A Model to Predict 1-Month Risk of Transplant or Death in Hepatitis A-Related Acute Liver Failure

Abstract

Acute liver failure (ALF) caused by hepatitis A is a rare but fatal disease. Here, we developed a model to predict outcome in patients with ALF caused by hepatitis A. The derivation set consisted of 294 patients diagnosed with hepatitis A-related ALF (ALFA) from Korea, and a validation set of 56 patients from Japan, India, and United Kingdom. Using a multivariate proportional hazard model, a risk-prediction model (ALFA score) consisting of age, international normalized ratio, bilirubin, ammonia, creatinine, and hemoglobin levels acquired on the day of ALF diagnosis was developed. The ALFA score showed the highest discrimination in the prediction of liver transplant or death at 1 month (c-statistic, 0.87: 95% confidence interval [CI], 0.84-0.92) versus King's College criteria (KCC: c-statistic, 0.56: 95% CI, 0.53-0.59), U.S. Acute Liver Failure Study Group index specific for hepatitis A virus (HAV-ALFSG: c-statistic, 0.70: 95% CI, 0.65-0.76), the new ALFSG index (c-statistic, 0.79: 95% CI, 0.74-0.84), Model for End-Stage Liver Disease (MELD: c-statistic, 0.79: 95% CI, 0.74-0.84), and MELD including sodium (MELD-Na: c-statistic, 0.78: 95% CI, 0.73-0.84) in the derivation set (all P < 0.01). In the validation set, the performance of the ALFA score (c-statistic, 0.84: 95% CI, 0.74-0.94) was significantly better than that of KCC (c-statistic, 0.65: 95% CI, 0.52-0.79), MELD (c-statistic, 0.74: 95% CI, 0.61-0.87), and MELD-Na (c-statistic, 0.72: 95% CI, 0.58-0.85) (all P < 0.05), and better, but not statistically significant, than that of the HAV-ALFSG (c-statistic, 0.76: 95% CI, 0.61-0.90: P = 0.28) and new ALFSG indices (c-statistic, 0.79: 95% CI, 0.65-0.93: P = 0.41). The model was well-calibrated in both sets. CONCLUSION: Our disease-specific score provides refined prediction of outcome in patients with ALF caused by hepatitis A.