Interactions between nociceptive and non-nociceptive afferent projections to cerebral cortex in humans

Interactions between nociceptive and non-nociceptive afferent projections to cerebral cortex in humans

Neuroscience Letters 248 (1998) 155–158 Interactions between nociceptive and non-nociceptive afferent projections to cerebral cortex in humans A. Ros...

127KB Sizes 0 Downloads 44 Views

Neuroscience Letters 248 (1998) 155–158

Interactions between nociceptive and non-nociceptive afferent projections to cerebral cortex in humans A. Rossi*, B. Decchi, V. Groccia, R. Della Volpe, R. Spidalieri Laboratory of Human Neurophysiology, Institute of Neurological Sciences, University of Siena, 53100 Siena, Italy Received 9 February 1998; received in revised form 30 March 1998; accepted 20 April 1998

Abstract We investigated the effect of a tonic discharge of muscle nociceptive afferents on somatosensory evoked potentials (SEPs) in humans in response to stimulation of non-nociceptive afferents arising from the same muscle. Conditioning nociceptive muscle stimulation was achieved by local injection of 50 mg levo-ascorbic acid (in a volume of 0.3 ml) in the body of the extensor digitorum brevis muscle (EDB). The test stimulus for SEPs was an electrical pulse applied to the EDB nerve at an intensity below the motor threshold. The main finding was that tonic muscle nociceptive stimulation strongly depressed the middle-latency P60N75 complex without modifying the size of the early P40-N50 complex of SEPs. Depression of the P60-N75 complex was correlated with the pain-induced loss of proprioception of the foot, making it plausible that this cortical complex reflects neuronal processes leading to perception.  1998 Elsevier Science Ireland Ltd. All rights reserved

Keywords: Muscle pain; Somatosensory evoked potentials; Sensory gating; Perception

A stimulus may be perceived differently if the circumstances surrounding its application are changed. For example, two stimuli differing in modality would be perceived differently from either stimulus alone. Separate afferent sources can, in fact, interact to produce spatial facilitation, occlusion or inhibition [6]. Somatosensory evoked potential (SEP) studies in humans have documented the suppressive effect of two different sensory modalities phasically activated (e.g. [2,12,13]). Under natural conditions, however, stimuli are perceived in a context that varies with the background afferent activity. Relatively few studies have focused on the effects of background activity of a particular modality on conventional SEPs to phasic activation of a nerve trunk. These studies have mainly concentrated on the interfering effects of tonic low-threshold cutaneous fibres [5]. Although it is a common experience that tonic pain can cancel out or distort the perception of an innocuous stimulus applied to the same site, there is no information available on interaction of a background nociceptive input * Corresponding author. Istituto di Scienze Neurologiche, Universita` degli Studi, 53100 Siena, Italy. Tel.: +39 577 585768; fax: +39 577 40327; e-mail: [email protected]

with SEPs to non-nociceptive stimulation. This probably reflects the objective difficulty of evoking selective tonic nociceptive input in humans. The recent introduction in man of a new method of evoking tonic pain discharge [8,9] has created new opportunities. The method is based on local injection (in 2–5 s) of 50 mg levo-ascorbic acid (L-AS) in a volume of 0.3 ml (pH 6.3 at 24°C). Due to a local drop in tissue pH, it evokes tonic pain stimulation which is fully reversible in about 20 min (Fig. 2A). Because of the small quantity of liquid injected, mechanical effects are relatively small and over within 2 min. After this interval there is a wide window in which discharge of nociceptive afferents is not contaminated by non-nociceptive volley. During recent experiments in our laboratory on interactions of muscle nociceptive input with group I spinal reflex pathways [9], it was observed incidently that tonic nociceptive stimulation applied to the body of the extensor digitor brevis muscle (EDB) systematically altered proprioception of the foot (see below). Proprioception was fully restored after the end of the pain sensation. In an attempt to identify some electrophysiological equivalent of it, we analysed the cortical components evoked by activation of EDB low-

0304-3940/98/$19.00  1998 Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940(98) 00354- 1

