Prediction of recurrence of early gastric carcinoma after endoscopic submucosal dissection

Abstract

Background: The aim of this retrospective study was to predict the post-endoscopic submucosal dissection (ESD) recurrence risk of early gastric carcinoma (EGC) using a nomogram model, and to provide valuable evidence for preventing the recurrence. Methods: The patients with EGC receiving ESD between March 2010 and February 2018 were enrolled. They were followed up once every three months after ESD. At the end of follow-up, they were assigned into recurrence and non-recurrence groups. The independent risk factors for post-ESD recurrence were identified using Cox regression analysis. A nomogram prediction model was established, and its predictive efficiency was assessed by plotting receiver operating characteristic (ROC) curve. Results: Helicobacter pylori (HP) infection, number of positive lymph nodes >3 and a large amount of intraoperative hemorrhage were risk factors for post-ESD recurrence. According to multivariate Cox regression analysis, HP infection and number of positive lymph nodes >3 were independent predictors. The nomogram model had a good fitting effect, and the area under ROC curve was 0.933 (95% confidence interval: 0.919-0.947), suggesting a high predictive efficiency. Conclusion: Positive lymph nodes and HP infection are closely correlated with the recurrence risk after ESD in EGC patients. The established model is a quantitative tool for predicting recurrence to improve the prognosis. Keywords: Endoscopy; gastric carcinoma; nomogram model; recurrence; submucosal dissection.