医学
内镜黏膜下剥离术
计分系统
多中心研究
放射科
外科
解剖(医学)
内窥镜检查
胃
试验预测值
作者
Zhilin Liu,Linfu Zheng,Xingjie Gao,B. Xu,S H Chen,Ying Sun,Gaozhen Deng,Zixing Yan,Jiaochun Liu,Ye Zhou,Kun Lin,Chushu Li,Wen Wang,Dazhou Li
标识
DOI:10.1080/00365521.2025.2610634
摘要
Background/Aim Endoscopic submucosal dissection (ESD) for early gastric cardiac cancer (EGCC) faces anatomical challenges. This study aims to develop and validate a predictive model for ESD difficulty in EGCC, providing a basis for matching difficulty levels with endoscopists’ experience.Methods This multicenter retrospective study included 514 EGCC patients from five tertiary hospitals (2017–2025). Patients from the 900th Hospital were randomized to training (n = 206) and internal validation (n = 164) cohorts; four other hospitals formed an external cohort (n = 144). Predictors with high collinearity (VIF ≥ 5) were excluded. Predictor selection was performed using LASSO regression, followed by multivariable logistic regression. Model performance was assessed using ROC curves, calibration plots, and decision curve analysis (DCA).Results The clinical scoring model incorporated four factors: Paris type III lesion (3 points), maximum diameter ≥3 cm (1 point), submucosal invasion (5 points), and lesion located on the anterior wall of the cardia (3 points). Model performance demonstrated an AUC of 0.784 (95% CI, 0.695–0.876) in the training cohort, 0.762 (95% CI, 0.653–0.884) in internal validation, and 0.740 (95% CI, 0.636–0.845) in external validation. Difficulty stratification was defined as: easy (0, 1, or 3 points), intermediate (4, 5, or 6 points), and difficult (8 or 11 points).Conclusion This validated system optimizes endoscopist matching, reducing costs and recurrence in EGCC ESD.
科研通智能强力驱动
Strongly Powered by AbleSci AI