A study of the reproducibility of three different normalisation methods in intramuscular dual fine wire electromyography of the shoulder.

The purpose of this study was to determine the most appropriate method of normalisation for dual fine wire electromyography of shoulder muscles. Five healthy subjects were studied, with one muscle investigated in each subject (2 supraspinatus, 2 infraspinatus, 1 subscapularis). Three dual fine wire electrodes were inserted 1 cm apart around the recognised insertion points. Each subject performed five types of cyclic exercise on an isokinetic muscle dynamometer with an isometric maximal voluntary contraction (MVC) being performed before and after the exercise protocol. The EMG signal was normalised using each of the MVC voltage, the peak voltage and the whole-cycle mean voltage. There was a considerable difference (5-143%) between the MVC signals pre- and post-protocol, although no systematic trend was demonstrable. The overall mean between electrode variation in the normalised signal measured at the peak of the cycle ranged from 48-71% when normalised to pre-protocol MVC, but only 4-13% when normalised to the peak voltage and 9-17% using the whole-cycle mean voltage. However the pattern of activation within the movement cycle, which was preserved by normalisation using the peak or mean signal, was consistent between different electrode positions. It was concluded that the EMG signal depended on electrode position even when near the recognised insertion point, and that the MVC signal is highly variable in magnitude between electrodes and between pre- and post-protocol measurements.