Prognostic Value of Model for End-Stage Liver Disease Scores in Patients with Fulminant Hepatic Failure

Abstract

BACKGROUND: This study was undertaken to investigate risk factors of mortality in patients with fulminant hepatic failure (FHF). METHODS: Fifty-three patients with FHF treated from January 2006 to April 2011 were allocated to a spontaneous survival group (group 1), a death without liver transplantation (LT) group (group 2), and an LT group (group 3). To analyze risk factors associated with mortality in FHF, we excluded group 3 patients. Clinical features, Model for End-Stage Liver Disease (MELD) scores, and King's College Hospital criteria at the time of hepatic encephalopathy in group 2 were compared with those of group 1. RESULTS: The causes of FHF were acute viral infection (n = 29, hepatitis A:B, 28:1), drugs (n = 18; including 4 acetaminophen and 14 herbal medication), autoimmune (n = 4), and miscellaneous (n = 2). Of the 53 patients, 19 were allocated to group 1, 18 to group 2, and 16 to group 3. According to univariate analysis, risk factors for mortality in group 2 were acute renal failure requiring renal replacement therapy and a MELD score ≥30 at the time of hepatic encephalopathy. However, by multivariate analysis, a MELD score ≥30 was the only independent risk factor for mortality in group 2 (P = .042; hazard ratio, 4.500). CONCLUSIONS: A MELD score ≥30 was found to be the only independent risk factor of mortality in FHF patients without LT. Therefore, the findings of this study suggest that these patients may need emergent LT for survival.