- Email: [email protected]
Pain thresholds and electromyographic features of periarticular muscles in patients with osteoarthritis of the knee
Pain, 10 (1981) 57--66
57
© Elsevier/North-Holland Biomedical Press
PAIN THRESHOLDS AND ELECTROMYOGRAPHIC FEATURES OF PERIARTICULAR MUSCLES IN PATIENTS WITH OSTEOARTHRITIS OF THE KNEE
M A U R O BRUCINI * ROBF_,RTO DURANTI * RENATO GALLETTI * T r r o P A N T A L E O ** and PIER LUIGI ZUCCHI *
* Cattedra di Patologia Spec'~te Medica B, and ** lstituto di Fisiologia Umana, Universitd degli 8tudi, Florence (Italy) (Received 25 June 1980, accepted 30 July 1980)
SUMMARY
Electrical stimulation pain thresholds and EMG activity were studied, using the vastus medialis muscle of healthy control subjects and of patients with osteoarthritis of the knee. Various categories of sensation elicited by progressive increases of the level of electrical stimulation (hlcluding one defined as pain threshold) were defined for control subjects. For patients, muscular pain thresholds differed significantly for the two sides of the body; they were usually lower in the more affecl~d side. Involuntary activity of certain motor units and delayed relaxation following voluntary contraction were consistently observed in patients. The involuntary activity was affected by limb position and by mechanical stimulation of tender areas of the muscle or joint. Differences in pain threshold between the two sides were significantly reduced and the EMG pattern became normal following treatment with injections of local anaesthetic into tender periarticular areas and systemic administration of lysine acetylsalicylate. In standing patients, abnormal EMG activity (which was characteristically sensitive 1,o body load and its variations) was found. Injection of a local anaesthetic into the joint cavity was able to induce a rapid subjective improvement and a consistent reduction of EMG activity.
INTRODUCTION
The tenderness and tension of periarticular muscles in osteoarthritis of the knee is well known: patients consistently report soreness, articular stiffness and a rapi d fatigue when it is necessary to maintain an upright position.
58
Muscular hypotrophy is often found in the more advanced stages of disease. Laboratory methods for inducing muscular pain have been described; yet our knowledge about muscle pain thresholds and their alterations in hyperalgesic states is scanty [13,25]. There have been a number of studies concerned with electromyographic (EMG) patterns in muscle pain syndromes, but they do not allow unequivocal conclusions [21]. For example spontaneous motor unit activity in tender muscles at rest has been reported by some investigators [2,8,24] but not by others [14]. TraveU [23] reported consistent EMG activity from muscular trigger points recorded by needle electrodes and Elliot [ 8] observed bursts of motor unit activity in muscular tender spots caused by deep palpation close to the needle. Abnormal responses of tender muscles to voluntary activity have also been described [18,22]. In the present stu:Jies, muscular pain thresholds and the EMG pattern of tender muscles were studie~ "n order to understand better the above reported pattern of clinical complaints and possibly to suggest mechanisms of their genesis. The experiments were carried out both in pain-free control subjects and in patients with osteoarthritis of the knee. Pain thresholds and EMG activity were ~udied chiefly in the vastus medialis muscle, since it :.Lsfrequently involved in this disease. The present study indicated that muscular pain threshohts of tender muscles associated with joint disease are altered and there are abnormalities in the EMG pattern of the muscles involved both at rest and during weightbearing in the upright position. The effects of local anaesthesia suggest that the peripheral afferent innervation plays a role in causing or maintaining these changes. METHODS
Eighteen patients with osteoarthritis of the knee were separated inr.o 2 groups: 3 men and 5 women (ranging in age from 45 to 76 years) and 4 men and 6 women (ranging in age from 38 to 72 years). In the first group, muscular pain thresholds and EMG activity were studied with the patient at rest in a supine comfortable position, before and after the following treatment lasting 20 days: periarticular injections of 5--8 ml 0.25% bupivacaine in tender muscular areas twice a week and i.v. administration of lysine acetylsalicylate 0.9 g twice a day. In the second group EMG activity only was recorded (with the patients both at rest in a supine position and standing barefoot) before and 30 min after the injection of 4--8 ml 0.25% or 0.50% bupivacaine into the joint cavity. The effects of changes in body weight d~stribution were investigated both by asking the patients to stand on one foot and to bend the body forward and backward. Eight healthy control subjects, 6 men and 2 women ranging in age from 24 to 36 years, were also studied.
