OBJECTIVE: This study aimed to determine the risk factors and establish a risk score for post-transjugular intrahepatic portosystemic shunt (TIPS) overt hepatic encephalopathy (OHE). METHODS: Altogether 299 and 62 cirrhotic patients receiving TIPS from January 2015 to March 2018 were divided into the derivation and validation cohorts, respectively. The data of the derivation cohort were analyzed for risk factors of post-TIPS OHE. A risk score was established from the derivation cohort and verified by the validation cohort. RESULTS: During a median follow-up of 112.6 weeks, 52 (17.4%) patients in the derivation cohort experienced post-TIPS OHE. Logistic regression showed that alcoholic cirrhosis (odds ratio [OR] 3.068, 95% confidence interval [CI] 1.423-6.613, P = 0.004), stent diameter of 10 mm (OR 12.046 [95% CI 2.308-62.862], P = 0.003), portal pressure gradient (PPG) decrement ≥60% (OR 3.548 [95% CI 1.741-7.230], P < 0.001), model for end-stage liver disease (MELD) score ≥10 (OR 2.695 [95% CI 1.203-6.035], P = 0.016), blood ammonia (OR 1.009 [95% CI 1.000-1.018], P = 0.043) and notable hydrothorax (OR 4.393 [95% CI 1.554-12.415], P = 0.005) were associated with an increased risk of post-TIPS OHE. The risk score reached a promising risk evaluation of post-TIPS OHE when verified by the validation cohort (sensitivity 71.4%, specificity 70.7%, accuracy 71.0%). CONCLUSIONS: Alcoholic cirrhosis and notable hydrothorax are independent risk factors for post-TIPS OHE in liver cirrhosis, together with the stent diameter of 10 mm, PPG decrement ≥60%, MELD score ≥10 and blood ammonia. The established risk score is reliable to identify high-risk individuals of developing post-TIPS OHE.
OBJECTIVE: This study aimed to determine the risk factors and establish a risk score for post-transjugular intrahepatic portosystemic shunt (TIPS) overt hepatic encephalopathy (OHE). METHODS: Altogether 299 and 62 cirrhotic patients receiving TIPS from January 2015 to March 2018 were divided into the derivation and validation cohorts, respectively. The data of the derivation cohort were analyzed for risk factors of post-TIPS OHE. A risk score was established from the derivation cohort and verified by the validation cohort. RESULTS: During a median follow-up of 112.6 weeks, 52 (17.4%) patients in the derivation cohort experienced post-TIPS OHE. Logistic regression showed that alcoholic cirrhosis (odds ratio [OR] 3.068, 95% confidence interval [CI] 1.423-6.613, P = 0.004), stent diameter of 10 mm (OR 12.046 [95% CI 2.308-62.862], P = 0.003), portal pressure gradient (PPG) decrement ≥60% (OR 3.548 [95% CI 1.741-7.230], P < 0.001), model for end-stage liver disease (MELD) score ≥10 (OR 2.695 [95% CI 1.203-6.035], P = 0.016), blood ammonia (OR 1.009 [95% CI 1.000-1.018], P = 0.043) and notable hydrothorax (OR 4.393 [95% CI 1.554-12.415], P = 0.005) were associated with an increased risk of post-TIPS OHE. The risk score reached a promising risk evaluation of post-TIPS OHE when verified by the validation cohort (sensitivity 71.4%, specificity 70.7%, accuracy 71.0%). CONCLUSIONS:Alcoholic cirrhosis and notable hydrothorax are independent risk factors for post-TIPS OHE in liver cirrhosis, together with the stent diameter of 10 mm, PPG decrement ≥60%, MELD score ≥10 and blood ammonia. The established risk score is reliable to identify high-risk individuals of developing post-TIPS OHE.