Abstract:Objective To explore the value of alpha fetal protein (AFP) and cholinesterase (ChE) in predicting clinical outcome of liver failure. Method A total of 179 patients with liver failure in our hospital was recruited. They were divided into the survival group and the death group according to the condition of the patients leaving the hospital. The laboratory index related to liver failure of two groups were collected. Multivariate Logistic regression analysis was used to explore the risk factors of clinical outcome of liver failure. ROC curves were used to evaluate the predictive ability of relevant indicators for clinical outcomes of liver failure. Results Multivariate logistic regression found that AFP [OlR?=?1.371, (95% CI: 1.075, 1.748)], ChE [OlR?=?5.276, (95% CI: 3.586, 7.762)], international normalized ratio (INR) [OlR?=?4.393, (95% CI: 2.174, 8.879)], and albumin (ALB) [OlR?=?1.956, (95% CI: 1.184, 3.231)] were independent factors influencing the prognosis of liver failure, and the differences were statistically significant (P?0.05). The AUC of AFP predicting the prognosis of liver failure was 0.808, the sensitivity was 86.6% (95% CI: 78.2%, 92.7%), and the specificity was 63.4% (95% CI: 52.0%, 73.8%). The AUC of ChE was 0.802, the sensitivity and specificity were 53.6% (95% CI: 43.2%, 63.8%) and 92.7% (95% CI: 84.8%, 97.3%). The AUC of the combination of AFP and ChE was 0.872, which was significantly higher than AFP and ChE (AFP: Z?=?3.084, P?= 0.002; ChE: Z?=?2.945, P?=?0.003), the sensitivity was 84.5% (95% CI: 75.8%, 91.1%), and the specificity was 76.8% (95% CI: 66.2%, 85.4%). Conclusion AFP and ChE can be used to predict the clinical outcomes of patients with liver failure, and the combination can make a more accurate prediction of the clinical outcome of liver failure.