59 All subjects were volunteers who gave their informed consent to participate in the experimental procedures. Pain thresholds were determined in the vastus medialis muscles of both sides by means of electrical stimulation. Stimuli were delivered by a constant current unit driven by a Grass $8 stimulator. Stimuli were applied to the muscle through two electrodes (Nichrome wire; O.D. 250 #m, insulated except for 2 mm at the tip). The electrodes were vertically inserted through hypodermic needles into the muscle along its longitadinal axis, about 3 cm above the patella;the distance between the electrodes was about 1 cm. Stimulation consisted of 18 msec trains of 0.5 msec square wave pulses at a frequency of 300 Hz. These stimulation parameters were selected on the basis of preliminary sessions to provide a suitable range of stimulation intensity for various sensations reported by the subjects (see Results); prolonged trah~s of stimuli (2 sec) were ~ s o employed with control subjects. Each train was repeated 5--10 times at fixed intervals of 20 sec. Stimulus current intensity was measured by an oscilloscope monitoring the voltage generated through a small series resistor (Tektronix 565 oscilloscope, Tektronix 3A74 4-trace amplifier). It was increased progressively and the sensations reported by the subjects were recorded. Threshold values (in mA) were obtained by measuring current intensities corresponding to the reported sensations. The current level which resulted in the first report of "cramp-like pain" by the subject was defined as the pain threshold. Each recorded threshold value was the average of the last 3 measurements after a series of 8--~2 preliminary trials to familiarize the subjects with the procedure. "False" trials with the stimulating current held at zero were also inserted to insure ~,hat the subject understood the task. The correct position of the electrode tips was checked by observing movements of the electrodes during voluntary muscular contraction and muscular twitching caused by low intensity (0.5 mA) stimuli. Some patients in whom no absolute certainty was achieved about electrode position were discarded. EMG recordings were made with concentric needle electrodes (HewlettPackard, No. 14051C) inserted into various points of the vastus medialis muscle of control and arthritic subjects; rectus femoris and vastus lateralis muscles were also tested in some. Muscular electrical activity was amplified (Tektronix 2A61 AC differential amplifier), displayed on a double beam oscilloscope (Tektronix 565) and photographed {Grass C4 Camera). EMG activity was also monitored by audiomonitor (Grass AM4B). Muscle action potentials were full-wave rectified and integrated over time (low-pass filter, time constant 100 msec) and displayed on the same oscilloscope after DC amplification {Tektronix 3A74 4-trace amplifier). The intended beginning or end of a voluntary contraction was signalled by the subject through a switch that caused a voltage display on the same oscilloscope. This switch was also used to signal changes in body weight distribution. All the experiments were performed in a quiet moderately warm room (23--25°C):
6O RESULTS
Control muscular pain thresholds Verbal reports of normal subjects exposed to brief trains of increasing current intensities resulted in the following range of sensations: (i) mild sensation of electrical shock and/or muscle twitch; (ii) uncomfortable mu Icle twitch, sometimes with sharply localized pain (pricking pain); (iii) c r a m p - ~ e pain. With prolonged trains somewhat different sensations resulted: (i) tingling; (ii) sustained muscle contraction with sensation of internal press!ire; (iii) uncomfortable sustained contraction of the muscle; (iv) cramp-like pain. These findings are summmdzed in Table I; for prolonged trains only the mean threshold values for the first and the last steps of the sensation scale h~ve been reported, because of the short range of current intensities employed and the wide variability found in the intermediate current range. No significant differences were found between the threshold values of the two limbs in these control subjects. The reliability of pain threshold values was examined in 5 subjects by repeating the tests 10 times after the patient had completed preliminary familiarizing trials during the same session, and by repeating the threshold tests 5--10 times on successive separate days. Under these conditions, the range of pain threshold values in the same subject was always less than 0.5 mA, indicating that in the same subject a reproducible measure of pain threshold could be obtained. During the experimental sessions it was found that by progressively decreasing the current intensity, soon after assessment of the pain threshold,
TABLE I S E N S A T I O N S A N D PAIN T H R E S H O L D V A L U E S F O R E L E C T R I C A L S T I M U L A T I O N O F T H E V A S T U S M E D I A L I S M U S C L E O F C O N T R O I S U B J E C T S (mean -+S.E.) Sensation
Threshold (mA)
Right Brief trains of stimuli (18 insect,
Prolonged trains of stimuli (2 s e c )
1. Mild electrical shock and/or muscle twitch 2. Uncomfortable muscle twitch sometimes with sharp localized pain (pricking pain) 3. Cramp-like pain (muscular pain threshold) 1. Tingling 2. Sustained muscle contraction with sensations of internal pressure 3. Uncomfortable sustained muscle contraction 4. Cramp-like pain
Left
0.28 + 0.03
0.3
+ 0.02
2.5
2.8
+ 0.07
+ 0.06
5.2 -+ 0.38 0.27 -+ 0.05
5.8 +- 0.24 0.28 + 0.03
--
2.1
+ 0.35
2.5
+ 0.28
61
we were able to elicit painful sensations with current intensities well under threshold value. This transitory hyperalgesia lasted about 1--2 min. In contrast, after a stimulation period of 2--5 rain with the same frequency and stimulus duration but with a current intensity capable of causing only muscle contraction (not pain) the observed pain thresholds increased by up to 100% or more (hypoalgesia) and recovered their original value in about an hour.
Control subjects EMG Our results are in agreement with the well established conclusion [ 1 ] that when a healthy voluntary muscle is relaxed, intramuscular electrodes record no ongoing electrical activity; and that, as a rule, vasti muscles are quiescent during relaxed standing. The vastus medialis muscle also remained inactive when the subject leaned forwards; it showed compensatory activity when the subject deliberately leaned backwards, as indicated by progressh,e recruitm e n t of m o t o r units with increasing bending. When the distribution of body weight on the feet was changed, by asking the subject to stand on one foot at a time, the vastus medialis muscle did not show activity. Patient muscular pain thresholds Pain threshold values attained with brief trains of stimuli are reported in Table II. Student's t tests of independent or paired samples were used. Pain thresholds differed more from one side to the other in patients than in control subjects. The difference in absolute value between the two sides {mean -+ S.E.) for controls was 1 + 0.24, for patients it was 3.3 -+ 0.88 (P < 0.05). In 6 patients pain thresholds (mean +- S.E. = 3.86 +- 0.5) were significantly lower (P < 0.02) than mean control values (Table I) in the affected limb (if both limbs were affected, in the side where the patient felt greater functional TABLE II PAIN T H R E S H O L D VALUES (in mA) FOR ELECTRICAL STIMULATION OF THE VASTUS MEDIALIS MUSCLES OF PATIENTS WITH OSTEOARTHRITIS OF THE KNEE Subjects
1 2 3 4 5 6 7 8
Before the treatment
After the treatment
Right
Left
Right
Left
2.5 5 5 5.5 6 5.5 6 15
5 7 ~ 12 3.3 3.4 7.3 7
4.7 5 5 6 5.5 5 7.5 10
5.4 5.3 5.4 7.3 4.8 5.5 4.8 5.5
62
impairment). Two patients showed higher threshold values in the more impaired limb (threshold values respectively 12 and 15). After treatment differences between the two sides (mean + S.E. = 1.38 + 0.52) appeared significantly reduced (P < 0.05) and n o t significantly different from control levels. Patient's EMGs EMG recordings were made from tense and tender muscular areas that showed the well-known "trigger p o i n t " characteristics. Involuntary persistent activity of at least some m o t o r units was found in tender vastus medialis muscles of 14 of the 18 patients (Fig. 1). At thfs time, the subjects were at rest in a supine position and considered their muscles to be completely relaxed. This activity was characterized by a typical regular discharge of some m o t o r uni'~s, firing at 5--16 Hz. Sudden interruptions in an otherwise regular discharge or variations of discharge were also observed. Sometimes the activity dropped to zero for a few seconds or minutes. Complete relaxation and EMG silence could be achieved by appropriate active or passive slight flexion and extrarotation of the limb in all patients. Digital compression of hyperalgesic areas of the knee joint, especially the medial or lateral borders of the patella, resulted in 8 patients in a m o t o r unit disch~ge during an EMG silence period or in an increase of pre-existing activity; this effect lasted up to 30 sec after the end of stimulation (Fig. 1B and C). In 5 of these patients the same effect could be obtained by deep
1 B . . . .
÷
~"',::~:~::~-.'";
; '
I ~'~ -:',:',: ',:
'
~ ;~'.I: I ]
+ C
.:
. ~ . i .I
.~.1.
