| | Joint-Angle–Dependent Neuromuscular Dysfunctions at the Wrist in Persons After StrokeAbstract Hu X, Tong K, Tsang VS, Song R. Joint-angle–dependent neuromuscular dysfunctions at the wrist in persons after stroke. ObjectiveTo evaluate the joint-angle–dependent neuromuscular functions at the affected wrist in hemiplegic subjects after stroke while doing isometric maximal voluntary wrist flexion and extension across different wrist angles. DesignWe investigated torques during isometric maximal voluntary wrist flexions and extensions at 8 different wrist angles, ranging from −45° to 60°. We used the associated electromyographic activities of 2 agonist and antagonist muscle pairs related to wrist and elbow joints for the analysis of muscular coactivations. We compared the data obtained from poststroke subjects’ affected and unaffected sides. SettingA research laboratory in a rehabilitation center. ParticipantsEleven subjects with hemiplegia after stroke with passive range of motion (ROM) in the wrist from −45° to 60°. InterventionsNot applicable. Main Outcome MeasuresDirectly measured torques, torques after normalization during maximal isometric wrist contractions, and normalized moving average electromyographic signals of each muscle at the tested positions. ResultsThe measured torques of the affected wrists were significantly lower than those of the unaffected wrists at all tested angles during wrist flexion and extension (P<.05). The angle-dependent patterns of the normalized torque across the tested wrist angles varied from those of the unaffected wrists (2-way analysis of variance, P<.05). There were decreases in normalized torques during both flexion and extension at the extended positions in the affected group (P<.05). Abnormal cocontractions were found in agonist and antagonist muscle pairs related to wrist and elbow joints, and between the elbow flexor and wrist extensor when subjects did the wrist contractions on the paretic side, especially at the wrist extended positions. ConclusionsWrist muscle weakness was distributed unevenly across the selected wrist ROM on the affected side, as represented by the varied patterns of the normalized torque-angle relationship, compared with the unaffected wrists. There were reductions in the selective control of muscle coactivating synergies both single-jointly and cross-jointly in the impaired nervous system during wrist contractions; the extent of these reductions was also related to the wrist angle configuration.
ALTERED NEURAL FUNCTIONS have been commonly observed in the contralesional neuromuscular pathways in stroke survivors with unilateral brain injury, causing consecutive neuromuscular changes such as decreased motor unit firing rates, stiffened and shortened muscle fibers, and muscle fiber type alterations.1, 2, 3, 4 Those changes are accompanied by clinical neuromuscular symptoms at the survivors’ paretic side, for example, weakness, spasticity, contracture, and loss of selective control of muscle coactivations. Rehabilitation therapies are useful in recovering motor functions in subjects after stroke. Some evolving therapeutic methods, such as robotic-aided training, constraint-induced movement therapy, and functional electric stimulation, also contribute to further recovery,5, 6, 7, 8 based on an effective adaptation of the treatment to the changes in the impaired neuromuscular systems. In the rehabilitation of the upper limb, however, many stroke survivors experience reasonable motor recovery of their proximal upper limb (shoulder and elbow) but limited recovery at the wrist (distal).9, 10 A possible reason for the relatively poorer recovery progress achieved in poststroke wrist rehabilitation is the limited understanding of the relation between the pathologic changes in stroke survivors (ie, the changes in neuromuscular properties and the coherent musculotendon variations) and the physical disabilities of the paretic wrist; an appropriate clinical measurement is important for exploring this relation during diagnoses and evaluations for rehabilitation treatments.
Many methods for evaluating poststroke upper-limb motor functions, including wrist functions, have been widely used clinically, including the Fugl-Meyer Assessment (FMA),11 the FIM instrument,12 the Motor Status Score test,13 the Modified Ashworth Scale (MAS)14 for grading spasticity during passive limb movements, and measurements taken during prescribed voluntary motions. These clinical evaluations, however, are accomplished through observation of unconstrained commonplace movements, and by the subjective judgment of the evaluator. Objective and quantitative measurements using designed apparatus have been introduced to evaluate upper-limb motor functions for muscle spasticity,15, 16 and for the abnormal voluntary functions with both static and dynamic motions such as weakness and slowness of muscle activation.17, 18, 19, 20 Most of these quantitative measurements were for the elbow and shoulder joints. Isometric strength measurement at a related joint is a simple and well-accepted clinical method for evaluating neuromuscular properties, such as muscle strength and weakness.21, 22 Four factors mainly affect wrist performance during isometric contractions: the muscle activation levels, muscle coactivations, moment arm of active muscles, and muscle architecture, for example, muscle fiber length and physiologic cross-sectional area (CSA). Chae et al9, 23 conducted preliminary studies on the muscle weakness and antagonist cocontraction in relation to wrist malfunctions in subjects after stroke. The root mean squares (RMS) of the electromyographic activity from wrist extensor and flexor were used to represent muscle strength during maximal isometric contractions at a neutral position of the wrist, where minimized electromyographic baselines could be obtained.9 The ratio of RMS of antagonist and agonist muscles was used as a measure of cocontraction. Muscle weakness and degree of cocontraction at the paretic wrist showed significant negative correlations with the upper-limb motor functional scores assessed by FMA and arm motor ability tests. Delays in the initiation and termination of muscle contraction during paretic wrist extension and flexion were also observed with electromyography.23 Abnormal muscle coactivations in stroke survivors usually not only existed in the antagonist and agonist muscles of the single-joint, but also could be found in the muscles attached to the neighboring joints. Correlations were found among muscular activities of shoulder and elbow of the affected side in poststroke subjects, but were absent in the unaffected side and in healthy controls.20 The cross-joint muscle coactivations associated with wrist contractions in stroke survivors have not yet been well studied.
