Elsevier

Clinical Biomechanics

Volume 18, Issue 6, July 2003, Pages 537-542
Clinical Biomechanics

Measurement of muscle activity with magnetic resonance elastography

https://doi.org/10.1016/S0268-0033(03)00070-6Get rights and content

Abstract

Objective. To non-invasively determine muscle activity.

Design. A correlation analysis study.

Background. Electromyography is traditionally used to measure the electrical activity of a muscle and can be used to estimate muscle contraction intensity. This approach, however, is limited not only in terms of the volume of tissue that can be monitored, but must be invasive if deep lying muscles are studied. We wished to avoid these limitations and used magnetic resonance elastography in an attempt to non-invasively determine muscle activity. This novel approach uses a conventional MRI system. However, in addition to the imaging gradients, an oscillating, motion sensitizing field gradient is applied to detect mechanical waves that have been generated within the tissue. The wavelength correlates with the stiffness of the muscle and hence with the activity of the muscle.

Methods. Six volunteers (mean age: 30.1 years, range: 27–36 years) without orthopedic or neuromuscular abnormalities, lay supine with their legs within the coil of a MRI scanner. The wavelengths of mechanically generated shear waves in the tibialis anterior, medial and lateral head of the gastrocnemius and the soleus were measured as the subjects resisted ankle plantar-flexing (8.2 and 16.4 nm) and dorsi-flexing (20.2 and 40.4 nm) moments. The findings were then compared to EMG data collected under the same loading conditions.

Results. Magnetic resonance elastography wavelengths were linearly correlated to the muscular activity as defined by electromyography. (TA, R2=0.89, P=0.02; MG, R2=0.82, P=0.05; LG, R2=0.88, P=0.03; S, R2=0.90, P=0.02)

Conclusions. Magnetic resonance elastography may be a promising tool for the non-invasive determination of muscle activity.

Relevance Magnetic resonance elastography has potential as the basis for a new non-invasive approach to study in vivo muscle function.

Introduction

It is often necessary to quantify muscular activity. Traditionally this has been done with electromyography (EMG) which, in conjunction with muscle cross-sectional area and muscle fiber direction, provides an estimation of muscle forces (Nieminen et al., 1995). EMG, although it is the gold standard, has a number of limitations. Surface EMG (sEMG), for example, is simple to implement but is adequate only for the study of superficial muscles. Unfortunately, the analysis of deeper muscles usually requires skilled personal, invasive needle examinations and the placement of fine wire electrodes.

Magnetic resonance elastography (MRE) is a new technique (Dresner et al., 2001; Kruse et al., 2000; Muthupillai et al., 1995; Muthupillai et al., 1996) that has already been used to assess the material properties of a number of tissues. This approach uses a conventional magnetic resonance imaging (MRI) system modified to permit the detection of tissue displacements produced by shear waves that have been generated in a tissue by a transducer attached to its surface. MRE is based on the concept that mechanical waves travel more rapidly, and hence have longer wavelengths, in stiffer, more rigid materials. In effect, MRE detects changes in material stiffness.

As a result MRE may be able to provide a measure of the muscle activity (i.e., the stronger the contraction, the stiffer the muscle and the longer the wavelength). Preliminary work in our laboratory provides further support for this idea in that MRE in and ex vivo experiments of single superficial muscles have shown that wavelength does increase with increases in passive loading or the intensity of an active muscle contraction (Dresner et al., 2001).

This study was designed to test the hypothesis that MRE can be used to measure muscle activity. This hypothesis has more than a theoretical basis. Muscles are complex (Moss and Halpern, 1977) but do become stiffer as they contract. In fact, in the case of an isometric contraction (in which muscle contracts without shortening) the stiffness of the muscle can be taken as a direct indication of the number of cross-bridges (Ettema et al., 1994).

This investigation had two phases. The first was to establish that MRE could collect load associated muscle activity data in multiple muscles simultaneously. The second was to correlate these findings with the EMG activity of the muscles under the same loading conditions.

Section snippets

Subjects

This study was reviewed and approved by our institution’s Institutional Review Board. The subjects consisted of six healthy volunteers (three men, three women) with an average age, weight, and height of 30.1 years (range: 27–36), 69.9 kg (range: 52.5–84) and 1.72 m range: 1.56–1.84) respectively. Participants were required to have a normal neuromuscular history and examination (including normal lower extremity range of motion, strength, sensation and deep tendon reflexes).

Apparatus

The experiment was

Results

Fig. 4A and B are typical MRE images and show shear waves propagating through muscles of the lower leg at rest (A) and as the subject resists a 20.2 nm dorsi-flexing torque (B). The increases of wavelength with load are apparent in the soleus as well as in the lateral head of the gastrocnemius.

Shear wave wavelengths differed little between the muscles when they were at rest. The shear wavelength (average (Standard error)) measured at rest was 2.3 (0.5) cm in the tibialis anterior, 3.0 (0.7) cm

Discussion

The goal of this study was to test the hypothesis that MRE can be used to measure muscle activity. We believe that the high linear correlation (R2=0.82–0.90) between the EMG and MRE findings support this conjecture. Our findings are limited to the lower extremity muscles we studied, but there is no reason to expect that similar correlations could not be obtained for all muscles or, at least, for those of the same or larger size.

One might ask “what is the point of all of this?” EMG is, after

Acknowledgements

This study was supported by NIH grant HD 37650-01 and NIH grant CA 75552.

References (16)

There are more references available in the full text version of this article.

Cited by (0)

View full text