AFTER A CEREBROVASCULAR ACCIDENT (CVA), Effect of an Electric Stimulation Facilitation Program on Quadriceps Motor Unit Recruitment After Stroke

Transcription

1 2040 Effect of an Electric Stimulation Facilitation Program on Quadriceps Motor Unit Recruitment After Stroke Craig J. Newsam, DPT, Lucinda L. Baker, PT, PhD ABSTRACT. Newsam CJ, Baker LL. Effect of an electric stimulation facilitation program on quadriceps motor unit recruitment after stroke. Arch Phys Med Rehabil 2004;85: Objective: To compare maximum voluntary isometric torque (MVIT) and motor unit recruitment of the quadriceps after an electric stimulation facilitation program in persons affected by cerebrovascular accident (CVA). Design: Three-week, randomized controlled trial with an electric stimulation facilitation program added to standard care. Setting: Inpatient rehabilitation center. Participants: Twenty patients receiving rehabilitation for first-time CVA ( y; days post-cva, d). Patients were randomly assigned to study and control groups. Interventions: All patients received standard physical therapy (PT) care. In addition, the study group received an electric stimulation facilitation program during weight-bearing and ambulatory activities of the PT program. Main Outcome Measures: MVIT and motor unit recruitment measured by interpolated twitch testing. A 2 4 repeatedmeasures analysis of variance was performed on measurements at 4 intervals: pretest, 1 week, 2 weeks, and 3 weeks. Results: MVIT increased by 77% in patients receiving electric stimulation, compared with a 31% increase for the control group. There was a significant effect for assessment time only. Motor unit recruitment increased from 35% to 53% for the study group, whereas the control group recorded no change in recruitment ability. A significant interaction was recorded, indicating improved motor unit recruitment for the study group. Conclusions: A brief and dynamic electric stimulation facilitation program significantly improved motor unit recruitment in persons after CVA. Key Words: Electric stimulation; Motor activity; Rehabilitation; Stroke by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation AFTER A CEREBROVASCULAR ACCIDENT (CVA), the ability to generate muscle force is often impaired. Reduced force generation has 2 sources. First, the ability to activate the muscle is compromised by poor descending motor control. A second complication is muscle weakness related to From the Pathokinesiology Laboratory, Rancho Los Amigos National Rehabilitation Center, Downey, CA (Newsam); and Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA (Newsam, Baker). 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(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Lucinda L. Baker, PT, PhD, Dept of Biokinesiology and Physical Therapy, University of Southern California, 1540 E Alcazar St, CHP 155, Los Angeles, CA 90033, llbaker@usc.edu /04/ $30.00/0 doi: /j.apmr disuse muscle atrophy after a period of decreased activity. A common therapeutic intervention designed to enhance muscle function after CVA is electric stimulation. Prolonged electric stimulation programs (lasting 4 6h daily) effectively produce muscle fiber hypertrophy in persons with central nervous system (CNS) lesions, resulting in increased muscle force generation. 1 This type of aggressive stimulation program, however, is not feasible in the clinical environment. By contrast, stimulation programs performed during shorter time periods also have shown strengthening benefits after stroke. 2 For persons in the chronic post-cva phase ( 3mo), 4 weeks of electric stimulation has produced muscle strength gains between 39% and 61%. 3,4 In persons between 1.5 and 2 months poststroke, Winchester et al 5 showed considerably greater effects after 4 weeks of stimulation of the quadriceps. Knee extension increased to 575% postintervention, compared with a 155% increase in the control group. Carnstam et al 6 reported increased isometric dorsiflexion strength after only 10 minutes of peroneal nerve stimulation in persons with stroke. The short time periods and moderate intensity of these stimulation programs 3-6 suggest that strength gains were primarily achieved through improved motor unit recruitment. To date, no studies have attempted to differentiate neural (ie, motor unit recruitment) and hypertrophic effects of electric stimulation in people post-cva. It was hypothesized that volitional strength gains after a brief electric stimulation facilitation program for patients in the early post-cva period would primarily result from an increased ability to recruit motor units. METHODS Participants Patients admitted for rehabilitation at Rancho Los Amigos National Rehabilitation Center (RLANRC) were recruited for participation. Recruitment was restricted to patients who had sustained a first-time, unilateral CVA. Twenty patients agreed to participate and signed a statement of informed consent approved by