Peaks or threshold-crossing: A comparative analysis of neuromuscular jitter measurement methods

Abstract

Introduction

Commercial electromyographic recording systems usually include two different methods for jitter measurement, based on peaks or threshold-crossing. There is reported evidence that the measurements obtained with both methods from discharges recorded with concentric needle electrodes offer comparable results, but this evidence is scarce. This study aimed to replicate such studies and extract conclusions related to the use of the two methods.

Methods

129 EMG recordings were obtained from 12 patients using concentric needle electrodes (0.02 mm²), filtered at 1000 Hz and oversampled to 200 kHz. The recordings were aligned using either peak or threshold-crossing methods. Jitter was measured using both methods and obtained as mean consecutive differences (MCD). Statistical analyses included Anderson-Darling, Wilcoxon, and linear regression tests.

Results

Of the 129 recordings, 49 (38 %) were excluded for presenting contaminated waveforms. MCD values were similar in the two methods, with a median difference of −0.92 µs. The Wilcoxon test confirmed this difference (p = 0.0002). The regression yielded a high coefficient of determination (R² = 0.994) and a slope of 0.989.

Conclusion

No significant differences were found between the two jitter measurement methods. Variations were only a few µs, not enough to affect pathological jitter diagnosis. Both methods are valid for clinical use.