Early identification using the referral system prolonged the time to onset for hepatic encephalopathy after diagnosing severe acute liver injury

Referral system has a diminished difference in the risk for hepatic encephalopathy development among each etiology in patients with acute liver injury

AIM

Hepatic encephalopathy (HE) development is crucial in liver transplantation for patients with acute liver injury (ALI) and failure (ALF); to predict HE development, the Japan Hepatic Encephalopathy Prediction (JHEP) model, calculated using age, etiology, prothrombin time (PT), and total bilirubin, was established in 2004, and a referral system to the liver center was implemented using the JHEP model from April 2004.

METHODS

The JHEP model's ability to predict HE development in 460 consecutive patients with ALI between April 2004 and January 2021 using data from the referral system was evaluated, and the JHEP model was revised.

RESULTS

During the observation period, 7.8% patients developed HE. There was no difference in the proportion of HE development among the etiologies. In the Hosmer-Lemeshow test for HE development prediction, the JHEP model, revised JHEP (rJHEP) model, which was calculated without etiology data, and the modified JHEP model, which used the PT international ratio instead of PT in the rJHEP model, were good fitting models. Upon 30% random sampling from the total patients 60 times, the receiver operating curve analysis of both JHEP and rJHEP models for HE development was performed in all the datasets. The area under the curve of the JHEP model was subtracted from that of the rJHEP model (95% confidential interval, 0.000516-0.01793).

CONCLUSIONS

The referral system using the JHEP model reduced the difference in the risk for HE development among each etiology; the rJHEP model had a better prediction ability for HE development than the JHEP model.

Serum alpha-fetoprotein increases prior to fibrosis resolution in a patient with acute liver failure

A 78-year-old woman who was diagnosed with acute liver failure due to an undetermined cause presented with liver atrophy. Coagulopathy was normalized at 35 days of hospitalization, although atrophy in the liver persisted. During the observation period, alpha-fetoprotein (AFP) bi-modally increased at 36 and 377 days. Around the second peak of AFP, the liver volume was regained within the normal range. Fucosylated AFP was found at the first peak but not at the second peak. Cytokines/chemokines were simultaneously evaluated, and the results were evaluated using PANTHER ( http://www.pantherdb.org/ ). Although transaminase and prothrombin time were within the normal range, cytokines/chemokines associated with angiogenesis and inflammation increased prior to the second peak of AFP. Our study suggests that the first peak of AFP occurs in response to acute insult, while the second peak may be associated with the resolution of liver fibrosis. The present case provides new insights into the mechanism of AFP elevation.