Disability refers to impairments, limitations in activity, and restricted participation (1, 2). The prevalence of disability among middle– and old-age individuals ranges from 19.4 to 33.3% (1–5), and it is positively correlated with age (5, 6). Disability significantly increases the rehospitalization risk and frailty, decreases quality of life, and can even lead to death (3, 7, 8). In comparison to older adults without disability, those with disability spend approximately $10,000 more in 2 years (9). Disability and sarcopenia are closely related; sarcopenia is one of the important causes of disability (10–12). Individuals with disability are prone to sarcopenia due to limited activities (13). Hence, people with disability and sarcopenia experience more personal, social, and economic burdens.

Sarcopenia is defined as loss of muscle mass, which is age-related, and it is also characterized by low muscle strength and/or low physical performance (14). Low muscle mass is an important element in the sarcopenia diagnosis (14). At present, common methods used for the measurement of muscle mass include bioelectrical impedance analysis (BIA) and dual energy X-ray absorption (DEXA) (14). However, the application of these two methods in individuals with disability is difficult. For instance, during the BIA, knee extension is required, but in a few people with disability, the knee joints are stiff and flexed, restricting extension. Furthermore, BIA is required to be performed empty stomach, which may not be possible for a few patients with disability. On the other hand, DEXA is a method in which radiation is used, and this method lacks portability and is inconvenient for older adults with disability and limited activity. Considering these shortcomings of available methods for measuring the muscle mass, the identification of an alternative method is necessary.

Ultrasonic measurement of the muscle mass is associated with advantages, such as no requirement of knee joint extension or flexion, no diet effect, no use of radiation, and portability. Multiple studies have demonstrated reliable performance of ultrasound method in assessing the muscle mass (15–18). A meta-analysis reported the area under the summary receiver operating characteristic curve [SROC] of 0.76 for ultrasonic measurement of gastrocnemius muscle (GM) thickness to diagnose low muscle mass, and its correlation with DEXA, which is a gold-standard method for diagnosing low muscle mass, was also high (r = 0.85–0.963) (19). GM thickness is an appropriate ultrasound parameter to assess muscle mass (19), and GM thickness of <15 mm can be used to suggest a low muscle mass (18).

The prone position is the most common position while performing the ultrasonic examination of the posterior muscles of the lower limbs (including the GM) because this position allows better observation of the longitudinal and transverse muscle planes (16, 18, 20–22). However, in clinical practice, we have observed that a few older adults with disability fail to complete the examination in the prone position, and hence, in such cases, the sonographer has to perform the examination with patients acquiring other postures. The effect of patients acquiring these non-prone positions during ultrasonic examination on the diagnosis of the low muscle mass remains unknown.

This study aimed to examine the accuracy and consistency of different ultrasound measurement protocols (method A, prone position; method B, lateral position with legs separated; and method C, lateral position with legs together) repeated two or three times to identify a low muscle mass in men and women. This study was directed toward identifying a suitable measurement scheme of ultrasonic measurement of the muscle mass that can be more feasible for older adults with disability.

This study included 489 individuals, of whom 221 (45.19%) were male. Table 1 demonstrates the baseline characteristics of all included participants. Women with age > 80 years were more in number, and a greater number of men had a high blood pressure, COPD, and disability (Table 1). The mean GM thickness was higher in men compared with that in women, regardless of the measurement protocol (Table 1).

The difference in the prevalence of low muscle mass identified between two and three replicates of the same measurement protocol ranged from 0 to 1.3% (Table 2). Among males, three repeated measurements of method C and A identified the highest (35.7%) and lowest (28.5%) prevalence of low muscle mass, respectively, whereas among females, two or three repeated measurements of method B and A identified the highest (41.8%) and lowest (34.7%) prevalence of low muscle mass, respectively (Table 2).

When the three repeated measurements of the method A was set as the reference, the sensitivity, specificity, AUC, and Kappa values (from low to high) were 93.70–98.40%, 88–98.70%, 0.977–1, and 0.773, 0.801, 0.829, 0.839, and 0.967, respectively, for males and 93.50–98.90%, 87.40–99.40%, 0.973–1, and 0.786, 0.794, 0.804, 0.819, and 0.984, respectively, for females (Table 3).