By modeling studies on unimpaired subjects, Gonzalez et al24 found that muscle architecture and moment arms affect wrist flexion-extension moments during isometric contractions. The output torques during either wrist flexion or extension differed across the joint angles and were associated with the changes in muscular moment arm related to the joint angle and the muscle’s force-length behavior.24 Koo et al19 observed angle-dependent muscle weakness at the elbow joint in the affected side of stroke survivors during maximal isometric voluntary contractions across the range of motion (ROM). Besides the lower elbow torques at the paretic limbs, in comparison with the healthy controls across the ROM, the normalized torques during elbow flexion at the flexed positions and during extension at the extended positions of the affected limb were significantly less than in the healthy controls,19 which suggests a nonuniform distribution of the muscle weakness at the affected elbow. Similar systematic studies on the effects of wrist angle and muscle architecture on wrist output torque for stroke survivors are still lacking. Previous studies of the affected wrist in stroke patients generally adopted a single-joint-angle configuration in the isometric contraction measurement,9, 23 which only provided an incomplete picture of the wrist function after stroke. Wrist and hand functions were posture related. For example, wrist postures affected the hand opening of poststroke subjects treated by electric stimulation in a study by Cameron et al.25 They demonstrated that maximal displacements between thumb tip and each fingertip could be evoked by the electric stimulation when the wrist was fully flexed. They also suggested that the grasp aperture of the paretic hand was strongly dependent on wrist posture and accompanying voluntary efforts.25 It is also known that muscle spasticity and atrophy usually coexist in a stroke survivor. These neuromuscular abnormalities resulted in increased joint stiffness, distortion in joint structure, and variation in muscle architecture.26 Increased stiffness and shortened resting length has been reported for spastic muscle fibers.1, 27 Atrophy from muscle disuse causes muscle fiber-type changes, especially atrophy of the fast-twitch fibers, and a reduction in the size of muscle cells.28, 29, 30, 31, 32 Hu et al33 found that atrophy of the fast-twitch fibers in a paretic muscle affected the amplitude of the muscle’s surface electromyographic signal. All of these pathologic changes may affect the moment arm and muscle architectures at the paretic wrist of stroke survivors.
Our purpose was to investigate poststroke wrist-angle–related muscle weakness and muscle coactivation patterns during maximal isometric wrist flexion and extension. We analyzed the effect of wrist joint angle on the wrist moment at the paretic wrist of stroke survivors, and compared that effect with their unaffected wrists; we also studied the cross-joint muscle coactivations from the elbow associated with wrist contractions.
Methods  Experiment After obtaining approval from the Human Subjects Ethics Subcommittee of the Hong Kong Polytechnic University, we recruited 11 poststroke subjects for the study. All had hemiplegia as a result of a single cerebral unilateral lesion, and all were in stable medical conditions (at least 6mo poststroke). We measured the motor impairment of their affected wrist at the paretic side with the FMA wrist and hand score (maximum score, 24),11 and spasticity of that wrist with the MAS14 (table 1). Passive ROM for the paretic wrist of all the subjects ranged from −45° to 60°. (The negative ROM represented the extended positions, the positive ROM represented the flexed positions, and the 0° ROM was denoted as the neutral position in this study.) We studied both the affected and unaffected wrists of the subjects. The experimental setup is shown in figure 1. During the experiment, subjects were comfortably seated, and the forearm to be tested was attached horizontally on a platform of a motora and torque-sensorb system (accuracy, .03Nm). The wrist joint was centered on the origin of the motor, and the palm was fixed on a manipulandum, which was rotated by the motor, with the axis of rotation in line with the wrist joint. The torque sensor measured wrist torque during contraction of wrist. The elbow angle was set at approximately 150°, with shoulder extension at 0°. Electromyographic signals were used to analyze the activities of the flexor carpi radialis, extensor carpi radialis, biceps brachii, and triceps brachii (long head), which were the representative agonist and antagonist muscles related to the wrist and elbow flexion and extension. A pair of surface electromyography electrodesc (diameter, 1.5cm; conductive area, 1.17cm2) was attached on the skin surface above the belly of each muscle of interest, with a center separation of 2cm. The center distances between different electromyography electrode pairs were at least 9cm. As suggested by Mogk and Keir,34 the crosstalk between adjacent surface electromyography electrode pairs attached circumferentially on the proximal forearm was mainly affected by the distance between the 2 electrode pairs; with 9cm spacing, the crosstalk was around 2.5% in the detected signals. A reference electromyography electrode was placed on the skin surface of the olecranon. The electromyographic signal was acquired with an amplification system,d and it was synchronized with the torque signal through an analog-to-digital acquisition interface card, with a sampling frequency at 1000Hz.e The collected signals were stored in a computer. | | |  | Subject | Sex | Age (y) | Affected Upper Limb | Wrist Ashworth Scale Score | FMA Wrist and Hand Score (max score, 24) | |
