Improved anatomical reproducibility in quantitative lower-limb muscle MRI.

PURPOSE

To compare the influence of two limb positions and slice prescription using scout-image-based and surface-anatomy-based methods on the reproducibility of quantitative MRI of lower-limb muscles.

MATERIALS AND METHODS

Ten healthy subjects were scanned at 3 Tesla with a two-dimensional turbo spin-echo T1-weighted acquisition. Imaging was performed at thigh and calf level in two subject limb positions and independently repeated by a second operator. Regions-of-interest (ROI) were drawn on three muscles at thigh and calf levels on axial slices at fixed distance from the knee joint and at a level determined by surface anatomy.

RESULTS

Test-retest reliability of muscle cross-sectional area and ROI area overlap were similar for both limb positioning methods. Changing limb position between scans reduced ROI overlap (P < 0.01). Scout-image-based slice prescription resulted in narrower limits of agreement and higher intraclass correlation coefficients compared with surface-anatomy-based slice prescription.

CONCLUSION

Slice prescription based on fixed distance from the knee joint provided superior reproducibility of slice location than a surface anatomy-based method and should be used for longitudinal quantitative MRI studies. Exact subject positioning will depend on scanner and coil configuration, but should be consistent through a longitudinal study.