Objective To analyze the risk factors of recurrence after endoscopic submucosal dissection (ESD) for early esophageal cancer by using Lasso-Logistic regression model.

Methods A total of 300 patients with early esophageal cancer from January 2019 to October 2020 were selected as research objects, and all of them were treated with ESD. The recurrence rate at 12 months after surgery was recorded, and the risk factors of recurrence after ESD for early esophageal cancer were analyzed by using Lasso-Logistic regression model.

Results At 12-month postoperative follow-up, 10 cases were lost due to poor follow-up awareness, and 5 cases were lost due to change of telephone number. A total of 28 patients experienced recurrence, with a recurrence rate of 9.82% (28/285), including 18 cases of mediastinal lymph node metastasis (64.29%) and 10 cases of distant metastasis (35.71%). In this study, we chose the optimal model at lambda. min=0.004, and a total of 3 variables (mean lesion diameter, degree of peri-traumatic mucosal defect, and pathological staging) entered the model; the fitting effect was assessed by using the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC), and the results indicated that the variables screened by the Lasso-Logistic regression model fitted and predicted relatively well.

Conclusion ESD has a significant efficacy in the treatment of patients with early esophageal cancer, but there is a risk of recurrence after surgery, and mean lesion diameter, degree of peri-traumatic mucosal defect and pathological staging are the influencing factors.