Muscle mass assessment by computed tomography in chronic kidney disease patients: agreement with surrogate methods

Background/objectives Patients with chronic kidney disease (CKD) are subjected to muscle wasting. Therefore, it is important to investigate surrogate methods that enable the assessment of muscle mass loss in the clinical setting. We aimed to analyze the agreement between computed tomography (CT) and surrogate methods for the assessment of muscle mass in non-dialysis CKD patients. Subjects/methods Cross-sectional study including 233 non-dialysis patients on CKD stages 3 to 5 (61 +/- 11 years; 64% men; glomerular filtration rate 22 (14-33) mL/min/1.73 m(2)). The muscle mass was evaluated by CT and bioelectrical impedance, skinfold thicknesses, midarm muscle circumference (MAMC), the predictive equations of Janssen and Baumgartner and the physical examination of muscle atrophy from the subjective global assessment. Results In males, the MAMC showed the best agreement with CT as indicated by the kappa test (k = 0.57, P < 0.01), sensitivity (S = 68%), specificity (S = 89%) and accuracy (area under the curve-AUC = 0.78), followed by the Baumgartner equation (kappa = 0.46, P < 0.01; sensitivity = 60%; specificity = 87% and AUC = 0.73). In female, the Baumgartner equation showed the best agreement with CT (kappa = 0.43, P < 0.01; sensitivity = 57%; specificity = 86% and AUC = 0.71). Conclusions The MAMC and Baumgartner equation showed the best agreement with CT for the assessment of muscle mass in non-dialysis CKD patients.