Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Eur J Appl Physiol (2001) 84: 213±220 Ó Springer-Verlag 2001 ORIGINAL ARTICLE Andrew B. Leger á Theodore E. Milner Motor impairment in the human hand following eccentric exercise Accepted: 19 September 2000 Abstract Motor impairment was induced by having subjects perform two sets of 50 maximal contractions, using the ®rst dorsal interosseus (FDI) muscle to abduct the index ®nger, while the muscle was being stretched. Tests were conducted prior to the exercise (pre-exercise) and 24 h following the exercise (post-exercise). There were declines of 19% in maximal abduction torque and 15% in maximal ¯exion torque at the metacarpaphalangeal joint, during isometric contraction post-exercise compared to pre-exercise. The ability to stabilize the metacarpophalangeal joint about the abduction/adduction axis was reduced by 14% post-exercise, and the variability in tracking an isometric torque target increased by 30%. There was a decrement of 7%±10% in the median frequency of the power density spectrum of FDI electromyogram (EMG) throughout a 60 s maintained abduction at 50% maximal voluntary contraction. The mean recti®ed EMG, on the other hand, increased by 100%±175% for torque levels below 40% of maximal voluntary contraction, post-exercise. The results were consistent with preferential injury of type II muscle ®bres in FDI. Although non-exercised synergist muscles appeared to be inhibited during maximal voluntary ¯exion, there was evidence that they compensated for injured FDI muscle ®bres during maintained contraction at sub-maximal ¯exion torque. Key words Muscle injury á Movement á Stability á Finger A. B. Leger (&) School of Rehabilitation Therapy, Queen's University, Kingston, Ontario K7L 3N6, Canada e-mail: [email protected] Fax: +1-613-5336776 T. E. Milner School of Kinesiology, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada Introduction Lengthening of active muscle, often referred to as eccentric contraction, occurs commonly in normal activities, for example, when slowly lowering an object held in the hand. Active muscle can develop much larger forces during lengthening, than during isometric or shortening contraction (Katz 1939). While this is advantageous for controlling the motion of objects which are being lowered under the in¯uence of gravity, it creates greater risk of injury to muscle or tendon than activities which involve only isometric or concentric contraction (Newham et al. 1983; Clarkson et al. 1986; Berry et al. 1990; Faulkner et al. 1993). A number of studies have shown that indicators of injury, induced by eccentric loading of muscle, occur more frequently in large, fast-twitch (type II) muscle ®bres than in the smaller, slow-twitch (type I) ®bres (FrideÂn et al. 1983; Lieber and FrideÂn 1988). While the selectivity of injury to type II muscle ®bres has received considerable attention, investigation of functional impairment, following eccentric exercise, has been limited. It is clear that eccentric exercise aects maximal force. Following a period of eccentric exercise, maximal isometric force has been found to decrease by as much as 50% (Clarkson et al. 1992) before recovering gradually over a period of days to weeks (Newham et al. 1987; Ebbeling and Clarkson 1989; Howell et al. 1993). However, most activities of daily living require sub-maximal muscle forces. Studies examining the eects of eccentric exercise on sub-maximal muscle forces have focused almost entirely on static tasks (Komi and Viitasalo 1977; Newham et al. 1983; Berry et al. 1990; Kroon and Naeije 1991). Our ®rst objective was to obtain a clear picture of motor impairment, following eccentric exercise, by examining dynamic tasks. While changes to the pattern of activation of synergist muscles have been reported following eccentric exercise (Hasson et al. 1993), previous protocols only 214 examined synergist muscles which had all been eccentrically exercised. No study has been designed to exercise a single muscle and then investigate activities in which both exercised and non-exercised muscles participate synergistically. Such situations may occur in activities such as weight training. For example, the anterior deltoid muscle is a prime mover during the bench press, as well as being activated eccentrically each time the weight is lowered. The posterior deltoid muscle, on the other hand, is relatively