|---|
| 1 | F | 42 | Left | 1 | 20 | |
| 2 | F | 51 | Left | 1+ | 15 | |
| 3 | F | 45 | Left | 2 | 10 | |
| 4 | M | 36 | Right | 2 | 16 | |
| 5 | M | 46 | Right | 2 | 13 | |
| 6 | F | 49 | Right | 1+ | 13 | |
| 7 | M | 57 | Right | 1+ | 13 | |
| 8 | M | 52 | Right | 1+ | 10 | |
| 9 | M | 49 | Left | 1+ | 19 | |
| 10 | M | 60 | Right | 1+ | 12 | |
| 11 | M | 53 | Left | 1+ | 10 | | | | |
During the test, subjects performed isometric maximal voluntary wrist extension and flexion at different wrist angles at both affected and unaffected sides. We selected 8 wrist positions, ranging from −45° to 60° at increments of 15°, and the testing sequence at the different angles was randomized for each wrist. At each wrist position, subjects were asked to conduct isometric maximum wrist flexion and extension twice. Before a contraction, subjects were given a visual starting signal (a programmed button lighting on a computer screen). Each contraction lasted for approximately 5 seconds, and the output torque level was stabilized for approximately 2 seconds. Subjects were given visual feedback of the torque. Subjects were allowed a 2-minute rest between the 2 consecutive contractions. Figure 2 shows the representative torque and electromyographic signals recorded from 1 subject during wrist extensions in the experiment. Signal Processing and Data Analysis The digitized electromyographic and wrist torque signals were processed offline. The torque signals were low-pass filtered (fourth-order Butterworth filter) at 10Hz. The electromyographic signals were band-pass filtered (fourth-order Butterworth filter) from 10 to 500Hz and rectified, and then the signal was processed by moving average with a 100-ms window. A fourth-order, zero-phase forward and reverse digital filter was used to avoid time delay of the filtered data. In each trial, the maximum torque value with respect to the initial resting offset level that a subject could generate at position θ was denoted as TTrialMax(θ):
where T(θ) is the torque value corresponding to the maximum torque amplitude generated during the wrist contraction; and T rest(θ) was the offset torque value during the initial resting state, as in the example in figure 2. In the experiment, we found that some subjects flexed or extended their affected wrists in a manner opposite to what they were instructed to do, especially when the subjects were instructed to perform the wrist flexions at more flexed positions (3 subjects) and the wrist extensions at more extended positions (7 subjects). These torques, T TrialMax(θ), obtained opposite to what was intended, would have negative value. For the respective wrist extension and flexion at each position, the maximal T TrialMax(θ) was selected from the repeated measurements, denoted as T Max(θ), for our analysis. As suggested in Bohannon’s test-retest study 35 (3 times repeated measure), the wrist maximal isometric torques of hemiparetic subjects (n=30) with neurologic disorders (including poststroke, n=16) could be reliably measured in a single session of strength assessment by using a commercial dynamometer (Pearson correlations between each measurement were r>.95). 35 The Pearson correlation coefficient between the 2 repeated measurements of torque values (T TrialMax[θ]) in this study was .92. The ensemble mean of the moving average electromyographic signal of muscle i over 100ms before the appearance of TMax(θ) was computed and denoted by mEMGi(θ), as Koo et al19 used in their study. The torque and electromyographic data were normalized for comparison across different subjects. The normalization procedure was: (1) for each instructed torque direction, the normalized mEMG of muscle i at position θ, that is,  , was computed by:
where offset i(θ) was the average electromyographic activities of muscle i during rest at wrist angle θ, mEMG imax was the maximum mEMG of muscle i recorded among the testing positions, and offset imax was the offset at the corresponding position; and (2) for each instructed torque direction, the normalized maximum torque magnitude at position θ, that is,  , was defined as:
where T Max was the maximum isometric torque generated among the testing positions. The analyses of variance (ANOVAs) with Bonferroni post hoc test and independent t test were used to evaluate the significant effects from the wrist groups (ie, the unaffected and affected groups) and different wrist positions on the output torque levels of the extension and flexion, respectively. Normalized mEMG/mEMG plots, which Dewald et al20 used, were also applied in this study to represent the coactivation between different muscle pairs. We analyzed the statistical significance of the coactivation of a muscle pair by the correlation between the normalized mEMG/mEMG coactivating representations of the muscles and its linear regression line; the correlation with a Pearson coefficient value of r greater than .273, and a probability of confidence for the correlation of P less than .05 were regarded as statistically significant.20 A linear regression line with a positive slope (s>0) suggested the muscle cocontraction (or coshortening) of the muscle pair, while a negative slope (s<0) implied a contraction-and-relaxation pattern of the muscle pair. We set the level of statistical significance at .05. Results Figure 3 shows the measured wrist extension and flexion torques from the unaffected and the affected groups, respectively. The mean values of flexion torques from the unaffected group first demonstrated a slight increase from −45° to −15° with no statistical significance, and then showed a monotonic decrease from −15° to 60°. This