Calf circumference refines sarcopenia in correlating with mortality risk

Abstract

Background: Recommendations in current guidelines on the use of calf circumference (CC) as a diagnostic measure in sarcopenia is conflicting. In this study, we incorporated CC into conventional models of sarcopenia and compared their predictive scores on mortality. Objective: To clarify the predictive capacity of CC-added models of sarcopenia on and establish the correlation of CC with mortality risk. Design, Setting and Subjects: From the National Health and Nutrition Examination Survey 1999-2002, 2,532 participants aged 50-85 and comprising 51.7% male were included. Methods: Our definitions contained four sarcopenic components, low CC (C), low muscle strength (S), low muscle mass (M) and low gait speed (G). Cox proportional hazard models and the receiver-operator characteristic curves for all-cause, cardiovascular (CV) and cancer mortality were conducted to compare the different sarcopenia definitions. Results: For all-cause mortality, the hazard ratio (HR) of C + S + M + G was 15.062, which was substantially higher than the conventional definition S + M + G (HR = 6.433). Other CC-added definitions such as C + M (HR = 2.260), C + G (HR = 4.978), C + S + M (HR = 5.761) also revealed higher HR than their without-CC counterparts. Similar patterns were observed in CV mortality, for instance, HR of C + S + M + G was 31.812 comparing to the conventional definition S + M + G (HR = 18.434). Concerning accuracy in predicting mortality, the area under the curve (AUC) of CC + S + G + M (AUC = 0.702) and C + S + G (AUC = 0.708) were higher than the conventional definition S + G + M (AUC = 0.697). A significant correlation was found between CC and each of the three conventional components of sarcopenia. Conclusions: CC-added definitions of sarcopenia correlated with higher all-cause and CV mortality risks. CC is potentially a simple but valuable screening tool for sarcopenia that could improve diagnostic accuracy when used with other parameters.