Effect of recording window and stimulation variables on the statistical technique of motor unit number estimation

A variety of methods are used for the selection of recording window sizes a nd stimulation current levels for statistical motor unit number estimation (MUNE). This study compares different recording window sizes and stimulatio n current levels within those windows in the same subjects to determine the effect on MUNE value and reproducibility. Four recording windows of 10% si ze were compared with four of 5%, with the stimulation current set in the l ower quarter, middle half, and upper quarter of the recording window. MUNE for stimulation current set in the lower quarter of the window was 81 (62-1 03) for 10% recording windows and 120 (108-135) for 5% recording windows, a nd 91 (61-123) and 133 (120-154) for stimulation current set in the middle half. Increasing the recording window size from 5 to 10% lowers the MUNE va lue in controls, but tends to improve reproducibility; and setting the stim ulation current in the lower quarter of the window, changes the MUNE value minimally, while tending to improve further reproducibility. Excellent repr oducibility of MUNE was obtained when applied to a pilot group of 10 amyotr ophic lateral sclerosis patients. Based on this study, we conclude that the ideal method for statistical motor unit estimation involves using 10% reco rding windows and setting the stimulation current in the lower quarter of t he recording window. (C) 2001 John Wiley & Sons, Inc.