To our knowledge, this study, for the first time, demonstrated high accuracy and consistency of different ultrasound protocols in identifying the low muscle mass in individuals with age > 60 years. The results presented in this study will be useful in clinical situations to manage cases where older adults with disability cannot complete the analysis of the muscle mass due to stiff and flexed joints. This study revealed that, in such cases, different body postures may be allowed during the ultrasound examination without compromising the effectiveness of detecting the low muscle mass.

In previous studies, the prevalence of the low muscle mass in males and females was reported to be 17–85% and 17–68%, respectively (24–26). Consistent with these results, the prevalence of the low muscle mass in males and females was found to be 28.5–35.7% and 34.7–41.8%, respectively. This data suggest that the prevalence of the low muscle mass in older adults (> 60 years) is high, and hence targeted screening and timely intervention are needed to minimize the occurrence of disabling comorbid sarcopenia.

Among different ultrasound protocols, method A is the most commonly used, and hence in the present study we used it as a reference method. Method B can be completed by most of the individuals with disability; however, a few older adults with limited leg separation, method C was selected during the measurement. In method C, the left limb may squeeze the muscles of the right limb, thereby affecting the accuracy of the measurement. In both males and females, method A demonstrated the lowest prevalence of the low muscle mass, whereas both methods B and C identified a slightly higher prevalence of the low muscle mass. Based on the result of the present study, we recommend the use of method B for patients who are not able to complete the evaluation using the method A, and in cases where both methods A and B are not feasible, the method C should be employed to identify the low muscle mass.

In previous studies, during the ultrasound evaluation of the GM thickness, measurements were repeated two (18, 27, 28) or three (23) times to achieve better accuracy. In the present study, we also compared the difference between measurements repeated two and three times to identify the low muscle mass. A small difference in the prevalence of the low muscle mass was found between measurements repeated two and three times (0–0.4% in women and slightly larger difference in men). This marginal difference may be related to the differences in the structure and thickness of the GM between men and women (29, 30). We also found that the accuracy and consistency of the muscle mass measurement repeated two times were also high, and hence it is possible to identify the low muscle mass by repeating the ultrasound measurement either two or three times.

This study had a few limitations. First, the number of participants was less. In addition, due to the presence of disabilities in the participants, nearly half were unable to complete height or weight measurements. Therefore, it was not possible to analyze the relationship between BMI and muscle mass. Second, the muscle thickness was not measured in sitting position. Although a few studies have conducted measurements in the sitting position (31), our aim was to determine feasible measurement conditions for older adults with disability, and hence we did not measure the muscle thickness in the sitting position. Third, as there is currently no consensus or guidelines on the protocol for ultrasound measurement of GM thickness, we could only refer to the protocols for ultrasound measurement of GM thickness in the published literature for use as the standard protocol in our study. Our next step will be to improve DEXA as the gold standard to delineate cutoff values for in-depth research. Finally, in the study, only the thickness of GM was measured. The thickness of other muscles, such as rectus femoris, vastus lateralis, tibialis anterior (32), was not measured, and further studies will be required in the future to explore this aspect.

All the different ultrasound protocols led to high accuracy and consistency in identifying a low muscle mass. We suggest that if method A cannot be used in an older adult with disability, method B should be employed for measuring the GM thickness. Repeating the measurements two or three times was found to be feasible.

The datasets generated and analyzed during the current study are not publicly available since it is part of a cohort study. The datasets will be available 2 years after publication and are currently available from the corresponding author on a reasonable request.

The studies involving humans were approved by Research Ethics Committee of Zigong Affiliated Hospital of Southwest Medical University, Zigong Mental Health Center (No.2021-06-01). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

HD: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft. XL: Writing – original draft, Conceptualization, Data curation, Formal analysis, Investigation, Methodology. SH: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft. JL: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing. ZL: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing. LC: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing. LZ: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing. YX: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing. QN: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing. XC: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Writing – review & editing.