Oscillatory motor cortex–muscle coupling during painful laser and nonpainful tactile stimulation

Abstract

Noxious stimulation activates–in addition to the brain structures related to sensory, emotional, and cognitive components of pain–also the brain's motor system. Effect of noxious input on the primary motor (MI) cortex remains, however, poorly understood. To characterize this effect in more detail, we quantified the ongoing oscillatory communication between the MI cortex and hand muscles during selectively noxious laser stimulation. The subjects maintained an isometric contraction of finger muscles while receiving the laser stimuli to the dorsum of the hand. Tactile stimuli with well-known effects on the MI cortex reactivity served as control stimuli. Cortex–muscle coherence was computed between magnetoencephalographic (MEG) signals from the contralateral MI and electromyographic (EMG) signals from the hand muscles. Statistically significant coherence at ∼20 Hz was found in 6 out of 7 subjects. The coherence increased phasically after both types of stimuli but significantly later after laser than tactile stimuli (mean ± SEM peak latencies 1.05 ± 0.12 s vs. 0.58 ± 0.06 s; P < 0.05), and the coherence increase lasted longer after laser than tactile stimuli (0.87 ± 0.09 s vs. 0.50 ± 0.06 s, P < 0.05). The observed coherence increase could be related to stabilization of the motor-cortex control after sensory input. Our findings add to the clinically interesting evidence about the cortical pain–motor system interaction.

Introduction

Sudden pain elicits avoidance reactions, including the spinal flexion reflex, but also more complex motor responses (Melzack and Wall, 1965). Although the mechanisms of the flexion reflex are well understood, little is known about the effect of pain on the cortical parts of the motor system. Interestingly, chronic pain is sometimes associated with motor dysfunction, and motor cortex stimulation alleviates chronic pain (Tsubokawa et al., 1991a, Tsubokawa et al., 1991b). A few brain imaging studies have shown pain-related activation of the primary motor cortex (Becerra et al., 2001, Casey et al., 1996, Davis et al., 2002, Gelnar et al., 1999, Peyron et al., 1999), and recent studies applying transcranial magnetic stimulation and magnetoencephalography (MEG) have shown modulation of the MI cortex function at a short latency (before 200 ms after the onset of painful stimulus to hand dorsum) (Raij et al., 2004, Valeriani et al., 1999).

Communication between the MI cortex and peripheral muscles can be studied by means of coherence between oscillatory signals arising from the MI cortex and the muscles. The 15- to 30-Hz MI oscillations establish a consistent coupling with surface electromyographic signals recorded from active muscle during various motor tasks (Brown et al., 1998, Conway et al., 1995, Feige et al., 2000, Gross et al., 2000, Halliday et al., 1999, Kilner et al., 1999, Kilner et al., 2000, Kilner et al., 2003, Kristeva-Feige et al., 2002, Marsden et al., 2000, Mima et al., 1999, Salenius et al., 1997a). Although precise neurophysiological mechanisms underlying the oscillatory cortex–muscle coupling are still debated (Grosse et al., 2002, Hari and Salenius, 1999, Mima and Hallett, 1999, Salenius and Hari, 2003), the somatotopical organization of the MEG–EMG coherence during movements of different limbs (Murayama et al., 2001, Salenius et al., 1997a) and the time lag between MI and muscle oscillations (Brown et al., 1998, Gross et al., 2000, Marsden et al., 2000, Salenius et al., 1997a) suggest that the MEG–EMG coherence provides a unique view into dynamic oscillatory interactions between the MI cortex and muscle.

Innocuous proprioceptive and tactile inputs to the MI cortex have an important and well-known role in the adjustment of voluntary movements. Accordingly, nonpainful median nerve stimulation modulates the motor cortex spontaneous oscillatory activity (Hari and Salenius, 1999, Salenius et al., 1997b, Stancak et al., 2003) and, possibly, the cortex–muscle coherence as well (Hari and Salenius, 1999). Spontaneous oscillatory activity of the MI cortex is suppressed after selectively noxious laser stimuli (Raij et al., 2004), but the effect of noxious input on the cortex–muscle coupling remains unknown. We therefore applied selectively noxious thulium-laser stimuli to the hand dorsum of subjects who maintained an isometric contraction of finger muscles, and we computed coherence between magnetoencephalographic and electromyographic signals. Tactile stimuli were applied to compare effects of the noxious and innocuous input.