Correction for the influence of background muscle activity on stretch reflex amplitudes

Background muscle activity (BGR) exerts a strong influence on stretch reflex amplitudes. While it is considered important to correct for this influence, which method best able to eliminate the effects of BGR remains unknown. We have therefore compared one previously described and one novel correction method which respectively consisted of calculating (1) the difference between reflex amplitude and BGR, and (2) the ratio of reflex amplitude to BGR. These correction methods were evaluated in a group of 23 healthy individuals. BGR and stretch reflexes were recorded from the gastrocnemius muscle of standing subjects who received sudden toe-up perturbations of a supporting platform upon which they were standing. Calculation of differences markedly reduced the influence of BGR on stretch reflex amplitudes in most, although not all, subjects. Calculation of ratios failed to correct for BGR in most subjects and caused a net increase in the influence of BGR on stretch reflex amplitudes. Because both correction methods insufficiently corrected for BGR in normal subjects, we introduce the use of analysis of covariance (ANCOVA) to reliably remove the influence of BGR on stretch reflex amplitudes. The use of ANCOVA is exemplified by showing that stretch reflex amplitudes are enhanced in patients with Parkinson's disease even if the influence of high BGR is completely taken into account.