S2502 Risk Stratification of Patients Admitted From the Emergency Department With Decompensated Cirrhosis

Introduction: Patients with acute-on-chronic liver failure (ACLF) have 30 d mortality between 25% to >40%, highlighting the need for early identification and aggressive management of decompensated cirrhotic patients with high short-term mortality. The LIV-4 and CLIF-C ACLF scores have been previously validated to predict mortality of patients admitted to the ICU. To date, these models have not been validated on patients presenting to the emergency department (ED). Methods: This single-center retrospective study of patients with decompensated cirrhosis admitted from the ED at Beth Israel Deaconess Medical Center between 2009 and 2022 utilized data available through the MIMIC-IV v3.1 database. Patients with decompensated cirrhosis were identified by ICD-9/10 diagnoses codes for ascites, esophageal variceal bleeding or hepatic encephalopathy. Patients were excluded with acute liver failure, no passage through the ED, or with insufficient lab or vital sign data to calculate LIV-4, CLIF-C ACLF or MELD 3.0 scores. Primary outcomes were 30 d and 90 d mortality. Categorical variables were analyzed with χ2 contingency tests. Performance of each risk score was measured by concordance (c)-index. Results: Of 32,019 admissions for patients with cirrhosis, 13,714 were identified with decompensated cirrhosis. 4,697 patients passed through the ED and had sufficient data for risk score calculation. Overall, 30 d mortality was 15.4% and 90 d mortality was 27.6%. Among patients transferred to the ICU after 48 h, mortality rates exceeded 38% after 30 d and 55% after 90 d, significantly higher than patients transferred within the first 48 h of admission (8-29% after 30 d and 20-41% after 90 d, P < 0.001, χ2 contingency test). The LIV-4, CLIF-C ACLF and MELD 3.0 models performed similarly in risk stratifying patients by mortality, with c-indices of 0.68, 0.75 and 0.66, respectively. The top 25% of patients ranked by the LIV-4, CLIF-C ACLF and MELD 3.0 scores capture up to 59% of all deceased patients after 30 days. Conclusion: All risk scores here were designed for the ICU setting but demonstrated efficacy in risk-stratifying patients with decompensated cirrhosis presenting to the ED. The CLIF-C ACLF model performed modestly better than MELD 3.0 and LIV-4 by c-index. Utilization of these risk scores would enable concentration of resources on patients with likely short-term mortality. Next steps include a prospective study to triage newly admitted cirrhotic patients for early ICU consultation.

Publication Details

Journal
The American Journal of Gastroenterology
Published
2025-10-01
DOI
https://doi.org/10.14309/01.ajg.0001137468.72854.41
Primary Topic
Liver Disease and Transplantation
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

S2502 Risk Stratification of Patients Admitted From the Emergency Department With Decompensated Cirrhosis

The American Journal of Gastroenterology
Liver Disease and Transplantation
article

S2502 Risk Stratification of Patients Admitted From the Emergency Department With Decompensated Cirrhosis

article en

Abstract

Introduction: Patients with acute-on-chronic liver failure (ACLF) have 30 d mortality between 25% to >40%, highlighting the need for early identification and aggressive management of decompensated cirrhotic patients with high short-term mortality. The LIV-4 and CLIF-C ACLF scores have been previously validated to predict mortality of patients admitted to the ICU. To date, these models have not been validated on patients presenting to the emergency department (ED). Methods: This single-center retrospective study of patients with decompensated cirrhosis admitted from the ED at Beth Israel Deaconess Medical Center between 2009 and 2022 utilized data available through the MIMIC-IV v3.1 database. Patients with decompensated cirrhosis were identified by ICD-9/10 diagnoses codes for ascites, esophageal variceal bleeding or hepatic encephalopathy. Patients were excluded with acute liver failure, no passage through the ED, or with insufficient lab or vital sign data to calculate LIV-4, CLIF-C ACLF or MELD 3.0 scores. Primary outcomes were 30 d and 90 d mortality. Categorical variables were analyzed with χ2 contingency tests. Performance of each risk score was measured by concordance (c)-index. Results: Of 32,019 admissions for patients with cirrhosis, 13,714 were identified with decompensated cirrhosis. 4,697 patients passed through the ED and had sufficient data for risk score calculation. Overall, 30 d mortality was 15.4% and 90 d mortality was 27.6%. Among patients transferred to the ICU after 48 h, mortality rates exceeded 38% after 30 d and 55% after 90 d, significantly higher than patients transferred within the first 48 h of admission (8-29% after 30 d and 20-41% after 90 d, P < 0.001, χ2 contingency test). The LIV-4, CLIF-C ACLF and MELD 3.0 models performed similarly in risk stratifying patients by mortality, with c-indices of 0.68, 0.75 and 0.66, respectively. The top 25% of patients ranked by the LIV-4, CLIF-C ACLF and MELD 3.0 scores capture up to 59% of all deceased patients after 30 days. Conclusion: All risk scores here were designed for the ICU setting but demonstrated efficacy in risk-stratifying patients with decompensated cirrhosis presenting to the ED. The CLIF-C ACLF model performed modestly better than MELD 3.0 and LIV-4 by c-index. Utilization of these risk scores would enable concentration of resources on patients with likely short-term mortality. Next steps include a prospective study to triage newly admitted cirrhotic patients for early ICU consultation.

The American Journal of GastroenterologyVol. 120(10S2)
Good health and well-being
Openalex Percentile: Top 33%
Liver Disease and Transplantation
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.