Objective To analyze the independent predictors for hypertriglyceridemia (HTG) remission one year after bariatric metabolic surgery, and to construct a nomogram prediction model. Methods A total of 136 obese patients with preoperative HTG, receiving bariatric metabolic surgery were enrolled, and they were randomly divided into training set (n=95) or validation set (n=41). According to the HTG remission status one year after surgery, patients in the training set were assigned to group A (HTG remission, n=80) or group B (HTG non⁃remission, n=15). Basic clinical data, body composition analysis indicators, biochemical metabolic indicators, and abdominal imaging data were analyzed in the two groups. Univariate and multivariate Logistic regression models were conducted to screen independent predictors associated with HTG remission 1 year after bariatric metabolic surgery. A nomogram prediction model was constructed based on the independent predictors. The predictive performance, degree of fitting, and clinical utility of the model were evaluated in the training and validation sets through receiver operating characteristic curves, calibration curves, decision curve analysis (DCA), and clinical impact curves (CIC). The SHAP algorithm was applied to conduct interpretability analysis of the prediction model through quantification of feature importance, single sample prediction influence analysis, and SHAP dependence analysis. Results The proportion of patients with type 2 diabetes mellitus, preoperative fasting blood glucose level, HbA1c level, triglyceride level, triglyceride⁃glucose index (TYG), and visceral adipose tissue area (VATA)⁃to⁃subcutaneous adipose tissue area (SATA) ratio in group B were higher than those in group A, while the preoperative HDL⁃C level was lower than that in group A (P<0.05). Logistic regression analysis results indicated that the TYG and VATA/SATA ratio were independent predictors for HTG remission 1 year after bariatric metabolic surgery (P<0.05). Model performance validation results revealed that in the training and validation sets, the areas under the curve of the nomogram prediction model constructed based on TYG and VATA/SATA ratio were 0.913 and 0.787, respectively. The calibration curve suggested that the model had favorable degree of fitting. DCA and CIC analysis indicated that the model could bring net benefits to clinical decision⁃making at different risk thresholds, with significant clinical utility. The average absolute SHAP values of TYG and VATA/SATA ratio were 1.354 and 0.551, respectively. Both features negatively correlated with the prediction results of HTG remission 1 year after bariatric metabolic surgery, with TYG having a more significant inhibitory effect. Conclusion TYG and VATA/SATA ratio are independent predictors for HTG remission 1 year after bariatric metabolic surgery, with TYG having a more significant impact on the prediction results. The combined application of the two can comprehensively reflect the preoperative glucose and lipid metabolism status and fat distribution characteristics of patients, providing a key basis for postoperative remission risk assessment.

无