Sex-specific predictive ability of SARC-F and SARC-CalF for ultrasound-based sarcopenia in older adults
摘要
The anterior thigh muscles are among the first to be affected in sarcopenia. This study aims to evaluate the sex-specific predictive value of SARC-F and SARC-Calf, in detecting sarcopenia using ultrasound-derived anterior thigh muscle thickness (ATMT) as the diagnostic reference.
MethodsThis cross-sectional study included 347 (227 female and 120 male) older adults recruited from a geriatrics outpatient clinic. Sarcopenia was diagnosed based on the presence of low handgrip strength and a low anterior thigh muscle thickness index (ATMT/BMI). The SARC-F and SARC-CalF questionnaires were administered, using both standard (< 31 cm) and population-specific (< 33 cm) calf circumference cut-offs. Anthropometric and ultrasound measurements were performed by trained geriatricians to minimize measurement bias. The screening performance of SARC-F and SARC-CalF was evaluated using receiver operating characteristic (ROC) curve analysis. Multivariate logistic regression models were applied to identify independent predictors of sarcopenia, adjusting for age, sex, BMI, and frailty. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported.
ResultsOf the participants, 21.3% were classified as sarcopenic. Sarcopenic individuals exhibited smaller arm circumferences, slower gait speeds, and a higher prevalence of frailty. SARC-F positivity was observed in 76 patients (21.9%), while SARC-CalF positivity was found in 41 patients (11.8%) using the population-specific cut-off (< 33 cm) and in 123 patients (35.4%) using the standard cut-off (< 31 cm). Although SARC-CalF < 33 cm improved sensitivity, the < 31 cm cut-off demonstrated relatively higher specificity, with no significant difference in overall diagnostic performance between the two thresholds. ROC analysis revealed that SARC-F had the highest diagnostic accuracy in the overall population (AUC = 0.693) and women (AUC = 0.708), with good sensitivity and specificity. SARC-CalF < 33 improved sensitivity in both women (72.09% vs. 58.11%) and the overall population (62.16% vs. 58.11%). In men, SARC-CalF < 33 achieved the highest specificity (73.03%), while SARC-F had the highest sensitivity (83.87%). Logistic regression analysis showed that older age, male sex, and frailty were independent predictors of sarcopenia, while SARC-F remained the most predictive screening tool overall. No significant differences in diagnostic superiority were found between tests, though SARC-F had the highest AUC overall.
ConclusionSARC-F demonstrated the highest diagnostic accuracy and negative predictive value, making it a reliable general screening tool, while population-specific calf circumference adjustments (< 33 cm) enhanced the sensitivity of SARC-CalF, particularly in women.