Exp Brain Res (1990) 83:67-78

Experimental BrainResearch 9 Springer-Verlag 1990

Functional organization of the nociceptive withdrawal reflexes I. Activation of hindlimb muscles in the rat J. Schouenborg and J. Kalliom~iki University of Lund, Institute of Physiology and Biophysics, S61vegatan 19, S-223 62 Lund, Sweden Received February 16, 1990 / Accepted June 30, 1990

Summary. 1. The organization of the nociceptive hindlimb withdrawal reflexes was investigated in 93 halothane/nitrous oxide anesthetized rats. Electromyographical techniques were used to record reflex activity in single motor units. 2. Most of the hindlimb muscles were activated by noxious mechanical stimulation of the skin of the ipsilateral hindlimb. These were the plantar flexors of the digits, the pronators of the paw, the dorsiflexors and the plantar flexors of the ankle, the flexors of the knee, the flexors of the hip and the adductors. By grading the stimulus intensity it was shown that all these muscles received input from cutaneous nociceptors. 3. Noxious stimulation of the skin failed to activate the obturator, knee extensors and m. tibialis posterior and, in most rats tested, m. semimembranosus and m. adductor magnus. The plantar flexors of the ankle, while exhibiting a clear nocireceptive field in all rats tested, had a high threshold and responded much more weakly than the dorsiflexors of the ankle. Thus, responses in muscles which oppose gravity in the standing position were either very weak or absent. 4. The present study shows that each of the activated hindlimb muscles has a highly organized nocireceptive field on the skin, which is related to the withdrawal movement caused by the muscle itself. Each of the muscles normally causes the withdrawal of its receptive field when the foot is on the ground. The skin area most effectively withdrawn, in this situation, corresponds to the most sensitive area of the nocireceptive field. However, with the exception of the plantar flexors of the digits and/or the ankle, each of the hindlimb muscles also withdraws the major parts of their receptive fields when the foot is off the ground. The locations of the nocireceptive fields were independent of the position of the hindlimb. These characteristics of the nociceptive withdrawal reflexes are the basis for their "local sign" (Sherrington 1906). 5. The threshold and the time course of reflex activation were different in different muscles. However, muscles with a similar action; the plantar flexors of the digits, the pronators of the paw, the dorsiflexors of Offprint requests to: J. Schouenborg (address see above)

the digits, the flexors of the knee and the adductors, respectively, had similar thresholds and time courses. Furthermore, the threshold and latency of activation of each muscle increased towards the border of its nocireceptive field, reflecting a decreasing sensitivity. These findings explain the progressive recruitment of muscles during increasing strength of noxious stimulation, termed "irradiation" (Sherrington 1906). 6. It is suggested that the nociceptive withdrawal reflexes are organized as separate reflex paths to individual muscles, each of which has a well organized cutaneous nocireceptive field.

Key words: Nociception - Withdrawal reflexes - Flexion reflex - Spinal cord - Pain - Rat

Introduction The nociceptive withdrawal reflex in cat hindlimb was described by Sherrington (1906, 1910) as an activation predominately of flexor muscles and inhibition of extensor muscles and, consequently, was termed the "flexion reflex". The receptive field of the "flexion reflex" included "the skin of the whole limb as far up as the groin in front, the perineum medially and the ischial region behind" (Sherrington 1910). The magnitude of the reflex activation of the flexor muscles was dependent on which nerve that was stimulated. Creed and Sherrington (1926) extended these observations on the "local sign" of the "flexion reflex" by comparing the simultaneous reflex activity in pairs of flexor muscles (three combinations of the muscles tensor fasciae femoris, semitendinosus and tibialis anterior) evoked by stimulation of various hindlimb nerves. Differences in thresholds and latencies of the reflex activation between these flexor muscles were noted. It was concluded, that "The term flexion reflex therefore, just as the term scratch reflex, denotes, strictly speaking, a 9 r o u p of reflexes all more or less alike and all using approximately the same motor apparatus in approximately the same way, yet from one afferent to another differing