,.I
l.i,J
,
1.
1.
i.
L
L
i.
1,
I,
l,i
,I
,1
~
I
I.
|,
!
,j
I !
÷
ls Fig. 1. EMG activity recorded from the left vastus medialis muscle of a patient affected by osteoarthritis of the knee, A: involuntary activity at rest in a supine position with delayed relaxation following voluntary contraction (the intended onset and offset of contraction are indicated respectively by the first and the second arrow). B a~d C, in succession: a short compression on the lateral edge of the patella (marked by the arrow) elicited persistent involuntary activity during a previous period of silence. D: following treatment, no involuntary activity was present at rest and rapid relaxation occurred following voluntary contraction (indicated by the arrow).
63 ÷
÷
B i_
|
~s Fig. 9-. IR,MG recorded from the v~stus lateralis muscle o f a patient w i t h osteo~thritis of the knee, A: involuntary activity o f one motor u n i t (at rest in a supine position). B: prolonged m o t o r unit discharge caused by digital compression of a tender area of the same muscle during a silent period. The arrow and the artefact respectively p~npoint the onset and the duration of the mechanical stimulation.
pressure on the muscle itself close to the needle. In two patients the vastus lateralis muscle was also tender and showed spots with "trigger point" characteristics. Spontaneous EMG activity with the same features reported above was found in these areas: digital compression on the tender spots or on other tender areas o f the same muscle evoked or increased EMG activity (Fig. 2). The effects of compression persisted even after skin anaesthesia caused b y subcutaneous injections of 0.50% bupivacaine. During brief voluntary contraction of the vastus medialis muscle, carried out in the supine position, a progressive recruitment of m o t o r units occurred; after intended arrest of contraction some m o t o r units failed to cease firing Lmmediately: E M G activity returned progressively to the precontraction level over a period ranging from 2 up to 30 sec (Fig. 1A). This delayed relaxation, not observed by us in the control subjects, was present in all tested patients. Following treatment, good clinical improvement was observed in all patients; at this time the E M G test showed no involuntary activity or delayed relaxation (Fig. 1D). With the patient standing, the E M G of arthritispatients showed an intense discharge of several motor units in the vastus medialis muscle. Voluntary contraction only slightly enhanced this activity.Delayed recovery to the precontraction discharge level was usually observed. W h e n the body weight was shifted, by asking the patient to stand on one foot, the E M G pattern changed characteri~stically.E M G activity consistently increased in the muscles of the weight-bearing side and decreased when the patient supported himself on the opposite foot. The effect of unloading the tested side was sometimes so intense as to induce an E M G silence (Fig. 3). Deliberate leaning backwards increased vastus medialis activity,whereas leaning forwards decreased it. Intra-articular anaesthesia (on the tested side) led to rapid subjective improvement of limb sensation and loss of involuntary E M G activity or delayed relaxation in the supine position. E M G activity observed in the upright position was also conshtently decreased; in a few instances it was eliminated almost entirely. This persisting activity could stillbe affected by changes in load (Fig. 3)and by voluntary contraction. The effect.,;of intraarticular anaesthesia were usually more intense when larger anlounts of
64 A
@
o
A
1
|
t
IEMG ~'~-.,.,. L _ " ~ . . ~ -
~'~..~
0.5 mV ÷
@
o
A
1
I
;
|
|
2s Fig. 3. EMG and integrated EMG (IEMG) of the left vastus medialis muscle of a standing patient. Changes in b o d y weight distribution are indicated by labelled arrows. A: the patient stands at first on b o t h feet, then on the left foot (loading) (e), on both feet (o), on the opposite foot (unloading) (A). B: the same sequence as in A, 20 min after an injection of 4 ml of 0.25% bupivacaine into the joint cavity.
