G O Matheson, L Maffey-Ward, M Mooney, K Ladly, T Fung, Y T Zhang
Department of Medicine, Stanford University School of Medicine, USA.
DOI: 10.2340/165019771997292935

This study was undertaken to investigate the use of vibromyography (VMG) as a tool for quantifying skeletal muscle force production. Fourteen healthy volunteers were pretested using a Cybex isokinetic dynamometer to determine their isometric quadricep maximum voluntary contraction (MVC) values. On the basis of these results, the subjects were separated into two groups: high-force ("HF" MVC mean = 289 ft. lb. , range 254-330) and low-force ("LF" MVC mean = 154 ft. lb. , range 101-198). A vibromyographic piezoelectric accelerometer (Dytran 3115A) and electromyographic (EMG) surface electrodes were affixed to the rectus femoris muscle and recordings were obtained at 20, 40, 60, 80, and 100% MVC. Root mean squares, median and mean values were computed from digitized data in the time domain while peak values were calculated from a fast Fourier transform for both the VMG and EMG data. A two-way repeated measures MANOVA using relative values and a linear regression model using absolute values were studied using BMDP and MiniTab software. Linear correlations were found between quadricept force and all EMG variables (R2 range 0. 71-0. 90) except peak (R2 = 0. 39). The relationship between VMG and force was less linear (R2 range 0. 19-0. 69) because VMG values reach a plateau or even drop at 80% and 100% MVC. The HF-LF group differences were significant (p < 0. 05), for all VMG values with the exception of root mean squares, but were not significant (p > 0. 05) for all four EMG values. This study shows that, while EMG can discriminate force production within a given subject, VMG is a better discriminator of absolute muscle force values between subjects, particularly up to 60% MVC.