THE EFFECT OF ACUTE JOINT DISTENSION ON MECHANORECEPTOR DISCHARGE IN THE KNEE OF THE CAT

Transcription

1 Quarterly Journal of Experimental Physiology (1987) 72, Printed in Great Britain THE EFFECT OF ACUTE JOINT DISTENSION ON MECHANORECEPTOR DISCHARGE IN THE KNEE OF THE CAT Institute of Physiology, W. R. FERRELL University of Glasgow, Glasgow G12 8QQ (RECEIVED FOR PUBLICATION 8 FEBRUARY 1987) SUMMARY Recordings were obtained from afferents travelling in the posterior articular nerve of the cat knee joint. These arose from tonically active slowly adapting mechanoreceptors which were located in the joint capsule and whose conduction velocities were in the group II range. Intraarticular fluid injection resulted in enhancement of receptor discharge. Both the static index (adapted discharge 2 s after ramp displacement) and the dynamic index (difference between peak discharge during joint displacement and that occurring 05 s later) increased after injection. However, whereas the dynamic index increased by three- to fivefold the static index never exceeded a twofold increase. Incremental rise of intra-articular volume resulted in progressive increase in both parameters at all angular velocities. Thus both position and velocity sensitivity are enhanced by the acute introduction of fluid into the synovial cavity. INTRODUCTION Although much is known regarding the behaviour of slowly adapting articular mechanoreceptors as a function of limb movement and position, little is known about the response of these receptors to alteration of intra-articular volume. Multi-unit recordings have demonstrated that with increasing intra-articular volume, joint afferent discharge increases (Andrew & Dodt, 1953; Wood & Ferrell, 1984) although the relationship is not linear (Wood & Ferrell, 1985). In other experiments it has been observed that the pattern ofjoint afferent discharge during movement is markedly influenced by injection of fluid into the synovial cavity (Ferrell, Nade & Newbold 1986). However, as yet there has been no systematic investigation of how individual afferents are influenced by increasing intraarticular volume. The object of the present study was to examine this aspect. METHODS The experiments were performed on five adult cats (2-4 kg) which were deeply anaesthetized by intraperitoneal injection of pentobarbitone (45 mg/kg), further doses being administered intravenously as required to maintain a deep level of anaesthesia as judged by the absence of the flexion withdrawal response to noxious stimuli applied to the forepaw. The left carotid artery was cannulated to provide a continuous measure of arterial blood pressure. The knee was fixed at an intermediate position ( deg) during identification of the afferent fibres, and thereafter was rotated through 5 deg at a velocity of 1-40 deg/s by a servo motor. The posterior articular nerve of the cat knee joint was sectioned proximally and dissected back towards the joint capsule to provide a length of about 2-3 cm of nerve, which was subdivided until filaments contained functionally 'single-unit' activity. Conduction velocity was measured by the technique depicted in Fig. 1. The single unit discharge was displayed on one channel of a digitizing oscilloscope (Tektronix 5223), the spikes being used to trigger the time-base amplifier. The nerve discharge recorded from more distally placed electrodes was

2 494 W. R. FERRELL A2 I0 ANp Posterior articular /I-L-I 0J D- Joint ( dcapsule 0-5 Ms A/II 25MV w v femoris Patella Fig. 1. Technique utilized to measure conduction velocity. The lower part of the Figure shows a lateral view of the knee joint with the stippled area representing the fluid-filled synovial cavity. The posterior articular nerve innervating the dorsal aspect of the joint capsule is placed over bipolar silver-wire electrodes connected to an amplifier and signal conditioner (A 1) whose output is displayed on the lower channel of an oscilloscope (trace A 1). The proximal end of the posterior articular nerve is subdivided until a single unit is isolated and the filament containing this unit is then placed over a second set of electrodes connected to a second amplifier system (A 2) whose output is displayed on the upper channel of the oscilloscope (trace A 2). Although the spike shapes differ due to the different recording conditions, measurement of the latency (L) from the upstroke of the two action potentials and the distance between the recording sites (D) allows the conduction velocity of the unit to be calculated from the equation: conduction velocity D then displayed on the second channel of the oscilloscope. Utilizing the pre-trigger viewing facility of the oscilloscope, the spike recorded from the more distally located electrodes (which precedes the trigger spike) is time-locked in advance of the trigger spike. Measurement of the distance between the recording electrodes and the interspike interval allowed the conduction velocity to be calculated. As the conduction distances and interspike intervals were both short, the conduction velocities obtained by this technique were not considered to be highly accurate, but were adequate for the purposes of the experiment as they served to indicate whether the afferents isolated had conduction velocities in the group II range. Intra-articular injections were administered by means of a 25 G needle inserted into the synovial cavity through the anteromedial aspect of the knee joint. The needle was advanced until it entered the posterolateral region of the joint. Volumes of 01-1 ml of dextran (Dextraven 110, Fisons) were administered by this route. As under normal conditions virtually no synovial fluid can be aspirated from the cat knee, it was assumed that the synovial cavity contained no fluid (intra-articular volume = 0 ml) prior to fluid injection. RESULTS Recordings were obtained from fifteen slowly adapting receptors, all of which had receptive fields clearly localized to the posterior aspect of the joint capsule. These units had conduction velocities ranging from 30 to 55 m/s, responded vigorously to gentle probing of the capsule but not to probing of surrounding structures, and all exhibited tonic discharges at intermediate angles ( deg). These characteristics are consistent with low-threshold slowly adapting mechanoreceptors. No receptors arising from the adjacent

