There is significant geographic variation in the etiologies and prognoses of acute liver failure (ALF). The aims of the present study were to determine the causes and short-term outcomes of ALF in Argentina, to evaluate the performance of prognostic criteria, and to identify clinical prognostic factors of death. We performed a retrospective analysis of 154 adult patients with ALF who were admitted to 6 liver transplantation (LT) programs between June 2005 and December 2011. The most frequent causes of ALF were viral hepatitis B (46 patients or 30%), autoimmune hepatitis (AIH; 40 patients or 26%), and indeterminate causes (40 patients or 26%). No acetaminophen (ACM) overdose was reported. One hundred and twenty one patients (78%) were included on the waiting list, and LT was performed for 83 patients (54%). Overall survival rate is now corected to 73%. Multivariate logistic regression identified 2 independent variables associated with adverse outcomes on admission: a Model for End-Stage Liver Disease (MELD) score ≥ 29 and an encephalopathy grade ≥ 3. In a direct comparison using a receiving operating characteristic curve analysis, the MELD score [C statistic = 0.830, 95% confidence interval (CI) = 0.73-0.93] had better prognostic accuracy for predicting outcomes than the Clichy criteria (C statistic = 0.719, 95% CI = 0.58-0.85) or the King's College criteria (C statistic = 0.631, 95% CI = 0.49-0.77). In conclusion, hepatitis B and AIH were the most frequent causes of fulminant hepatic failure in our series, and no cases of ACM overdosing were identified. A MELD score ≥ 29 and an encephalopathy grade ≥ 3 at admission were associated with death. The MELD score at admission showed the highest prognostic accuracy.
© 2014 American Association for the Study of Liver Diseases.