Modification of motor control of wrist extension by mesh-glove electrical afferent stimulation in stroke patients

252

Modification of Motor Control of Wrist Extension by MeshGlove Electrical Afferent Stimulation in Stroke Patients Meta M. Dimitrijevid, MD, Dobrivoje S. Stokid, MD, Artur W. Wawro, MS, Chuan-Chuan C. Wun, PhD ABSTRACT. Dimitrijevi6 MM, Stoki6 DS, Wawro AW, Wun C-CC. Modification of motor control of wrist extension by mesh-glove electrical afferent stimulation in stroke patients. Arch Phys Med Rehabil 1996;77:252-8. Objective: To study the effect of mesh-glove afferent stimulation on motor control of voluntary wrist movement in stroke patients who have chronic neurological deficits. Design: Case series. Motor control was evaluated by surface EMG of the arm muscles and kinematics of voluntary wrist movements on 3 occasions: before and immediately after the initial session of mesh-glove stimulation, and then after a daily mesh-glove stimulation program conducted over several months. Setting: Tertiary care center. Patients: The inclusion criteria were: a history of stroke lasting longer than 6 months; completion of a rehabilitation program during early recovery; and preserved cognitive and communicative ability. Fourteen referred patients (age 63 _+ 9yr; time since stroke 31 _+ 22mo) fulfilled the criteria and completed the daily stimulation program. Intervention: A single initial and then daily mesh-glove electrical afferent stimulation was applied to the hand of the involved upper limb for 20 to 30min. Main Outcome Measures: Surface EMGs from the affected biceps brachii and wrist extensor muscles and amplitudes of wrist movements were analyzed. Results: The single, initial mesh-glove application had no effect on outcome measures. Following a daily mesh-glove stimulation program, however, both the amplitude of wrist extension movement and wrist extensor integrated EMG were significantly increased while coactivation of biceps brachii decreased. These findings were most prominent in subjects with partially preserved voluntary wrist movements. Conclusion: We conclude that daily mesh-glove stimulation can modify altered motor control and improve voluntary wrist extension movement in stroke subjects with chronic neurological deficits.

© 1996 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation

From the Division of Restorative Neurology and Human Neurobiology, Baylor College of Medicine (Drs. Dimitrijevi6 and Stokir), Houston, TX; the Department of Neurology, Warsaw Medical School (Mr. Wawro), Warsaw, Poland; and the Department of Biometry, University of Texas School of Public Health (Dr. Wun), Houston, TX. Submitted for publication March 27, 1995. Accepted in revised form August 15, 1995. Supported by the Vivian L. Smith Foundation for Restorative Neurology. 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 authors or upon any organization with which the authors are associated. Reprint requests to Meta M. Dimitrijevir, MD, Division of Restorative Neurology and Human Neurobiology, Baylor College of Medicine ($800), One Baylor Plaza, Houston, TX 77030. © 1996 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/96/-/703-346353.00/0

Arch Phys Med Rehabil Vol 77, March 1996

N ACUTE cerebrovascular accident (CVA) can result in an initial loss of motor functions followed by abnormally active proprioceptive responses that are clinically recognized as more brisk tendon reflexes and increased resistance to passive stretch. ~ Recovery of voluntary movements occurs gradually, accompanied by an increase in muscle strength and a progressive decrease of spasticity. Movements of distal limb joints and independent finger movements are the last to be recovered. Motor recovery is a progressive process that may cease at any time and patients can be categorized in different functional performance groups according to the level of motor improvement. Major recovery in arm motor functions occurs within the first 3 months, 2'3 slowly reaching a plateau during the sixth month after the onset of hemiplegia.l'4-9 It has been reported, however, that specific training procedures can result in the further improvement of sensory m'~ and motor functions. ~z'~3Several electrical stimulation paradigms have been found to be effective. For example, Fields TM reported a beneficial effect when combining electromyographically initiated electrical stimulation of the paretic wrist extensor muscles with conventional therapy. Functional electrical stimulation (FES) has been found to reduce cocontraction in spastic biceps brachii. '5 More recently, Levin and Hui-Chan '6 demonstrated improvement in reflex and voluntary motor functions of the affected leg in stroke patients following daily transcutaneous electrical nerve stimulation (TENS) applied to the peroneal nerve. Kraft and colleagues ~7 showed that chronic stroke patients who reach a plateau in recovery can achieve additional functional improvement of the affected arm and hand, especially when low-intensity electrical stimulation of wrist extensor muscles is combined with voluntary efforts. In these studies, electrical stimulation was generally applied either to paretic muscle groups or to specific nerves of the affected limb. We recently described a new method, "mesh-glove" wholehand electrical afferent stimulation. ~8 In our preliminary study, we reported that the application of mesh-glove stimulation to the affected hand after stroke resulted in the reduction of muscle tone and facilitation of volitional finger movements. 19 The purpose of this study was to determine whether electrical stimulation of hand afferents could modify electromyography (EMG) output of wrist extensors and biceps brachii muscles as well as the amplitude of volitional wrist extension movement. We studied the effects of mesh-glove stimulation in 14 subjects, 8 months and more after the onset of stroke, before and after weeks or months of daily mesh-glove stimulation.

