内镜黏膜下剥离术后异时性早期胃癌的内镜及病理特征分析

Endoscopic and pathological characteristics of metachronous early gastric cancer after endoscopic submucosal dissection

摘要:

目的 探讨早期胃癌内镜黏膜下剥离术（endoscopic submucosal dissection，ESD）后异时性早期胃癌的内镜及病理特征。方法 回顾性收集2005年1月1日至2022年12月31日于北京大学第三医院消化科住院接受ESD治疗的451例早期胃癌连续病例，其中252例符合研究要求且内镜随访超过1年者纳入回顾性分析。采用多因素Cox回归分析早期胃癌患者ESD术后出现异时性早期胃癌的独立危险因素，采用Pearson列联系数分析早期胃癌初发病灶与异时性病灶的内镜下相关性，采用独立样本t检验、χ2检验或Fisher精确概率法分析早期胃癌初发病灶与异时性病灶在内镜及病理特征上的差异以及每年2次随访的异时性早期胃癌组与每年1次随访的异时性早期胃癌组在异时性早期胃癌符合ESD绝对适应证的占比和最大径方面的差异，采用Kaplan‑Meier曲线分析异时性早期胃癌的累积发病风险。结果 在中位随访时间40个月中，26例［10.3%（26/252）］出现异时性早期胃癌，异时性早期胃癌平均出现时间43.9个月。多因素Cox回归分析发现，早期胃癌初发病灶垂直部位位于胃中1/3（HR=3.783，95%CI：1.300~11.011，P=0.015）、水平部位位于胃前壁（HR=3.934，95%CI：1.113~13.904，P=0.033）、最大径<15 mm（HR=3.034，95%CI：1.074~8.571，P=0.036）是出现异时性早期胃癌的独立危险因素。Pearson列联系数分析显示，早期胃癌初发病灶与异时性病灶在垂直部位（C=0.375，P=0.372）、水平部位（C=0.508，P=0.434）、大体形态（C=0.287，P=0.675）和有无溃疡（C=0.194，P=0.313）方面并无明显相关性。相比早期胃癌初发病灶，异时性病灶多位于胃后壁（胃小弯/胃大弯/胃前壁/胃后壁：11/2/1/12比96/49/46/61，P=0.031）、多为分化型（分化型/未分化型：26/0比214/38，P=0.032）、最大径更小［（8.08±5.99）mm比（13.95±10.26）mm，t=4.383，P<0.001］。每年1次随访的异时性早期胃癌组与每年2次随访的异时性早期胃癌组比较，异时性早期胃癌符合ESD绝对适应证的占比组间差异无统计学意义（14/16比9/9，P=0.520），且异时性胃癌的最大径组间差异也无统计学意义［（8.11±6.94）mm比（6.67±4.35）mm，t=-0.275，P=0.535］。异时性早期胃癌累积发病风险曲线显示，其累积发病风险在10年后不再明显增加。结论 对胃中1/3、胃前壁以及最大径偏小的早期胃癌ESD切除患者，术后尤要加强内镜随访，随访过程中需重点关注胃后壁，且每年1次的内镜随访方案最好维持至术后10年。

Abstract:

Objective To investigate the endoscopic and pathological characteristics of metachronous early gastric cancer (EGC) after endoscopic submucosal dissection (ESD) for EGC. Methods Data of 451 consecutive EGC patients treated with ESD at the Department of Gastroenterology, Peking University Third Hospital between 1 January, 2005 and 31 December, 2022 were retrospectively collected, of which 252 patients who met the criteria and had endoscopic follow-up ≥ 1 year were enrolled in the retrospective dynamic cohort. Multivariate Cox regression analysis was used to identify independent risk factors for metachronous EGC after ESD. Pearson''s contingency coefficient was applied to analyze endoscopic correlation between the index and metachronous lesions. T-test, χ2 test, and Fisher exact test were used to compare endoscopic pathological features between index and metachronous lesions, the proportion of lesions meeting absolute ESD indication and their maximum diameters between patients undergoing annual vs bi-annual follow-up. Kaplan-Meier analysis assessed the cumulative incidence of metachronous EGC. Results During a median follow-up of 40 months, 26 patients [10.3% (26/252)] developed metachronous EGC, with a mean interval of 43.9 months. Multivariate Cox regression identified the independent risk factors of index lesions including location in the middle third of the stomach (HR=3.783, 95%CI: 1.300-11.011, P=0.015), in the anterior wall (HR=3.934, 95%CI: 1.113-13.904, P=0.033), and the maximum diameter <15 mm (HR=3.034, 95%CI: 1.074-8.571, P=0.036). Pearson''s contingency coefficient showed no significant concordance between index and metachronous lesions for vertical location (C=0.375, P=0.372), horizontal location (C=0.508, P=0.434), gross morphology (C=0.287, P=0.675), or ulcer presence (C=0.194, P=0.313). Compared to index lesions, metachronous lesions were more frequently located on the posterior wall (lesser curvature/greater curvature/anterior wall/posterior wall: 11/2/1/12 VS 96/49/46/61, P=0.031), more often differentiated (differentiated/undifferentiated: 26/0 VS 214/38, P=0.032), and smaller in maximum diameter (8.08±5.99 mm VS 13.95±10.26 mm, t=4.383, P<0.001). No significant differences were observed between patients undergoing annual vs bi-annual follow-up in the proportion of metachronous lesions meeting absolute ESD indication (14/16 VS 9/9, P=0.520) or in maximum diameter (8.11±6.94 mm VS 6.67±4.35 mm, t=-0.275, P=0.535). The cumulative incidence curve of metachronous EGC plateaued after 10 years. Conclusion Patients with EGC located in the middle third of the stomach, in the anterior wall, or of smaller diameter need intensive endoscopic surveillance after ESD. Posterior wall deserves particular attention during follow-up, with annual endoscopy recommended for at least 10 years post-ESD.