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

We appreciate Tan et al.'s interest and comment on our recent manuscript, "Associations of CT Muscle Area and Density With Functional Outcomes and Mortality Across Anatomical Regions in Older Men" [1]. They raise an important question about the role of age in the relationship of muscle area and density with risk of mortality. Although all models were adjusted for age, we acknowledge that standard covariate adjustment may not fully address or disentangle the confounding influence of age. In response to the comments, we performed additional analyses to assess age as a potential effect modifier. We examined interactions between age and both muscle area and muscle density at two trunk levels (Trunk-L1 and Trunk-L3) and in the right (Thigh-RT) and left (Thigh-LT) thigh, using Cox proportional hazards models with age modeled both continuously and in tertiles. We also conducted age-stratified subgroup analyses based on age tertiles of our sample. Figure 1 presents hazard ratios (HRs) (95% CI) per standard deviation increase in muscle density and area for 10-year all-cause mortality, stratified by age tertile and anatomical region. No significant age interactions were observed when age was modeled as a continuous variable (p-interaction > 0.05), and similar HRs were found across all the age tertiles, as indicated by overlapping 95% CIs. However, some region-specific patterns emerged. Trunk muscle density at both Trunk-L1 and Trunk-L3 was most strongly associated with lower mortality risk in the youngest age tertile, with weaker associations in older tertiles, although no statistically significant interaction was detected (p-interaction > 0.05). In contrast, thigh muscle density was most strongly associated with mortality in the middle tertile, with a significantly greater effect than in the youngest tertile for the Thigh-RT (p-interaction = 0.006) and a borderline difference for the Thigh-LT (p-interaction = 0.06). Unlike muscle density, muscle area was not significantly associated with mortality in the overall sample or within most age tertiles, except for the Thigh-RT, where a significant association was observed only in the youngest tertile and the full cohort. These findings suggest that the impact of muscle density on mortality risk varies by both anatomical region and stage of aging. The stronger association observed in the youngest-old age tertile for trunk muscles, but in the middle-old age tertile for thigh muscles, may reflect differences in the onset and progression of age-related muscle quality decline across regions. Lumbar paraspinal muscles are particularly vulnerable to early degenerative changes, such as fat infiltration, even in relatively healthy adults, likely due to their sensitivity to disuse and neuromuscular dysfunction [2, 3]. As a result, lower muscle density in the trunk, primarily driven by decreased muscle density of the paraspinal muscles, may serve as an early marker of systemic decline, which could explain its stronger association with mortality in the younger-old men (i.e., 65–69 years). With advancing age, as fat infiltration becomes more prevalent and widespread in trunk muscles, variability narrows, reducing the discriminatory value of muscle density measures. Unlike paraspinal muscles, thigh muscle quality tends to decline later in the aging process [3], typically aligning with age-related reductions in physical activity and emerging functional impairments [4]. As a result, greater between-person variability and stronger associations with mortality are observed in middle-old men (ages 70–75), whereas in younger-old men, thigh muscle density remains relatively preserved. In the oldest-old group (i.e., 76+ years), muscle deterioration is more widespread, which may reduce variability in muscle quality and limit its ability to stratify mortality risk. These results underscore that the prognostic significance of muscle metrics, particularly those reflective of muscle quality, on survival is not uniform across muscle groups for all older adult ages. Accordingly, both research and clinical practice should consider region-specific aging patterns when evaluating muscle-related mortality risk. In line with this, and emphasized by Tan et al., further examination of muscle size and density in relation to shorter-term health outcomes (e.g., hospitalizations and sick days), particularly in diverse populations, can help refine age-tailored approaches to assessment and intervention. M.H.-R.: formal analysis, writing. A.A.W.: writing. P.M.C.: supervision, writing. Financial sponsors played no role in study design, methods, subject recruitment, collection, analysis, and interpretation of data, or preparation of the paper. The authors declare no conflicts of interest. This publication is linked to a related reply by Tan et al. To view this article, visit https://doi.org/10.1111/jgs.70133.