inactive during this manoeuvre. However, during other movements, such as shoulder abduction, the two muscles act as synergists. Thus, the two muscles may be used synergistically after only one of them has been eccentrically exercised. Our second objective was to provide a more comprehensive description of motor impairment by investigating the eects of eccentric exercise of one muscle on the function of a non-exercised synergist muscle. To achieve the ®rst objective, we examined the eects of eccentric exercise on the ability to perform activities such as ballistic movement, postural stabilization and force tracking. Subjects performed a period of strenuous eccentric exercise, during which the ®rst dorsal interosseus (FDI) muscle was stretched while attempting to abduct the metacarpophalangeal (MP) joint of the index ®nger. Because FDI contributes to both abduction and ¯exion of the MP joint, we were able to quantify motor impairment for the two actions. This allowed us to achieve our second objective by examining tasks involving ®nger ¯exion. In this way, we were able to determine whether eccentric exercise of FDI, in abduction, aected synergist ¯exor muscles, which did not contribute to abduction. Methods Fig. 1 Diagram of apparatus used for metacarpophalangeal joint abduction of the index ®nger Fig. 2 Diagram of apparatus used for metacarpophalangeal joint ¯exion of the index ®nger. EMG (FDS) Electromyogram (¯exor digitorum super®cialis muscle) Subjects Ten healthy male subjects, with no history of neuromuscular disease or injury involving the index ®nger, participated in this study (age range 22±31 years). Nine of the subjects were right-handed, the other left-handed. Each gave written informed consent to participate prior to the experiment. None had previously participated in any studies involving eccentric exercise. The subjects were asked not to participate in any weight-training activities, speci®cally for the upper extremities, for the duration of the study. The experiment was approved by the University Research Ethics Review Committee at Simon Fraser University. All experiments were conducted in compliance with Canadian Research Council guidelines for experimentation using human subjects. General design The subjects were tested on two separate occasions, prior to eccentric exercise (pre-exercise), and 24 h later (post-exercise). To ensure reproducible placement of electromyogram (EMG) electrodes on both days, indelible ink was used to mark the skin. To ensure accurate positioning in the test apparatus, the subject's forearm and hand were supported and strapped in a sti elastoplast splint, moulded to the shape of the hand (Figs. 1 and 2). All tests were performed on the left hand. Apparatus A torque motor (PMI U16M4) was used to generate loads, which were computer controlled. The maximal torque that could be produced by the motor was 5 N á m. Position and velocity were measured by a potentiometer and tachometer, respectively. The torque was measured by a linear strain gauge mounted on a cylinder, coupling the motor shaft to a lever arm. The resolution of the torque sensor was 0.004 N á m. Recording Surface EMG was recorded from the FDI, ¯exor digitorum super®cialis (FDS) and extensor digitorum communis (EDC) muscles. The FDS is an extrinsic ¯exor and EDC an extrinsic extensor of the MP joint. The electrode was initially positioned over the centre of the muscle in alignment with the muscle ®bres, approximately midway between origin and insertion. It was then moved incrementally in dierent directions to obtain the largest signal while the subject performed brisk test movements of the index ®nger. The ®nal placement varied from subject to subject, depending on where the largest signal was obtained. Test movements comprised isolated MP abduction, ¯exion and extension. 