the institutional review board. Nine men and 11 women were randomly assigned to the study and control groups. Mean age of participants was 51.8 years (range, 24 78y), and the average time since CVA was 38.4 days (range, 2 180d). All participants showed at least minimal voluntary control of knee extension of the hemiparetic lower extremity (manual muscle test score, 1/5 or greater). Patients were excluded if they showed severe spasticity of the quadriceps femoris muscle as determined by the Ashworth Scale. Persons with documented history of peripheral nerve injury or disease also were excluded from participation. Therapeutic Protocol All patients received the standard physical therapy (PT) care on the inpatient stroke service at RLANRC. Functional goals focused on improving transfer ability to a stand-by assistance

2 EFFECT OF ELECTRIC STIMULATION ON MOTOR UNIT RECRUITMENT, Newsam 2041 level and on improving ambulation to limited community or household status. PT treatments were planned on an individual basis by the treating therapist. Therapeutic sessions incorporated functional training with treatments directed at impairments such as weakness, range of motion limitations, impaired motor control, and decreased balance, with an emphasis on improving functional status. All participants received a minimum of 1 hour of PT, 6 days a week. Patients assigned to the study group (n 10) received an electric stimulation facilitation program during the standard PT care, 5 days per week. Stimulation was applied during gait training or weight shifting activities only. When necessary, upper-extremity assistive devices and ankle-foot orthoses were used to provide stability and to ensure safety. Stimulation was delivered via 2 self-adhesive electrodes a (3 5in) placed over the motor points of the vastus lateralis and the vastus medialis oblique. The treatment stimulator b provided a train of symmetric biphasic square wave pulses (35 pulses/s; phase duration, 220 s). Administration of the stimulation train was controlled by a hand trigger, activated by the investigator during the stance phase of gait or during weight-shifting activities. Ramps of the stimulation train were adjusted for patient comfort during weight-shifting activities ( 2s). During gait activities, ramps were minimized to provide a more immediate muscle response ( 0.5s). Stimulation amplitude was adjusted on an individual patient basis, as determined by the treating investigator. Stimulation intensity was targeted at the minimal amplitude necessary to provide control of knee extension during weight-bearing activities. Initially, some patients were unable to tolerate the intensity necessary to achieve the desired motion. In these instances, stimulation amplitude was set to produce a maximal tolerated muscle contraction. As tolerance to the stimulation developed, intensity was adjusted daily until the desired response was attained. As the patient s functional capabilities improved, more difficult activities were introduced (eg, stair climbing). In addition, stimulation intensity was continually assessed, to determine the minimal amplitude necessary to provide controlled knee extension during function. By the end of the 3-week treatment intervention, most patients in the study group required only sensory cues for muscle timing from the stimulator. Testing Protocol Maximal voluntary isometric torque (MVIT) and interpolated twitch of the hemiparetic knee extensors were measured 4 times during the investigation: a pretreatment assessment, 2 interim assessments, and a posttreatment assessment. Two pretreatment assessments were performed on successive days to determine the influence of a novel testing situation on a voluntary effort. Interim assessments were conducted after 1 week and 2 weeks of treatment. Posttreatment assessment was performed after 3 weeks of treatment on 2 successive days, similar to pretreatment testing. MVIT and interpolated twitch were performed with the patient seated comfortably on the chair of a stationary dynamometer. c The back support was reclined to position the hip in 80 of flexion. A padded cuff was secured proximal to the malleoli and was attached to a lever arm that placed the knee in 60 of flexion. A self-adhesive strap was fastened around the patient and the back support, to secure the pelvis and to maintain the testing posture during maximal effort. Before recording the maximal effort, 3 practice trials at a moderate effort were performed, to provide warm-up for the tested muscle. Next, 3 maximal effort tests were performed, Fig 1. Schematic representation of the interpolated twitch test. each lasting 4 seconds. Patients received visual feedback related to the torque produced from a computer monitor and verbal encouragement throughout the maximal effort. A minimum of 1 minute rest was provided between moderate and maximal effort trials. For the purpose of the interpolated twitch, a short-duration, supramaximal