156

A. Rossi et al. / Neuroscience Letters 248 (1998) 155–158

threshold afferents before, during and after selective tonic nociceptive stimulation applied to the body of the same muscle. Data was obtained from 12 experimental sessions performed with seven healthy adult volunteers (22–40 years of age, two females and five males). The experimental procedure was approved by the local Ethical Committee. All subjects were members of the laboratory staff and were very familiar with the experimental protocol. They were seated in a reclining chair with the stimulated foot fixed to a rigid platform at 10° to the horizontal. Test stimuli for SEPs were applied through a monopolar electrode (hemiball electrode 1 cm in diameter) to the EDB nerve (branch of the deep peroneal nerve) at the ankle joint with the cathode fixed above the Achille’s tendon. The position of the active electrodes was adjusted so as to obtain selective contraction of the EDB, in order that test and conditioning (see below) stimuli activated afferents from the same site, i.e. mainly deep receptors from the dorsal foot. The stimulation condition was judged selective for the EDB nerve if, at an intensity 1.1 times the motor threshold, it evoked an isolated twitch of this muscle without any irradiating cutaneous sensation. The strength of the test stimulus was then decreased and kept constant just below the motor threshold. The effectiveness of the test stimulus in activating muscle afferents was verified by the post-stimulus time histogram which revealed a clear-cut monosynaptic facilitation of the EDB. Under this experimental condition, however, some coactivation of low-threshold cutaneous fibres could not be avoided [10]. The attempt to stimulate afferents from the EDB as selectively as possible was mainly made to obtain test and conditioning (see below) volleys from the same anatomical site. SEPs were recorded using stainless steel needle electrodes inserted subcutaneously, 2 cm behind Cz on the midline (Cz′) referenced to linked mastoid. The evoked activity (1.5–2 Hz rate) was amplified, filtered (1– 6000 Hz) and averaged over 200 ms after EDB nerve stimulation. Usually, a series of 150–200 responses were averaged. The conditioned SEPs (by nociceptive stimulation) were measured in terms of area (analysis windows: P40N50 for the early complex and P60-N85 for the middlelatency complex (see Fig. 1) and expressed as a percentage of their unconditioned size. The experimental protocol was as follows: a sterile butterfly needle was inserted in the body of the extensor digitorum brevis muscle for the L-AS injection (50 mg L-AS in 0.3 ml) and the flexible connector and the syringe with L-AS were secured to the skin. After the disappearance of any local sensation due to the needle, two or three series of 200 evoked responses were recorded. LAS was then injected and the mean value of cortical responses was computed every 5 min. Subjective pain sensation induced by L-AS injection in the body of the EDB, was monitored using a 11-point box scale: 0, absence of pain; 10, worst possible pain [3]. L-AS injection in the body of the EDB resulted in compression-like pain (see [9] for further details) the time-

course of which is reported in Fig. 2A. As already mentioned, during pain the subjects lost the sense of position of the stimulated foot (i.e. they had difficulty in determining the position of the foot in space), accompanied by loss or distortion of the sensation of foot contact with the platform and loss of sensation of the electrical stimulus to the EDB nerve. Since these sensory changes were about parallel in their time-course, we chose to score the subjective sensation evoked by electrical stimulus to the EDB nerve (i.e. the test stimulus for SEPs). Subjective sensation evoked by it was quantitatively scored with an arbitrary scale, 10 being the intensity in control conditions (Fig. 2B). The time-course of the stimulus sensation is illustrated in Fig. 2B. Comparison with Fig. 2A shows that there was a clear-cut inverse relationship between the intensity of subjective pain sensation and of perception of the electrical stimulus. Representative waveforms of the SEPs after low-threshold stimulation of the EDB nerve are shown in Fig. 1. Five main peaks P40, N50, P60, N75, P85 were identified, the mean latencies (±SD) of which were 38.7 (5.3), 49.4, (6.3),

Fig. 1. Somatosensory evoked potentials induced by stimulation of the deep peroneal nerve at the ankle (extensor digitorum brevis nerve), recorded over the cortical foot area. Each trace is the average of 100 sweeps (three superimposed traces). On the left: time intervals from the start of nociceptive stimulation to extensor digitorum brevis muscle (lasting 19 min in this case). Data in brackets are the size (area) of the P60-N75 complex expressed as a percentage of its control value (pre-pain). Post-pain traces are recorded 30 min after the end of any subjective pain sensation.