215 Active bipolar surface electrodes were used (Liberty Mutual MYO 111) consisting of stainless steel contacts, 4 mm in diameter, at a ®xed spacing of 13 mm. The active electrode circuit had a passband of 45±550 Hz, gain of 2,400 and common mode rejection greater than 90 dB. The EMG, position, velocity and torque were simultaneously recorded, digitized at 2 kHz and stored on disk for later analysis. Exercise protocol The forearm and wrist were immobilized in a pronated position by the splint. The tip of the index ®nger was clamped and secured to the lever arm of the motor and the ®nger was splinted to prevent MP ¯exion. The ®nger adduction-abduction axis was aligned with the axis of rotation of the motor (Fig. 1). The subjects performed two sets of ®fty maximal eccentric contractions, using the FDI muscle to abduct the index ®nger. During each repetition, the subjects performed a maximal isometric contraction, then resisted maximally as a 20° displacement was imposed by the torque motor to stretch the muscle over a period of 3 s. There was a 3 s break between successive repetitions and 5 min rest between the two sets. The subjects were provided with visual feedback of their torque levels and encouraged verbally to produce maximal eort. Test procedures Each subject was seated comfortably in a chair with the left forearm and hand splinted and supported in a comfortable position, as described above. To avoid fatigue during the test session, the subjects relaxed for at least 30 s after each trial and for up to 5 min between dierent tests. The total duration of the session was 60±75 min. For tests performed during MP ¯exion, the forearm was orientated midway between pronation and supination with the wrist, hand and three remaining ®ngers splinted to allow only movement of the index ®nger in ¯exion/extension (Fig. 2). The hand was positioned such that the axis of rotation of the MP joint was directly over the shaft of the motor. A clamp at the end of the lever arm was used to hold the distal phalanx of the ®nger. The ®nger was taped to prevent any movement of the proximal and distal interphalangeal joints. The tests were performed in the sequence listed below for all subjects. MP abduction 1. Ballistic movement. A target window, which represented the neutral angle of the index ®nger, was displayed on the computer screen. The subject was instructed to move a cursor representing the position of the ®nger into the target window. Once in the target window for 1 s, the cursor changed colour and the subject abducted the ®nger as rapidly as possible, contacting a mechanical stop placed at the end of the range of movement. The subject was encouraged to accelerate throughout the entire movement. The subject was provided with feedback of his maximal velocity following each trial. Five trials were collected. 2. Postural stability. To determine the limit of postural stability of the MP joint, the subject was required to move the index ®nger into a 1° target window, located at the neutral angle, while the torque motor produced a destabilizing load. The load increased in proportion to the distance from the centre of the target window and was always directed away from the target. The limit of stability was determined by systematically increasing the proportionality constant of the load, i.e. the gain. A trial was considered successful if the ®nger could be positioned in the target window within 20 s of the cue to start and held there for a least 1 s. If successful, the gain was incremented by 0.01 N á m á °)1 for the next attempt. 3. Maximal voluntary contraction (MVC). The motor was locked at the neutral angle of the index ®nger. The subject warmed up in preparation for maximal eort by performing several submaximal isometric contractions, building from about 50% to 80% of maximal eort. Two or three warm-up contractions were generally performed, although the number varied according to the preference of each subject. Three trials at maximal eort, each lasting 3 s, were then performed and stored for later analysis. 4. Torque tracking. While the motor was locked at the neutral angle of the index ®nger, the subject contracted the FDI muscle isometrically to produce MP abduction torque. The target was initially 0 and increased at a constant rate over a period of 5 s to 65% of the torque recorded during pre-exercise MVC. The subject was instructed to match the increasing target torque as closely as possible. Five trials were collected. 