• Kamei K
• Kumar DK
• Polus BI

• Electrocardiography
• Electromyography / methods
• Muscle, Skeletal / physiology
• Posture
• Reproducibility of Results

Design: Test-retest experimental design.

Participants: Twenty (20) healthy participants who had no recorded history of acute, severe, low back pain having occurred within the last 12 months.

Method: Each participant was tested on 2 occasions. They were required to maintain an unsupported sitting posture. The SEMG signal was recorded from lumbar paraspinal muscles for 60 sec during the unsupported sitting posture. The ECG artefact was removed from the raw data (EMG signal + ECG artefact) using a manual ECG cleaning technique.

Results: All subjects had an ECG artefact present for at least one electrode site. After ECG removal, the amplitude of the cleaned data (only EMG signal) was significantly (p < 0.05) smaller than the raw data in 10 subjects where the SEMG signal was contaminated by the ECG artefact for all electrode sites on both recording days. intraclass correlation coefficients (ICC) were improved after ECG artefact removal.

Conclusion: The ECG artefact is a large component of the raw EMG signal under conditions where the required level of muscle activation is low. This reduces the level of reliability with respect to the magnitude of the signal before the ECG artefact is removed. Although our ECG cleaning technique improved reliability, reproducibility of the EMG signal with respect to EMG amplitude remains only low to moderate. We conclude that SEMG may not be a reliable and valid method to measure lumbar muscle activity during execution of the unsupported sitting posture.

This abstract is reproduced with the permission of the publisher; full text (print only) by subscription.