Reliability of magnetic resonance imaging measurements of the cross-sectional area of the muscle contractile and non-contractile components

Abstract

Magnetic resonance imaging (MRI) is increasingly used to study skeletal muscles patients with muscular disorders. We report an MRI technique for evaluating the trunk muscles. This technique takes both the component surface area (CSA) and the density of the muscles on MRI axial slices into account . Using a computer-based image analysis system, we combined MRI data measuring the muscle CSA, which was separated into the contractile component (CCSA) and the non-contractile component (NCCSA). The purpose of this study was to analyze the reliability of this method of measuring the CSA, CCSA and NCCSAin trunk muscles on MRI axial slices through L4 and T12. Thirty volunteer subjects were enrolled in this study. Two acquisitions were performed. For the reliability analyses, each of the two slices (T12 and L4) from 30 subjects was measured by three raters trained in this technique, on two occasions 2weeks apart. Each muscle was surrounded and its CSA, NCCSA and CCSA were recorded. For each muscle, the agreement between the two sets of 30 measurements performed by three observers was evaluated by calculating an intra-class correlation coefficient (ICC). Regarding the slice through L4 and T12, the reliabilities of the measurement of CSA, CCSA were very good for all the muscles except the parietal muscles. MRI measurements of the trunk muscle cross-sectional areas and of the CCSA and NCCSA are reliable.