Premium
CLIF ‐ SOFA scoring system accurately predicts short‐term mortality in acutely decompensated patients with alcoholic cirrhosis: a retrospective analysis
Author(s) -
Lee Minjong,
Lee JeongHoon,
Oh Sohee,
Jang Yoonhyuk,
Lee Wonik,
Lee Hyung Joo,
Yoo Jeongju,
Choi WonMook,
Cho Young Youn,
Cho Yuri,
Lee Dong Hyeon,
Lee Yun Bin,
Yu Su Jong,
Yi NamJoon,
Lee KwangWoong,
Kim Yoon Jun,
Yoon JungHwan,
Suh KyungSuk,
Lee HyoSuk
Publication year - 2015
Publication title -
liver international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.873
H-Index - 110
eISSN - 1478-3231
pISSN - 1478-3223
DOI - 10.1111/liv.12683
Subject(s) - medicine , cirrhosis , model for end stage liver disease , receiver operating characteristic , alcoholic liver disease , sofa score , population , liver disease , mechanical ventilation , chronic liver disease , gastroenterology , intensive care medicine , liver transplantation , intensive care unit , transplantation , environmental health
Abstract Background & Aims Accurate prognostication of acute‐on‐chronic liver failure ( ACLF ) is essential for therapeutic decisions. Our aim was to validate a novel scoring system for predicting mortality, the chronic liver failure‐sequential organ failure assessment ( CLIF ‐ SOFA ), in a population of Asian patients with ACLF . Methods A total of 345 patients with acutely decompensated alcoholic cirrhosis were selected for study, comparing areas under the receiver operating characteristic ( AUROC ) curves of CLIF ‐ SOFA and five existing scoring systems for end‐stage liver disease [model for end‐stage liver disease ( MELD ), MELD ‐Na, Refit‐ MELD , Refit‐ MELD ‐Na, and Child‐Turcotte‐Pugh]. Results CLIF ‐ SOFA displayed the highest AUROC of 0.943 significantly outperforming all five reference methods in predicting short‐term mortality at Week 4 (all P < 0.001) by competing risk analysis. In 262 patients given supportive care only, the power of CLIF ‐ SOFA to predict short‐term mortality was high ( AUROC : 0.952 at Week 1; 0.959 at Week 4), again surpassing the other methods (all P < 0.001). For the remaining 83 liver transplant recipients, CLIF ‐ SOFA also excelled in predicting 12‐week mortality ( AUROC : 0.978); and high‐grade ACLF by CLIF ‐ SOFA was associated with prolonged postoperative mechanical support (i.e. mechanical ventilation and renal replacement therapy) and ICU stays (all P < 0.05). Conclusions CLIF ‐ SOFA enables more accurate prediction of short‐term mortality in patients with acutely decompensated alcoholic cirrhosis than other available scoring systems and is useful in predicting both 12‐week mortality and the need for mechanical support after liver transplantation.