A

MATERIAL Criteria for participation in this study were: (1) a history of stroke of more than 6 months; (2) completion of a rehabilitation program during early recovery; and (3) preserved cognitive and communicative ability. Five women and nine men were included after giving informed consent. The time from the onset of stroke ranged from 8 to 69 months (34.3 _+ 21.8) (table 1). All subjects underwent a comprehensive rehabilitation program that was initiated immediately after the stroke. This program focused on the prevention of complications due to paralysis such as glenohumeral subluxation, shoulder-hand syndrome, and soft tissue contractures. Spasticity management included positioning techniques

MESH-GLOVE STIMULATION IN CHRONIC STROKE, Dimitrijevi(~

253

Table 1: Characteristics of 14 Patients Divided Into 4 Categories According to Recovery of Voluntary Wrist Movements Sex Category M M M Category F F M F Category F F Category M M M M M

Age(yr)

Location

Side

Type

Deficit

Onset(too)

Follow-Up(mo)

56 66 55

Hemis Bstem Hemis

R L L

Inf Inf Inf

Mot Sen-mot Mot

69 8 18

10 7 5

62 49 50 77

Both Bstem Hemis Hemis

R R R R

Hem Hem Inf Inf

Sen-mot Sen-mot Mot Mot

64 58 12 32

2 2 10 2

65 66

Bstem Hemis

R R

Inf Inf

Sen-mot Sen-mot

45 19

3 3

57 62 77 78 68

Hemis Hemis Bstem Hemis Hemis

L L L R L

Inf Hem Inf Inf Hem

Mot Mot Mot Sen-mot Sen-mot

69 9 17 33 27

4 2 3 5 4

Protocol*

1: Good IV II IV

2: Partial

3: Trace

4: Absent

Mean age, 63 _+ 9yr; mean time since onset, 31 _+ 22mo; mean follow-up, 3.9 _+ 2.4mo. Abbreviations: Hemis, hemisphere; Bstem, brainstem; Inf, infarction; Hem, hemorrhage; Mot, motor; Sen-mot, sensorimotor. * Stimulation protocol at the time of the third assessment.

and static splinting, and passive and active assistive range of motion exercises were added to the program. Occupational therapy interventionsl were tailored individually to enhance upper extrenuty funcuoqs; these included tradittonal motor strengthening and functional regimens as well as facilitation/inhibition techniques. None of ~ e subjects had received prior electrical stimulation or biofeedback treatment. Subjects selected were free of any secondary changes such as contractures, fixed deformities, or pain. According to the extent of recovery of voluntary wrist motor functions, 4 categories were established: (1) good wrist extension movement, greater than 75% of normal active range of motion (aROM) (3 subjects); (2)partial wrist extension movement, between 25% to 75% of normal aROM (4 subjects); (3) trace of wrist extension, which measured less than 25% of normal aROM (2 subjectS); and (4) absent wrist extension movement (5 subjects). According to neurological findings, the subjects were further separated into 2 groups, the first group with motor hemiparesis only and th6 second with sensory and motor dysfunction (table 1). In the ~ategories with good and partial recovery of wrist extension movement, the affected hand was functional in 3 subjects and partially functional in 4 subjects, respectively. In the remaining 7 subjects the affected hand was nonfunctional when the subject was performing daily activities. •

.

I

.

.

.

METHODS Recording Technique Motor control in the paretic arm was evaluated by simultaneous recording of surface EMG activity from several arm muscles and kinematic measures of voluntary wrist movements. Each subject was comfrrtably seated at a table that was adjusted to the subject's height. The forearm of the paretic limb was positioned on the tablet in pronation with the shoulder in slight abduction and flexion, the elbow in approximately 60 ° flexion and the hand relaxed over the edge of the table. The metacarpophalangeal joint (MCP) of the fifth digit and wrist joint were marked and positioned on the center of the side-view camera. Video recording was displayed on a monitor and recorded on video tape. Pairs of surface Ag-AgC1 ~lectrodes (lcm in diameter) were placed 3cm apart (cathode proximal, anode distal) and oriented parallel to the long axis of the following muscles or muscle groups bilaterally: (1) Biceps brachii was identified with the forearm flexed in supination, and the surface electrodes were placed over the muscle bulk in the mid-arm. (2) The lateral head of the triceps brachii

