Motor unit synchronous firing as revealed by determinism of surface myoelectric signal

https://doi.org/10.1016/j.jneumeth.2006.01.003Get rights and content

Abstract

Information on motor strategies can be extracted from the surface electromyogram (EMG) by non-linear methods. The percentage of determinism (%DET) obtained from recurrence quantification analysis (RQA) may be a sensitive variable to detect synchronous motor unit behaviour. The purpose of the present study was to validate this methodology by comparing it with an established technique estimating the degree of synchronization of pairs of voluntary activated motor units from the correlation of their firing in the time-domain. Single motor unit activity was recorded in extensor carpi radialis (ECR) muscle by pairs of tungsten microelectrodes inserted into the muscle belly. Cross-correlation analysis was performed in order to determine synchronization peak area by computing synchronous impulse probability. Surface EMG activity was recorded in parallel by electrodes placed over the skin of the same muscle and %DET was used as a measure of synchronous activity. The %DET appeared to be a valid measure of synchronization yielding results comparable to those obtained with cross-correlation analysis. Increases in %DET (t = 64.59, P < 0.0001) highly correlated (r2 = 0.70, P = 0.0013) with pharmacologically induced increases in the synchronization activity of pairs of ECR motor units (t = 8.71, P < 0.0001). RQA may be used as an alternative methodology for testing synchronous motor unit behaviour from surface EMG under physiological and pathological conditions.

Introduction

Human single motor unit recordings have been extensively used to study the functional coupling (synchronization) of pairs of motor units (Adam et al., 1978, Datta et al., 1991, Kim et al., 2001, Mattei et al., 2003, Schmied et al., 1993, Schmied et al., 1994, Sears and Stagg, 1976, Turker and Powers, 2001). However, this technique has a number of drawbacks: (1) it is invasive and causes discomfort, (2) recordings from pairs of voluntary activated motor units can be reliably obtained only for weak or moderate levels of tonic muscle contraction, and (3) motor unit discrimination is time-consuming. In addition, correlating the firing of pairs of voluntary activated motor units in the time-domain (cross-correlation analysis) (Mattei et al., 2003, Schmied et al., 1993) yields findings which may not reflect the properties of the entire motoneurone population.

Although the surface electromyogram (EMG) is a global measure of motor unit activation in humans, there are limits to the information that can be extracted from this signal using conventional techniques (Disselhorst-Klug et al., 2000, Farina et al., 2002, Farina et al., 2004b, Semmler and Nordstrom, 1999, Yue et al., 1995). It has been proposed that motor strategies can be examined with variables extracted from the surface EMG by non-linear recurrence methods (Farina et al., 2004b, Filligoi and Felici, 1999, Nieminen and Takala, 1996, Webber et al., 1995). Recurrence quantification analysis (RQA) (Eckmann et al., 1987) is a technique for the detection of state changes in drifting dynamic systems that does not necessitate any a priori constraint on data size, stationarity, and statistical distribution (Fattorini et al., 2005, Filligoi and Felici, 1999). Subtle changes in surface EMG can be detected by different indexes extracted from RQA such as the percentage of determinism (%DET), which reflects the amount of rule-obeying structure in the signal dynamic (Farina et al., 2002, Farina et al., 2004b, Webber et al., 1995). The %DET may be a sensitive variable to detect motor unit synchronization, because it reveals embedded determinisms in an apparently stochastic signal (Eckmann et al., 1987).

The purpose of the present study was to investigate whether RQA is effective in detecting motor unit synchronization from the surface EMG. This methodology was compared with findings obtained in parallel from single muscle fibre recordings and cross-correlation analysis.

Section snippets

Material and subjects

The data for the present study was retrieved from original recordings obtained in our lab and used in a previous publication (Mattei et al., 2003). In that work, an increased synchronization of pairs of voluntarily activated motor units recorded from extensor carpi radialis (ECR) muscle was demonstrated following the acute injection of l-acetylcarnitine (l-Ac), a cholinergic substance shown to potentiate spinal recurrent inhibition (Mazzocchio and Rossi, 1997), but not after saline injection.

Results

Fig. 2 shows the group data relative to the amount of synchronization between pairs of voluntary activated ECR motor units before, during and after l-Ac (A) and saline (B) injection. There was a significant difference (t = 8.71, P < 0.0001) in the synchronization peak area, as expressed by SIP, between the pre-injection period (0.053 ± 0.002) and the l-Ac injection period (0.071 ± 0.001), whereas no significant difference in the synchronization peak area was found between the pre-injection period (0.053

Discussion

This study demonstrates that the application of RQA to the surface EMG signal (Farina et al., 2004b, Filligoi and Felici, 1999, Nieminen and Takala, 1996, Webber et al., 1995) is as effective as the cross-correlation analysis of single motor unit activity (Adam et al., 1978, Datta et al., 1991, Kim et al., 2001, Mattei et al., 2003, Schmied et al., 1993, Schmied et al., 1994, Sears and Stagg, 1976, Turker and Powers, 2001) in extracting information on the synchronous behaviour of motoneurones

Acknowledgements

This research was supported by grants from the University of Siena (Piano di Ateneo per la Ricerca: PAR), Italian Ministry for Scientific Research and Association Française contre les Myopathies (AFM), la Fondation pour la Recherche Médicale (FRM), the Délégation Générale à L’Armement (DGA).

References (24)

  • J.P. Eckmann et al.

    Recurrence plots of dynamical systems

    Europhys Lett

    (1987)
  • D. Farina et al.

    Nonlinear surface EMG analysis to detect changes of motor unit conduction velocity and synchronization

    J Appl Physiol

    (2002)
  • Cited by (0)

    View full text