A limited number of slices yields comparable results to all slices in foot intrinsic muscle deterioration ratio on computed tomography and magnetic resonance imaging

Abstract

Diagnostic imaging modalities, like computed tomography (CT) and magnetic resonance imaging (MRI), can be used to assess in vivo muscle quality. Quantitative assessment using these techniques is time-intensive and costly due in part to extensive post-processing needs. The purpose of this study was to identify whether a subset of slices on CT and MRI would yield comparable results to the full number of slices for a measure of muscle quality (muscle deterioration ratio = fat volume/muscle volume) in the foot intrinsic muscles of people with diabetes and peripheral neuropathy. CT (0.6 mm slice thickness) and MRI (3.5 mm slice thickness) scans were obtained using previously described methods. The total number of slices acquired during the scan was compared to several conditions using a portion of slices. Bland-Altman plots and Lin’s concordance correlation coefficient were used to test agreement. Any condition using at least three slices yielded substantial to almost perfect agreement with the total number of slices on both CT and MRI (Range of Lin’s concordance correlation coefficient: 0.947–0.999). Using a single slice in the middle of the region of interest demonstrated poor to moderate agreement with the total number of slices. The findings of this study suggest that using a limited number of slices to quantify muscle deterioration ratio on CT or MRI is a viable way to balance the combined need for measurement accuracy with feasibility in research and clinical settings.