was identified with the ann adducted and extended; the cathode was placed two fingerbreadths posterior to the insertion of the deltoid muscle and the anode 3cm distally. (3) For the wrist extensor muscle group (extensor carpi radialis longus and brevis), the forearm was fully pronated and the cathode was placed two fingerbreadths distal to the lateral epicondyle over the mid-dorsal aspect of the forearm, and the anode 3cm distally. (4) For the wrist flexor muscle group (flexor carpi radialis), the forearm was fully supinated and the cathode was positioned four fingerbreadths distal to the midpoint of a line connecting the medial epicondyle and the biceps tendon with the anode 3cm distally toward the base of the second metacarpal bone. The electrode placement area was cleaned with an alcohol pad and then the skin was gently robbed about the diameter of the electrode, using a cotton swab moistened with the abrasive solution,a Surface electrodes, filled with a conductive jelly, were placed over the prepared area and secured with tape. The electrode pair impedance was maintained below 5kfl throughout the recording. The same investigator conducted all assessment sessions and verified the appropriate placement of the electrodes prior to each recording. EMG signals were amplified (gain 5,000), bandpass filtered (50 to 1,000Hz) and displayed on an inkjet stripchart recorder along with a manually triggered event marker. Simultaneously, full bandwidth EMG data were recorded on the computer at 2,000samples/sec. The whole EMG recording was integrated off-line using a root-mean-square (RMS) algorithm and integrated EMG (IEMG) was used for later analysis. During each assessment session, the subject was instructed to perform 3 consecutive repetitions of isolated maximal wrist extension-wrist flexion movement of the affected hand at a comfortable speed (approximately 3sec each)• Peak wrist extension was a starting point for the wrist flexion movement. The examiner manually triggered the event marker to indicate the onset and the end of each repetition of the extension and flexion movements, respectively. If the subject was unable to generate a visible wrist movement, the event marker indicated the onset and end of the verbal command for the movement. Only wrist extension movement was analyzed and reported here because it is likely that, due to gravity, the passive wrist flexion following wrist extension confounded the EMG output of the flexion movement. Stimulation Protocols Details of the mesh-glove stimulation method and protocols have been previously described) 8 Briefly, a mesh-glove b (fig 1)

Arch Phys Med Rehabil Vol 77, March 1996

254

MESH-GLOVE STIMULATION IN CHRONIC STROKE, Dimitrijevi~

stimulation could induce recovery of movement. Protocol IV was carried out only in the subjects with preserved voluntary hand movements (table 1). After the completion of the daily mesh-glove stimulation program, all 14 subjects underwent the third assessment session.

Data Analysis

Fig 1. Mesh glove stimulation method (healthy subject shown): Mesh glove (A) is connected as a common anode (B); two carbon rubber electrodes, placed over the dorsal (C) and ventral (hidden) aspects of the forearm, respectively, serve as cathodes of the two-channel stimulator (D).

made of conductive wire was fitted over the affected hand and a pair of 4 × 3cm surface electrodes were placed 2cm proximal to the wrist, over the volar and dorsal aspects of the forearm, respectively. The mesh-glove was connected as a common anode while surface electrodes served as cathodes of a twochannel stimulator,c Following first (baseline) assessment of EMG activity and voluntary wrist movements, each subject underwent an initial 20-minute treatment session of continuous synchronous stimulation (stimulation from both channels simultaneously) at 50Hz (pulse duration 300/zsec) with current adjusted just below the threshold for sensory perception, during which the recording electrodes were kept in place. To examine any immediate effect of this initial mesh-glove stimulation, the assessment protocol was subsequently repeated (second assessment session). Following these two assessment sessions, the subject was enrolled in a daily mesh-glove stimulation program which consisted of t8: Protocol I: 20 to 30min continuous synchronous two-channel stimulation just below the sensory threshold; Protocol II: 20 to 30min continuous synchronous two-channel stimulation at the sensory threshold; Protocol III: reciprocal two channel stimulation (duty cycle 5 to 15sec) at the motor twitch level gradually increasing stimulation from 5 to 30rain; Protocol IV: stimulation triggered with a hand switch to induce reciprocal finger extension and flexion movements synchronized with the residual volitional finger movements. The whole stimulation program was designed in such a way that each protocol should be applied daily (once or twice) between 2 and 10 months (table 1). Initially, all subjects carried out Protocol I for 4 to 6 weeks. However, several clinical observations were taken into consideration during the course of this study when deciding the duration and protocol for stimulation. For example, if spasticity in the hand and forearm was the main finding, then Protocol I would be used throughout the treatment. In attempt to further enhance motor functions, in Protocol II we added cutaneous to kinaesthetic input for subjects in whom muscle hypertonia was successfully controlled using Protocol I. Protocol III was applied only in one instance, in the subject who had no movement (category 4) in an attempt to determine whether this level of

Arch Phys Med Rehabil Vol 77, March 1996

EMG (RMS) data were analyzed using customized software that calculated the integrated EMG (IEMG) over a window manually selected from the event marker to indicate the onset and the end of the wrist extension movement (see Recording Technique). For comparison between the three assessment sessions, IEMG over the entire movement window (#Vs) was time averaged (divided by movement duration) and, therefore, the IEMG values were reported in terms of the average RMS voltage (/~V) for each of 3 movement repetitions within the same assessment session. This procedure allowed us to eliminate the difference in IEMG caused by a slight discrepancy in the duration of the active movements within and between the subsequent assessment sessions. EMG analysis was based on the following: (1) wrist extensor (EX) IEMG; (2) biceps hrachii (BI) IEMG; and (3) the coactivation index (CI), which was calculated as BI IEMG divided by the BI and EX 1EMGs, ie, CI = BIJ(BI + EX). The coactivation index was used to describe a well-known synergy pattern that occurs after stroke, an unintentional coactivation of the proximal muscles that accompanies the movement of distal limb segments. The usefulness of such defined coactivation index for quantitative EMG evaluation was shown by Hammond and associates 2° in their needle EMG study of cocontraction of the antagonistic forearm muscles after stroke. The analysis of movement focused on two kinematic parameters: angle of maximal wrist extension (maxE), and the range of active wrist extension/flexion (rE/F) motion. By definition, the angle of maximal wrist extension represented the angle between the line that connected the markers on the fifth MCP joint and wrist joint in relation to a vertical line passing through the wrist (fig 2). The range of wrist extension/flexion was measured as the angle between the full wrist extension and the full wrist flexion movements for each repetition. After reading the positions of the markers using a coordinate system displayed on the monitor screen, both parameters were calculated using simple trigonometric functions. Before the initial application of mesh-glove stimulation, in the first assessment session, we recorded baseline EMG and kinematic data for 14 subjects. We then applied 20 minutes of mesh-glove stimulation according to Protocol I, ie, below sensory threshold. Immediately after the stimulation, in the second assessment session, we repeated the EMG and kinematic assessments in all but three subjects (subjects 1, 9, and 10; table 1). Several months after initiation of the mesh-glove program (3.9 _+2.4 months [mean _+ SD]), we carried out the third assessment session in all 14 subjects. The long-term effect was determined by comparing the results of the first and the third assessment sessions. The various stimulation protocols that the subjects used at the time of the third assessment are listed in table 1. Statistical Analysis Each of three movement repetitions recorded during the first, second, and third assessment sessions was treated as an independent measurement in order to account for intrarecording individual variation. Repeated measures analysis of variance (ANOVA) was used to test the difference in EMG and kinematic parameters between the three assessment sessions. Where appropriate, Student's t test was used for paired data. The Pearson's product-moment correlation coefficient and intraclass correlation coefficient (ICC) 2~ were used to correlate the changes