5. Maintained contraction. While the motor was locked at the neutral angle of the index ®nger, the subject maintained a constant isometric MP abduction torque for 60 s, equal to 50% of the value recorded during pre-exercise MVC. The subject's torque and the target torque were displayed on the oscilloscope. Only one trial was performed. MP ¯exion 6. MVC. The same protocol as test 3 with the subject ¯exing the index ®nger. 7. Maintained contraction. The same protocol as test 5 with the subject ¯exing the index ®nger. 8. Ballistic movement. The same protocol as test 1 with the subject ¯exing the index ®nger. Analysis Ballistic movement Trials for each subject were analysed individually to obtain peak velocity and time of peak velocity. Movement onset was de®ned as the time when velocity exceeded 5° á s)1. The values for each parameter were averaged for the ®ve trials to produce a mean value for each subject. Postural stability The limit of stability was de®ned as the maximal gain for which the subject was able to successfully complete the task. The mean recti®ed EMG (MEMG) was determined for each muscle over a 250 ms interval while the subject was stable in the target window. Torque tracking The ®ve trials for each subject were individually analysed to determine the MEMG during each 1 s interval of the 5 s tracking task. The variance about the best-®t straight line from beginning to end of the torque tracking was also determined (expressed in Newton metres squared). The mean parameters for the ®ve trials were then calculated. Maintained contraction During the 60 s contraction, the median frequency (MDF) of the EMG power density spectrum for each muscle was computed over 5 s intervals, using a computer program written by one of the authors. A 5 s interval was divided into nine segments of 1,024 points each, which overlapped by half their length. The data was windowed using a Hanning window prior to computing the 216 fast Fourier transform. The power density spectrum, obtained by averaging the nine resulting spectra, reduced the variance of the spectrum estimation by a factor of about 7.5 compared to using a single segment. The slope of MDF plotted against time was then determined. The MEMG was computed for the ®rst and last 5 s intervals. The eects of eccentric exercise were evaluated by comparing parameter values pre- and post-exercise for each task using paired Student's t-tests. Results There was no signi®cant dierence in the maximal velocity of ballistic MP abduction (P > 0.6) or latency from movement onset to peak velocity (P > 0.5) postexercise [mean 487 (SD 98)° á s)1; mean 76 (SD 8) ms] from pre-exercise [mean 513 (SD 123)° á s)1; mean 74 (SD 8) ms]. The maximal velocity of ballistic ®nger ¯exion showed a small, but statistically signi®cant decrement (P < 0.05) post-exercise [mean 992 (SD 176)° á s)1] compared to pre-exercise [mean 1,040 (SD 142)° á s)1], but there was no signi®cant dierence in latency to peak (P > 0.3) post-exercise [mean 74 (SD 7) ms] compared to pre-exercise [mean 71 (SD 6) ms]. Postural stability decreased following eccentric exercise. The mean limit of stability, which was 0.118 (SD 0.017) N á m á °)1 pre-exercise, was reduced to 0.101 (SD 0.019) N á m á °)1 post-exercise (P < 0.001). The MEMG during the period of stabilization was not signi®cantly dierent for the FDI (P > 0.7), FDS (P > 0.8) or EDC (P > 0.5) muscles, pre-exercise [FDI mean 25.0 (SD 15.7) lV; FDS mean 26.8 (SD 20.3) lV; EDC mean 26.8 (SD 20.3) lV], compared to post-exercise [FDI mean 27.2 (SD 18.2) lV; FDS mean 28.8 (SD 19.8) lV; EDC mean 34.0 (SD 30.4) lV]. Maximal torque during MVC in abduction was reduced from a mean of 2.15 (SD 0.35) N á m pre-exercise to 1.74 (SD 0.28) N á m post-exercise, a decline of 19% (P < 0.001). This was accompanied by a signi®cant decrease in FDI MEMG from a mean of 75.2 (SD 35.7) lV pre-exercise to 61.7 (SD 22.5) lV postexercise (P < 0.05). Maximal torque during MVC in ¯exion also decreased, from a mean of 4.84 (SD 1.35) N á m to 4.11 (SD 1.24) N á m, a decline of 15% (P < 0.001). However, there was no signi®cant change in MEMG of FDI (P > 0.7), FDS (P > 0.9) or EDC (P > 0.8) pre-exercise [FDI mean 57.9 (SD 39.8) lV; FDS mean 61.3 (SD 50.3) lV; EDC mean 14.3 (SD 34.1) lV] compared to post-exercise [FDI mean 53.1 (SD 29.6) lV; FDS mean 58.7 (SD 52.6) lV; EDC mean 11.4 (SD 19.2) lV]. The reduction in maximal voluntary torque after 24 h is unlikely to have been due to prolonged eects of fatigue, i.e. diminished force generating capacity of healthy muscle ®bres. This was tested by means of repeated maximal isometric abduction of the ®nger, which reduced the maximal voluntary torque by approximately 30% immediately after exercise, but showed no prolonged eect. Maximal volun- tary isometric torque was normal when measured 24 h later (unpublished observations). The torque tracking performance pre- and postexercise is