68 in detailed distribution of the m o t o r units employed, while yet always conforming to the general type "flexion reflex" (Creed and Sherrington 1926). Lloyd (1943a, b) found, using graded nerve stimulation and recording of the afferent nerve volley, that impulses in group I I - I I I muscle afferents and cutaneous afferents triggered di- or trisynaptic reflex discharges in flexor muscles and inhibition of extensor muscles. These reflex discharges were assumed to be part of the "flexion reflex". Intracellular recordings f r o m flexor and extensor alpha motoneurones by Eccles and Lundberg (1959), demonstrated a large excitatory convergence f r o m group I I - I I I muscle afferent fibres, joint and cutaneous nerves onto single flexor alpha motoneurones and inhibitory effects onto extensor alpha motoneurones (see also Lundberg et al. 1987a). Similar convergences have been demonstrated on numerous spinal interneurones and ascending neurones (Holmqvist et al. 1960; H o n g o et al. 1966; Oscarsson 1973). These afferent convergences were assumed to be characteristic for the "flexion reflex". Consequently, the afferent fibres which m a y activate the "flexion reflex" were collectively termed the "flexion reflex afferents" ( F R A ) (Lundberg 1959; Holmqvist et al. 1960). However, it was emphasized that each of these receptor systems m a y have other specialized reflex connections as well that do not belong to the c o m m o n reflex pathway referred to above (Holmqvist and Lundberg 1961 ; see also Lundberg et al. 1987b). M o r e recently, it has been suggested that F R A serve as a general feedback system regulating active movements and that the effects of the F R A should be differentiated from those of the nociceptive afferent fibres (Lundberg 1979, 1982; Lundberg et al. 1987b). H a g b a r t h (1952), Megirian (1962) and Engberg (1964) using adequate stimulation of cutaneous nociceptors in spinalized or decerebrate cats and Kugelberg et al. (1960), using painful electrical stimulation of h u m a n feet, found that extensor muscles also have an excitatory nocireceptive field on the skin. Hence the term "flexion reflex" is, strictly speaking, a misnomer (see also Grah a m - B r o w n and Sherrington 1912). However, the results obtained by these researchers differed regarding the location of the receptive fields of the flexor and the extensor muscles and, consequently, they drew different conclusions on the organization of these reflexes. H a g b a r t h (1952) found an excitation of extensor muscles f r o m the overlying and adjacent skin and an excitation o f flexor muscles from most of the hindlimb skin with the exception of the area activating the antagonistic extensor muscles. Megirian (1962) obtained similar results to Hagbarth for knee extensors and flexors but got a variable result for ankle extensors and flexors. Engberg (1964) found a nociceptive input to some foot muscles from almost the opposite skin areas as those depicted by Hagbarth (1952). The discrepancies between the results were tentatively explained by Engberg (1964) as dependent on the differences in the preparations used since Holmqvist and Lundberg (1961) had shown that decerebration and/or spinalization can drastically affect the receptive fields of hindlimb nociceptive reflexes. F r o m studies on humans, Kugelberg and collaborators (1960) suggested

yet another organizational principle, that the withdrawal reflexes consist of an extension distal to the noxious stimulus and a flexion proximal to the stimulus. The spatial organization of the cutaneous nociceptive input to hindlimb alpha motoneurones is thus not well known. Therefore, the cutaneous excitatory nocireceptive fields of most hindlimb muscles were m a p p e d in the intact rat.

Methods

Animals used 95 Wistar rats weighing 300-500 g were used. These rats were housed in large cages (bottom 700*500 mm; height 450 ram; 5 rats in each cage), where they could climb on the walls and the roof. The cages were, in addition, equipped with treadmills to permit running exercise. The animals housed in this way appeared to be in excellent physical condition.

Preparation 93 rats were anesthetized with halothane (1.0-2.0 % during surgery) in 65% nitrous oxide and 35% oxygen. Two rats were anesthetized with pentobarbitone (60 mg/kg i.p. followed by supplementary injections of 10 mg/kg i.v. when needed). An infusion of 30-50 gl/min of 5% glucose in Ringer acetate (pH = 7.0) was continuously given through a cannula inserted into the right jugular vein. A cannula was inserted in the trachea to facilitate breathing. The blood pressure (90-130 mmHg) was monitored in the right brachial artery. The rectal temperature was kept at 36-38 ~ C, using a warm table. Following surgery the halothane concentration was reduced to 0.5~.6% in the same gas mixture as before. The animals were unparalyzed and unrestrained allowing a continuous and adequate control of the anesthesia. The skeletal muscles were completely atonic. Weak withdrawal reflexes, but no other motor responses, were evoked by the noxious stimulation of the skin employed. The level of anesthesia was further characterized by an adequate breathing (the end-expiratory CO 2 was 3.5-5.0%) and a normal blood pressure (90-130 mmHg), which remained stable during noxious pinch of the skin. Rats at the same level of anesthesia do not exhibit EEG arousal following noxious stimulation of the skin (cf. LeBars et al. 1979; Schouenborg et al. 1986). The level of anesthesia used during recordings appears to correspond to stage 3, plane 1-2 in humans (Grimby et al. 1966).