255

MESH-GLOVE S T I M U L A T I O N IN CHRONIC STROKE, Dimitrijevid

markedly reduced BI IEMG and CI (p < . 06). At the same time, kinematic analysis revealed that those EMG findings were accompanied by an increase in maximal wrist extension movement (p < .01) and range of wrist extension/flexion motion (p < .06).

Reference I

Extension

,'

Kinematics of Wrist Movements After Daily Mesh-Glove Stimulation P r o g r a m

Wrist Joint

m

/ii! !exi" n :::

.::

,

I

::iiii!!i~: ::~iiiii~:

|

....~:iii~i::.!iiiiiii!~::=:~::::?i~ ................?::iiii;~

| I

Fig 2. Amplitude of maximal wrist extension (maxE) represents the angle between t h ~ l i n e t h a t connected MCP and wrist joints and the reference; range of qctive wrist extension/flexion (rE/F) depicts the range between full wrist extension and full wrist flexion.

between two kindmatic parameters and to assess the repeatability of EMG a n d kinematic data within the same assessment session. Group ayeraged IEMG and kinematic data were expressed as standard error of means (mean _-2-SE). The probability level of p < . 05 was considered significant. RESULTS The results presented are based on a comparison between the first (baseline) agsessment and second and third assessments, carried out after the initial treatment session and after the longterm daily mesh-glove stimulation program, respectively. Initial T r e a t m e n t Session Versus Daily Mesh-Glove Stimulation Program Eleven out of 14 subjects completed all three assessment sessions, as described above. IEMG values for EX and BI, as well as CI and kinematics of wrist movement, were not significantly different between the first and second assessment sessions (table 2). However, the third assessment session, carried out after a lo0g-term program of daily mesh glove stimulation, revealed a s!gnificant increase of EX IEMG (p < . 05) and

During the first assessment session, volitional wrist movements were found to be present only in 9 of 14 subjects and in those, maxE ranged between 17° to 139 °, and rE/F between 8 ° to 104 °. The long-term effect in these 9 subjects was determined in the third assessment session, which showed a significant mean increase of 14% _+ 5% in maxE, from 79.5 ° _+ 8.2 ° to 86.8 ° ___ 8.1 ° (F = l l . 9 ; df = 1,26; p < .01). This effect was significantly different among the 3 categories of wrist movements (F = 9.8; df = 2,24; p < .01). The most prominent difference was found in subjects with partial and trace wrist movements, categories 2 and 3, respectively (group mean increase of 20% and 25%, respectively). MaxE remained unchanged in the category of good wrist extension movement (category 1). In these 9 subjects, a 12% mean increase in rE/F reached borderline significance (p < .06), and the differences observed in maxE and rE/F following a long term program of daily mesh-glove stimulation correlated significantly (linear fit, r = .72; p < .01). This result supports the presence of a consistent improvement in maximal wrist extension that is primarily responsible for the difference in the range of wrist extension/ flexion motion. The five subjects in whom wrist movement was initially absent did not respond to the long-term program of daily mesh glove stimulation. W r i s t Extensor I E M G For the subjects in categories with volitional movements (categories 1, 2, and 3), as expected, the mean amplitude of EX IEMG differed markedly during the first assessment session. After a daily mesh-glove stimulation program, there was a significant mean increase in EX IEMG activity for all 14 subjects (EX~ = 43.3 _+ 8.6 and EX3 = 55.9 - 9.2~V; F = 6.5; df = 1,40; p < .05). This effect was significantly different among the four categories (F = 6.2; df = 3,37; p < .01). Further analysis revealed a significant increase of IEMG in the category with partial movement (category 2) and trace movement (category 3), mean increase of 125% and 77%, respectively (p < .01) (fig 3).

