Development and validation of modified Liaoning score for predicting the prognosis of liver cirrhosis: a retrospective, international multicenter, observational study

Abstract

ABSTRACT Background Liaoning score has been developed and validated to predict the risk of esophageal varices in liver cirrhosis. This study aimed to further modify the Liaoning score by combining clinical and laboratory parameters to predict the long-term outcome of cirrhotic patients. Methods First, 474 cirrhotic patients were retrospectively enrolled from Shenyang, China as the training cohort. Independent predictors for death were identified by competing risk analyses, and then a new prognostic model, called as modified Liaoning score, was developed. Its performance was externally validated at three centers from Fuzhou, China (n = 1944), Jinan, China (n = 485), and São Paulo, Brazil (n = 221). Results Age, total bilirubin (TBIL), albumin (ALB), serum creatinine (SCr), and Liaoning score were independently associated with death in the training cohort. Modified Liaoning score = 0.159×Liaoning score + 0.010×TBIL(µmol/L)+0.029×age(years)+0.011×SCr(µmol/L)-0.037×ALB(g/L). The area under curve of modified Liaoning score was 0.714 (95%CI = 0.655–0.773), which was higher than that of Child-Pugh score (0.707, 95%CI = 0.645–0.770), MELD score (0.687, 95%CI = 0.623–0.751), and Liaoning score (0.583, 95%CI = 0.513–0.654). A modified Liaoning score of ≥ 1.296 suggested a higher cumulative incidence of death in liver cirrhosis (p < 0.001). Modified Liaoning score still had the highest prognostic performance in Chinese and Brazilian validation cohorts. Conclusions Modified Liaoning score can be considered for predicting the long-term outcome of cirrhotic patients.