decrease was significant (1-way ANOVA, P<.001; post hoc tests, P<.05). The unaffected extension torques increased from −45° to 0° (1-way ANOVA, P=.003; post hoc tests, P<.05), and then varied across the wrist angles from 0° to 60° with no statistical significance. In the unaffected group, the wrist flexion torques were larger than those for the extension at the positions of −45°,−30°, −15°, and 0° (t tests, P<.05), while the mean value of extension torques was greater than the flexion at 60° (t test, P<.05). The mean values of wrist torques for extension and flexion of the affected group were significantly lower than those of the respective torques from the unaffected group at all corresponding wrist angles (t tests, P<.05). The affected flexion torque reached the maximum at 0° (1-way ANOVA, P<.021; post hoc tests, P<.05), and showed a decreasing tendency from 0° to 60° (1-way ANOVA, P=.004; post hoc tests, P<.05). The affected wrist extension torque did not have a significant angle-dependent variation. We also found that in the affected group, the wrist flexion torque values were larger than the extension at the wrist angles at −45°, −30°, −15°, and 0° (t tests, P<.05). Figure 4 shows the torque-angle pattern by the normalized wrist flexion and extension torques of the unaffected and affected groups. Group differences resulted in significant effects on the torque-angle patterns during both wrist flexion and extension (2-way ANOVA tests, for both flexion and extension, P<.05; interactions between the wrist angle and group difference: flexion, P=.004; extension, P=.006). The normalized unaffected flexion torque had a plateau phase from −45° to −15°, and then a statistically significant steady decrease from 0° to 60° (1-way ANOVA with post hoc tests, P<.05; fig 4A); the maximal mean value was observed at −15°. The normalized affected flexion torque had an ascending limb from −45° to 0° (P<.05), and a descending limb from 0° to 60° (1-way ANOVA with post hoc tests, P<.05; see fig 4A); the maximal mean value was at 0°. The normalized unaffected flexion torques were significantly higher than those of the affected from −45° to −15° (t tests, P<.05). The normalized unaffected extension torque increased from −45° to 0° (P<.05), then decreased to 30° (P<.05), and finally remained almost unchanged from 30° to 60° (1-way ANOVA with post hoc tests, P <.001; fig 4B). There was no significant variation in the normalized extension torques from the affected group across wrist ROM (see fig 4B). The peak mean value of the normalized affected wrist extension torque was observed at 30°, but was not statistically significant. The normalized unaffected extension torques were significantly higher than those of the affected torques, from −45° to 15°, and at 45° (t tests, P<.05). The electromyographic activities, represented by the normalized mEMG, of the wrist flexor and extensor from the unaffected and affected groups are plotted in figure 5. During wrist flexion (see fig 5A), the activation level of the unaffected wrist flexors increased steadily from −45° to 0° (P<.05), and then varied (not significantly) from 0° to 60° (1-way ANOVA with post hoc tests, P=.000); the activity of the unaffected wrist extensors also increased from −45° to 60° (1-way ANOVA, P<.001). The electromyographic activation levels of the flexor were significantly higher than the extensor during flexion over all wrist angles in the unaffected group (t tests, P<.05). There was no significant change in the electromyographic activation level of the unaffected wrist flexor versus the different wrist angles during the wrist extension (see fig 5C). The electromyographic level of the unaffected wrist extensor, however, decreased monotonically from −45° to 60° (1-way ANOVA, P<.001; see fig 5C). The electromyographic activation levels of the extensor were significantly higher than those of the flexor from −45° to 30° for the unaffected wrists (t tests, P<.05; see fig 5C). During the flexion of the affected wrists (see fig 5B), different wrist angles did not affect the electromyographic activations of the wrist flexors. The electromyographic activities of the affected extensor were relatively lowered at the wrist angles from −45° to 0°, then gradually elevated and reached the maximum at 45° with the wrist angle increased (1-way ANOVA with post hoc tests, P<.001; see fig 5B). The electromyographic activation levels of the flexor were significantly higher than the extensor at −30°, −15°, and 0°, but significantly lower than the extensor at 45° during flexion of the affected wrists (t tests, P<.05; see fig 5B). In figure 5D, variations of the electromyographic activity levels for the affected extensors and flexors were not significant across the selected wrist ROM during the extensions of the affected wrist. Figure 6 shows the overall coactivations of different pairs of muscles during the wrist contractions across the whole ROM, represented by the normalized mEMG/mEMG plot and the corresponding linear regression. Coactivations associated with significance (r>.273; probability of confidence for the correlation, P<.05) were found between the wrist flexor and biceps brachii, between the triceps brachii and wrist flexor (slope of the linear regression line, s<0), and between the triceps brachii and wrist extensor in the unaffected group. Cocontractions at significant levels (r>.273, P<.05) were observed in all pairs of muscles in the affected group, except between the triceps brachii and wrist flexor. Table 2 shows the cocontractions of muscle pairs with statistical significance (r>.273, P<.05, s>0) at different wrist angles for both the unaffected and affected groups. |
⁎
Statistically significant.
