Fatigue of the erector spinae muscles. A quantitative assessment using "frequency banding" of the surface electromyography signal.

STUDY DESIGN

The authors investigated fatigue-induced changes in the frequency content of the surface electromyographic (EMG) signal from the erector spinae muscles.

OBJECTIVES

The objective of the study was to understand the EMG changes in fatiguing muscle and to obtain a reliable index of fatigue.

SUMMARY OF BACKGROUND DATA

Power spectral analysis has been used increasingly in recent years to monitor muscle fatigue, but parameters other than the mean or median frequency have received little attention.

METHODS

Thirty-five healthy volunteers participated. They pulled upward with constant force on a handlebar attached to a floor-mounted load cell while the EMG signal from the erector spinae was recorded at the levels of T10 and L3 at 1024 Hz; 1.0-sec "windows" of the signal were analyzed using fast Fourier transforms, and the resulting power spectra were divided into 10 frequency bands between 5 Hz and 300 Hz. The median frequency, total power, and peak amplitude of the spectra were also calculated. Changes in the frequency content of the EMG signal were examined during submaximal contractions of different intensity and duration.

RESULTS

Median frequency decreased steadily during the contractions, whereas total power and peak amplitude increased. The most repeatable and linear index of change was the increase in the EMG signal in the 5-30 Hz frequency band. The middle-to-high frequency component of the EMG signal increased during the early stages of the contractions, but decreased as the endurance limit was approached.

CONCLUSIONS

Changes in the 5-30 Hz band of the EMG power spectrum provide a more reliable and linear index of fatigue in the erector spinae muscles than do changes in median frequency. In the erector spinae, the early effects of fatigue appear to be delayed by the recruitment of additional motor units.