Predictive value of endoscopic features of early gastric cancer for non-curative outcome of endoscopic resection
10.3760/cma.j.cn321463-20200601-00488
- VernacularTitle:早期胃癌内镜下特征对内镜下切除术非治愈性切除的预测意义
- Author:
Ruohan GUO
1
;
Xi WU
;
Long ZOU
;
Weixun ZHOU
;
Tao GUO
;
Qiang WANG
;
Yunlu FENG
;
Qingwei JIANG
;
Kun ZHANG
;
Ruinan LIU
;
Luolin WANG
;
Aiming YANG
Author Information
1. 北京协和医院消化内科 100730
- Keywords:
Risk factors;
Early gastric cancer;
Endoscopic resection;
Non-curative resection
- From:
Chinese Journal of Digestive Endoscopy
2021;38(10):806-810
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the endoscopic features of early gastric cancer (EGC) related to non-curative endoscopic resection, and to construct an assessment model to quantify the risk of non-curative resection.Methods:From August 2006 to October 2019, 378 lesions that underwent endoscopic resection and were diagnosed pathological as EGC in the Department of Gastroenterology, Peking Union Medical College Hospital were included in this case-control study.Seventy-eight (20.6%) non-curative resection lesions were included in the observation group, and 234 lesions which selected from 300 lesions of curative resection were included in the control group according to the difference of operation year ±1 with the observation group, and the ratio of 1∶3 of the observation group to the control group. Univariate and multivariate logistic regression analysis were performed to explore the risk factors for non-curative resection. The independent risk factor with the minimum β coefficient was assigned 1 point, and the remaining factors were scored according to the ratio of their β coefficient to the minimum. A predictive model was established to analyze the 378 lesions.The non-curative resection rates of lesions of different scores were calculated. Results:Univariate analysis showed that the lesion diameter, the location, redness, ulcer or ulcer scar, fold interruption, fold entanglement, and invasion depth observed with endoscopic ultrasonography (EUS) were associated with non-curative resection of EGC lesions ( P<0.05), and contact or spontaneous bleeding may be associated with non-curative resection ( P=0.068). Multivariate logistic regression analysis showed that submucosal involvement (VS confined to the mucosa: β=0.901, P=0.011, OR=2.46, 95% CI: 1.23-4.92), lesion diameter of 3-<5 cm (VS <3 cm: β=0.723, P=0.038, OR=2.06, 95% CI: 1.04-4.09), lesion diameter of ≥5 cm (VS <3 cm: β=2.078, P=0.003, OR=7.99, 95% CI: 2.02-31.66), location in the upper 1/3 of the stomach (VS lower 1/3: β=1.540, P<0.001, OR=4.66, 95% CI: 2.30-9.45), and fold interruption ( β=2.287, P=0.008, OR=1.93, 95% CI: 0.95-3.93) were independent risk factors for non-curative resection of EGC lesions. The factor of lesion diameter of 3-<5 cm and submucosal involvement were assigned 1 point respectively, location in the upper 1/3 of the stomach was assigned 2 points, diameter of ≥5 cm and fold interruption were assigned 3 points respectively, and other factors were assigned 0 point. Then the analysis of 378 lesions showed that the probability of non-curative resection at ≥2 points was 41.9% (37/93), 4 times as much as that at 0 [11.5% (25/217)]. Conclusion:EGC lesions with diameter ≥3 cm, located in the upper 1/3 of the stomach, interrupted folds or submucosal involvement are highly related to non-curative resection. The predictive model based on these factors achieves satisfactory efficacy, but it still needs further validation in larger cohorts.