†
Muscles from the affected group. |
Discussion The measured unaffected wrist flexion and extension torques demonstrated a clear characteristic of the wrist-angle dependent (see Fig 3, Fig 4). The patterns of the wrist torques during flexion and extension across the wrist angles for the unaffected group were similar to those reported by Garner and Pandy36 in experiments with unimpaired subjects and reproduced by the modeling study for understanding the muscle architecture in relation to the wrist moments. The significantly higher peak torque (see fig 3, appeared at −15°) during unaffected wrist flexion than that seen during extension (appeared at 0°) that we found is also consistent with the previous reports.37, 38, 39 This phenomenon was caused by the larger summed physiologic CSA of the wrist flexors than that of the extensors.39 The characteristics of the unaffected wrist torque over the selected ROM of wrist could well be explained by the combined changes in moment arm and maximal isometric force of individual active muscles at different wrist angles.36, 39 Scaled Hill-type models24 revealed that the wrist torques at different wrist angles were also affected by the musculotendon architecture parameters of individual muscle, such as muscle fiber length, physiologic CSA, muscle fiber pennation angle, and tendon length, for wrist contractions.36, 39 Poststroke subjects experience changes in the biomechanic properties of their muscles, such as a decreased number of functional motor units, increased muscle stiffness, and shortening of muscle fibers,1, 2, 3 because of immobilization and altered control of limb positions and movements. These changes will affect the joint torque production during isometric contractions. Significantly lowered affected wrist torques (see fig 3), when compared with those from the unaffected group, suggests that there is general muscular weakness in the affected wrists. The higher peak torque of the affected wrist flexion (at 0°) than at extension (at 15°) implies that the summed physiologic CSA of the affected wrist flexors is still larger than that of the extensors, although paretic muscles after stroke could undergo several pathologic changes, as mentioned previously. Clinically, stroke survivors commonly have a flexed wrist at the affected side, mainly caused by the muscular contractures in the upper limb after the upper motoneuron lesion. Therefore, subjects usually would have lengthening wrist extensors and shortening wrist flexors. It had been found that skeletal muscles adapt to a shortened (lengthened) position by reducing (increasing) the number of sarcomeres.40 Spastic muscles with shortening fiber length result in a shortening resting fiber length.26, 41 These changes can bring shifts in the muscle length-force curve of each active muscle during affected wrist contractions. As compared with the unaffected wrist contraction over the wrist ROM, the peak values of the affected normalized torque during wrist flexion and extension shifted to more flexed positions (see figs 4A, 4B). Shortened wrist flexor muscles brought the operational ranges to the right in the length-force curve. Experimental and modeling studies suggest that the prime wrist flexors—the flexor carpi radialis and flexor carpi ulnaris—work in the ascending limb of the length-force curve with the wrist ROM from −45° to 50° for unimpaired subjects.24 Therefore, wrist flexor shortening in stroke survivors increases the flexion torque at the flexed positions but decreases flexion torque at extended positions. Figure 4A shows relatively lowered normalized flexion torques in the affected group, from −45° to −15°, compared with those from the unaffected group. The normalized flexor torques between the affected and unaffected groups did not differ significantly at wrist-flexed positions (0°−60°) in our study. The operational ranges of the major wrist extensors (ie, extensor carpi radialis, extensor carpi ulnaris) for unimpaired subjects with wrist angle from −45° to 50° were mainly in the middle of the length-force curve.24 In this area, muscle force was relatively less affected by length changes. Lengthening of the wrist extensors brought the operational range of the extensors to the left of the force-length curve, which resulted in a decrease in the wrist extension torques at extended positions, but an increase at flexed positions. Compared with the unaffected group, the normalized affected extension torques were lower than those of the unaffected from −45° to 15° (see fig 4B). Increased wrist flexor stiffness in the affected group caused extra passive mechanical restraint at the extended positions during wrist extension, which also contributed to the lowered extension torques of the affected wrist when they were extended. The wrist torque distribution over the range of wrist angles was affected not only by the biomechanic properties of the muscles, but also controlled by the neural inputs. Electromyographic activity was used as a reference of the nervous control on individual muscles. We found it interesting that the electromyographic activation levels of the represented wrist agonist muscles also demonstrated the angle-dependent characteristic in the unaffected group, although at each wrist position subjects were asked to perform the maximal voluntary contractions. This phenomenon could be related to the hypothesis that the muscle with the larger mechanical advantage receives the larger activation.21, 42 With the increase of the moment arm related to the wrist position changes, the electromyographic activation level of the agonist muscles also increased (see figs 5A, 5C). Excessive activation of antagonists and/or reduced activation of the agonists has been found in the paretic neuromuscular systems of stroke survivors.43 In our work, the affected wrist flexor and extensor did not demonstrate similar changes in electromyographic level as in the unaffected group with the corresponding increase in the moment arms during wrist flexion and extension, respectively (see figs 5B, 5D). This possibly suggests a reduction in the activation of the agonist muscles in the affected wrists. Excessive activation of antagonists could be found in both affected wrist flexion and extension. The electromyographic activity levels of the extensor were equivalent, for the most part, to those of the flexor of the flexed positions during affected wrist flexions. There was overexcitation of the affected flexor during extension across the entire wrist ROM. Abnormal muscle coactivations during isometric torque generation at cross-joint positions (elbow and shoulder) has been reported in hemiplegic subjects after stroke, due mainly to the loss of selective control of muscle activities from the nervous system after the cortical lesion.20 Varied electromyographic coactivations among the agonist and antagonist muscles for the wrist and elbow were also observed in the affected group, compared with those in the unaffected group (see fig 6). Unlike the coactivation patterns of muscle pairs in the unaffected group, additional significant cocontractions were found between the wrist extensor and