Biceps Brachii IEMG and Coactivation Index (CI) Baseline assessment showed that coactivation of the biceps brachii muscle during the wrist extension movement, as measured by CI, was inversely related to the recovery of wrist motor function. After long term mesh-glove stimulation, we found a significant reduction in biceps IEMG for all 14 subjects (group mean BL = 27.3 _+ 5.4 and BI3 = 20.8 _+ 3.8#V; F = 4.5; df

Table 2: EMG and Kinematic Results (mean _+ SE) in 11 Subjects Seen on 3 Occasions Integrated EMG (/zV)

Wrist Kinematics (°)

Assessment Session

Extensor

Biceps

CI (%)

maxE

rE/F

1 2 3

50.7 _+ 10.6 56.5 + 10.8 64.7 _+ 11.2"

25.1 _+ 6.9 19.9 _+ 3.9 17.9 _+ 4.5 t

39.4 + 5.1 38.3 _+ 4.9 34.3 _+ 5.0 t

79.5 _+ 8.2 75.9 _+ 8.6 86.8 _+ 8.2 ~

59.1 +_ 7.7 55.9 _+ 7.7 64.1 _+ 7.4 t

Second and third assessment sessions w e r e c o m p a r e d to the first (baseline) assessment. Abbreviations: CI, C0activation Index BI/(BI + EX); maxE, a m p l i t u d e of m a x i m a l w r i s t extension; rE/F, range of active wrist extension/flexion. * p < .05. t p < .06. p < .01.

Arch Phys Med Rehabil Vol 77, March 1996

256

MESH-GLOVE STIMULATION IN CHRONIC STROKE, Dimitriievi~

160 140

>v•

120

8

~oo

z

80

B

60 40

Fz_ 20

GOOD

PARTIAL

TRACE

ABSENT

Fig 3. Integrated EMG (IEMG) recorded from surface electrodes over the wrist extensor muscle group in 4 categories of recovery of wrist movement, before (1st session, 0) and after (3rd session, I ) the completion of daily mesh glove stimulation program (**p < .01).

= 1,40; p < .05). Accordingly, CI was significantly decreased (CI~ = 46.6% _+ 4.5% and CI3 = 40.2% ___ 4.3%; F = 8.9; df = 1,40; p < .01). Again, the effect was significantly different among the four categories (F = 6.7; df = 3,37; p < .01): although the CI was unchanged in categories of pre-existing nearly complete recovery or wrist paralysis, it was noticeably decreased in subjects with partial movements (CI, = 45.8% _ 6.4% and CI3 = 28.1% _ 6.3%; p < .07) and to a lesser extent in the category with trace wrist movements (fig 4). To summarize, after a daily mesh-glove stimulation program, there was a remarkable decrease in BI IEMG and CI in subjects with partial and trace of wrist extension movement (fig 4) and both categories showed a significant increase in EX IEMG activity (fig 3). These results clearly illustrate that the increase in EX IEMG is accompanied by reduced coactivation of the proximal, biceps muscle. In addition, the behavior of CI in the third recording differed significantly according to the type of neurological deficit (F = 7.2; df = 1,39; p < .01). Although CI was unchanged in subjects with motor deficits, this index decreased from 47.4% +_ 6.3% to 35.7% _ 6.1% in the subjects with sensory-motor deficits.

DISCUSSION This study shows that in stroke subjects with chronic deficits, a long-term program of dally mesh-glove stimulation of the affected hand can elicit an increase in EMG activity in the prime mover wrist extensor muscles and a decrease in EMG in coactivated, synergistic elbow flexor muscles. These findings were only observed in subjects with some volitional control of wrist movement and not in paralyzed subjects. Moreover, changes in EMG output were paralleled by the results in kinematic analysis, which showed a significant increase in the amplitude of maximal wrist extension and borderline increase in the range of wrist extension/flexion motion. With the improvement of the wrist extension movement, unintentional coactivation of the brachii muscle was decreased (figs 3 and 4). Coactivation of antagonistic and more proximal muscles is a well-recognized clinical and neurophysiological finding after upper motor neuron dysfunction, stroke in particular. Knutsson 22 pointed out that there are 3 distinct features of altered motor control in subjects with spastic hemiparesis: decreased stretch reflex threshold, decreased muscle activation, and stereotypical coactivation of muscles. Coactivation occurs as a part of synergistic patterns during early recovery of voluntary movements and gradually diminishes as the recovery process evolves. Therefore, an increase in agonist IEMG and decrease in CI, as demonstrated in our study, suggest a modification in motor control of the paretic upper limb during performance of wrist extension movement. A remarkable increase in EX IEMG and a significant decrease in CI after long-term daily mesh-glove stimulation was detected in a group of subjects with clinical evidence of sensorimotor deficit, whereas such effects were lacking in those with pure motor impairment. It is obvious that, in the population of subjects with sensorimotor deficit, it was necessary to use stronger stimulation to reach sensory threshold than in subjects with motor deficits only. Therefore, it can be argued that larger afferent input, due to stronger electrical stimulation, might have caused a more pronounced effect in the group with sensorimotor deficit. On the other hand, application of different stimulation protocols that, in fact, consisted of a progressive increase in current intensity had no similar effect in subjects with motor impairment only. Thus, it is more likely that pathophysiological conditions were a major factor contributing to the efficacy of stimulation. To summarize, it appears that a long-term dally mesh-glove stimulation of the affected hand can improve the

