Endoscopic sleeve gastroplasty reintervention score using supervised machine learning

Reintervention following Endoscopic sleeve gastroplasty (ESG) can be indicated due to postprocedural complications from various pre-interventional or postprocedural comorbidities. We developed and internally validated an ESG reintervention scoring system (ESGRS) that determines the individualized risk of reintervention within the first 30 days following ESG.We used data from a sample of 3583 patients who underwent ESG in the Metabolic and Bariatric Surgery Accreditation Quality Improvement Program (MBSAQIP) database (2016 to 2021). The Least Absolute Shrinkage and Selection Operator (LASSO) penalized regression was utilized to select the most promising predictors of reintervention following ESG within 30 days. The predictive variables extracted by LASSO regression were entered into multivariate analysis to generate ESGRS by using the coefficients of the statistically significant variables. The Model performance was assessed using receiver operator curves by 10-fold cross-validation.Eleven variables were selected by LASSO regression and used in the final multivariate analysis. ESGRS was inferred using the above five factors (history of foregut surgery, preoperative AC use, female gender, ASA score >=2, and hypertension.) weighted by their regression coefficients in the multivariable logistic regression model. The area under the curve of ESGRS was 0.74 (95%CI: 0.70, 0.78). For ESGRS, the optimal cut point was 67.9 (high risk versus low risk), with a sensitivity of 0.76 and specificity of 0.71.ESGRS serves the purpose of aiding clinicians in preoperative risk stratification of patients undergoing ESG while guiding about factors contributing to a higher risk of reintervention.