Ultrasound Muscle Measures Improve Identification of Low Physical and Social Activity in Community-Dwelling Older Adults

Title: Ultrasound Muscle Measures Improve Identification of Low Physical and Social Activity in Community-Dwelling Older Adults
JQ Chia1,2, A Yeo2, CN Tan2, A Tandon3, WS Lim1,2, J Chew1,2

1Department of Geriatric Medicine, Tan Tock Seng Hospital, Singapore
2Institute of Geriatrics and Active Ageing, Tan Tock Seng Hospital, Singapore 
3Department of Diagnostic Radiology, Tan Tock Seng Hospital, Singapore

Introduction
Current sarcopenia diagnosis emphasises reduced muscle quantity. Ultrasound (US), which captures both muscle quantity and quality, may offer additional diagnostic value, but its optimal application remains unclear. This study examined the best-performing combination of US-derived muscle measures for sarcopenia and evaluated whether US improves the identification of individuals with low physical and social activity levels compared with existing clinical criteria.

Methods
We analysed 122 community-dwelling older adults (mean age 67.2 ± 5.8 years). US assessment of the quadriceps included quantity measures—muscle thickness (MT) and cross-sectional area (CSA)—and quality measures—pennation angle (PA), echo-intensity (EI), and shear-wave velocity (SWV). Associations between US parameters and physical performance (Short Physical Performance Battery, SPPB) were examined using multivariable linear regression. Diagnostic performance of individual and combined US parameters was evaluated using area under the receiver operating characteristic curve (AUC) analyses against the Asian Working Group for Sarcopenia (AWGS) 2025 criteria. We compared the proportion of individuals with low physical activity (International Physical Activity Questionnaire, IPAQ) and low social activity (Frenchay Activities Index, FAI) identified by US-based classification, self-report screening (SARC-F ≥ 2), and AWGS 2025 possible sarcopenia criteria.

Results
SWV was significantly associated with SPPB performance (β = 0.523, p = 0.024). MT (β = 0.597, p = 0.055) and CSA (β = 0.123, p = 0.055) demonstrated moderate associations. The combined model integrating SWV, MT and CSA achieved the highest AUC (0.637), outperforming individual parameters. US-derived measures identified a greater proportion of individuals with low FAI (59%) and low IPAQ (50%) compared with SARC-F (10.3% and 7.7%) and AWGS 2025 possible sarcopenia (30.8% and 18.5%).

Conclusions
Integrated US-based muscle quantity and quality measures may enhance early detection of risk profiles associated with reduced physical and social activity. These findings provide preliminary support for incorporating muscle US into sarcopenia assessment, warranting further validation in larger clinical cohorts.