illustrated in Fig. 3. Variability in tracking a target torque was greater post-exercise [mean 27.4 (SD 16.5) N á m2] than pre-exercise [mean 19.2 (SD 12.6) N á m2; P < 0.05]. This was accompanied by signi®cantly higher FDI MEMG for the ®rst 4 s of the tracking (P < 0.001, P < 0.001, P < 0.01 and P < 0.05, respectively), as illustrated in Fig. 4. There was a strong linear correlation of MDF related to time for FDI EMG during maintained abduction at 50% MVC, as has been previously reported in muscle fatigue tests (Kranz et al. 1985). The mean correlation coecient for all subjects was 0.96 (SD 0.03) pre-exercise and 0.95 (SD 0.05) post-exercise. In all cases, the slope was signi®cantly less than 0, indicating a gradual decrease in MDF throughout the contraction (Fig. 5). However, there was no signi®cant dierence between slopes, pre- and post-exercise. This suggests that the FDI muscle did not fatigue at a faster rate post-exercise than pre-exercise. Post-exercise, MDF was lower Fig. 3 Metacarpophalangeal joint abduction torque and electromyogram of ®rst dorsal interosseus muscle (FDI EMG) while tracking the same ramp increase in target torque, pre-exercise (A, B) compared to post-exercise (C, D). The increase in torque is less smooth postexercise, particularly at higher torque levels. Traces are single trials for the same subject 217 Fig. 4 Mean recti®ed electromyogram (MEMG), calculated for 1 s intervals while tracking a ramp increase in metacarpophalangeal joint abduction torque. MEMG is expressed as a percentage of maximal voluntary contraction (%MVC) electromyogram (EMG) on the day of recording. Values are means and standard deviations for ten subjects. *Statistically signi®cant dierences pre- and post-exercise. FDI First dorsal interosseus muscle throughout the contraction. Initial MEMG was signi®cantly greater than pre-exercise, but there was no signi®cant dierence in ®nal MEMG. Parameter values are listed in Table 1. In the case of maintained ¯exion at 50% MVC, there was again a strong linear correlation between MDF and time for FDI EMG. The mean correlation coecient for all subjects was 0.95 (SD 0.04) pre-exercise and 0.96 (SD 0.03) post-exercise. In all cases, the slope was signi®cantly less than 0. However, there was no signi®cant dierence between pre-exercise and Fig. 5 Median frequency of power density spectrum of ®rst dorsal interosseus muscle electromyogram (FDI EMG) calculated at 5 s intervals during maintained abduction of metacarpophalangeal joint at 50% of pre-exercise maximal voluntary contraction. Values are means and standard deviations for nine subjects. Statistical tests were conducted only on initial and ®nal values. *Statistically signi®cant dierences pre- and post-exercise post-exercise values of initial MDF, ®nal MDF or slope. Nor were there any signi®cant dierences in initial and ®nal MEMG post-exercise compared to pre-exercise (Table 1). Since FDS is a synergist of FDI in ¯exion, we compared FDS EMG pre- and postexercise. The linear correlation between MDF and time was weaker for FDS than FDI. Pre-exercise, the slope was signi®cantly dierent from 0 for only ®ve out of the ten subjects, although post-exercise it was signi®cantly less than 0 for seven subjects. The mean correlation coecient for these subjects was 0.84 (SD 0.10) pre-exercise and 0.92 (SD 0.06) postexercise. There were no signi®cant dierences between pre- and post-exercise values of mean initial MDF, ®nal MDF, slope or initial MEMG for FDS. However, mean ®nal MEMG was greater post-exercise. We also compared EDC MEMG pre- and post-exercise, but found no signi®cant dierence in either initial or ®nal mean values. Discussion At 24 h following a period of strenuous eccentric exercise, involving index ®nger abduction, both maximal voluntary abduction torque and maximal voluntary ¯exion torque at the MP joint were signi®cantly lower than pre-exercise. There were also impairments in postural stability and the ability to regulate muscle force precisely. As a whole, the impairments were consistent with preferential injury of type II muscle ®bres in the FDI and suggest that functional impairment was not limited to the exercised muscle. One sign of impaired control was a de®cit in the ability to stabilize the MP joint against an externally imposed mechanical instability. Mechanical stability has usually been achieved by cocontraction of antagonistic muscles to increase