A. Rossi et al. / Neuroscience Letters 248 (1998) 155–158

59.4 (7.0), 74.1 (7.1), 85.9 (6.8) ms, respectively. During nociceptive stimulation applied to the EDB, significant changes were selectively observed in the middle-latency P60-N75 wave, the size of which underwent marked monotonic depression, with a maximum at about 5 min (P , 0.001 at 5 and 10 min; Student’s t-test) and recovered to its control value in parallel with the decay phase of the pain sensation curve (Fig. 2C). The time-course of this depression closely resembled that of perception of the electrical test stimulus (Fig. 2B). On the contrary, the early P40N50 wave showed non-significant changes: a slight potentiation (roughly coinciding with maximum inhibition of the P60-N75) was followed by a slight depression (Figs. 1 and 2C). The pain-induced depression of the middle-latency P60-N75 wave, with a constant early N40-P50 wave, prob-

Fig. 2. (A) Grand mean curve of subjective pain sensation evoked by 50 mg (in 0.3 ml) levo-ascorbic acid injected into the body of the extensor digitorum brevis muscle (EDB). (B) Grand mean curve of subjective sensation evoked by electrical stimulation to the EDB nerve (test stimulus for SEPs) during tonic EDB nociceptive stimulation. Note the inverse time-course of the intensity of pain sensation and the intensity of perception of the electrical stimulus. (C) Grand mean curves of the P40-N50 (white circles) and of the P60-N75 (black circles) components evoked by EDB nerve stimulation during tonic EDB nociceptive stimulation. Vertical bars represent one SD.

157

ably reflects an inhibitory process at cortical or precortical level. In fact, gating at subcortical level (e.g. at dorsal column nuclei level) was expected to depress the early wave too. Although the exact site of this interaction cannot be established from the present study, the thalamic reticular nucleus, where several functionally related cortical areas and thalamic nuclei interact, is a possible candidate. In particular, nociceptive discharge (see [11] for nociceptive supraspinal projections in humans), through the inhibitory connections that go from the somatosensory sector of this nucleus to the thalamic relay cells [4], could selectively gate thalamocortical transmission to the P60-N75 generator site. As detailed above, EDB nociceptive stimulation was accompanied by loss of perception of the electrical test stimulus and by a general loss, or severe reduction, of the sense of foot position. It is plausible, therefore, that the same neuronal chain leading to proprioception [1] also contribute to generation of the middle latency P60-N75 cortical wave, which is assumed to take place from multiple generator sources in areas 1, 2 and 4 ([15] and cf. [7]). Functionally, the pain-induced gating of this middle-latency component could act to prevent the neuronal processes responsible for perception of unpainful stimuli, so as to avoid any orientation of attention away from the nociceptive input. On the contrary, the absence of any significant changes in the early P40-N50 wave suggests that the pathways to perception largely bypass the site of generation of early components, (areas 3 and possibly 2) of the primary somatosensory cortex [15]. The input to these areas would act as an event marker for the timing of the sensation [12], and/or is involved in motor control mechanisms. In fact, neurones in area 4 mainly receive (via areas 3a and 2) [14] kinesthetic feedback from peripheral afferents which contributes to the control and monitoring of the movement. In summary, the present findings indicate that: (1) tonic muscle pain discharge strongly interacts with non-nociceptive input arising from the same site; (2) this interaction manifests with an inhibition of the middle-latency P60N75 cortical complex without modifying the early P40N50 component in the primary somatosensory cortex, suggesting that gating must occur at cortical or precortical level; (3) depression of the middle-latency complex correlates with the loss of proprioception of the foot, suggesting that this cortical complex reflects neuronal processes leading to perception. As a corollary, the different behaviour of the early and middle-latency components observed in the present paper clearly demonstrates that their generating mechanism is different. This study was financed by C.N.R. and M.U.R.S.T. (60%) and by a grant from the Istituto di Riabilitazione Fisiomedica Loretana (CB) to B.D. The authors thank Dr. Arrigucci, Dr. Di Troia, Dr. Ginanneschi and Dr. Zalaffi for technical assistance. [1] Aloisi, A.M., Decchi, B., Fontani, G., Rossi, A. and Carli, G.,