anaesthetic were used. The intra-articular injection of the same volume of control isotonic saline did not affect the EMG pattern. DISCUSSION
Electrical stimulation was chosen as a method of measuring muscle pain thresholds: it allows quantifiable and reasonably reproducible measurements. Electrical stimulation has been widely used in studies of cutaneous pain [ 13, 25 ]; trains of electrical stimuli, similar to those used in the present research, have been employed in the study of skin pain thresholds with apparently satisfactory results [9,19 ]. As with cutaneous tests [ 9 ], electrical stimulation of muscle with increasing intensities elici~ted various sensations: cramp-like discomfort, defined as the muscular pain threshold for our studies, appeared to be readily recognized. The observed increase in pain threshold after prolonged non-painful stimulation resembles certain effects of electroacupuncture or transcutaneous electrical stimulation: muscle afferents with lower thresholds (large fibres) and their central actions may be involved in this effect [17]. On the other hand, transitory hyperalgesia aftei reaching the pain threshold is not easy to explain although the rapid onset and the brief duration of the effect seem to suggest a nervous mechanism, possibly central, mediated by smaller afferent fibres (groups III and IV). Another possibility is the local release of pain threshold lowering substances (perhaps substance P) in hyperalgesic muscle areas s;_milar to that proposed for cutaneous axon reflex flare [4]. The presence of significantly different threshold values on the two :;ides of patients and their absence after therapy is not surprising, but needs further consideration. Alterations of cutaneous pain thresholds have been noted in patients with reflex sympathetic distrophies; sympathetic efferent activity
65
apparently is implicated [19~20]. A similar autonomic component may be present in hyperalgesic muscle are~. The present results agree with previous findings on EMG activity in tender muscle areas at rest [2,8] and dmd.ng voluntary activity [18,22]. The involunta~l EMG activity and delayed relaxation found in the vastus medialis muscle could result from a central excitatory process, while the effect of changes in limb position seems to suggest that the afferent input from muscle, tendon and joint receptors plays a significant role. A major involvement of joint receptors may easily be hypothesized in patients with chronic degenerative joint disease. The pattern of EMG activity in the upright position and its sensitiveness to changes in load and to intra-articular anaesthesia support the hypothesis that reflexes involving knee joint receptors may contribute to the observed muscular changes. In this connection, studies on animals, recently reviewed by McCloskey [15] are pertinent: inputs from knee joint receptors of the cat [3,5,6,10] or of the monkey [11] appear to be maximum at the extremes of both flexion and extension. Moreover, evidence has been reported for a positive feed-back mechanism involving the knee extensor motoneurones (of the cat), that are excited by knee extension [ 12]. Considering the observed effects of deliberate leaning, postural alterations related to the knee joint disease probably contribute to the EMG changes. The motor unit discharge rate in tender muscle areas does not reach the level observed with cramph~g i 7]: neve~Leless the level of EMG activity was so intense and persistent as to account for the muscular tension, hyperalgesia and early fatigue reported by the patients when standing. Also, an abnormal neuronal input arising from tender muscles or trigger points may combine with the input from the diseased joint to produce referred pain in the periarticular areas by means of various central nervous mechanisms [ 16 ]. These results seem also to indicate that abnormal afferent activity from a tender muscle or other periarticular tender axeas may contribute to the establishment of a vicious circle, by maintaining sustained contraction; this appears to be at least partially susceptible to interruption by injections of local anaesthetic into tender periarticular areas or into the joint cavity. ACKNOWLEDGEMENTS
We are grateful to Professor Loring F. Chapman for valuable suggestions and criticism of the manuscript. REFERENCES 1 Basmajian, J.V., Muscles Alive, Williams and Wilkins, Baltimore, Md., 1978. 2 Bayer, H., Die rheumatische Muskelharte -- eine Eigenreflextetanus, Klin. Wschr., 27
(1949) 122--126. 3 Burgess, P.R. and Clark, F.J., Characteristics of knee joint receptors in the cat, J. Physiol. (Lond.), 203 (1969) 317--335.
66 4 Chapman, L.F.j Mechanisms of the flare reaction in human skin, J. invest. Dermatol., 69 (1977) 8 8 - ~ 7 . 5 Clark, F.J., Information signaled by sensory fibers in medial articular nerve, J. Neurophysiol., 38 (1975) 1464--1472. 6 Clark, F.J. and Burgess, P.R., Slowly adapting receptors in the cat knee joint: can they signal joint angle?, J. Neurophysiol., 38 (1975) 1448--1463. 7 Denny-Brown, D. and Foley, J.M., Myokymia and the benign fasciculations of muscular cramps, Trans. Ass. Amer. Phys., 61 (1948) 88--96. 8 Elliot, F.A., Tender muscles in sciatica; electromyographic studies, La~cet, i (1944) 47--49. 9 Francini, F., Zoppi, M., Maresca, M. and Procacci, P., Skin potential and EMG changes induced by cutaneous electrical stimulation. I. Normal man in arousing and non-arousing environment, Appl. Neurophysiol., 42 (1979) 113--124. 10 Grigg, P., Mechanical factors influencing response of joint afferent neurons from cat knee, J. Neurophysiol., 38 (1975) 1473--1484. 11 Grigg, P. and Greenspav, B.J., Response of primary joint afferent neurons to mechanical stimulation of knee joint, J. Neurophysiol., 40 (1977) 1--8. 12 Grigg, P., Harrigan, E.P. and Fogarty, K.E., Segmental reflexes mediated by joint afferent neurons in cat knee, J. Neurophysiol., 41 (1978) 9--14. 13 Hilgard, E.R., Pain perception in man. In: R. Held, H.W. Leibowitz and H.-L. Teuber (Eds.), Perception, Handbook of Sensory Physiology, Vol. 8. Springer, Berlin, 1978, pp. 849--874. 14 Kraft, G.H., Johnso:~, E.W. and LaBan, M.M., The fibrositis syndrome, Arch. phys. Med., 49 (1968) 155--162. 15 McCloskey, D.I., Kinesthetic sensibility, Physiol. Rev,, 58 (1978) 763--820. 16 Melzack, R., Stillwell, D.M. and Fox, E.J., Trigger points and acupuncture points for pain: correlations and implications, Pain, 3 (1977) 3--23. 17 Melzack, R. and Wall, P.D., Pain mechanism: a new theory, Science, 150 (1965) 971-979. 18 Miehlke, K., Schulze, G. und Eger, W., Klinische und experimentelle Untersuchungen zum Fibrositissyndrom, Z. Rheumaforsch., 19 (1960) 310--330. 19 Procacci, P., Francini, F., Maresca, M. and Zoppi, M., Skin potential and EMG changes induced by cutaneous electrical stimulation. II. Subjects with reflex sympathetic distrophies, Appl. Neurophysiol., 42 (1979) 125--134. 20 Procacci, P., Francini, F., Zoppi, M., Maresca, M. and Giovannini, L., Role of sympathetic system in reflex dystrophies. In: J.J. Bonica and D. Albe-Fessard (Eds.), Advances in Pain Research and Therapy, Vol. 1, Raven Press, New York, 1976, pp. 953--957. 21 Simons, D.G., Muscle pain syndromes, part II, Amer. J. phys. Med., 55 (1976) 15-42. 22 Taverner, D., Muscle spasm as a cause of somatic pain, Ann. rheum. Dis., 13 (1954) 331--335. 23 Travell, J., Symposium on mechanism and management of pain syndromes, Proc. Rudolf Virchow Med. Soc., 16 (1957) 128--136. 24 Travell, J., Berry, C. and Bigelow, N., Effects of referred somatic pain on structures in the reference zone, Fed. Proc., 3 (1944) 49. 25 Wolff, B.B., The role of laboratory pain induction methods in the systematic study of human pain, Acupunct. electro-ther. Res. int. J., 2 (1977) 271--305.