flexor muscles (the antagonist cocontraction related to the wrist joint), between the biceps brachii and triceps brachii (the antagonist cocontraction related to the elbow joint), and between the biceps brachii and extensor (the cocontraction related to the wrist and elbow joints). There was no significant cocontraction of antagonist muscle pairs in the unaffected group. The electromyographic activities of the triceps brachii and the wrist flexor coactivated with a negative slope of the regression line, which suggests a contraction-and-relaxation pattern of these 2 muscles during the maximal wrist contractions. This pattern, however, was not seen in the affected group, which indicates the disruption of this regulation mechanism after stroke. Increased single-joint and cross-joint cocontractions reduced the possible muscle synergies in the affected limb to perform separated movements in the upper extremity. It was often observed clinically that the attempts at wrist movement in the affected limb were always associated with the elbow motions. The cross-joint cocontractions also could be the reason for the unsatisfied recovery of the wrist functions in poststroke rehabilitation, inasmuch as the wrist movements could be greatly affected by the elbow-contracting muscles in the affected limb. Table 2 shows that the cocontractions between different muscle pairs were mainly distributed at wrist-flexed positions in the unaffected group. While cocontracting muscle pairs were almost evenly distributed at the extended and flexed wrist positions in the affected group, the muscle cocontracting pairs, with statistical significance across the tested wrist ROM for the affected group, were much more than the unaffected group. This suggests that the affected wrist impairment possibly was more severe at the extended positions for the stroke survivors in this study. More spastic flexors compared with the extensors in the upper limb after stroke had been reported.8, 44 Significant cocontractions related to the flexors of the wrist flexor and biceps brachii were more frequently observed in the affected group, especially at the joint-extended positions, where the flexors were stretched. The muscle coactivation patterns obtained in this study were from the isometric experimental configuration. Further study is warranted of the muscle coactivation patterns during dynamic motions of the paretic wrist after stroke. The unaffected side of hemiplegic subjects after stroke may not be normal. Some studies have proposed that the unilateral damage to motor cortex has bilateral effects on the control of movement,45, 46 although other studies have suggested that the ipsilateral components of the corticospinal system were small and had little influence over the control of distal movements.47, 48 In this work, the unaffected limb was compared with the affected limb of the hemiplegic subjects, as in many other studies to study muscle coactivating patterns at the elbow and shoulder,20 upper-limb recovery procedure,49 and muscle isotonic strength evaluation.50 Our results provide additional information on paretic wrist functions in studies of this type, where the unaffected side of hemiplegic subjects after stroke is used as the reference to the affected side. Conclusions In this study, wrist torque and the related muscle coactivating patterns during wrist flexion and extension in hemiplegic subjects poststroke were assessed by the wrist maximal isometric contraction over a common wrist ROM. The significantly higher torque in the unaffected wrist during both flexion and extension at all wrist angles suggests general muscle weakness in paretic wrists. The wrist muscle weakness was distributed unevenly across the selected wrist ROM, which was represented by the varied patterns of normalized torque-angle relationship as compared with the unaffected wrists. Abnormal cocontractions were commonly observed in the affected limb between the muscle pairs of the wrist flexor and extensor, the wrist flexor and biceps brachii, and the wrist extensor and biceps brachii, especially in wrist-extended positions. That suggests a wrist-position–related reduction in the selective control of muscle coactivating synergies from the impaired nervous system, which could be single-joint and cross-joint; the extent of this reduction was also related to the wrist angle configuration. Our results provide a new method with which to evaluate poststroke wrist functions, which may provide a more complete understanding of the poststroke malfunctions of the affected wrist related to wrist postures. The muscle coactivation patterns observed in the affected limb during wrist isometric contractions has potential for use in the design of myoelectricity involved rehabilitation treatment. Suppliers
References 1.
1
Friden J
, Lieber RL
.
Spastic muscle cells are shorter and stiffer than normal cells
.
Muscle Nerve
. 2003;26:157–164
.
2.
2
Hara Y
, Akaboshi K
, Masakado Y
, Chino N
.
Physiologic decrease of single thenar motor units in the F-response in stroke patients
.
Arch Phys Med Rehabil
. 2000;81:418–423
.
Abstract | Full Text |
Full-Text PDF (129 KB)
|
CrossRef
3.
3
Hara Y
, Masakado Y
, Chino N
.
The physiological functional loss of single thenar motor units in the stroke patients: when does it occur? Does it progress?
.
Clin Neurophysiol
. 2004;115:97–103
.
Abstract | Full Text |
Full-Text PDF (298 KB)
|
CrossRef
4.
4
Thomas CK
, Butler JE
, Zijdewind I
.
Patterns of pathological firing in human motor units
.
Adv Exp Med Biol
. 2002;508:237–244
.
MEDLINE 5.
5
Volpe BT
, Krebs HI
, Hogan N
, Edelstein OL
, Diels C
, Aisen M
.
A novel approach to stroke rehabilitation
(robot-aided sensorimotor stimulation)
.
Neurology
. 2000;54:1938–1944
.
MEDLINE 6.
6
Fasoli SE
, Krebs HI
, Hogan N
.
Robotic technology and stroke rehabilitation
(translating research into practice)
.
Top Stroke Rehabil
. 2004;11:11–19
.
MEDLINE |
CrossRef
7.
7
Sterr A
.
Training-based interventions in motor rehabilitation after stroke
(theoretical and clinical considerations)
.
Behav Neurol
. 2004;15:55–63
.
MEDLINE 8.
8
Powell J
, Pandyan AD
, Granat M
, Cameron M
, Stott DJ
.
Electrical stimulation of wrist extensors in poststroke hemiplegia
.
Stroke
. 1999;30:1384–1389
.
MEDLINE 9.
9
Chae J
, Yang G
, Park BK
, Labatia I
.
Muscle weakness and cocontraction in upper limb hemiparesis
(relationship to motor impairment and physical disability)
.
Neurorehabil Neural Repair
. 2002;16:241–248
.
MEDLINE 10.
10
Chae J
, Hart R
.
Intramuscular hand neuroprosthesis for chronic stroke survivors
.
Neurorehabil Neural Repair
. 2003;17:109–117
.
MEDLINE 11.
11
Fugl-Meyer AR
, Jaasko L
, Leyman I
, Olsson S
, Steglind S
.
The post-stroke hemiplegic patient. I: A method for evaluation of physical performance
.
Scand J Rehabil Med
. 1975;7:13–31
.
MEDLINE 12.
12
Keith RA
, Granger CV
, Hamilton BB
, Sherwin FS
.
The functional independence measure
(a new tool for rehabilitation)
.
Adv Clin Rehabil
. 1987;1:6–18
.
MEDLINE 13.
13
Ferraro M
, Demaio JH
, Krol J
.
Assessing the motor status score
(a scale for the evaluation of upper limb motor outcomes in patients after stroke)
.