joint stiness (Hogan 1984; DeSerres and Milner 1991). The ®nding that maximal joint stability was reduced post-exercise suggests that the decrement in maximal joint torque was also accompanied by a decrement in maximal joint stiness. The ability to activate the exercised muscle did not appear to be compromised since FDI EMG was comparable, pre- and post-exercise. The results of a parallel study, in which we found that short range stiness decreased 24 h following eccentric exercise (Leger and Milner 2000), suggest that the de®cit in stiness was due to preferential injury of type II muscle ®bres. Greater MEMG at the onset of torque tracking postexercise, indicated increased activity of low threshold type I motor units. The elevated level of activity of type I motor units was probably maintained throughout the tracking task, accounting for the increased MEMG. This is supported by the overall shift to lower MDF during maintained isometric contraction at 50% MVC. This shift in MDF could not have been due to a change in the recording con®guration or electrode location 218 Table 1 Median frequency (MDF) and mean recti®ed electromyogram (MEMG) during maintained contraction at 50% maximal voluntary contraction. FDI First dorsal interosseus, FDS ¯exum digitorum super®cialis, EDC extensor digitorum communis muscles Task Abduction Abduction Abduction Abduction Abduction Flexion Flexion Flexion Flexion Flexion Flexion Flexion Flexion Flexion Flexion Flexion Flexion a Muscle FDI FDI FDI FDI FDI FDI FDS FDI FDS FDI FDS FDI FDS EDC FDI FDS EDC Parameter Initial MDF Final MDF MDF Slope Initial MEMG Final MEMG Initial MDF Initial MDF Final MDF Final MDF MDF Slope MDF Slope Initial MEMG Initial MEMG Initial MEMG Final MEMG Final MEMG Final MEMG Units Hz Hz Hz á s)1 lV lV Hz Hz Hz Hz Hz á s)1 Hz á s)1 lV lV lV lV lV lV Pre-exercise Signi®cancea Post-exercise Mean SD Mean SD 126 84 )0.83 43.2 35.8 132 180 98 168 )0.65 )0.25 29.3 15.9 4.3 35.3 25.2 5.5 22 18 0.36 19.1 18.8 30 27 29 24 0.39 0.14 21.9 17.1 2.5 23.4 15.9 3.3 113 75 )0.71 56.4 29.7 127 176 89 160 )0.73 )0.31 32.4 15.6 4.7 35.8 35.8 8.8 28 13 0.34 27.6 9.8 22 30 32 40 0.37 0.48 17.6 7.6 3.7 26.7 21.8 6.8 P P P P P P P P P P P P P P P P P < < > < > > > > > > > > > > > < > 0.05 0.05 0.4 0.05 0.3 0.6 0.7 0.5 0.5 0.6 0.7 0.7 0.95 0.7 0.95 0.05 0.1 Signi®cance levels are for Student's t-tests of paired comparisons since these parameters were the same pre- and post-exercise. Nor could it have been due to broadening of the motor unit action potentials since we found that the compound action potential (M-wave) did not increase in duration (Leger et al. 1996). Therefore, it was most likely indicative of greater activity of type I motor units since type I muscle ®bres have lower conduction velocities than type II muscle ®bres. The lower conduction velocities have been shown to result in broader action potentials, reducing the MDF of the EMG power density spectrum (De Luca 1984). Felici et al. (1997), investigating MDF of the biceps muscle during maintained contraction at 50% and 80% MVC, also found a decrease in MDF for several days following eccentric exercise. A reduction in the MDF of the surface EMG power density spectrum normally occurs during a maintained fatiguing contraction, which is probably also due to broader action potentials, resulting from decreased muscle ®bre conduction velocity (De Luca 1984). It has been reported that mean frequency drops at a faster rate during maintained contraction, following eccentric exercise, due to increased fatigability of the muscle (Kroon and Naeije 1991). However, we found that MDF of FDI EMG declined at the same rate, during a 60 s maintained contraction at 50% MVC, pre- and post-exercise, indicating that fatigability was the same. This again suggests that fatigue resistant type I motor units make a greater contribution to the total muscle force postexercise, which is more in agreement with the ®ndings of Balnave and Thompson (1993) for the quadriceps muscles. In the post-exercise tracking test, torque was more variable than pre-exercise. This impairment was most evident at the higher torque levels where type II motor units would have been recruited. Although several factors could have been responsible, including decreased perception of muscle force or tremor (Saxton et al. 1995), injury to type II muscle ®bres is likely to have contributed signi®cantly. Orderly