Age-related loss of muscle mass and function is a prevalent phenomenon. Muscle strength declines by 20%-40% in the seventh and eighth decades of life, with this decline gradually exacerbating over time11. It is noteworthy that older adults with T2DM have approximately three times higher risk of low muscle mass compared to those without diabetes12. Elevated blood glucose levels ultimately result in diminished protein synthesis and metabolic abnormalities, consequently leading to further deterioration in muscle mass, strength, and function5. Therefore, increased attention should be devoted to the risk of sarcopenia among older adults with T2DM.
In our study, the prevalence of sarcopenia was higher in male patients with T2DM (68.1%) compared to female patients (31.9%). This difference may be attributed to the reduced secretion of testosterone in males, which could lead to a more rapid decline in skeletal muscle mass. Additionally, older men generally have relatively less muscle mass compared to females, who tend to have a higher proportion of adipose tissue within their muscles13. Moreover, the decrease in estrogen levels among postmenopausal women can also contribute to a decline in muscle strength14, 15. Therefore, it is essential to actively screen for sarcopenia among both older men and women.
We selected the medial head of the gastrocnemius for assessment due to two primary reasons. Firstly, MRI studies have confirmed that there is an uneven decline in muscle mass associated with aging, with lower limb muscles experiencing greater loss compared to upper limb muscles16, 17. The impact of sarcopenia tends to be more pronounced in anti-gravity muscle groups. Kuyumcu et al.18 discovered that among a cohort of 100 older individuals, the thickness and length of the muscle bundle in the medial head of the gastrocnemius exhibited high sensitivity and negative predictive value in detecting sarcopenia. Secondly, this particular site offers several advantages for evaluation purposes: it is superficial and provides clear visualization of the muscle structure, making it ideal for assessing both muscle bundle length and pinnate angle; furthermore, its simplicity of operation and high repeatability help minimize measurement errors.
The pinnate angle and muscle bundle length of the medial head of the gastrocnemius are highly influenced by ankle angle and isometric autonomous contraction during ultrasound evaluation19. Therefore, we opted to measure muscle bundle length and pinnate angle in a standardized posture with the ankle in a neutral position under relaxed conditions. Our study revealed that compared to the control group, the sarcopenia group exhibited a decrease in pinnate angle, which aligns with Naric et al.'s findings20 indicating a correlation between pinnate angle and contracted muscle fiber count. Peripheral neuropathy is one of the most prevalent chronic complications among patients with T2DM. Reduced motor neuron count leads to decreased motor unit numbers, resulting in gradual skeletal muscle fiber atrophy and denervation changes. Additionally, peripheral microvascular lesions can cause inadequate perfusion of muscle tissue and impaired blood circulation, further accelerating muscle atrophy, hindering skeletal muscle cell regeneration, reducing overall muscle fiber count21, and subsequently decreasing the pinnate angle.
The biomechanical properties under pathological or physiological conditions can be quantitatively measured in real time by SWE. When the ankle joint is in the neutral position, the gastrocnemius muscle remains relaxed and there is no change in muscle hardness. However, when the ankle joint reaches maximum dorsiflexion, the gastrocnemius muscle becomes constricted, resulting in an increase in muscle hardness and subsequently increasing Young's modulus. Our study revealed that the Young's modulus value was lower in the relaxed state of the sarcopenia group compared to the control group, while there was no significant difference between both groups in terms of Young's modulus value during constriction. This discrepancy may be attributed to individual variations in maximum dorsiflexion of the ankle joint during operation and inconsistent ankle angles during measurement. To address this issue, a simple device can be developed to fixate the ankle angle at a consistent level.
ASMI is a well-established method for assessing muscle mass. Consistent with previous studies, our research demonstrated positive correlations between ASMI and various parameters including medial head thickness of gastrocnemius muscles, pinnate angle, and Young's modulus values during relaxation state - all reflecting muscle mass and function to some extent. Among these parameters, medial head thickness exhibited strongest correlation where greater thickness indicated higher muscle mass. Grip strength also showed correlation with muscle thickness as reported previously22. However, we did not observe any significant correlation between 6m pace (walking speed) and ultrasonic characteristics which has not been confirmed by relevant literature23. This could potentially be explained by mechanical activities during exercise being primarily carried out by tendons rather than muscles themselves; tendons stretch and recoil while storing and releasing elastic energy for forward propulsion24.
Compared to the control group, the sarcopenia group exhibited a significant reduction in both the thickness and elasticity of the medial head of gastrocnemius muscle. This can primarily be attributed to the fact that muscle serves as the primary consumer of glucose25. The decline in muscle mass exacerbates insulin resistance, thereby impeding muscle protein synthesis and establishing a detrimental cycle that ultimately leads to diminished muscle thickness and hardness, consequently increasing susceptibility to sarcopenia26. Additionally, individuals with T2DM are prone to peripheral nerve and microvascular damage, resulting in limb pain and discomfort along with restricted muscular activity. Consequently, this diminishes exercise levels which subsequently contribute to neuropathic muscular atrophy. Simultaneously, ectopic fat deposition within muscle tissue negatively impacts muscle metabolism, insulin sensitivity, and circulating glucose levels27, 28, eventually leading to adipose infiltration into muscle fibers accompanied by reduced muscular contraction strength as well as significantly decreased thickness and hardness.
Based on multivariate regression analysis, this study developed a diagnostic nomogram model by integrating clinical, US, and SWE features. Following internal validation, the model demonstrated a robust diagnostic ability with an AUC of 0.882. The calibration curve closely approximated the ideal curve, indicating excellent calibration and diagnostic consistency of the nomogram model. However, external verification is required to further validate the effectiveness of combining US and SWE in the diagnostic model. Several limitations were identified in our study: firstly, the small sample size limited to older adults with T2DM from a specific region; secondly, potential influence of hypoglycemic drugs on muscle mass was not excluded; thirdly, measurement errors may arise due to variations in probe position, pressure application, and inclination.
In conclusion, both US and SWE demonstrate potential value in quantifying skeletal sarcopenia and are useful for screening sarcopenia in T2DM patients. Among older adults with T2DM, those with sarcopenia exhibit reduced muscle thickness, pinnate angle, and hardness compared to non-sarcopenic patients. The combination of two-dimensional US and SWE model can effectively diagnose the presence of sarcopenia in older T2DM patients with high sensitivity, specificity, and diagnostic efficacy. In future research, large-scale multi-center studies should be conducted to establish reference ranges for the severity of sarcopenia in T2DM patients. Additionally, deep learning algorithms can be employed to extract muscle structure features from two-dimensional US images as well as elastic image features separately. These extracted features can then be fused at the feature level to develop a more cost-effective, convenient, and efficient machine learning model for diagnosing and screening older sarcopenia. This will help guide the development of targeted exercise and treatment programs aimed at reducing or reversing muscle mass loss and strength decline while significantly improving the quality of life among older adults.