Which Confounders Have the Largest Impact in Shear Wave Elastography of Muscle and How Can They be Minimized? An Elasticity Phantom, Ex Vivo Porcine Muscle and Volunteer Study Using a Commercially Available System

Abstract

The goal of the study was to investigate the quantitative impact of region of interest (ROI), software choice, muscle fiber orientation and preload tension on shear wave velocity (SWV). First, SWV was assessed in an isotropic elasticity phantom and ex vivo porcine muscle using a commercially available clinical ultrasound system. Secondly, SWV was acquired in relaxed and stretched calf muscles of healthy volunteers (dorsal extension of the talocrural joint), for both parallel and transverse probe direction to the fibers, as well as for different ROIs and software versions. The effect of intermediate probe-fiber alignments was also analyzed. Finally, the impact of confounding factors on SWV reproducibility was minimized with a second force-controlled volunteer study, in which the calf was isometrically loaded, and fiber orientation and ROI were well-defined. 2046 in vivoSWE images were acquired to analyze SWV reproducibility with different confounder settings. In healthy volunteers, the main variance-contributing factors were in order of importance muscle tension, fiber orientation, horizontal ROI size and insertion depth. Regression analysis showed significantly reduced SWV with increasing insertion depth for each study material. Parallel probe-fiber orientation, muscle stretch and increasing horizontal ROI size led to significantly higher SWV. Based on the results of the study, we provide recommendations to minimize the impact of confounders in musculoskeletal elastography and discuss the main confounding mechanisms and trade-offs between confounding variables. Coefficients of variation can be significantly reduced with a controlled protocol, if the confounders are clinically taken into account.