3 VELOCITY SENSITIVITY OF JOINT RECEPTORS [ MX' : ~~~~~~~~~~~~~40 deg/s 50 I 25 EOL 75-50,- ZX, S;; Z5 ~ 20 A 25-5t 751 The joint was displaced from 135 to 130 deg andback10 75i 50._,_, _ X :, - ~~~~~~~~~~~~~20 25/ so~~ ~ ~ ~~~~~i 75 T Dynamic - v50 v..p 4 25 p iv Static ( & index l Fig. 1980).~~ 2. Response of a slowly adapting joint receptor to a 5 deg ramp displacement at different angular velocities. The joint was displaced from 135 to 130 deg and back again. popliteus muscle (Burgess & Clark, 1969; McIntyre, Proske & Tracey, 1978; Ferrell, 1980) were encountered in this study. Succinylcholine was not used to identify the units isolated as it has been shown that the enhancement of afferent discharge induced by this agent occurs not only in muscle spindle afferents, but also in true joint afferents (Ferrell, 1980). In accordance with previous investigations (Boyd & Roberts, 1953; Skoglund, 1956), it was found that these receptors exhibited velocity sensitivity, an example of which is shown in Fig. 2. With increasing velocity, the discharge 'overshoot' at the end of the ramp displacement becomes progressively greater. A measure of the velocity sensitivity of afferents is the dynamic index (Matthews, 1972) which is the difference between the peak 1s

4 496 W. R. FERRELL A 50 impulses/s[' 130 deg 135 deg B 50 impulses/s. 130 deg 135 degec 2 s Fig. 3. Response of a slowly adapting receptor to ramp displacement under control conditions (A), and after injection of 0-5 ml dextran into the synovial cavity (b). The angular velocity (15 deg/s) was the same in both traces. l;; 30 -\ x 20 '0 zlol~~i 10 I I I '0' Intra-articular volume (ml) Fig. 4. The effect of increasing intra-articular volume on the static index of a slowly adapting receptor. 0 ml intraarticular volume represents control conditions, whilst 'O' represents the situation prevailing after fluid aspiration. discharge at the end of the ramp and that occurring 05 s later. The greater the angular velocity, the greater is the dynamic index. However, the receptor discharge 2 s after ramp displacement (static index) is the same irrespective of the velocity of displacement. As can be seen, receptor adaptation is virtually complete at 2 s. The static index can be considered as providing a measure of the position sensitivity of the receptor (Matthews, 1972). Boyd & Roberts (1953) and Ferrell (1980) have shown that slowly adapting joint receptors exhibit position sensitivity as many of these receptors show a systematic change in adapted discharge rate with joint angle. Enhancement of velocity sensitivity can also be achieved at a fixed velocity by increasing the intra-articular volume. When 0 5 ml of dextran is injected into the synovial cavity, the dynamic index shows a marked increase compared to control (Fig. 3). The static index is

