# Development and validation of a preoperative clinical parameter-based nomogram to predict overt hepatic encephalopathy within 1 year after transjugular intrahepatic portosystemic shunt

## TL;DR

This study creates a model to predict the risk of hepatic encephalopathy after a TIPS procedure in cirrhosis patients using preoperative clinical data.

## Contribution

A new nomogram using LASSO and logistic regression improves OHE prediction over existing scores like MELD and CTP.

## Key findings

- The model achieved an AUC of 0.8835 in the training cohort and 0.858 in the validation cohort.
- Five predictors were selected: PNI, age, previous hepatic encephalopathy, serum ammonia, and creatinine.
- The model outperformed MELD and CTP scores in predicting OHE risk.

## Abstract

Transjugular intrahepatic portosystemic shunt (TIPS) is an important intervention for relieving portal hypertension-related complications in patients with decompensated cirrhosis. However, over-hepatic encephalopathy (OHE) after TIPS is common and significantly impacts patients’ prognosis and quality of life. There is an urgent need for an effective predictive model to evaluate the risk of OHE. This study aims to develop and validate a practical, accessible, and high-performance predictive model for OHE based on preoperative clinical parameters.

A total of 440 patients with decompensated cirrhosis who underwent their first TIPS procedure between January 2017 and December 2023 were retrospectively enrolled and randomly divided into training (n = 310) and validation (n = 130) cohorts in a 7:3 ratio. Least absolute shrinkage and selection operator (LASSO) regression was used for variable selection, followed by multivariate logistic regression to construct the predictive model, which was visualized as a nomogram. The model’s performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, decision curve analysis (DCA), and clinical impact curves (CIC).

LASSO regression selected five predictors from 34 variables: prognostic nutritional index (PNI), age, previous history of hepatic encephalopathy, serum ammonia, and creatinine. The model achieved an AUC of 0.8835 (95% CI: 0.8408–0.9262) in the training cohort, outperforming MELD (AUC: 0.7204) and CTP scores (AUC: 0.6576). In the validation cohort, the AUC was 0.858, indicating good discrimination. Calibration curves, DCA, and CIC also demonstrated strong model accuracy and clinical utility.

The prediction model based on preoperative clinical parameters accurately assesses the 1-year risk of OHE after TIPS in patients with cirrhosis and may serve as a practical tool for clinical decision-making.