Repeatability of IEMG and Kinematic Data During each assessment session, as mentioned before, all subjects performed three successive repetitions of maximal wrist extension movement. The repeatability of those three repetitions within the first and third assessment sessions, respectively, was assessed using Pearson's correlation coefficient and intraclass correlation coefficient (ICC). For all 14 subjects, linear correlation coefficients (r) between any pair of 3 repetitions (R1-R2, R2-R3 and R1-R3) within the same assessment session ranged from .92 to .99 for the above reported IEMG values (EXl, EX3, CI., CI3). A similar correlation was found for both kinematic parameters in 9 subjects with volitional wrist movements (r = .90-.99). Because correlation coefficient indicates the degree of linear interdependence between two repetitions but does not manifest possible offset between them, intraclass correlation coefficient (ICC) was also calculated for the same EMG and kinematic parameters. The ICCs ranged from 92% to 98%, indicating that only 2% to 8% of the total variance was caused by trial-to-trial variability, thus confirming the high repeatability of the data recorded within the first and third assessment sessions, respectively. Arch Phys Med Rehabil Vol 77, March 1996

80 ~" 70 _~ 60 ~ ~ 50 _z _~ 40 ~ 30 < 20 8 10 0

GOOD

PARTIAL

TRACE

ABSENT

Fig 4. Coactivation Index in 4 categories of wrist movement recovery, before (1st session, O) and after (3rd session, I ) the complstion of daily mesh glove stimuiation program. Borderline significance (p < .07) was found in the category with partial wrist movement.

MESH-GLOVE STIMULATION IN CHRONIC STROKE, Dimitrijevi~

pattern of EMG activity in the paretic upper limb and increase the amplitude of voluntary wrist extension in stroke subjects with chronic neurological deficit. The beneficial effect of stimulation was especially seen in subjects with sensorimotor dysfunction. The validity of our results, obtained with surface EMG recording, seems to be supported by the following observations: (1) consistent findings of reciprocal IEMG output between EX and BI and (2)iparallel findings for the improvement in EX IEMG motor output and amplitude of wrist extension movement. For correct interpretation of our EMG results, it was vital to ensure reproducible placement of the surface electrodes throughout the gntire study. The single investigator who conducted all assessment sessions ensured such reproducibility. Other researchers have confirmed high reliability of the data when surface electrodes are removed and then properly reapplied. 23 The stability of surface EMG recordings in our laboratory has been previously reported 24 and 5% to 10% of variation is in agreement with othersY We believe that with our reproducible electrode placement and high intrarecording repeatability, it was possible t o separate the systematic effects induced by the treatment intervention from the unsystematic error of measurement. In this study, we did not normalize the IEMG to any particular refererlce, eg, isometric maximal voluntary contraction of the studied muscles. We believed that such normalization technique, particularly in our stroke subjects, might be another source of error d~e to motivation, fatigue, or paralysis itself. In addition, such ngrmalization procedures have been questioned when studying multilayer muscles that can participate in different movements. 2~ The purpose of mesh-glove stimulation is to simultaneously depolarize large diameter afferent fibers of the entire volar and dorsal aspects of the hand. At a stimulation level below the threshold for sensory perception (Protocol I), it is likely that primary activatirn occurs in musculotendinous and joint afferents that are known to mediate kinaesthetic sensation. 26 In general, afferent stimulation was a common feature in all subjects regardless of the iactual current intensity. Considering supraspinal and peripheral afferent convergence onto spinal interneurons, 27depolarization of large diameter afferent fibers (primarily activated by eledtrical stimulation) has been found effective in modifying the s~gmental and suprasegmental excitability levels z8-3~and mighl be responsible for increased presynaptic inhibition, if presyaaptic inhibition is indeed diminished after stroke. 32 If the irifluence of afferent stimulation is mediated by those mechanisms, it is reasonable to expect that such an effect can be seen during or immediately after the stimulation. Indeed, that was the case in our preliminary report, in which we described a successful decrease in muscle hypertonia in the paretic upper limb durin' a single session of mesh-glove stimulation at g 19 • • the subsensory level. In the experiments of Levln and C h a n ,3"~ a single session of TENS applied over the nerve trunk resulted in segmentally Or plurisegmentally prolonged increase of the stretch reflex thrgshold up to 60 minutes, without resulting in an obvious cliniqal reduction of spasticity. It is likely that these effects are mediated predominantly through the spinal reflex circuits. On theother hand, it should be noted that weeks of daily mesh-glove stimulation were necessary to improve wrist extension movement, which suggests that the underlying mechanisms facilitatil)g the improvement in voluntary wrist motor control are different from those that are involved in the suppression of spasticity. Plasticity of flae CNS has been recognized as a part of the structural and pl~ysiological substrate for recovery of function after brain injury134'35Dynamic remodelling processes occurring in the sensorimo~or cortex of humans and experimental animals

