Your browser doesn't support javascript.
loading
Construction and evaluation of a predictive model for esophageal stenosis after endoscopic submucosal dissection of early esophageal cancer and precancerous lesions / 中华消化内镜杂志
Chinese Journal of Digestive Endoscopy ; (12): 728-732, 2021.
Article in Chinese | WPRIM | ID: wpr-912166
ABSTRACT

Objective:

To investigate the risk factors of esophageal stenosis after endoscopic submucosal dissection (ESD) of early esophageal cancer and precancerous lesions, and to construct and assess a predictive model for esophageal stenosis.

Methods:

Data of 421 patients with early esophageal cancer or precancerous lesions who underwent ESD and were confirmed by pathology in the First Affiliated Hospital of Zhengzhou University between January 2015 and April 2020 were retrospectively analyzed.Eighty-nine cases developed postoperative esophageal stenosis (stenosis group) and 332 cases did not (non-stenosis group). Risk factors of esophageal stenosis were investigated by univariate and multivariate logistic regression analysis. Independent risk factors were used as predictors to construct a nomogram model by using the lasso algorithm.The accuracy of the model was evaluated by the consistency index (C-index) and the calibration curve. Bootstrap was applied to internal verification to avoid over-fitting of the model.

Results:

Univariate analysis showed that postoperative pathology, depth of infiltration, median long and short diameters of the specimen, circumferential resection range, and muscularis propria injury were related to postoperative esophageal stenosis( P<0.05). Further multivariate logistic regression analysis showed that the resection range≥1/2 of the circumference (VS <1/2 circumference P<0.01, OR=48.453, 95% CI 11.288-207.983), muscularis propria injury( P<0.01, OR=4.671, 95% CI 2.283-9.557)and longitudinal length≥50 mm (VS <50 mm P=0.008, OR=2.741, 95% CI 1.299-5.785) were independent risk factors for esophageal stenosis after ESD. The nomogram model was constructed through the lasso algorithm by taking the longitudinal length, circumferential resection range, and muscularis propria injury as the predictive factors. The C-index of the nomogram was 0.934 (95% CI 0.909-0.959)and was 0.931 after 100 times of Bootstrap internal sampling, which meant the prediction probability of the model in the calibration curve was in good agreement with the actual observation probability.

Conclusion:

Circumferential resection range ≥1/2, muscularis propria injury, and longitudinal length≥50 mm are independent risk factors for post-ESD esophageal stenosis. The nomogram model constructed by the above three indicators as predictors shows an ideal prediction effect on esophageal stenosis after ESD for early esophageal cancer and precancerous lesions, which is helpful to establish a standard plan for early intervention in patients at high risk of stenosis after ESD.

Full text: Available Index: WPRIM (Western Pacific) Type of study: Prognostic study / Risk factors Language: Chinese Journal: Chinese Journal of Digestive Endoscopy Year: 2021 Type: Article

Similar

MEDLINE

...
LILACS

LIS

Full text: Available Index: WPRIM (Western Pacific) Type of study: Prognostic study / Risk factors Language: Chinese Journal: Chinese Journal of Digestive Endoscopy Year: 2021 Type: Article