57
© Elsevier/North-Holland Biomedical Press
PAIN THRESHOLDS AND ELECTROMYOGRAPHIC FEATURES OF PERIARTICULAR MUSCLES IN PATIENTS WITH OSTEOARTHRITIS OF THE KNEE
M A U R O BRUCINI * ROBF_,RTO DURANTI * RENATO GALLETTI * T r r o P A N T A L E O ** and PIER LUIGI ZUCCHI *
* Cattedra di Patologia Spec'~te Medica B, and ** lstituto di Fisiologia Umana, Universitd degli 8tudi, Florence (Italy) (Received 25 June 1980, accepted 30 July 1980)
SUMMARY
Electrical stimulation pain thresholds and EMG activity were studied, using the vastus medialis muscle of healthy control subjects and of patients with osteoarthritis of the knee. Various categories of sensation elicited by progressive increases of the level of electrical stimulation (hlcluding one defined as pain threshold) were defined for control subjects. For patients, muscular pain thresholds differed significantly for the two sides of the body; they were usually lower in the more affecl~d side. Involuntary activity of certain motor units and delayed relaxation following voluntary contraction were consistently observed in patients. The involuntary activity was affected by limb position and by mechanical stimulation of tender areas of the muscle or joint. Differences in pain threshold between the two sides were significantly reduced and the EMG pattern became normal following treatment with injections of local anaesthetic into tender periarticular areas and systemic administration of lysine acetylsalicylate. In standing patients, abnormal EMG activity (which was characteristically sensitive 1,o body load and its variations) was found. Injection of a local anaesthetic into the joint cavity was able to induce a rapid subjective improvement and a consistent reduction of EMG activity.
INTRODUCTION
The tenderness and tension of periarticular muscles in osteoarthritis of the knee is well known: patients consistently report soreness, articular stiffness and a rapi d fatigue when it is necessary to maintain an upright position.
58
Muscular hypotrophy is often found in the more advanced stages of disease. Laboratory methods for inducing muscular pain have been described; yet our knowledge about muscle pain thresholds and their alterations in hyperalgesic states is scanty [13,25]. There have been a number of studies concerned with electromyographic (EMG) patterns in muscle pain syndromes, but they do not allow unequivocal conclusions [21]. For example spontaneous motor unit activity in tender muscles at rest has been reported by some investigators [2,8,24] but not by others [14]. TraveU [23] reported consistent EMG activity from muscular trigger points recorded by needle electrodes and Elliot [ 8] observed bursts of motor unit activity in muscular tender spots caused by deep palpation close to the needle. Abnormal responses of tender muscles to voluntary activity have also been described [18,22]. In the present stu:Jies, muscular pain thresholds and the EMG pattern of tender muscles were studie~ "n order to understand better the above reported pattern of clinical complaints and possibly to suggest mechanisms of their genesis. The experiments were carried out both in pain-free control subjects and in patients with osteoarthritis of the knee. Pain thresholds and EMG activity were ~udied chiefly in the vastus medialis muscle, since it :.Lsfrequently involved in this disease. The present study indicated that muscular pain threshohts of tender muscles associated with joint disease are altered and there are abnormalities in the EMG pattern of the muscles involved both at rest and during weightbearing in the upright position. The effects of local anaesthesia suggest that the peripheral afferent innervation plays a role in causing or maintaining these changes. METHODS
Eighteen patients with osteoarthritis of the knee were separated inr.o 2 groups: 3 men and 5 women (ranging in age from 45 to 76 years) and 4 men and 6 women (ranging in age from 38 to 72 years). In the first group, muscular pain thresholds and EMG activity were studied with the patient at rest in a supine comfortable position, before and after the following treatment lasting 20 days: periarticular injections of 5--8 ml 0.25% bupivacaine in tender muscular areas twice a week and i.v. administration of lysine acetylsalicylate 0.9 g twice a day. In the second group EMG activity only was recorded (with the patients both at rest in a supine position and standing barefoot) before and 30 min after the injection of 4--8 ml 0.25% or 0.50% bupivacaine into the joint cavity. The effects of changes in body weight d~stribution were investigated both by asking the patients to stand on one foot and to bend the body forward and backward. Eight healthy control subjects, 6 men and 2 women ranging in age from 24 to 36 years, were also studied.
