HEV-LF_S : A novel scoring model for patients with hepatitis E virus-related liver failure

A noninvasive assessment method for acute or acute-on-chronic liver failure in patients with hepatitis E virus (HEV) infection is urgently needed. We aimed to develop a scoring model for diagnosing HEV patients who developed liver failure (HEV-LF) at different stages. A cross-sectional set of 350 HEV-LF patients were identified and enrolled, and the Guidelines for Diagnosis and Treatment of Liver Failure in China and the Asian Pacific Association for the Study of the Liver were adopted as references. HEV-LF_S , a novel scoring model that incorporates data on cholinesterase (CHE), urea nitrogen (UREA), platelets and international normalized ratio was developed using a derived dataset. For diagnosing HEV-LF stages F1 to F3, the HEV-LF_S scoring model (F1: 0.87; F2: 0.90; F3: 0.92) had a significantly higher AUROC than did the CLIF-C-ACLFs (F1: 0.65; F2: 0.56; F3: 0.51) and iMELD (F1: 0.70; F2: 0.57; F3: 0.51) scoring models, of which the HEV-LF_S scoring model had the best sensitivity and specificity. In addition, the HEV-LF_S scoring model was correlated with mortality, length of hospitalization and ICU stay. As the GDTLF score increased, the CHE level decreased and the UREA increased gradually. Encouragingly, a calibration curve showed good agreement between the derivation and validation sets. Notably, we also established a nomogram to facilitate the practical operability of the HEV-LF_S scoring model in clinical settings. In conclusion, both CHE and UREA may be indicators for HEV-LF patients. The HEV-LF_S scoring model is an efficient and accessible model for classifying HEV-LF at different stages.