早期胃癌内镜黏膜下剥离术后非治愈性切除的危险因素

Risk Factors of Non-Curative Resection after Endoscopic Submucosal Dissection for Early Gastric Cancer

摘要: 目的：探讨早期胃癌(Early Gastric Cancer, EGC)患者行内镜黏膜下剥离术(Endoscopic Submucosal Dissection, ESD)后非治愈性切除的危险因素，并构建术前预测模型。方法：回顾性分析2020年1月1日至2024年1月1日于烟台毓璜顶医院行ESD治疗且病理确诊为EGC的302例患者的临床资料。根据术后病理结果将患者分为治愈性切除组(Curative Resection, CR)和非治愈性切除(Non-curative Resection, NCR)组。对比分析两组患者的临床特征、内镜下表现、肿瘤病理特征及8项炎症营养指标，采用单因素和多因素Logistic回归分析筛选NCR的独立危险因素，并基于术前可获取的变量构建预测模型。结果：302例患者中，NCR组55例(18.2%)。单因素分析显示，年龄、肿瘤大小、肿瘤部位、分化程度、浸润深度、溃疡、黏膜红斑、萎缩与NCR相关(P < 0.05)。多因素分析证实，肿瘤大小(OR = 1.729, P < 0.001)、溃疡(OR = 7.379, P < 0.001)、黏膜红斑(OR = 3.624, P < 0.001)和萎缩(OR = 0.162, P < 0.001)是NCR的独立危险因素。预后营养指数(Prognostic Nutritional Index, PNI)、淋巴细胞/单核细胞比值(Lymphocyte-to-Monocyte Ratio, LMR)和中性粒细胞/淋巴细胞比值(Neutrophil-to-Lymphocyte Ratio, NLR)等炎症营养指标与NCR无显著相关性。基于年龄、肿瘤大小、肿瘤部位、溃疡、黏膜红斑、萎缩构建的诺莫图预测NCR的AUC为0.853，内部验证校准曲线显示模型具有良好的准确性与区分度。结论：肿瘤大小、溃疡、黏膜红斑和萎缩早期胃癌ESD术后NCR的独立危险因素。基于术前临床及内镜特征构建的诺莫图对NCR具有良好的预测效能，有助于术前识别高风险患者并指导个体化治疗决策。

Abstract: Objective: To investigate the risk factors for noncurative resection (NCR) after endoscopic submucosal dissection (ESD) in patients with early gastric cancer (EGC) and to develop a preoperative prediction model. Methods: A retrospective analysis was conducted on 302 EGC patients who underwent ESD at Yantai Yuhuangding Hospital between January 1, 2020, and January 1, 2024. Patients were divided into curative resection (CR) and non-curative resection (NCR) groups on the basis of postoperative pathology. Clinical characteristics, endoscopic findings, tumor pathological features, and eight inflammatory/nutritional indicators were compared between the groups. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for NCR. A predictive nomogram was constructed using preoperative variables. Results: Among the 302 patients, 55 (18.2%) were in the NCR group. Univariate analysis revealed that age, tumor size, tumor location, differentiation degree, depth of invasion, ulceration, mucosal redness, and atrophy were associated with NCR (P < 0.05). Multivariate analysis confirmed that tumor size (OR = 1.729, P < 0.001), ulceration (OR = 7.379, P < 0.001), mucosal redness (OR = 3.624, P < 0.001), and atrophy (OR = 0.162, P < 0.001) were independent risk factors for NCR. Inflammatory/nutritional indices, such as the prognostic nutritional index (PNI), lymphocyte-to-monocyte ratio (LMR) and neutrophil-to-lymphocyte ratio (NLR), were not significantly correlated with the NCR. The nomogram constructed on the basis of age, tumor size, tumor location, ulceration, mucosal redness, and atrophy achieved an AUC of 0.853 for predicting NCR. Internal validation via bootstrap resampling revealed good calibration and discrimination. Conclusion: Tumor size, ulceration, mucosal redness, and atrophy are independent risk factors for NCR after ESD for EGC. The preoperative nomogram based on clinical and endoscopic features demonstrated good predictive performance for NCR, which can assist in identifying high-risk patients preoperatively and guiding individualized treatment strategies.