Imaging without biology: limitations of CT-derived psoas muscle volume as a standalone diagnostic criterion for sarcopenia Authors: Harsh Jain, Samarth Mittal, Buddhadev Chowdhury Published: 2026 ABSTRACT AbstractWe comment on the study by Choi et al., which proposes CT-derived psoas muscle volume (PV) as a diagnostic indicator for sarcopenia. The study is well designed, using automated deep-learning-based CT segmentation for psoas muscle volume measurement, and adds to the growing interest in opportunistic CT-based muscle assessment. However, a few points may warrant consideration before wider clinical application. These include differences from current consensus definitions of sarcopenia, limitations related to the use of BIA as a reference standard, and the lack of validation against clinical outcomes. Furthermore, volumetric measures alone may not fully capture the histopathological changes associated with sarcopenia, including muscle fibre atrophy, fat infiltration, and neuromuscular denervation. We also note that applying T-score thresholds from osteoporosis to muscle volume is not straightforward, as no comparable outcome data currently exist to anchor these cut-offs in sarcopenia. To the editor We recently read the study by Choi et al. [1] with great interest. The authors propose CT-derived psoas muscle volume (PV) as a diagnostic indicator for sarcopenia in a large Korean screening cohort. The study is well conducted, using automated CT-based measurements in a large sample, and it contributes to the growing interest in imaging-based muscle assessment. We would like to highlight a few points that may help in interpreting these findings in a clinical context. Alignment with current sarcopenia definitions The study primarily evaluates muscle size, which differs from current consensus recommendations. Both EWGSOP2 [2] and AWGS 2019 [3] emphasise muscle strength as the primary diagnostic component, with muscle mass serving a supporting role. In this setting, PV thresholds—however carefully derived—cannot by themselves establish a diagnosis of sarcopenia. It may therefore be more appropriate to present these criteria as a tool for identifying low muscle mass rather than as a standalone diagnostic measure. Reference standard and outcome validation The use of BIA-derived ASM as the reference standard also deserves some consideration. BIA is known to be influenced by hydration status and device-related variability, which can affect its reliability. When an imaging-based parameter is compared against such a method, this may influence the observed performance of the derived thresholds. Validation against DXA-derived ASM would provide a more stable reference. In addition, the proposed cut-offs have not yet been linked to clinical outcomes. In routine practice, sarcopenia is important because of its association with falls, fractures, postoperative complications, and functional decline [4]. Demonstrating that individuals classified using PV criteria actually experience these outcomes at a higher rate would strengthen the clinical relevance of the findings. At present, these thresholds should perhaps be viewed as preliminary. Muscle quality and histopathological considerations Sarcopenia is not only a reduction in muscle size. At a microscopic level, it is characterised by type II fibre atrophy, fibre size variability, fatty infiltration, fibrosis, and progressive denervation. These changes reflect deterioration in muscle quality that volumetric measurements alone are unable to capture. It is therefore possible for muscle to appear preserved on CT while significant structural changes have already occurred. CT attenuation values (Hounsfield Units) offer a useful indirect measure of muscle composition. Since these can be obtained from the same scan without additional cost, incorporating muscle attenuation alongside PV may provide a more complete assessment. This becomes particularly relevant in sarcopenic obesity, where muscle volume can appear relatively preserved despite significant qualitative deterioration. T-score thresholds and generalizability Adapting T-score thresholds from the osteoporosis framework is a reasonable starting point, but it requires cautious interpretation. Bone density thresholds are supported by extensive data linking them to fracture risk [5], whereas similar outcome-based validation is lacking for muscle volume. The authors’ observation that a − 2.0 threshold aligns more closely with expected prevalence than − 2.5 suggests that this analogy may not directly translate to sarcopenia. Furthermore, as the study population is derived from a single Korean screening cohort, it would be important to validate these findings in other populations before broader clinical application. Conclusion These points are intended to build upon the work of Choi et al. The study provides a valuable foundation for CT-based assessment of sarcopenia. Further work incorporating functional measures, muscle quality indices, and longitudinal outcomes will be important in determining how these imaging-based criteria can be applied in clinical practice. Author contributions Author Contributions: Harsh Jain : Conceptualization, literature review, drafting of the manuscript.Samarth Mittal : Critical revision of the manuscript for important intellectual content.Buddhadev Chowdhury: Supervision and final approval of the manuscript.All authors reviewed and approved the final version of the manuscript. Data availability No datasets were generated or analysed during the current study. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.