Triglyceride Glucose Index predicts the risk of hepatic encephalopathy after transjugular intrahepatic portosystemic shunt

Background Hepatic encephalopathy (HE) represents a frequent and clinically significant complication following transjugular intrahepatic portosystemic shunt (TIPS) placement, profoundly affecting both patient quality of life and long-term prognosis. Early assessment and timely identification of potential post-TIPS HE are critically important for improving clinical outcomes. Methods Cirrhotic patients undergone TIPS were retrospectively enrolled from January 2018 to November 2021 in this study. Demographic information and clinical materials were retrospectively collected. Based on the occurrence of overt hepatic encephalopathy (OHE) postoperatively, patients were stratified into no-OHE and OHE groups. Univariate and multivariate logistic regression were performed to identify predictors of post-TIPS OHEs. And a nomogram was developed using independent risk factors. ROC curve analysis was plotted to assess the predictive ability of the triglyceride glucose (TyG) index. Calibration plots and Decision curve analysis (DCA) was used to evaluate the model's fit and predictive benefits. Results A total of 146 patients were retrospectively enrolled in this study, of which 67 suffered from OHE. Compared with the no-OHE group, TyG index was significantly higher in the OHE group (8.748 (8.289,9.128) vs 8.107 (7.724,8.395), P  < 0.001). Multivariate logistic regression analysis showed that TyG index, neutrophil count, and SBP incidence were independent risk factors for post-TIPS OHE (OR 9.655, 95% CI 3.957–23.555 P  = 0.001; OR 1.272, 95% CI 1.000–1.656 P  = 0.05; OR 0.212, 95% CI 0.061–0.736 P  = 0.015, respectively). Then a nomogram model for the incidence of OHE was constructed and the ROC curve of the new model was plotted with the AUC of the new model 0.704, indicating a good predictive performance. Calibration plots and Decision curve analysis (DCA) also showed the model had good fit. Conclusions TyG index serves as an independent predictor and may represent a novel prognostic biomarker for predicting post-TIPS HE.