Validation of shear wave elastography in skeletal muscle

Abstract

Skeletal muscle is a very dynamic tissue, thus accurate quantification of skeletal muscle stiffness throughout its functional range is crucial to improve the physical functioning and independence following pathology. Shear wave elastography (SWE) is an ultrasound-based technique that characterizes tissue mechanical properties based on the propagation of remotely induced shear waves. The objective of this study is to validate SWE throughout the functional range of motion of skeletal muscle for three ultrasound transducer orientations. We hypothesized that combining traditional materials testing (MTS) techniques with SWE measurements will show increased stiffness measures with increasing tensile load, and will correlate well with each other for trials in which the transducer is parallel to underlying muscle fibers. To evaluate this hypothesis, we monitored the deformation throughout tensile loading of four porcine brachialis whole-muscle tissue specimens, while simultaneously making SWE measurements of the same specimen. We used regression to examine the correlation between Young's modulus from MTS and shear modulus from SWE for each of the transducer orientations. We applied a generalized linear model to account for repeated testing. Model parameters were estimated via generalized estimating equations. The regression coefficient was 0.1944, with a 95% confidence interval of (0.1463-0.2425) for parallel transducer trials. Shear waves did not propagate well for both the 45° and perpendicular transducer orientations. Both parallel SWE and MTS showed increased stiffness with increasing tensile load. This study provides the necessary first step for additional studies that can evaluate the distribution of stiffness throughout muscle.

Keywords: Elastic moduli; Materials testing; Passive stiffness; Shear wave elastography; Ultrasonography.

Conflict of interest statement

Conflict of interest statement: The authors do not have any financial or personal relationships to disclose that could have inappropriately biased this work.

Copyright © 2013 Elsevier Ltd. All rights reserved.

Figures

Figure 1

Experimental set-up. Ultrasound transducer aligned with long axis of the brachialis. Bone segments for proximal and distal attachment sites are embedded in bone cement and fixed to the MTS base and load cell, respectively. L0 and L1 measured between arrowheads indicating origin and insertion, with the elbow at 90° and 180°, respectively. Axial tensile testing begins as the MTS base is translated away from the load cell, stretching the brachialis from L0 to L1.

Figure 2

Ultrasound transducer orientations over brachialis muscle, with sample B-mode images acquired using the Aixplorer ultrasound system in the MSK/Superficial MSK mode to demonstrate underlying muscle structure. Ultrasound transducer coming out of the page, white box indicates ultrasound transducer footprint, black dashed line indicates imaging plane. (A) parallel, demonstrating muscle fibers running right-to-left. (B) 45°, (C) perpendicular, demonstrating muscle fibers visible in cross-section. Solid bars indicate focusing range. Scale (on right) in cm.

Figure 3

Shear wave elastography and time-to-peak fitting. A) Focused ultrasound “push” beam generates shear waves in the muscle tissue specimen. The same transducer monitors the time-to-peak of shear wave displacements at different lateral distances, indicated by vertical bars. B) Example of shear waveform propagation across a lateral range of approximately 6 mm. (Slow time indicates time-direction of shear wave propagation; x represents lateral distance to the initial “push” beam.) C) Example of time-to-peak fitting of waveforms, obtained via linear regression.

Figure 4

Moduli data from a representative sample for specimen #3. A) Stress-strain curve obtained from MTS test. B) Mean Young's moduli data, measured based on the stress-strain curve in (A) throughout tensile test from all 15 loading trials. Error bars indicate ±SD. C) Mean shear moduli data throughout tensile test for each of the three transducer orientations. Error bars indicate ±SD.

Figure 5

Scatterplots of elastic moduli (Shear modulus, G; Young's modulus, E) for parallel ultrasound transducer trials. Generalized linear model regression line obtained from generalized estimating equations analysis. A) Specimen #1. G = 0.2602 E – 9.1963; R2= 0.9651. B) Specimen #2. G = 0.1539 E +0.1928; R2= 0.9884. C) Specimen #3. G = 0.1750 E – 5.4259; R2= 0.9577. D) Specimen #4. G = 0.2044 E – 1.0085; R2= 0.9160. E) All trials from all 4 specimens combined. G = 0.1944 E – 3.6760.