5 VELOCITY SENSITIVITY OF JOINT RECEPTORS : >10A * '0' Intra-articular volume (ml) Fig. 5. Alteration of the dynamic index of a receptor at different angular velocities with increasing intra-articular volume. A = 1 deg/s; A = 2 5 deg/s; 0 = 5 deg/s; 0 = 10 deg/s. also increased, but less so than the dynamic index. This is reflected in the data for all the units (n = 15) where the mean dynamic index (±S.E.M.) at a velocity of 15 deg/s prior to fluid injection was impulses/s which increased to impulses/s after 0 5 ml of dextran was instilled into the synovial space. The static index under control conditions was impulses/s and rose to impulses/s after injection. Over-all, the dynamic index showed a three- to fivefold rise when fluid was present in the joint, whilst the maximal observed increase in the static index was twofold. Incremental increase in intra-articular volume resulted in progressive augmentation in both dynamic and static indices. As shown in Fig. 4, the static index progressively rises as the intra-articular volume increases. The same is true for the dynamic index, which shows the greatest rise at the highest velocities (Fig. 5). The data in Figs 4 and 5 are from the same unit. 0 ml intra-articular volume denotes the control situation prior to fluid injection, whilst '0' indicates the situation prevailing after as much of the injected fluid as possible (045 ml) had been aspirated. It is noticeable that both the dynamic and static indices after fluid aspiration declined below control values. Also observed was that successive cycles of displacement (I5 s intervals) at a given velocity resulted in a gradual decline in the static and dynamic indices. As the runs at a given intra-articular volume were always performed starting with the lowest velocity, with the greatest velocity being tested last, the dynamic indices in Fig. 5 are most accurate at the slowest velocity but tend to be underestimated (by - 20%) at the higher velocities. The values for the static indices plotted in Fig. 4 were measured after the first run at each velocity and are therefore not underestimated. The mean static and dynamic indices described earlier were all measured immediately after dextran injection. DISCUSSION The present results show that an acute increase in intra-articular volume results in enhancement of slowly adapting articular mechanoreceptor discharge. These single-unit studies demonstrate that both the velocity and position sensitivities of these receptors are increased, but not to the same extent by similar intra-articular volumes. The underlying

6 498 W. R. FERRELL mechanism for this enhancement is probably mechanical in origin. Increasing intraarticular volume leads to increasing stretch of the joint capsule with consequent increased receptor excitation. However, the disproportionate increase in dynamic index compared to the static index suggests that there may be separate mechanisms giving rise to the velocity and position sensitivities of these receptors, perhaps related to non-linearities in the mechanical properties of the joint capsule. It has been shown that Ruffini endings in the joint capsule give rise to slowly adapting discharges (Boyd, 1954). The 'spray' terminals of these endings are attached to collagen fibres embedded in an extracellular matrix (Halata, 1977). Mechanically, these fibres have been likened to a spring arrangement attached to a dashpot (representing the matrix), the latter giving rise to viscous 'creep' (Wood, 1985). When the capsule is stretched, tension is developed in the 'springs' initially, but with maintained stretch the dashpot arrangement results in 'creep' of the capsule. With a ramp displacement, the initial velocity-dependent overshoot of these endings may be a property of the stretch of collagen fibres whilst the gradual fall in discharge during the adaptation phase may be due to viscous creep of the extracellular matrix. This model assumes linearity i.e. that all parallel collagen fibres at a given spray terminal are tensioned by stretch, and with increasing stretch, greater tension is developed in all these fibres. An alternative nonlinear arrangement has been described where increasing stretch results in tensioning of progressively more collagen fibres (Viidik, 1973), which would result in increasing stiffness of the 'springs' with increasing stretch of the capsule. It is less likely that the properties of the extracellular matrix would change with stretch and thus viscous 'creep' of the capsule would be relatively less affected. Such divergence in the properties of the joint capsule could underly the observed disproportionate changes in the dynamic and static indices with fluid infusion. The gradual decline in the dynamic and static indices observed with repetitive movement could have resulted from plastic deformation of the capsule or from dispersion of the dextran to other areas of the synovial cavity. It has been observed that the discharge of slowly adapting joint receptors gradually declines (concurrently with intra-articular pressure) after injection of non-absorbable fluid into the synovial cavity (Wood & Ferrell, 1984), and this may be partly attributed to fluid shift as the cat knee joint is not hydraulically continuous under physiological conditions (Wood & Ferrell, 1985). The anteromedial and posterolateral regions of the synovial cavity are linked by small channels (Wood, 1985) which remain closed, effectively isolating these two regions, unless intraarticular pressure rises above its normal subatmospheric level. In the present experiments, injection of fluid into the posterolateral region of the knee elevated intra-articular pressure in that region, thus forcing fluid into the anterolateral region. An interesting observation from these experiments was that after fluid aspiration, both the dynamic and static indices fell below control values, suggesting that some degree of plastic deformation of the capsule occurred. There is evidence to suggest that viscous creep of the cat knee joint capsule occurs when distended by experimental effusion (Wood, 1985), and this can result in gradual decline of receptor discharge despite maintained intraarticular volume (Wood & Ferrell, 1984). Thus, although an acute intra-articular fluid infusion results in enhanced joint afferent discharge (Ferrell et al. 1986), chronic joint effusions could perhaps result in diminished articular afferent discharge. Either of these would significantly disrupt the normal pattern of joint-mediated reflex activity which may have a protective role (Baxendale & Ferrell, 1981). Loss of this could result in joint injury and perhaps further effusion. The enhanced dynamic index when fluid is present in the joint could perhaps explain the

