Work-related upper extremity disorders of the tendons, nerves, and musdes associated with. repetitive motion represent an increasing proportion of occupational illnesses, medical services, and workers' compensation costs. Severe local muscle fatigue may be a precursor of repetition strain injuries. The demand for objective and quantitative assessment of local muscle fatigue is now increasing. The purpose of this study is to determine the usefulness of power spectral analysis of individual muscle as one of the objective spectral parameters in quantification of local muscle fatigue.

From 10 normal male without neuromuscular dysfuction, motor unit action potentials (MUAPs) of three muscles (upper trapezius, infraspinatus, and anterior deltoid) of left shoulder girdle were obtained. The surface electrodes were used and the changes of power spectrum according to frequency components were calculated per minute during the keyboard task for 30-minute period by power spectral analysis of MUAPs signal.

There was a linear relationship between median frequency (MF) and time flow in the upper trapezius muscle, however the same was not observed in other muscles. The MF slope of the upper trapezius muscle showed a decline of the MF over time and reflected the fatigue rate. Other spectral parameters, such as mean frequency and total power were not significantly correlated to time flow during the keyboard task in all three muscles. And no significant correlation was observed between keying rate and slope of median frequency of upper trapezius muscle.

Above results suggested that the power spectral analysis on median frequency in upper trapezius muscle would be useful in quantification of local muscle fatigue during the keyboard task and local muscle fatigue of shoulder was affected by static posture, not by keying rate.