158

[2]

[3]

[4]

[5]

[6]

[7]

[8]

A. Rossi et al. / Neuroscience Letters 248 (1998) 155–158 Response of cat cortical neurons to position and movement of the femur, Somatosens. Motor. Res., 13 (1996) 263–271. Burke, D., Gandevia, S.C., McKeon, B. and Skuse, N.F., Interactions between cutaneous and muscle afferent projections to cerebral cortex in man, Electroenceph. clin. Neurophysiol., 53 (1982) 349–360. Ekblom, A. and Hansson, P., Pain intensity measurements in patients with acute pain receiving afferent stimulation, J. Neurol. Neurosurg. Psychiatry, 51 (1988) 481–486. Guillery, R.W., Fieg, S.L. and Lozsadi, D.A., Paying attention to the thalamic reticular nucleus, Trends Neurosci., 21 (1998) 28– 32. Kakigi, R. and Jones, S.J., Influence of concurrent tactile stimulation on somatosensory evoked potentials following posterior tibial nerve stimulation in man, Electroenceph. clin. Neurophysiol., 65 (1986) 118–129. Kang, R., Herman, D., MacGillis, M. and Zarzecki, P., Convergence of sensory inputs in somatosensory cortex: interactions from separate afferents sources, Exp. Brain Res., 57 (1985) 271–278. Mima, T., Ikeda, A., Terada, K., Yazawa, S., Mikuni, N., Kunieda, T., Taki, W., Kimura, J. and Shibasaki, H., Modalityspecific organization of cutaneous and proprioceptive sense in human primary sensory cortex studied by chronic epicortical recording, Electroenceph. clin. Neurophysiol., 104 (1997) 103–107. Rossi, A. and Decchi, B., Cutaneous nociceptive facilitation of lb heteronymous pathways to lower limb motoneurones in humans, Brain Res., 700 (1995) 164–172.

[9] Rossi, A. and Decchi, B., Changes in lb heteronymous inhibition to soleus motoneurones during cutaneous and muscle nociceptive stimulation in humans, Brain Res., 774 (1997) 55– 61. [10] Rossi, A., Mazzocchio, R. and Parlanti, S., Cortical projection of putative group lb afferent fibres from the human forearm, Brain Res., 547 (1991) 62–68. [11] Svensson, P., Beydoun, A., Morrow, T.J. and Casey, L., Nonpainful and painful stimulation of human skin and muscle: analysis of cerebral evoked potentials, Electroenceph. clin. Neurophysiol., 104 (1997) 343–350. [12] Tinazzi, M., Zanette, G., La Porta, F., Polo, A., Volpato, D., Fiaschi, A. and Mauguiere, F., Selective gating of lower limb cortical somatosensory evoked potentials during passive and active foot movements, Electroenceph. clin. Neurophysiol., 104 (1997) 312–321. [13] Jones, S.J., An ‘interference’ approach to the study of somatosensory evoked potentials in man, Electroenceph. clin. Neurophysiol., 52 (1981) 517–530. [14] Jones, E.G. and Friedman, D.P., Projection pattern of functional components of thalamic ventrobasal complex on monkey somatosensory cortex, J. Neurophysiol., 48 (1982) 521–543. [15] Xiang, J., Ryusuke, K., Hoshiyama, M., Kaneoke, Y., Naka, D., Takeshima, Y. and Koyama, S., Somatosensory evoked magnetic fields and potentials following passive toe movement in humans, Electroenceph. clin. Neurophysiol., 104 (1997) 393– 401.