Neurorehabil Neural Repair
. 2002;16:283–289
.
MEDLINE 14.
14
Ashworth B
.
Preliminary trials of carisoprodol in multiple sclerosis
.
Practitioner
. 1964;192:540–542
.
MEDLINE 15.
15
Pisano F
, Miscio G
, Conte CD
, Pianca D
, Candeloro E
, Colombo R
.
Quantitative measures of spasticity in post-stroke patients
.
Clin Neurophysiol
. 2000;111:1015–1022
.
Abstract | Full Text |
Full-Text PDF (254 KB)
|
CrossRef
16.
16
Schmit BD
, Dhaher Y
, Dewald JP
, Rymer WZ
.
Reflex torque response to movement of the spastic elbow
(theoretical analyses and implications for quantification of spasticity)
.
Ann Biomed Eng
. 1999;27:815–829
.
MEDLINE |
CrossRef
17.
17
Canning CG
, Ada L
, O’Dwyer NJ
.
Abnormal muscle activation characteristics associated with loss of dexterity after stroke
.
J Neurol Sci
. 2000;176:45–56
.
|
CrossRef
18.
18
Patten C
, Kothari D
, Whitney J
, Lexell J
, Lum PS
.
Reliability and responsiveness of elbow trajectory tracking in chronic poststroke hemiparesis
.
J Rehabil Res Dev
. 2003;40:487–500
.
MEDLINE |
CrossRef
19.
19
Koo TK
, Mak AF
, Hung LK
, Dewald JP
.
Joint position dependence of weakness during maximum isometric voluntary contractions in subjects with hemiparesis
.
Arch Phys Med Rehabil
. 2003;84:1380–1386
.
Abstract | Full Text |
Full-Text PDF (203 KB)
|
CrossRef
20.
20
Dewald JP
, Pope PS
, Given JD
, Buchanan TS
, Rymer WZ
.
Abnormal muscle coactivation patterns during isometric torque generation at the elbow and shoulder in hemiparetic subjects
.
Brain
. 1995;118:495–510
.
21.
21
Beer RF
, Given JD
, Dewald JP
.
Task-dependent weakness at the elbow in patients with hemiparesis
.
Arch Phys Med Rehabil
. 1999;80:766–772
.
Abstract |
Full-Text PDF (1596 KB)
|
CrossRef
22.
22
Colebatch JG
, Gandevia SC
.
The distribution of muscular weakness in upper motor neuron lesions affecting the arm
.
Brain
. 1989;112:749–763
.
23.
23
Chae J
, Yang G
, Park BK
, Labatia I
.
Delay in initiation and termination of muscle contraction, motor impairment, and physical disability in upper limb hemiparesis
.
Muscle Nerve
. 2002;25:568–575
.
CrossRef
24.
24
Gonzalez RV
, Buchanan TS
, Delp SL
.
How muscle architecture and moment arms affect wrist flexion-extension moments
.
J Biomech
. 1997;30:705–712
.
Abstract |
Full-Text PDF (1836 KB)
|
CrossRef
25.
25
Cameron T
, McDonald K
, Anderson L
, Prochazka A
.
The effect of wrist angle on electrically evoked hand opening in patient with spastic hemiplegia
.
IEEE Trans Rehabil Eng
. 1999;7:109–111
.
MEDLINE |
CrossRef
26.
26
Lieber RL
, Steinman S
, Barash IA
, Chambers H
.
Structural and functional changes in spastic skeletal muscle
.
Muscle Nerve
. 2004;29:615–627
.
CrossRef
27.
27
Mirbagheri MM
, Barbeau H
, Ladouceur M
, Kearney RE
.
Intrinsic and reflex stiffness in normal and spastic, spinal cord injured subjects
.
Exp Brain Res
. 2001;141:446–459
.
MEDLINE |
CrossRef
28.
28
Maier A
, Cocket JL
, Simpson DR
, Saubert CI
, Edgerton VR
.
Properties of immobilized guinea pig hindlimb muscles
.
Am J Physiol
. 1976;231:1520–1526
.
MEDLINE 29.
29
Lieber RL
, Jacobson MD
, Fazeli BM
, Abrams RA
, Botte MJ
.
Architecture of selected muscles of the arm and forearm
(anatomy and implications for tendon transfer)
.
J Hand Surg [Am]
. 1992;17:787–798
.
Abstract |
Full-Text PDF (1391 KB)
|
CrossRef
30.
30
Lieber RL
, Fazeli BM
, Botte MJ
.
Architecture of selected wrist flexor and extensor muscles
.
J Hand Surg [Am]
. 1990;15:244–250
.
Abstract |
Full-Text PDF (731 KB)
|
CrossRef
31.
31
Toffola ED
, Sparpaglione D
, Pistorio A
, Buonocore M
.
Myoelectric manifestations of muscle changes in stroke patients
.
Arch Phys Med Rehabil
. 2001;82:661–665
.
Abstract | Full Text |
Full-Text PDF (88 KB)
|
CrossRef
32.
32
Slager UT
, Hsu JD
, Jordan C
.
Histochemical and morphometric changes in muscles of stroke patients
.
Clin Orthop Relat Res
. 1985;(199):159–168
.
33.
33
Hu XL
, Tong KY
, Li L
.
The vibromyography of persons after stroke during isometric voluntary contractions
.
Proceedings of the 4th World Congress for NeuroRehabilitation
. 2006;
.
34.
34
Mogk JP
, Keir PJ
.
Crosstalk in surface electromyography of the proximal forearm during gripping tasks
.