recruitment of new motor units, according to the size principle, tends to produce increments in force, which are a ®xed percentage of the force at the time of recruitment. Recruitment of an injured motor unit would produce a smaller increment in force than normal, thereby reducing the force at which each higher threshold motor unit was recruited. Uninjured higher threshold motor units would produce their usual increment in force at recruitment, but superimposed on a lower force than normal. Thus, the percentage increment in force provided by an uninjured motor unit would be greater than normal after eccentric exercise, resulting in more uneven development of muscle force. Surprisingly, the decrease in maximal voluntary torque did not result in a signi®cant decrease in peak velocity during ballistic MP abduction. This is in contrast to the study of Miles et al. (1993), investigating control of elbow ¯exion. They found that peak velocity decreased and latency from movement onset to peak velocity increased, 24 h after eccentric exercise. The failure to ®nd a signi®cant decrease in peak velocity might have been a consequence of the novelty of the task. Rapid ®nger abduction is not a commonplace activity, especially when performed by the non-dominant hand. Any decrement in velocity post-exercise may have been oset by an increment in performance, due to practice. Such an interpretation is consistent with the ®nding that the peak velocity of MP ¯exion did decrease signi®cantly post-exercise. Ballistic ®nger ¯exion is more familiar, associated with activities such as ®nger tapping 219 and typing. Consequently, it would be less aected by practice. The decline in maximal voluntary ¯exion torque was greater than expected from the estimated decline in FDI ¯exion torque. Thomas et al. (1986) have shown that motor units in FDI are recruited in the same order for MP ¯exion and abduction. Therefore, any decrement in abduction torque would be expected to aect ¯exion torque. Based upon experiments employing muscle block with lidocaine by Ketchum et al. (1978), the estimated contribution of FDI to total MP ¯exion torque would be approximately 39%. Zijdewind and Kernell (1994), on the other hand, concluded from anatomical and biomechanical measurements, that the FDI muscle moment potential was only 15% of the total ¯exion torque. Assuming that FDI is the only muscle contributing to abduction of the index ®nger and using the estimate of Ketchum et al. (1978) the 19% decrement in maximal abduction torque which we observed, should have produced a decrement in maximal ¯exion torque of 7.4%. Using the value suggested by Zijdewind and Kernell (1994), the expected decline would have been even lower. Thus, the decrement in MP ¯exion torque was at least twice that predicted by the impaired force production of FDI. This suggests that the eect of eccentric exercise of the FDI muscle was to inhibit the extrinsic synergist muscles during MP ¯exion. It is unlikely that the extrinsic ®nger muscles could have been injured by the eccentric exercise since they produce adductor moments at the MP joint. Thus, the eccentric exercise in abduction would have caused them to shorten rather than lengthen. Since EDC EMG during maximal isometric ¯exion did not increase post-exercise, it is unlikely that the reduction in maximal ¯exion torque could have been due to increased contraction of antagonist muscles. During MP ¯exion at 50% MVC there was greater FDS activity late in the maintained contraction, following eccentric exercise. In addition, more subjects showed a signi®cant correlation between MDF and time post-exercise, which suggests that FDS muscle ®bres were being activated for a suciently long time to show greater evidence of fatigue. It would, therefore, appear that increased activity of the extrinsic ®nger ¯exor muscles was used to compensate for the FDI force de®cit during MP ¯exion. The results of this study indicate that a period of eccentric exercise, which is suciently strenuous, can produce de®cits in muscle mechanics that aect the ability to stabilize a joint and the ability to control muscle force smoothly. They also suggest that nonexercised synergist muscles can compensate for de®cits in joint torque at sub-maximal activation levels, but that these muscles exhibit inhibition during maximal activation. Because strenuous eccentric exercise has been shown to result in