59 All subjects were volunteers who gave their informed consent to participate in the experimental procedures. Pain thresholds were determined in the vastus medialis muscles of both sides by means of electrical stimulation. Stimuli were delivered by a constant current unit driven by a Grass $8 stimulator. Stimuli were applied to the muscle through two electrodes (Nichrome wire; O.D. 250 #m, insulated except for 2 mm at the tip). The electrodes were vertically inserted through hypodermic needles into the muscle along its longitadinal axis, about 3 cm above the patella;the distance between the electrodes was about 1 cm. Stimulation consisted of 18 msec trains of 0.5 msec square wave pulses at a frequency of 300 Hz. These stimulation parameters were selected on the basis of preliminary sessions to provide a suitable range of stimulation intensity for various sensations reported by the subjects (see Results); prolonged trah~s of stimuli (2 sec) were ~ s o employed with control subjects. Each train was repeated 5--10 times at fixed intervals of 20 sec. Stimulus current intensity was measured by an oscilloscope monitoring the voltage generated through a small series resistor (Tektronix 565 oscilloscope, Tektronix 3A74 4-trace amplifier). It was increased progressively and the sensations reported by the subjects were recorded. Threshold values (in mA) were obtained by measuring current intensities corresponding to the reported sensations. The current level which resulted in the first report of "cramp-like pain" by the subject was defined as the pain threshold. Each recorded threshold value was the average of the last 3 measurements after a series of 8--~2 preliminary trials to familiarize the subjects with the procedure. "False" trials with the stimulating current held at zero were also inserted to insure ~,hat the subject understood the task. The correct position of the electrode tips was checked by observing movements of the electrodes during voluntary muscular contraction and muscular twitching caused by low intensity (0.5 mA) stimuli. Some patients in whom no absolute certainty was achieved about electrode position were discarded. EMG recordings were made with concentric needle electrodes (HewlettPackard, No. 14051C) inserted into various points of the vastus medialis muscle of control and arthritic subjects; rectus femoris and vastus lateralis muscles were also tested in some. Muscular electrical activity was amplified (Tektronix 2A61 AC differential amplifier), displayed on a double beam oscilloscope (Tektronix 565) and photographed {Grass C4 Camera). EMG activity was also monitored by audiomonitor (Grass AM4B). Muscle action potentials were full-wave rectified and integrated over time (low-pass filter, time constant 100 msec) and displayed on the same oscilloscope after DC amplification {Tektronix 3A74 4-trace amplifier). The intended beginning or end of a voluntary contraction was signalled by the subject through a switch that caused a voltage display on the same oscilloscope. This switch was also used to signal changes in body weight distribution. All the experiments were performed in a quiet moderately warm room (23--25°C):
6O RESULTS
Control muscular pain thresholds Verbal reports of normal subjects exposed to brief trains of increasing current intensities resulted in the following range of sensations: (i) mild sensation of electrical shock and/or muscle twitch; (ii) uncomfortable mu Icle twitch, sometimes with sharply localized pain (pricking pain); (iii) c r a m p - ~ e pain. With prolonged trains somewhat different sensations resulted: (i) tingling; (ii) sustained muscle contraction with sensation of internal press!ire; (iii) uncomfortable sustained contraction of the muscle; (iv) cramp-like pain. These findings are summmdzed in Table I; for prolonged trains only the mean threshold values for the first and the last steps of the sensation scale h~ve been reported, because of the short range of current intensities employed and the wide variability found in the intermediate current range. No significant differences were found between the threshold values of the two limbs in these control subjects. The reliability of pain threshold values was examined in 5 subjects by repeating the tests 10 times after the patient had completed preliminary familiarizing trials during the same session, and by repeating the threshold tests 5--10 times on successive separate days. Under these conditions, the range of pain threshold values in the same subject was always less than 0.5 mA, indicating that in the same subject a reproducible measure of pain threshold could be obtained. During the experimental sessions it was found that by progressively decreasing the current intensity, soon after assessment of the pain threshold,
TABLE I S E N S A T I O N S A N D PAIN T H R E S H O L D V A L U E S F O R E L E C T R I C A L S T I M U L A T I O N O F T H E V A S T U S M E D I A L I S M U S C L E O F C O N T R O I S U B J E C T S (mean -+S.E.) Sensation
Threshold (mA)
Right Brief trains of stimuli (18 insect,
Prolonged trains of stimuli (2 s e c )
1. Mild electrical shock and/or muscle twitch 2. Uncomfortable muscle twitch sometimes with sharp localized pain (pricking pain) 3. Cramp-like pain (muscular pain threshold) 1. Tingling 2. Sustained muscle contraction with sensations of internal pressure 3. Uncomfortable sustained muscle contraction 4. Cramp-like pain
Left
0.28 + 0.03
0.3
+ 0.02
2.5
2.8
+ 0.07
+ 0.06
5.2 -+ 0.38 0.27 -+ 0.05
5.8 +- 0.24 0.28 + 0.03
--
2.1
+ 0.35
2.5
+ 0.28
61
we were able to elicit painful sensations with current intensities well under threshold value. This transitory hyperalgesia lasted about 1--2 min. In contrast, after a stimulation period of 2--5 rain with the same frequency and stimulus duration but with a current intensity capable of causing only muscle contraction (not pain) the observed pain thresholds increased by up to 100% or more (hypoalgesia) and recovered their original value in about an hour.
Control subjects EMG Our results are in agreement with the well established conclusion [ 1 ] that when a healthy voluntary muscle is relaxed, intramuscular electrodes record no ongoing electrical activity; and that, as a rule, vasti muscles are quiescent during relaxed standing. The vastus medialis muscle also remained inactive when the subject leaned forwards; it showed compensatory activity when the subject deliberately leaned backwards, as indicated by progressh,e recruitm e n t of m o t o r units with increasing bending. When the distribution of body weight on the feet was changed, by asking the subject to stand on one foot at a time, the vastus medialis muscle did not show activity. Patient muscular pain thresholds Pain threshold values attained with brief trains of stimuli are reported in Table II. Student's t tests of independent or paired samples were used. Pain thresholds differed more from one side to the other in patients than in control subjects. The difference in absolute value between the two sides {mean -+ S.E.) for controls was 1 + 0.24, for patients it was 3.3 -+ 0.88 (P < 0.05). In 6 patients pain thresholds (mean +- S.E. = 3.86 +- 0.5) were significantly lower (P < 0.02) than mean control values (Table I) in the affected limb (if both limbs were affected, in the side where the patient felt greater functional TABLE II PAIN T H R E S H O L D VALUES (in mA) FOR ELECTRICAL STIMULATION OF THE VASTUS MEDIALIS MUSCLES OF PATIENTS WITH OSTEOARTHRITIS OF THE KNEE Subjects
1 2 3 4 5 6 7 8
Before the treatment
After the treatment
Right
Left
Right
Left
2.5 5 5 5.5 6 5.5 6 15
5 7 ~ 12 3.3 3.4 7.3 7
4.7 5 5 6 5.5 5 7.5 10
5.4 5.3 5.4 7.3 4.8 5.5 4.8 5.5
62
impairment). Two patients showed higher threshold values in the more impaired limb (threshold values respectively 12 and 15). After treatment differences between the two sides (mean + S.E. = 1.38 + 0.52) appeared significantly reduced (P < 0.05) and n o t significantly different from control levels. Patient's EMGs EMG recordings were made from tense and tender muscular areas that showed the well-known "trigger p o i n t " characteristics. Involuntary persistent activity of at least some m o t o r units was found in tender vastus medialis muscles of 14 of the 18 patients (Fig. 1). At thfs time, the subjects were at rest in a supine position and considered their muscles to be completely relaxed. This activity was characterized by a typical regular discharge of some m o t o r uni'~s, firing at 5--16 Hz. Sudden interruptions in an otherwise regular discharge or variations of discharge were also observed. Sometimes the activity dropped to zero for a few seconds or minutes. Complete relaxation and EMG silence could be achieved by appropriate active or passive slight flexion and extrarotation of the limb in all patients. Digital compression of hyperalgesic areas of the knee joint, especially the medial or lateral borders of the patella, resulted in 8 patients in a m o t o r unit disch~ge during an EMG silence period or in an increase of pre-existing activity; this effect lasted up to 30 sec after the end of stimulation (Fig. 1B and C). In 5 of these patients the same effect could be obtained by deep
1 B . . . .
÷
~"',::~:~::~-.'";
; '
I ~'~ -:',:',: ',:
'
~ ;~'.I: I ]
+ C
.:
. ~ . i .I
.~.1.
