Comment on: Associations of CT Muscle Area and Density With Functional Outcomes and Mortality Across Anatomical Regions in Older Men

We read with great interest the article by Hetherington-Rauth et al., which highlights the prognostic utility of automated CT-derived muscle area and density across anatomical regions in older men [1]. The findings underscore the value of automated opportunistic imaging for assessing sarcopenia and mortality risk in aging populations [2]. We commend the study's analysis and emphasis on functional outcomes, including grip strength, leg power, and walking speed. Nonetheless, the interpretation of results may benefit from a more thorough consideration of age as a potential residual confounder. Older participants naturally face a higher 10-year mortality risk. While age was included as a covariate, muscle size and density are strongly inversely correlated with age. Without accounting for age through matching or stratification, especially with the benefit of having a large cohort, the observed association between muscle metrics and mortality may be confounded by age. Propensity score matching or age-stratified subgroup analyses could have further strengthened causal inference and minimized residual confounding. This is especially pertinent since the survival analyses showed consistent associations across regions—but without deeper exploration of whether these held across narrower age bands. It remains unclear whether poorer muscle metrics are independently predictive of mortality or simply reflect age-related risk of mortality. From the health economic angle, where the cost of healthcare consumption is a pertinent marker of health in a given population, the authors could also consider analyzing, by decade bands of age, the association of muscle size and density with 1-year mortality, hospital attendance, or mean sick days, which would allow readers to better appreciate the impact of muscle size and quality on shorter term health-related outcomes. These parameters, with the strength of sample size, would allow more emphasis on muscle health in health promotion policies. It was good of the authors to highlight differences with previous studies. The contrary findings regarding the association between muscle density and leg power in Korean older adults [3] compared with the MrOS study participants highlight important ethnic and regional differences in the assessment, cut-offs, and variations in muscle strength, density, and intermuscular adipose tissue (IMAT) in different populations. Validation in diverse populations, especially older women and multi-ethnic Asian cohorts, is crucial for generalizability. It would have been interesting for the authors to consider and highlight the associations between muscle mass, muscle function, muscle quality, and health-related outcomes [4]. With increasing emphasis on muscle quality as a measurement metric for muscle health [5, 6], future studies pertaining to muscle health should be encouraged to assess and present these components of muscle health to allow readers a holistic appreciation of the complexity of this increasingly recognized health state, which implicates a myriad of diseases. While the manner of assessments of muscle quality still remains debatable [7], having more reports of it in literature would allow retrospective assessments to be done to aid researchers in this field to ascertain which would have the highest accuracy. Nonetheless, this study makes a valuable contribution by demonstrating that opportunistic, automated assessment using CT scans taken can yield clinically relevant muscle health insights. Integrating automated segmentation into radiologic workflows holds promise for earlier identification of sarcopenia, and further work is needed to ensure these associations are robust across sex, ethnicity, and age strata. Li Feng Tan and Frederick H. Koh conceptualized and wrote the initial manuscript. All authors revised and edited the manuscript. There was no sponsor for this manuscript. The authors declare no conflicts of interest. This publication is linked to a related letter by Megan Hetherington-Rauth et al. To view this article, visit https://doi.org/10.1111/jgs.70129.