。

方法

在这项回顾性研究中,我们纳入了2017年1月至2023年1月期间在中南大学湘雅医院因食管静脉曲张出血接受首次EVL的患者。将患者分为训练队列和验证队列。系统收集临床数据,并分析EVL后6周和6个月内的再出血发生率。采用逻辑回归分析影响术后再出血的高危因素,并建立可视化模型,用验证集进行评估。

结果

训练集纳入629例患者,验证集纳入312例患者。初次EVL后6周和6个月内再出血的患者分别占34.8%(219/629)和80.1%(504/629)。食管静脉曲张严重程度和血小板计数是6周和6个月再出血的共同独立危险因素。列线图模型在验证队列中显示出强大的预测准确性,6周再出血风险的曲线下面积(AUC)为0.942(95%CI 0.901-0.983),6个月再出血风险的AUC为0.852(95%CI 0.759-0.945)。

结论

本研究提出了一种可视化预测模型,旨在评估食管静脉曲张出血患者初次EVL后再出血的风险,以优化临床决策并改善患者管理策略。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验