,.I
l.i,J
,
1.
1.
i.
L
L
i.
1,
I,
l,i
,I
,1
~
I
I.
|,
!
,j
I !
÷
ls Fig. 1. EMG activity recorded from the left vastus medialis muscle of a patient affected by osteoarthritis of the knee, A: involuntary activity at rest in a supine position with delayed relaxation following voluntary contraction (the intended onset and offset of contraction are indicated respectively by the first and the second arrow). B a~d C, in succession: a short compression on the lateral edge of the patella (marked by the arrow) elicited persistent involuntary activity during a previous period of silence. D: following treatment, no involuntary activity was present at rest and rapid relaxation occurred following voluntary contraction (indicated by the arrow).
63 ÷
÷
B i_
|
~s Fig. 9-. IR,MG recorded from the v~stus lateralis muscle o f a patient w i t h osteo~thritis of the knee, A: involuntary activity o f one motor u n i t (at rest in a supine position). B: prolonged m o t o r unit discharge caused by digital compression of a tender area of the same muscle during a silent period. The arrow and the artefact respectively p~npoint the onset and the duration of the mechanical stimulation.
pressure on the muscle itself close to the needle. In two patients the vastus lateralis muscle was also tender and showed spots with "trigger point" characteristics. Spontaneous EMG activity with the same features reported above was found in these areas: digital compression on the tender spots or on other tender areas o f the same muscle evoked or increased EMG activity (Fig. 2). The effects of compression persisted even after skin anaesthesia caused b y subcutaneous injections of 0.50% bupivacaine. During brief voluntary contraction of the vastus medialis muscle, carried out in the supine position, a progressive recruitment of m o t o r units occurred; after intended arrest of contraction some m o t o r units failed to cease firing Lmmediately: E M G activity returned progressively to the precontraction level over a period ranging from 2 up to 30 sec (Fig. 1A). This delayed relaxation, not observed by us in the control subjects, was present in all tested patients. Following treatment, good clinical improvement was observed in all patients; at this time the E M G test showed no involuntary activity or delayed relaxation (Fig. 1D). With the patient standing, the E M G of arthritispatients showed an intense discharge of several motor units in the vastus medialis muscle. Voluntary contraction only slightly enhanced this activity.Delayed recovery to the precontraction discharge level was usually observed. W h e n the body weight was shifted, by asking the patient to stand on one foot, the E M G pattern changed characteri~stically.E M G activity consistently increased in the muscles of the weight-bearing side and decreased when the patient supported himself on the opposite foot. The effect of unloading the tested side was sometimes so intense as to induce an E M G silence (Fig. 3). Deliberate leaning backwards increased vastus medialis activity,whereas leaning forwards decreased it. Intra-articular anaesthesia (on the tested side) led to rapid subjective improvement of limb sensation and loss of involuntary E M G activity or delayed relaxation in the supine position. E M G activity observed in the upright position was also conshtently decreased; in a few instances it was eliminated almost entirely. This persisting activity could stillbe affected by changes in load (Fig. 3)and by voluntary contraction. The effect.,;of intraarticular anaesthesia were usually more intense when larger anlounts of
64 A
@
o
A
1
|
t
IEMG ~'~-.,.,. L _ " ~ . . ~ -
~'~..~
0.5 mV ÷
@
o
A
1
I
;
|
|
2s Fig. 3. EMG and integrated EMG (IEMG) of the left vastus medialis muscle of a standing patient. Changes in b o d y weight distribution are indicated by labelled arrows. A: the patient stands at first on b o t h feet, then on the left foot (loading) (e), on both feet (o), on the opposite foot (unloading) (A). B: the same sequence as in A, 20 min after an injection of 4 ml of 0.25% bupivacaine into the joint cavity.
