Letter to the Editor: “Altered Serum Bile Acid Pattern as an Independent Factor for Covert and Overt Hepatic Encephalopathy in Patients With Cirrhosis”

We read with great interest the study by Miwa et al. [1], which reports that serum bile acid (BA) profiles independently predict covert and over hepatic encephalopathy (HE) and mortality in patients with cirrhosis. The results offer an important perspective on HE pathophysiology and raise the possibility that BAs may serve as biomarkers or therapeutic targets. We wish to highlight two methodological issues that could affect interpretation. First, BA concentrations are influenced by diet, circadian rhythm, and transient changes in liver function [2]. The study relied on a single fasting serum measurement; without information on sampling timing, recent food intake, or short-term liver function variability, a single timepoint may not capture an individual's usual BA exposure or distinguish persistent alterations from transient fluctuations. Serial sampling, standardized sampling protocols (including time of day and presampling diet), or reporting of postprandial levels could strengthen the assessment of BA–HE associations. Second, ammonia is a well-established mediator in HE. Although Miwa et al. adjusted for ammonia in multivariable models, they did not report formal interaction or mediation analyses between ammonia and BA species. BAs and ammonia could interact via the gut microbiome, intestinal permeability, or blood–brain barrier effects; testing for interaction or conducting mediation analyses would help clarify whether BAs contribute independently to HE risks or act partly through ammonia-related pathways [3]. Integrating microbiome or metabolomic data could further elucidate mechanistic links. We commend the authors for advancing understanding of the gut–liver–brain axis in cirrhosis. Future studies that include dynamic BA measurements, standardized sampling, and joint analyses with ammonia and omics data will help determine the clinical utility of BA profiling for risk stratification and for identifying potential therapeutic targets. The authors have nothing to report. The authors have nothing to report. The authors declare no conflicts of interest. The authors have nothing to report.