The relationship between cognitive function and muscle mass in older adults: a longitudinal study based on CLHLS

Shi Yin, Yu Zhang, Xinyu Yang, Jiali Yang, Shuguang Wang, Yun Jeong HongFrontiers in Psychiatry 2025 · DOI

This study, published in Frontiers in Psychiatry, investigates the longitudinal link between skeletal muscle mass and cognitive decline in older Chinese adults. By analyzing data from over 2,500 participants, the researchers found that muscle mass is a significant predictor of future cognitive health—particularly in women—and established specific muscle-index thresholds that could serve as early warning signs for dementia risk.

TL;DR

Low muscle mass isn't just a physical frailty issue; it’s a primary driver of cognitive decline. This longitudinal study identifies a "muscle-to-brain" predictive pathway and sets specific thresholds for the Appendicular Skeletal Muscle Index (ASMI) to help prevent cognitive impairment in older adults.

What Problem Does This Paper Address?

As global populations age, the dual challenges of sarcopenia (age-related muscle loss) and cognitive impairment are becoming major public health burdens. While researchers have long noted that these two conditions often occur together, several gaps remained:

  1. Directionality: Does losing muscle cause cognitive decline, or does a declining brain lead to a less active, weaker body?
  2. Gender Differences: Men and women have different hormonal profiles and body compositions; do they share the same risk thresholds?
  3. Linearity: Most studies assume a straight-line relationship, but biological systems often have "tipping points" or non-linear dynamics.

The authors aimed to move beyond simple correlations to establish a causal temporal relationship and provide actionable clinical thresholds.

Methods

The researchers utilized a Cross-Lagged Panel Model (CLPM). Unlike standard regression, CLPM allows researchers to look at two variables (muscle mass and cognition) across multiple time points to see which one predicts the other more effectively over time.

  • Cognition Measurement: Evaluated using the Mini-Mental State Examination (MMSE), a 30-point scale covering memory, orientation, and language.
  • Muscle Mass Measurement: Calculated as the Appendicular Skeletal Muscle Index (ASMI). Instead of expensive DXA scans, they used a validated anthropometric equation based on weight, height, gender, and age—making the findings more applicable to low-resource clinical settings.
  • Non-linear Analysis: They employed Restricted Cubic Splines (RCS) to identify specific "inflection points" where the risk of cognitive impairment significantly increases.

Datasets

The study drew from the Chinese Longitudinal Healthy Longevity Survey (CLHLS), specifically Phase III (2011–2018).

  • Sample Size: 2,536 participants aged 60 and older.
  • Demographics: Roughly equal split between males (50.6%) and females (49.4%), with an average age of 77.5 years.

Key Insights and Ideas

  • The "Muscle-First" Pathway: The study found a strong one-way prediction effect. Muscle mass at Time 1 significantly predicted cognitive function at Time 2, but cognitive function was a much weaker predictor of future muscle mass. This suggests that maintaining muscle is a proactive way to protect the brain.
  • Gender Dimorphism: The protective effect of muscle was more pronounced in women. The authors suggest this may be due to the role of estrogen and its impact on both muscle satellite cell proliferation and hippocampal plasticity.
  • The 60-69 "Golden Window": Subgroup analysis showed the strongest association between muscle and brain health in the 60–69 age cohort. This suggests that interventions (like resistance training) might be most effective if started in early seniority before mitochondrial decline accelerates.
  • Biological Crosstalk: The paper highlights the role of "myokines"—hormone-like proteins released by muscles during contraction—which can cross the blood-brain barrier to influence neuroplasticity and reduce inflammation.

Key Results

  • Thresholds for Protection: The study identified specific ASMI levels below which cognitive risk rises sharply:
  • Men: 7.45 kg/m²
    • Women: 5.68 kg/m²
  • Hazard Ratios: In the 60–69 age group, higher muscle mass was associated with a significantly lower risk of cognitive impairment (HR = 0.44).
  • Statistical Significance: The path coefficients for ASMI $\rightarrow$ MMSE were statistically significant ($P < 0.05$) across all time intervals for the general population.

Limitations and Future Work

While the study provides a strong longitudinal framework, the authors acknowledge a few hurdles:

  • Estimation vs. Measurement: ASMI was calculated via an equation rather than measured directly via gold-standard imaging (DXA or MRI), which may introduce some calculation error.
  • Recall Bias: The MMSE, while standard, can be influenced by a participant's education level and the "practice effect" of taking the test multiple times.
  • Generalizability: The thresholds were derived from a Chinese cohort. Because body composition varies significantly by ethnicity, these specific numbers (7.45 and 5.68) need to be validated in Western or other Asian populations before being used globally.

Future research should focus on whether specific types of exercise (e.g., hypertrophy-focused resistance training vs. endurance) provide different levels of neuroprotection.

Summary updated

Abstract

Muscle mass had a major predictive effect on cognitive trajectory, especially in females. Maintaining ASMI above gender-specific thresholds may help slow cognitive decline, suggesting that muscle mass can serve as an adjustable biomarker for dementia prevention. Longitudinal studies should verify the validity of these thresholds in different populations.
BibTeX
@inproceedings{yin2025relationship,
  author = {Shi Yin and Yu Zhang and Xinyu Yang and Jiali Yang and Shuguang Wang and Yun Jeong Hong},
  title = {{The relationship between cognitive function and muscle mass in older adults: a longitudinal study based on CLHLS}},
  booktitle = {Frontiers in Psychiatry},
  year = {2025},
  doi = {10.3389/fpsyt.2025.1595625},
}
Figure