An Anthropometric Prediction Equation for Appendicular Skeletal Muscle Mass in Combination With a Measure of Muscle Function to Screen for Sarcopenia in Primary and Aged Care

Abstract

OBJECTIVES: Sarcopenia is the presence of low muscle mass and poor physical function. We have developed an anthropometric prediction equation (PE). We compared the accuracy of our previously developed anthropometric prediction equation (PE) to dual absorptiometry x-ray (DXA) in predicting low muscle mass and sarcopenia. DESIGN: Cross-sectional study design. SETTING: Community dwelling. PARTICIPANTS: Men and women aged 65 years and older. MEASUREMENTS: Gender-specific low muscle mass cutoffs were identified using the lowest 20% of the skeletal muscle index (SMI) where muscle mass was determined using PE in 611 men and 375 women aged 65 years and older. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of PE derived low muscle mass were compared with DXA-derived low muscle mass. The cohort was randomized into a development and validation group to identify various cutoffs for low muscle mass via the PE method and test its performance against the DXA method. RESULTS: The PE cutoff for low muscle mass was less than 8.05 kg/m(2) in men and less than 5.35 kg/m(2) in women. On validation of various cutoffs with improving sensitivity values from 70% to 97%, specificity increased from 45.5% to 85.7%, PPV increased from 31.3% to 56.9%, and NPV increased from 93.0% to 98.6% in men. In women, specificity improved from 42% to 72%, PPV reduced from 56.9% to 31.3%, and NPV improved from 93.0% to 98.6%. When the PE method was combined with a measure of muscle performance, a similar pattern of performance was observed. CONCLUSION: The PE when combined with a measure of muscle function to create a screening tool performs as a "rule-out" test with high sensitivity values and NPVs.