A diffusion tensor-based method facilitating volumetric assessment of fiber orientations in skeletal muscle

Background The purpose of this study was to develop a DTI-based method to quantitatively assess fiber angles and changes therein in leg muscles in order to facilitate longitudinal studies on muscle fiber architectural adaptations in healthy subjects. Methods The upper legs of five volunteers were scanned twice on the same day. The right lower legs of five volunteers were scanned twice with the ankle in three positions, i.e. -15 degrees dorsiflexion, 0 degrees neutral position, and 30 degrees plantarflexion. The MRI protocols consisted of a noise scan, a 3-point mDixon scan and a DTI scan. Fiber-angle color maps were generated for four muscles in the upper legs and two muscles in the lower leg. Voxel-wise fiber angles (theta) were calculated from the angle between the principal eigenvector of the diffusion tensor and a reference line defined between the origo and insertion points of each muscle. Bland-Altman analysis, intraclass correlation coefficient (ICC), coefficient of variation (CV%), minimal detectable change (MDC), standard error (SE) and Friedman test were used for assessing the feasibility of this method and in order to have an indication of the repeatability and the sensitivity. Results Bland-Altman analysis showed good repeatability (CV%<10 and 0.7 <= ICC <= 0.9) with exception of the Tibialis Anterior (TA) muscle in dorsiflexion position(CV%: 12.2) and the Semitendinosus (ST) muscle (left leg) (CV%: 11.4). The best repeatability metrics were found for the SOL muscle in neutral position (CV%: 2.6). Changes in average theta in TA and SOL with ankle positions were observed in accordance with expected agonist and antagonist functions of both muscles. For example, for the anterior left compartment the change in fiber angle Delta theta with respect to the neutral position Delta theta = -1.6 degrees 0.8 degrees and 2.2 degrees +/- 2.8 degrees (p = 0.008), for dorsiflexion and plantarflexion, respectively. Conclusion Our method facilitates fast inspection and quantification of muscle fiber angles in the lower and upper leg muscles in rest and detection of changes in lower-leg muscle fiber angles with varying ankle angles.