Electromyographic recordings The hairy skin of the right hindlimb was carefully shaved to permit localized mechanical stimulation of the skin. A pause of 2-4 h was allowed between surgery and the studies of reflexes, since there was usually an increased reflex threshold immediately following the surgery (see also Clarke 1985). A small opening was made in the skin overlying the investigated muscle. A fine steel needle electrode, which was isolated with the exception of 50 gm at the tip, was inserted into the middle of the muscle belly. A reference electrode was placed in the adjacent skin. The electrodes were soldered to thin moveable wires, by which they were suspended. The muscle was identified by observing the action of the muscle contraction caused by trains (100 Hz, for 200 ms) of weak cathodal stimulation through the recording electrode. Criteria used to identify a muscle were: 1) the threshold current to activate the examined muscle was less than 20 p.A (usually it was below 10 gA), 2) the threshold for activating nearby muscles was at least 10 times higher (usually it was more

69 than 25 times higher) than for the examined muscle. The position of the recording electrode was checked repeatedly throughout the experiment and readjusted when necessary. The activity in a few motor units (usually 1-5 units per muscle) in the skeletal muscle was recorded. Only units which were clearly distinguishable from the background noise were considered to belong to the investigated muscle (see Figs. 1D, 2D and 3D). The recordings were stored on tape and analyzed off-line using a computer. The magnitude of the responses evoked in the motor units studied was determined by counting the number of action potentials during the first 5 s from the stimulus artifact or, in some cases when the latency of the reflex discharges was long, during the first 10 s (stated in Results).

Stimulation of the skin in two different ways In order to map the receptive field of the motor units studied, manual pinch using fine forceps with flattened tips (area about 0.5 mm 2 on each side) was applied to a large number of places on the skin (more than 200 points). Care was taken to avoid pinch of subcutaneous tissue. Although not directly measured when applied on the skin, an estimate of the range of forces applied is 0.1-3 N. These stimuli caused sensations ranging from mild pressure to moderate, well-localized, pain when tested on our hands. The location of the borders of the receptive fields was marked on photographs of the hindlimb. The location of a border was determined with an accuracy of about + 0.5 mm on the paw and • 1 mm on the more proximal parts of the hindlimb. An electrically triggered mechanical pincher, which was firmly attached to a metal frame to avoid movements while pinching, was used for controlled stimulation (Schouenborg and Dickenson 1988). A fold of skin, approximately 4 mm 2, was placed within the grip of the pincher. The area of the pincher exerting pressure on the skin was about 0.7 mm 2 on each side. The intensity of the pincher was changed by exchanging calibrated springs. The forces exerted on the skin were: 0.3 N, 0.6 N, 1.1 N, 1.5 N and 2.0 N. When tested on the dorsum of our own hands and feet for a period of 5 s, pinch at an intensity of 0.3 N evoked a sensation of moderate pressure. Pinch at 0.6 N evoked a faint pain sensation in most, but not all, tests made. Hence, this intensity was close to the threshold of pain. The pain threshold in humans with a comparable mechanical stimulus has been shown to be about 0.8 N/ram 2 (Lindblom 1985). Pinch at 1.1-2.0 N generated weak to moderate pain. Confirmations of the sensations provoked were made by three other researchers in our laboratory. When applied on the skin of the rat, pinch at an intensity of 0.3-0.6 N was innocuous, i.e. there was neither rupture of the skin nor visible tissue reactions such as redness. Following pinch at 1.1 N the skin was transiently reddened, but intact. At intensities of 1.5-2.0 N small ruptures of the skin were sometimes observed. The delay between the onset of the pincher and the actual pinch was less than 10 ms. The duration of the pinch was 5-10 s. Pauses of 5-10 rain were made between each stimulation during quantified pinch in order to reduce possible changes in sensitivity of the reflex pathway studied following a pinch (such as "wind-up", Mendell 1966, see also Results). Up to 26 points on the skin were stimulated with the triggered pincher in a varying order.

Nomenclature The hindlimb muscles were named according to "Anatomy of the Rat" by Green (1955).

Results The entire ipsilateral h i n d l i m b , hip, c a u d a l p e r i n e u m a n d back were systematically searched for responses in each muscle recorded. P i n c h o f the c o n t r a l a t e r a l h i n d l i m b a n d

Table 1. Number of experiments and the thresholds (Thr) for activation are given for the different muscles. The thresholds were inferred to be between two intensities if the higher, but not the lower, of these intensities regularly evoked a discharge Responding muscles

No. of exp. Threshold (N)

M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M.

17 rats 2 rats 2 rats 10 rats 5 rats 5 rats 6 rats 11 rats 2 rats 7 rats 10 rats 5 rats 2 rats 8 rats 4 rats 4 rats 5 rats 5 rats 5 rats 6 rats

interossei of digits 2-5 interosseus of digit 1 lumbrical of digit 5 peroneus longus peroneus brevis peroneus digiti quinti extensor digitorum brevis extensor digitorum longus extensor hallucis longus tibialis anterior gastrocnemius - soleus flexor digitorum longus plantaris biceps posterior semitendinosus iliopsoas gracilis posticus adductor brevis adductor longus pectineus

0.3 < Thr < 0.6 manual tests only manual tests only 0.6

Functional organization of the nociceptive withdrawal reflexes. I. Activation of hindlimb muscles in the rat.

1. The organization of the nociceptive hindlimb withdrawal reflexes was investigated in 93 halothane/nitrous oxide anesthetized rats. Electromyographi...
1MB Sizes 0 Downloads 0 Views