EASL-CLIF criteria outperform NACSELD criteria for diagnosis and prognostication in ACLF

Abstract

There is no consensus on the best definition for acute-on-chronic liver failure (ACLF). In this study, we compared the prevalence and 30-day all-cause and transplant-free mortality of patients with ACLF identified by European Association for the Study of the Liver-Chronic Liver Failure Consortium (EASL-CLIF) and North American Consortium for the Study of End-stage Liver Disease (NACSELD) criteria. We performed this comparative analysis using the United Network for Organ Sharing (UNOS) data from January 11, 2016 to August 31,2020. A total of 10,198 (21%) adult patients had EASL-CLIF ACLF grade 1-3, but of these only 15.3% had ACLF by NACSELD. Of the 2,562 with EASL-CLIF ACLF grade 3, only 48.8% had NACSELD-ACLF, 16.8% had no organ failure (OF) and 34.4% had 1 OF. The 30-day all-cause mortality was 1.5%, 7.7%, 13.3% and 25.8% for EASL-CLIF grade 0-3, respectively, and it was 15.4% and 28.1% in those without and with NACSELD-ACLF. When EASL-CLIF grade 3 patients were stratified by NACSELD OF, the mortality ranged from 18.6% with no OF to 41.0% with 4 OFs. The 30-day transplant-free mortality in those with no OF by NACSELD was 2.7%, but when the same group is stratified by EASL-CLIF grades 0-3, the mortality rates were 1.5%, 10.5%, 43.5% and 86%, respectively; the mortality rates ranged from 3.0% to 75.7% in those with 1 OF by NACSELD. There is a clear discordance in the prevalence and 30-day mortality rates of patients with ACLF identified by the EASL-CLIF and NACSELD criteria. EASL-CLIF criteria have a better sensitivity to detect ACLF and have a better prognostic capability. There is no consensus on the definition of acute-on-chronic liver failure. European (EASL-CLIF) and North American (NACSELD) consortia have each proposed a commonly used definition. In this study, we compared the prevalence and short-term (30-day) mortality based on these definitions. Using a very large data set, we observed that there was a significant discordance in the prevalence and mortality based on these criteria. EASL-CLIF criteria appeared to be more sensitive to identify acute-on-chronic liver failure, and were better at predicting all-cause and short-term mortality.