Association between serum albumin and psychometric hepatic encephalopathy score in cirrhosis without overt hepatic encephalopathy

Abstract Background: Hepatic encephalopathy is the primary cause of mortality in cases of liver cirrhosis. The Psychometric Hepatic Encephalopathy Score (PHES) is widely acknowledged as a gold standard for diagnosis of covert hepatic encephalopathy. However, PHES is burdensome due to its strict indications and is not readily embraced by clinicians. Consequently, there is an imperative to identify simpler clinical indicators for PHES. Methods: This retrospective cross-sectional study utilized data from a single center. Multiple linear regression analysis and restricted cubic spline (RCS) models were employed to investigate the relationship between albumin levels and PHES. Stratified analyses based on albumin levels were conducted to assess the trend of this association. Additionally, subgroup analyses of covariates were performed. Results: A total of 33 liver cirrhosis participants without overt hepatic encephalopathy were included in this study. After adjusting for covariates, a positive association was observed between albumin levels and PHES (β = 0.38, 95%CI: 0.08-0.67, P=0.02). This linear relationship was consistent with the findings obtained from the RCS model (P for non-linearity =0.57). The trend analysis revealed an exponential increase in the strength of the association across different albumin groups (β = 2.61, 95%CI: 0.63-4.59, P = 0.02). Subgroup analyses confirmed this positive association and indicated no interactions between albumin and other covariates (all P for interaction > 0.1). Conclusions: Serum albumin, as one of the commonly used clinical indices for liver cirrhosis patients, shows potential as a simple predictor for PHES.