J Electromyogr Kinesiol
. 2003;13:63–71
.
Abstract | Full Text |
Full-Text PDF (237 KB)
|
CrossRef
35.
35
Bohannon RW
.
Test-retest reliability of hand-held dynamometry during a single session of strength assessment
.
Phys Ther
. 1986;66:206–209
.
MEDLINE 36.
36
Garner BA
, Pandy MG
.
Musculoskeletal model of the upper limb based on the visible human male dataset
.
Comput Methods Biomech Biomed Eng
. 2001;4:93–126
.
37.
37
Jacobson MD
, Raab RR
, Fazeli BM
, Abrams RA
, Botte MJ
, Lieber RL
.
Architecture design of the human intrinsic hand muscles
.
J Hand Surg [Am]
. 1992;17:804–809
.
Abstract |
Full-Text PDF (615 KB)
|
CrossRef
38.
38
Lehman SL
, Calhoun BM
.
An identified model for human wrist movements
.
Exp Brain Res
. 1990;81:199–208
.
MEDLINE 39.
39
Conzalez RV
, Buchanan TS
, Delp SL
.
How muscle architecture and moment arms affect wrist flexion-extension moments
.
J Biomech
. 1997;30:705–712
.
Abstract |
Full-Text PDF (1836 KB)
|
CrossRef
40.
40
Tardieu C
, Tabary JC
, Tardieu G
.
Adaptation of sarcomere numbers to the length imposed on muscle
.
In:
Guba F
, Marechal G
, Takacs O
editor.
Mechanism of muscle adaptation to functional requirements
. Elmsford: Pergamon Pr; 1981;p. 99–114
.
41.
41
Friederich JA
, Brand RA
.
Muscle fiber architecture in the human lower limb
.
J Biomech
. 1990;23:91–95
.
MEDLINE |
CrossRef
42.
42
Nakazawa K
, Kawakami Y
, Fukunaga T
, Yano H
, Miyashita M
.
Differences in activation patterns in elbow flexor muscles during isometric, concentric and eccentric contractions
.
Eur J Appl Physiol
. 1993;66:214–220
.
43.
43
Gowland C
, deBruin H
, Basmajian JV
, Plews N
, Burcea I
.
Agonist and antagonist activity during voluntary upper-limb movement in patients with stroke
.
Phys Ther
. 1992;72:624–633
.
MEDLINE 44.
44
Pandyan AD
, Granat MH
.
Effects of electrical stimulation on flexion contractures in the hemiplegic wrist
.
Clin Rehabil
. 1997;11:123–130
.
MEDLINE |
CrossRef
45.
45
Carey JR
, Baxter TL
, Di Fabio RP
.
Tracking control in the nonparetic hand of subjects with stroke
.
Arch Phys Med Rehabil
. 1997;79:435–441
.
Abstract |
Full-Text PDF (805 KB)
|
CrossRef
46.
46
Desrosiers J
, Bourbonnais D
, Bravo G
, Roy PM
, Guay M
.
Performance of the ‘unaffected’ upper extremity of elderly stroke patients
.
Stroke
. 1996;27:1564–1570
.
MEDLINE 47.
47
Dum RP
, Strick PL
.
The origin of corticospinal projections from the premotor areas in the frontal lobe
.
J Neurosci
. 1991;11:667–689
.
MEDLINE 48.
48
Dum RP
, Strick PL
.
The corticospinal system: a structural framework for the central control of movement. Section A: Neural control of movement
.
In:
Rowell LB
, Shepard JT
editor.
Handbook of exercise physiology
(integration of motor, circulatory, respiratory and metabolic control during exercise)
. New York: Oxford; 1996;p. 217–254
.
49.
49
Turton A
, Wroe S
, Trepte N
, Fraser C
, Lemon RN
.
Contralateral and ipsilateral EMG responses to transcranial magnetic stimulation during recovery of arm and hand function after stroke
.
Electroencephalogr Clin Neurophysiol
. 1996;101:316–328
.
MEDLINE 50.
50
Hsu AL
, Tang PF
, Jan MH
.
Test-retest reliability of isokinetic muscle strength of the lower extremities in patients with stroke
.
Arch Phys Med Rehabil
. 2002;83:1130–1137
.
Abstract | Full Text |
Full-Text PDF (179 KB)
|
CrossRef
Department of Health Technology and Informatics, Hong Kong Polytechnic University, Hong Kong.  Reprint requests to Kaiyu Tong, PhD, Dept of Health Technology and Informatics, Hong Kong Polytechnic University, Rm ST417, Core S, 4/F, Kowloon, Hong Kong.
Supported by the Hong Kong Polytechnic University (grant nos. G-T598, G-YX65) and the Research Grants Council, Hong Kong Special Administrative Region, China (grant no. PolyU 5320/03E). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. PII: S0003-9993(06)00106-7 doi:10.1016/j.apmr.2006.02.003 © 2006 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT00106-7/journalimage?img=PIIS0003999306001067.gr1.lrg.gif&fig=fig1&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
00106-7/journalimage?img=PIIS0003999306001067.gr2.lrg.gif&fig=fig2&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
00106-7/journalimage?img=PIIS0003999306001067.gr3.lrg.gif&fig=fig3&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
00106-7/journalimage?img=PIIS0003999306001067.gr4.lrg.gif&fig=fig4&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
00106-7/journalimage?img=PIIS0003999306001067.gr5.lrg.gif&fig=fig5&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
00106-7/journalimage?img=PIIS0003999306001067.gr6.lrg.gif&fig=fig6&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)