Turns-amplitude analysis at different sampling frequencies

In the EMG interference pattern the number of potential reversals of more than 100 μV (i.e., turns) measured may be low if a low sampling frequency is used. This might be more pronounced in a myopathic muscle with small short potentials than in a control muscle, resulting in a decreased diagnostic yield. The influence of sampling frequencies from 6 to 200 kHz on the turns, mean amplitude and ratio of turns to mean amplitude analysed on the interference pattern at a force of 30% of maximum was examined at 10 sites in each of 5 control muscles, 5 myopathic muscles and 5 neurogenic muscles (i.e., 150 sites in all). In this small group of muscles low sampling frequency did not tend to reduce the diagnostic yield of the measurements. Proper reproducibility was obtained with sampling frequencies down to 17 kHz. However, due to systematic errors, measurements obtained with sampling frequencies below 25 kHz require comparable control values or compensations for errors. As time intervals between turns (not studied here) have a skew distribution with many small time intervals a sampling frequency of 50 kHz or higher is probably required.