Elsevier

Clinical Neurophysiology

Volume 118, Issue 8, August 2007, Pages 1877-1888
Clinical Neurophysiology

Scalp electrical recording during paralysis: Quantitative evidence that EEG frequencies above 20 Hz are contaminated by EMG

https://doi.org/10.1016/j.clinph.2007.04.027Get rights and content

Abstract

Objective

To identify the possible contribution of electromyogram (EMG) to scalp electroencephalogram (EEG) rhythms at rest and induced or evoked by cognitive tasks.

Methods

Scalp EEG recordings were made on two subjects in presence and absence of complete neuromuscular blockade, sparing the dominant arm. The subjects undertook cognitive tasks in both states to allow direct comparison of electrical recordings.

Results

EEG rhythms in the paralysed state differed significantly compared with the unparalysed state, with 10- to 200-fold differences in the power of frequencies above 20 Hz during paralysis.

Conclusions

Most of the scalp EEG recording above 20 Hz is of EMG origin. Previous studies measuring gamma EEG need to be re-evaluated.

Significance

This has a significant impact on measurements of gamma rhythms from the scalp EEG in unparalysed humans. It is to be hoped that signal separation methods will be able to rectify this situation.

Introduction

High frequency EEG rhythms above 30 Hz (gamma rhythms) can be recorded reliably in experimental animals using intracerebral and subdural electrodes. In preparation for epilepsy surgery, surface and depth EEG recordings are sometimes required for localisation of the epileptogenic zone. These recordings often reveal spontaneous high frequency EEG components (sustained gamma), which increase with visual, memory or sensorimotor tasks, (induced gamma) reviewed by Lachaux et al., 2000, Tallon-Baudry, 2003. Use of indwelling electrodes is not practical or ethical in the routine clinical setting, so assessing the ability of the scalp EEG to evaluate gamma EEG is important. There is debate as to whether or not information in human scalp EEG reliably follows animal brain recordings (Juergens et al., 1999). Many suggest that changes in the EEG/magnetoencephalography (MEG) with cognitive tasks, despite being in areas of the scalp prone to muscle activity, show task specificity and a different profile to perceived muscle activity (Tallon-Baudry et al., 1998). It has long been assumed that scalp EEG is heavily influenced by EMG activity and so methods that attempt to remove its influence have been devised, for example, evoked response methodology. Specifically, there are well-recognised time-locked responses to stimuli that, when averaged, eradicate the influence of spontaneous EEG and artefact. These evoked-responses frequently reveal components of synchronous time-locked activity within the gamma frequency range (Herrmann and Knight, 2001) (evoked gamma) as well as non-synchronous, loosely time-locked gamma range activity (induced gamma) occurring around the time of the P3 (Herrmann and Knight, 2001) though even here there are not widely concordant data. The former is believed to reflect attention to the task and the latter generated only when a degree of cognitive reflection on the task is required. Generally, the more complex the task, the larger the amplitude of the P3 (Senkowski and Herrmann, 2002).

The scalp EEG has not been completely evaluated for routine use in measurement of high frequency rhythms. We have previously reported that sustained or induced gamma rhythms in the EEG can be recorded via scalp electrodes (Fitzgibbon et al., 2004) and that there are changes in gamma EEG power in primary generalised epilepsies (Willoughby et al., 2003). Because it is non-invasive, the scalp recorded EEG will remain an important experimental, diagnostic and research methodology in humans. Given the established relationship of various forms of gamma oscillations to conscious and sub-conscious cognitive processing, it has become an area of intense interest in states where cognition and mentation are disturbed. However, it is not yet clear how reliably induced gamma EEG can be recorded from humans because recordings obtained from ‘relaxed’ humans may be confounded by low voltage EMG activity close to the scalp electrodes (Goncharova et al., 2003), or from high voltage EMG activity from distant muscles, for example in the neck. This possibility has more than just a theoretical basis because the spectrum of frequencies in the EMG overlaps the spectrum of gamma EEG frequencies (Goncharova et al., 2003, Kumar et al., 2003).

We aimed to identify the influence of EMG on EEG by taking scalp recordings from subjects before and after complete neuromuscular blockade, thus enabling a direct comparison to be made between recordings when EMG activity might be a significant confounding component of the EEG and recordings when all skeletal muscle activity was blocked. Given the view that valid gamma EEG can be extracted from time-locked signal averaging of EEG, eg during evoked responses, an auditory odd-ball task was also undertaken, so providing, under normal and paralysed conditions, a comparison of evoked and induced EEG changes during pre-cognitive and cognitive processing.

Section snippets

Methods

The study was approved by the Clinical Research Ethics Committee of Flinders University and Flinders Medical Centre. Two male individuals (one right-, one left-handed) associated with the EEG research group were subjects. They were familiar with the scientific background and purpose of the study and gave written, informed consent for the procedures.

Results

The subjects tolerated the procedure without adverse effects and experienced the procedure with feelings of alertness and anticipation. They felt muscularly relaxed and comfortable in both non-paralysed and paralysed phases of the experiment and there were no changes in posture or facial expression due to paralysis. All tasks were performed accurately in both phases of the experiment and the subjects found no perceived difference in the execution of the tasks in the presence of paralysant or

Discussion

This study reveals that EMG contaminates the majority of the routine scalp EEG at frequencies above 20 Hz. The most striking feature of the power spectrum after paralysis is the power reduction of frequencies above 20–30 Hz (Fig. 1). In the unparalysed state, high frequency power was most evident in lateral and posterior leads, consistent with a localisation close to cranial and cervical muscles. However, even centrally and frontally placed electrodes exhibited high power relative to the

Acknowledgement

The EEG Research Unit was entirely funded by an equipment grant from The Wellcome Trust, London, UK. Project support has been provided by the National Health and Medical Research Council, the Epilepsy Foundation of South Australia, and Flinders Medical Centre Foundation. Dr. Whitham was also supported by a Pfizer Neuroscience Research Fellowship and an Epilepsy Society of Australia Fellowship from UCB Pharma. We also thank Professor Simon Gandevia for helpful advice.

References (24)

  • S.N. Baker et al.

    EEG oscillations at 600 Hz are macroscopic markers for cortical spike bursts

    J Physiol

    (2003)
  • Fitzgibbon SP, Powers DMW, Pope KJ, Clark CR. Removal of EEG noise and artefact using blind source separation. J Clin...
  • Cited by (377)

    • Graph neural networks in EEG spike detection

      2023, Artificial Intelligence in Medicine
    View all citing articles on Scopus
    View full text