damage to muscle ®bres (FrideÂn et al. 1983; Lieber and FrideÂn 1988) leading to pain (Newham et al. 1983), the eects of eccentric exercise on motor control may also serve as a model for impaired control following musculoskeletal injury. Acknowledgement This work was supported by the Natural Sciences and Engineering Research Council of Canada. References Balnave CD, Thompson MW (1993) Eect of training on eccentric exercise-induced muscle damage. J Appl Physiol 75: 1545±1551 Berry CB, Moritani T, Tolson H (1990) Electrical activity and soreness in muscles after exercise. Am J Phys Med Rehabil 69: 60±66 Clarkson P, Byrnes W, McCormick KM, Turcotte LP, White JS (1986) Muscle soreness and serum creatine kinase activity following isometric, eccentric and concentric exercise. Int J Sports Med 7: 152±155 Clarkson PM, Nosaka K, Braun B (1992) Muscle function after exercise-induced muscle damage and rapid adaptation. Med Sci Sports Exerc 24: 512±520 De Luca CJ (1984) Myoelectrical manifestations of localized muscular fatigue in humans. CRC Crit Rev Biomed Eng 11: 251±279 De Serres SJ, Milner TE (1991) Wrist muscle activation patterns and stiness associated with stable and unstable mechanical loads. Exp Brain Res 86: 451±458 Ebbeling C, Clarkson P (1989) Exercise-induced muscle damage and adaptation. Sports Med 7: 207±234 Faulkner JA, Brooks SV, Opiteck JA (1993) Injury to skeletal muscle ®bers during contractions: conditions of occurrence and prevention. Phys Ther 73: 911±921 Felici F, Colace L, Sbriccoli P (1997) Surface EMG modi®cations after eccentric exercise. J Electromyogr Kinesiol 7: 192± 202 FrideÂn J, SjoÈstroÈm M, Ekblom B (1983) Myo®brillar damage following intense eccentric exercise in man. Int J Sports Med 4: 170±176 Hasson SM, Daniels JC, Divine JG, Niebuhr BR, Richmond S, Stein PG, Williams JH (1993) Eect of ibuprofen use on muscle soreness, damage, and performance: a preliminary investigation. Med Sci Sports Exerc 25: 9±17 Hogan NJ (1984) Adaptive control of mechanical impedance by coactivation of antagonist muscles. IEEE Trans Autom Control 29: 681±690 Howell JN, Chelboun G, Conatser R (1993) Muscle stiness, strength loss, swelling, and soreness following exercise-induced injury in humans. J Physiol (Lond) 464: 183±196 Katz B (1939) The relation between force and speed in muscular contraction. J Physiol (Lond) 96: 45±64 Ketchum LD, Thompson D, Pocock G, Wallingford DA (1978) A clinical study of forces generated by the intrinsic muscles of the index ®nger and the extrinsic ¯exor and extensor muscles of the hand. J Hand Surg 3: 571±578 Komi PV, Viitasalo JT (1977) Changes in motor unit activity and metabolism in human skeletal muscle during and after repeated eccentric and concentric contractions. Acta Physiol Scand 100: 246±254 Kranz H, Cassell JF, Inbar GF (1985) Relation between electromyogram and force in fatigue. J Appl Physiol 59: 821± 825 Kroon GW, Naeije M (1991) Recovery of the biceps electromyogram after heavy eccentric, concentric or isometric exercise. Eur J Appl Physiol 63: 444±448 Leger AB, Milner TE (2000) The eect of eccentric exercise on intrinsic and re¯ex stiness in the human hand. Clin Biochem l5: 574±582 Leger AB, Milner TE, Bawa P (1996) M-wave depression and potentiation following eccentric muscle contraction. Soc Neurosci Abstr 22: 1098 220 Lieber RL, FrideÂn J (1988) Selective damage of fast glycolitic muscle ®bers with eccentric contraction of the rabbit tibialis anterior. Acta Physiol Scand 133: 587±588 Miles MP, Ives J, Vincent KR (1993) Neuromuscular control following maximal eccentric exercise. Med Sci Sports Exerc 25: S176 Newham DJ, Mills KR, Quigley BM, Edwards HT (1983) Pain and fatigue after concentric and eccentric muscle contractions. Clin Sci 64: 55±62 Newham DJ, Jones DA, Clarkson PM (1987) Repeated high-force eccentric exercise: eects on muscle pain and damage. J Appl Physiol 63: 1381±1386 Saxton JM, Clarkson PM, James R, Miles M, Westerfer M, Clark S, Donnelly AE (1995) Neuromuscular dysfunction following eccentric exercise. Med Sci Sports Exerc 27: 1185± 1193 Thomas CK, Ross BH, Stein RB (1986) Motor-unit recruitment in the human ®rst dorsal interosseous muscle for static contractions in three dierent directions. J Neurophysiol 55: 1017±1029 Zijdewind I, Kernell D (1994) Index ®nger position and force of the human ®rst dorsal interosseous and its ulnar nerve antagonist. J Appl Physiol 77: 987±997