7 VELOCITY SENSITIVITY OF JOINT RECEPTORS 499 observation in human experiments that intra-articular injection of dextran resulted in enhanced proprioceptive performance (Ferrell, Gandevia & McCloskey, 1987). In these experiments it was observed that the ability of a subject to correctly nominate the direction of passive displacement of the terminal phalanx of the third finger at slow angular velocities (2-5 deg/s) was superior to control detections immediately after a small volume of dextran was injected into the distal inter-phalangeal joint of that finger. More recently, it has been observed that injection of saline into the knee joint cavity produces systematic distortion of position matching (W. R. Ferrell & R. H. Baxendale, unpublished observations), suggesting that the altered position sensitivity of knee joint afferents resulting from fluid injection is responsible for the distorted perception of knee position. The author is indebted to Dr L. Wood for critical review of the manuscript. Support from the Medical Research Funds of Glasgow University is gratefully acknowledged. REFERENCES ANDREW, B. L. & DODT, E. (1953). The deployment of sensory nerve endings at the knee joint of the cat. Acta physiologica scandinavica 28, BAXENDALE, R. H. & FERRELL, W. R. (1981). Modulation of transmission in flexion reflex pathways by knee joint afferents in the cat. Journal of Physiology 315, BoYD, I. A. (1954). The histological structure of the receptors in the knee-joint of the cat correlated with their physiological response. Journal of Physiology 124, BOYD, I. A. & ROBERTS, T. D. M. (1953). Proprioceptive discharges from stretch receptors in the knee joint of the cat. Journal of Physiology 122, BURGESS, P. R. & CLARK, F. J. (1969). Characteristics of knee joint receptors in the cat. Journal of Physiology 203, FERRELL, W. R. (1980). The adequacy of stretch receptors in the cat knee for signalling joint angle throughout a full range of movement. Journal of Physiology 299, FERRELL, W. R., GANDEVIA, S. C. & MCCLOSKEY, D. I. (1987). The role of joint receptors in human kinaesthesia when intramuscular receptors cannot contribute. Journal of Physiology 386, FERRELL, W. R., NADE, S. & NEWBOLD, P. J. (1986). The interrelation of neural discharge, intraarticular pressure, and joint angle in the knee of the dog. Journal of Physiology 373, HALATA, Z. (1977). The ultrastructure of the sensory nerve endings in the articular capsule of the knee joint of the domestic cat (Ruffini corpuscles and Pacinian corpuscles). Journal of Anatomy 124, MATTHEWS, P. B. C. (1972). Mammalian Muscle Receptors and Their Central Actions. London: Edward Arnold. MCINTYRE, A. K., PROSKE, U. & TRACEY, D. (1978). Afferents from muscle receptors in the posterior nerve of the cat's knee joint. Experimental Brain Research 33, SKOGLUND, S. (1956). Anatomical and physiological studies of the knee joint innervation in the cat. Acta physiologica scandinavica 36, suppl. 124, VIIDIK, A. (1973). Functional properties of collagenous tissues. International Review of Connective Tissue Research 6, WOOD, L. (1985). Pressure-volume relationships in the knee joint of the cat and their effect on the discharge of articular mechanoreceptors. Ph.D. Thesis, Glasgow University. WOOD, L. & FERRELL, W. R. (1984). Response of slowly adapting articular mechanoreceptors in the cat knee joint to alterations in intra-articular volume. Annals of the Rheumatic Diseases 43, WOOD, L. & FERRELL, W. R. (1985). Fluid comparmentation and articular mechanoreceptor discharge in the cat knee joint. Quarterly Journal of Experimental Physiology 70, EPH 72