257

have been recently demonstrated. It has been shown that internal topography of the somatosensory cortical representation of stimulated skin surfaces of monkey's digits was enlarged after behaviorally controlled tactile stimulation) 6 Brasil-Neto and colleagues 37 showed that cortical motor output in intact humans can be rapidly modulated by reduced input from large myelinated afferents. Conversely, increased sensory input in blind subjects who were proficient Braille readers resulted in an expansion of the sensorimotor cortical representation of the reading index finger, as demonstrated by larger scalp area for evoking the early components of somatosensory evoked potentials when compared with the opposite side) 8 All these findings provide neurophysiological evidence that cortical reorganization may ensue as a consequence of prolonged sensory input. For our study in particular, it is appropriate to point out the experimental findings of Strick and Preston, 39'4° who found that motor representation areas for evoking a wrist extension movement receive cutaneous input from wide receptive fields located over the volar aspect of the fingers. The evidence for cortical plasticity, in the presence of such functional organization, may provide some insight into the neurophysiological basis for improved wrist movements and underlying neural control following a daily program of mesh-glove stimulation, as observed in this study. The purpose of this study was to explore whether we could provide supportive laboratory evidence for our preliminary observations that movement, when present, can be augmented by whole-hand stimulation) 9 Therefore, in this present study, we designed strict laboratory protocols for the measurement of EMG output and movement of the hand, and applied different modalities and length of mesh-glove stimulation purposefully in subjects with a wide variety of neurological deficit. The described laboratory results have confirmed our previously reported clinical observations that motor control of the volitional hand movements can be altered by mesh-glove stimulation. The results obtained in this study warrant further examination of the effect of proprioceptive and proprio-exteroceptive sensory stimulation in group of subjects after stroke, selected according to the site and nature of the lesion. More in depth knowledge must be acquired about the effectiveness and optimal duration of the stimulation protocols for the enhancement of residual motor control in stroke patients. In conclusion, this study demonstrates that daily mesh-glove, whole-hand electrical afferent stimulation can modify voluntary motor control and increase wrist extension motion in the paretic limb in stroke subjects with chronic neurological deficit. We hope that the encouraging results in this study will promote further interest in studying the effect of externally controlled afferent input to the CNS in an attempt to develop new restorative procedures in stroke patients with various neurological impairments. Acknowledgments: The authors thank the Vivian L. Smith Foundation for Restorative Neurology, which provided financial support for this work, and express their gratitude to Milan R. Dimitrijevir, MD, DSc, for his guidance and critical comments in preparing the manuscript. References

I. Twitchell TE. The restoration of motor function following hemiplegia in man. Brain 1951;74:443-80. 2. Wade DT, Wood VA, Langton Hewer R. Recovery after stroke-the first 3 months. J Nenrol Neurosurg Psychiatry 1985;48:7-13. 3. Heller A, Wade DT, Wood VA, Sunderland A, Langton Hewer R, Ward E. Arm function after stroke: measurement and recovery over the first three months. J Neurol Neurosurg Psychiatry 1987;50:7149. 4. Bard G, Hirschberg GE. Recovery of voluntary motion in upper

Arch Phys Med Rehabil Vol 77, March 1996

258

5.

6. 7. 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

23.

24.

MESH-GLOVE STIMULATION IN CHRONIC STROKE, Dimitrijevi(~

extremity following hemiplegia. Arch Phys Med Rehabil 1965;64: 567-72. De Souza LH, Lengton Hewer R, Miller S. Assessment of recovery of ann control in hemiplegic stroke patients. 1. Arm function tests. Int Rehabii Med 1980;2:3-9. Andrews K, Brocklehurst JC, Richards B, Laycock PJ. The rate of recovery from stroke and its measurement. Int Rehabil Med 1981;3:155-61. Skilbeck CE, Wade DT, Lengton Hewer R, Wood VA. Recovery after stroke. J Neurol Neurosurg Psychiatry 1983;46:5-8. Wade DT, Langton-Hewer R, Wood VA, Skilbeck CE, Ismail HM. The hemiplegic arm after stroke: measurement and recovery. J Neurol Neurosurg Psychiatry 1983;46:521-4. Parker VM, Wade DT, Lengton Hewer R. Loss of arm function after stroke: measurement, frequency and recovery. Int Rehabil Med 1986;8:69-73. Goldman H. Improvement of double simultaneous stimulation perception in hemiplegic patients. Arch Phys Med Rehabil 1966;47: 681-7. Guttman YE. A controlled trial of the retraining of the sensory function of the hand in stroke patients. J Neurol Neurosurg Psychiatry 1993;56:241-4. Prevo AJH, Visser SL, Vogelaar TW. Effect of EMG feedback on paretic muscles and abnormal co-contraction in the hemiplegic arm, compared with conventional physical therapy. Scand J Rehabil Med 1982; 14:121-31. Tangeman PT, Banaitis DA, Williams AK. Rehabilitation of chronic stroke patients: changes in functional performance. Arch Phys Med Rehabil 1990;71:876-80. Fields RW. Electromyographically triggered electric muscle stimulation for chronic hemiplegia. Arch Phys Med Rehabil 1987;68:40714. Lagasse PP, Roy MA. Functional electrical stimulation and the reduction of co-contraction in spastic biceps brachii. Clin Rehabil 1989;3:111-6. Levin MF, Hui-Chan CWY. Relief of hemiparetic spasticity by TENS is associated with improvement in reflex and voluntary motor functions. Electroencephalogr Clin Neurophysiol 1992; 85:131-42. Kraft GH, Fitts SS, Hammond MC. Techniques to improve function of the arm and hand in chronic hemiplegia. Arch Phys Med Rehabil 1992;73:220-7. Dimitrijevi6 MM. Mesh-glove. 1. A method for the whole-hand electrical stimulation in upper motor neuron dysfunction. Scand J Rehabil Med 1994;26:183-6. Dimitrijevi6 MM, Soroker N. Mesh-glove. 2. Modulation of residual upper limb motor control after stroke with whole-hand electrical stimulation. Scand J Rehabil Med 1994; 26:187-90. Hammond MC, Fitts SS, Kraft GH, Nutter PB, Trotter MJ, Robinson LM. Co-contraction in the hemiparetic forearm: Quantitative EMG evaluation. Arch Phys Med Rehabil 1988;69:348-51. Bartko JJ. The intraclass correlation coefficient as a measure of reliability. Psychol Rep 1966; 19:3-11. Knutsson E. Analysis of gait and isokinetic movements for evaluation of antispastic drugs on physical therapies. In: Desmedt JE, editor. Motor control mechanisms in health and disease. New York: Raven Press, 1983:1013-34. Giroux B, Lamontange M. Comparisons between surface electrodes and intramuscular wire electrodes in isometric and dynamic conditions. Electromyogr Clin Neurophysiol 1990;30:397-405. Sherwood AM, Dimitrijevic MR, McKay WB. Stability of multichannel surface EMG recordings for studies of motor control in

