I read with great interest the study by Utakata et al. 1, which identified amino acid imbalance, specifically decreased branched-chain amino acids (BCAAs) and increased tyrosine levels, as independent prognostic factors for mortality in patients with liver cirrhosis. While the study provides valuable insights into metabolic derangements in liver disease, I have two major concerns regarding the validity of the independence of these factors and the interpretation of the results. First, the study raises a critical concern regarding unmeasured confounding by sarcopenia. BCAAs are primarily metabolised in skeletal muscle, and low serum BCAA levels are inextricably linked to sarcopenia 2. Sarcopenia itself is a well-established, potent predictor of mortality in cirrhosis. Although the authors adjusted for Body Mass Index (BMI) in their multivariable analysis, BMI is notoriously unreliable for assessing muscle mass in cirrhotic patients due to fluid retention. Notably, 35.1% of the study cohort had ascites, which artificially elevates body weight and masks muscle depletion 3. Without adjusting for skeletal muscle mass (e.g., using the Skeletal Muscle Index via CT), it remains unclear whether low BCAA levels are truly an ‘independent’ driver of mortality or merely a biochemical surrogate for sarcopenia. Second, the reporting of hazard ratios (HR) in multivariable analysis obscures the clinical relevance of the findings. In table 3, the HRs for BCAA and Tyrosine are reported as 1.00 (95% CI 1.00–1.00) and 1.01 (1.00–1.01), respectively. While statistically significant (p < 0.05), presenting HRs per 1 μmol/L change yields values that are mathematically indistinguishable from the null effect. This ‘unit bias’ makes it impossible for clinicians to gauge the magnitude of the risk. To demonstrate that the statistical significance translates into a meaningful clinical effect size rather than an artefact of sample size, I urge the authors to report standardised HRs (e.g., per 1 standard deviation change or per 100 μmol/L decrement). I believe that clarifying the relationship between BCAA, muscle mass, and mortality, along with providing standardised effect sizes, is essential to validate the clinical utility of amino acid profiling in cirrhosis. The author takes full responsibility for this article. The author has nothing to report. The author declares no conflicts of interest. This article is linked to Utakata et al papers. To view these articles, visit https://doi.org/10.1111/apt.70525 and https://doi.org/10.1111/apt.70588. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Zhanbo Qu (Wed,) studied this question.