LAUSR

BACKGROUND The purpose of this research was to determine the index that contributes the most to assess the effectiveness of weight loss 1 year following bariatric surgery, and to implement it as the clinical outcome to develop and confirm a nomogram to predict whether bariatric surgery would be effective. METHODS Patient information was extracted from the Chinese Obesity and Metabolic Surgery Database for this retrospective study. The most contributing weight loss effectiveness evaluation index was created using canonical correlation analysis (CCA), and the predictors were screened using logistic regression analysis. A nomogram for estimating the likelihood of effectiveness of weight loss was constructed, and its performance was further verified. RESULTS Information was obtained for 540 patients including 30 variables. According to CCA, ≥ 25 percentage total weight loss (%TWL) was found to be the most correlated with patient information and contributed the most as a weight loss effectiveness evaluation index. Logistic regression analysis and nomogram scores identified age, surgical strategy, abdominal circumference, weight loss history, and hyperlipidemia as predictors of effectiveness in weight loss. The prediction model's discrimination, accuracy, and clinical benefit were demonstrated by the consistency index (C-index), calibration curve, and decision curve analysis (DCA). CONCLUSIONS We determined 25% TWL as an index for weight loss effectiveness assessment by CCA, and next established and validated a nomogram, which demonstrated promising performance in predicting the probability of effectiveness of weight loss in bariatric surgery. The nomogram might be a valuable tool in clinical practice.