anaesthetic were used. The intra-articular injection of the same volume of control isotonic saline did not affect the EMG pattern. DISCUSSION
Electrical stimulation was chosen as a method of measuring muscle pain thresholds: it allows quantifiable and reasonably reproducible measurements. Electrical stimulation has been widely used in studies of cutaneous pain [ 13, 25 ]; trains of electrical stimuli, similar to those used in the present research, have been employed in the study of skin pain thresholds with apparently satisfactory results [9,19 ]. As with cutaneous tests [ 9 ], electrical stimulation of muscle with increasing intensities elici~ted various sensations: cramp-like discomfort, defined as the muscular pain threshold for our studies, appeared to be readily recognized. The observed increase in pain threshold after prolonged non-painful stimulation resembles certain effects of electroacupuncture or transcutaneous electrical stimulation: muscle afferents with lower thresholds (large fibres) and their central actions may be involved in this effect [17]. On the other hand, transitory hyperalgesia aftei reaching the pain threshold is not easy to explain although the rapid onset and the brief duration of the effect seem to suggest a nervous mechanism, possibly central, mediated by smaller afferent fibres (groups III and IV). Another possibility is the local release of pain threshold lowering substances (perhaps substance P) in hyperalgesic muscle areas s;_milar to that proposed for cutaneous axon reflex flare [4]. The presence of significantly different threshold values on the two :;ides of patients and their absence after therapy is not surprising, but needs further consideration. Alterations of cutaneous pain thresholds have been noted in patients with reflex sympathetic distrophies; sympathetic efferent activity
65
apparently is implicated [19~20]. A similar autonomic component may be present in hyperalgesic muscle are~. The present results agree with previous findings on EMG activity in tender muscle areas at rest [2,8] and dmd.ng voluntary activity [18,22]. The involunta~l EMG activity and delayed relaxation found in the vastus medialis muscle could result from a central excitatory process, while the effect of changes in limb position seems to suggest that the afferent input from muscle, tendon and joint receptors plays a significant role. A major involvement of joint receptors may easily be hypothesized in patients with chronic degenerative joint disease. The pattern of EMG activity in the upright position and its sensitiveness to changes in load and to intra-articular anaesthesia support the hypothesis that reflexes involving knee joint receptors may contribute to the observed muscular changes. In this connection, studies on animals, recently reviewed by McCloskey [15] are pertinent: inputs from knee joint receptors of the cat [3,5,6,10] or of the monkey [11] appear to be maximum at the extremes of both flexion and extension. Moreover, evidence has been reported for a positive feed-back mechanism involving the knee extensor motoneurones (of the cat), that are excited by knee extension [ 12]. Considering the observed effects of deliberate leaning, postural alterations related to the knee joint disease probably contribute to the EMG changes. The motor unit discharge rate in tender muscle areas does not reach the level observed with cramph~g i 7]: neve~Leless the level of EMG activity was so intense and persistent as to account for the muscular tension, hyperalgesia and early fatigue reported by the patients when standing. Also, an abnormal neuronal input arising from tender muscles or trigger points may combine with the input from the diseased joint to produce referred pain in the periarticular areas by means of various central nervous mechanisms [ 16 ]. These results seem also to indicate that abnormal afferent activity from a tender muscle or other periarticular tender axeas may contribute to the establishment of a vicious circle, by maintaining sustained contraction; this appears to be at least partially susceptible to interruption by injections of local anaesthetic into tender periarticular areas or into the joint cavity. ACKNOWLEDGEMENTS
We are grateful to Professor Loring F. Chapman for valuable suggestions and criticism of the manuscript. REFERENCES 1 Basmajian, J.V., Muscles Alive, Williams and Wilkins, Baltimore, Md., 1978. 2 Bayer, H., Die rheumatische Muskelharte -- eine Eigenreflextetanus, Klin. Wschr., 27
(1949) 122--126. 3 Burgess, P.R. and Clark, F.J., Characteristics of knee joint receptors in the cat, J. Physiol. (Lond.), 203 (1969) 317--335.
66 4 Chapman, L.F.j Mechanisms of the flare reaction in human skin, J. invest. Dermatol., 69 (1977) 8 8 - ~ 7 . 5 Clark, F.J., Information signaled by sensory fibers in medial articular nerve, J. Neurophysiol., 38 (1975) 1464--1472. 6 Clark, F.J. and Burgess, P.R., Slowly adapting receptors in the cat knee joint: can they signal joint angle?, J. Neurophysiol., 38 (1975) 1448--1463. 7 Denny-Brown, D. and Foley, J.M., Myokymia and the benign fasciculations of muscular cramps, Trans. Ass. Amer. Phys., 61 (1948) 88--96. 8 Elliot, F.A., Tender muscles in sciatica; electromyographic studies, La~cet, i (1944) 47--49. 9 Francini, F., Zoppi, M., Maresca, M. and Procacci, P., Skin potential and EMG changes induced by cutaneous electrical stimulation. I. Normal man in arousing and non-arousing environment, Appl. Neurophysiol., 42 (1979) 113--124. 10 Grigg, P., Mechanical factors influencing response of joint afferent neurons from cat knee, J. Neurophysiol., 38 (1975) 1473--1484. 11 Grigg, P. and Greenspav, B.J., Response of primary joint afferent neurons to mechanical stimulation of knee joint, J. Neurophysiol., 40 (1977) 1--8. 12 Grigg, P., Harrigan, E.P. and Fogarty, K.E., Segmental reflexes mediated by joint afferent neurons in cat knee, J. Neurophysiol., 41 (1978) 9--14. 13 Hilgard, E.R., Pain perception in man. In: R. Held, H.W. Leibowitz and H.-L. Teuber (Eds.), Perception, Handbook of Sensory Physiology, Vol. 8. Springer, Berlin, 1978, pp. 849--874. 14 Kraft, G.H., Johnso:~, E.W. and LaBan, M.M., The fibrositis syndrome, Arch. phys. Med., 49 (1968) 155--162. 15 McCloskey, D.I., Kinesthetic sensibility, Physiol. Rev,, 58 (1978) 763--820. 16 Melzack, R., Stillwell, D.M. and Fox, E.J., Trigger points and acupuncture points for pain: correlations and implications, Pain, 3 (1977) 3--23. 17 Melzack, R. and Wall, P.D., Pain mechanism: a new theory, Science, 150 (1965) 971-979. 18 Miehlke, K., Schulze, G. und Eger, W., Klinische und experimentelle Untersuchungen zum Fibrositissyndrom, Z. Rheumaforsch., 19 (1960) 310--330. 19 Procacci, P., Francini, F., Maresca, M. and Zoppi, M., Skin potential and EMG changes induced by cutaneous electrical stimulation. II. Subjects with reflex sympathetic distrophies, Appl. Neurophysiol., 42 (1979) 125--134. 20 Procacci, P., Francini, F., Zoppi, M., Maresca, M. and Giovannini, L., Role of sympathetic system in reflex dystrophies. In: J.J. Bonica and D. Albe-Fessard (Eds.), Advances in Pain Research and Therapy, Vol. 1, Raven Press, New York, 1976, pp. 953--957. 21 Simons, D.G., Muscle pain syndromes, part II, Amer. J. phys. Med., 55 (1976) 15-42. 22 Taverner, D., Muscle spasm as a cause of somatic pain, Ann. rheum. Dis., 13 (1954) 331--335. 23 Travell, J., Symposium on mechanism and management of pain syndromes, Proc. Rudolf Virchow Med. Soc., 16 (1957) 128--136. 24 Travell, J., Berry, C. and Bigelow, N., Effects of referred somatic pain on structures in the reference zone, Fed. Proc., 3 (1944) 49. 25 Wolff, B.B., The role of laboratory pain induction methods in the systematic study of human pain, Acupunct. electro-ther. Res. int. J., 2 (1977) 271--305.