Arch Phys Med Rehabil Vo177, March 1996

25. 26. 27.

28.

29. 30. 31. 32. 33. 34.

35.

36.

37.

38. 39. 40.

upper motor neuron dysfunction. In: Pedotti A, editor. Electrophysiological kinesiology. Proceeding of the ISEK '92 Congress; 1992 June 28-July 2; Florence, Italy. Amsterdam: lOS Press, 1993:95. Kadaba MP, Wootten ME, Gainey J, Cochran GVB. Repeatability of phasic muscle activity: performance of surface and intramuscular wire electrodes in gait analysis. J Orthop Res 1985;3:350-9. Gandevia SC. Illusory movements produced by electrical stimulation of low-threshold muscle afferents from the hand. Brain 1985; 108:965-81. Baldissera F, Hultborn H, Illert M. Integration in spinal neuronal systems. In: Brooks VB, editor. Handbook of physiology. Section 1: The nervous system, vol. II. Motor control, part I. Bethesda (MD): American Physiological Society, 1981:509-95. Delwaide, PJ, Crenna P, Fleron MH. Cutaneous nerve stimulation and motoneuronal excitability: I. Soleus and tibialis anterior excitability after ipsilateral and contralateral sural nerve stimulation. J Neurol Neurosurg Psychiatry 1981;44:699-707. Delwaide, PJ, Crenna P. Cutaneous nerve stimulation and motoneuronal excitability: II. Evidence for non-segmental influences. J Neurol Neurosurg Psychiatry 1984;47:190-6. Malmgren K, Pierrot-Deseilligny E. Inhibition of neurones transmitting non-monosynaptic Ia excitation to human wrist flexor motoneurones. J Physiol (Lond) 1988;405:765-83. Malmgren K, Pierrot-Deseilligny E. Evidence for non-monosynaptic Ia excitation of human wrist flexor motoneurones, possibly via propriospinal neurones. J Physiol (Lond) 1988;405:747-64. Ashby P, Verrier M. Neurophysiologic changes in hemiplegia. Neurology 1976;26:1145-51. Levin MF, Chan CWY. Stretch reflex latency changes following repetitive reciprocal and hetero-segmental stimulation in spastic hemiplegic subjects. Soc Neurosci Abst 1989; 15:916. Marshall JF. Brain function: neural adaptations and recovery from injury. In: Rosenzweig MR, Porter LW, editors. Annual review of psychology. Palo Alto (CA): Annual Reviews Inc, 1984;35:277308. Jenkins WM, Merzenich MM. Reorganization of neocorticai representation after brain injury: a neurophysiologieal model of the bases of recovery from stroke. In: Seil FJ, Herbert E, Carlson BM, editors. Progress in brain research. New York: Elsevier Science Publishers BV, Biomedical Division, 1987;71:249-66. Jenkins WM, Merzenich MM, Ochs MT, Allard T, Guic-Robles E. Functional reorganization of primary somatosensory cortex in adult owl monkeys after behaviorally controlled tactile stimulation. J Neurophysiol 1990;63:82-103. Brasil-Neto JP, Vails-Sole J, Pascual-Leone A, Cammarota A, Amassian VE, Cracco R, et al. Rapid modulation of human cortical motor outputs following ischaemic nerve block. Brain 1993; 116:511-25. Pascual-Leone A, Torres F. Plasticity of the sensorimotor cortex representation of the reading finger in Braille readers. Brain 1993; 116:39-52. Strick PL, Preston JB. Two representations of the hand in area 4 of a primate. I. Motor output organization. J Neurophysiol 1982; 48:139-49. Strick PL, Preston JB. Two representations of the hand in area 4 of a primate. II. Somatosensory input organization. J Neurophysiol 1982;48:150-9.

Suppliers a. Omniprep; D. O. Weaver & Co., 565-C Nucla Way, Aurora, CO 80011. b. Electro-Mesh glove electrode; Prizm-Medical Inc. Norcross, GA 30071. c. Respond II; Empi Inc., PO Box 64640, St Paul, MN 55164-0640.