electric stimulus was delivered at approximately 3 seconds into each of the maximal tests. The supramaximal stimulus was delivered by using a portable electric stimulator, d and electrodes were placed over the motor points for the vastus lateralis and vastus medialis oblique. The stimulus consisted of a 200-ms train of symmetric biphasic square wave pulses. Pulse frequency was 99 pulses per second (total pulses, 20) and phase duration was set at 300 s. Amplitude was adjusted to provide a supramaximal contraction of the quadriceps (80mA). Data Management Isometric torque data produced on the dynamometer were stored on a dedicated personal computer and printed in graphic form. The sampling rate of the dynamometer computer was 100 samples per second. Graphical output representative of the torque produced during testing were digitized on a personal computer by using SigmaScan software. e Peak torque measured before the delivery of the supramaximal stimulus was the dependent variable for MVIT. Digitization for the measurement of peak torque from a single trial was repeated 5 separate times by the same investigator. The high and low values were discarded, and the remaining 3 digitization measurements were averaged, to represent the peak MVIT for that trial. The dependent variable for interpolated twitch testing was the percentage of motor unit recruitment. This was calculated by dividing the peak torque produced during the MVIT by the peak torque produced during the supramaximal stimulus (MVIT superimposed twitch supramaximal contraction). This ratio was then multiplied by 100, to be expressed as a percentage of motor units active during the voluntary effort (fig 1). A similar procedure of repeated digitization was used to determine the torque produced during the supramaximal contraction. Data Analysis The age and time since CVA of study and control groups were compared by using a 2-sample t test. To assess the

3 2042 EFFECT OF ELECTRIC STIMULATION ON MOTOR UNIT RECRUITMENT, Newsam Table 1: Demographic Data for Study and Control Groups Groups Age (y) Time Since CVA (d) Side of CVA (L/R) Gender (M/F) Study /5 4/6 Control /3 4/5 Total /8 8/11 NOTE. Values are mean SD. Abbreviations: F, female; L, left; M, male; R, right. stability of the MVIT and interpolated twitch, Pearson productmoment correlation coefficients were calculated to determine the association between day 1 and day 2 values from the preand posttest assessments. A 2 4 repeated-measures analysis of variance (ANOVA) was used to compare changes in MVIT and motor unit recruitment between groups and assessment intervals. Group assignment (study, control) was the betweengrouping factor, and assessment time (pre, 2 interim, post) was the within-group factor. Tukey honestly significant difference post hoc testing was used to identify significant comparisons when ANOVA revealed significance. Additionally, a nonparametric t test was used to compare the change in the supramaximal contraction torque between the study and control groups. Comparisons were considered significant when a P value less than.05 was obtained. RESULTS Preliminary screening of the initial torque and motor unit recruitment data revealed that 1 patient in the control group had considerably greater knee extensor function than all other patients. This patient had an initial MVIT that was more than 2 standard deviations (SDs) greater than the group mean. In addition, motor unit recruitment ability for this patient was twice that of the group s mean. At the posttreatment assessment, knee extensor MVIT and motor unit recruitment also were more than double that of the remaining patients. Data from this individual, therefore, were excluded from the final statistical analysis. The remaining 19 patients in the study group (n 10) and in the control group (n 9) had similar age (52.5y and 50.7y, respectively). The study group had a longer time since CVA (46.9d) than the control group (31.7d); however, the statistical comparison was not significant (table 1). The repeatability of a maximal voluntary effort recorded on days 1 and 2 of the pretreatment assessment was excellent. Mean MVIT standard error of the mean (SEM) for all 19 subjects on day 1 of the pretest was ft-lb and ft-lb on day 2. Statistical comparison between repeated testing did not differ significantly, and there was an excellent correlation at r equal to.89. Similar results were obtained for the comparison of MVIT recorded on days 1 and 2 of the posttreatment assessment (r.96). The greatest MVIT from 6 trials (3 trials over 2 testing sessions) was recorded as the maximal value for both the pretreatment and posttreatment assessments and was subsequently used for repeated-measures analysis. The repeatability of the supramaximal contraction torque recorded during interpolated twitch was excellent across the 2 pretest and posttest assessments. Mean supramaximal torques at days 1 and 2 of the pretest were ft-lb and ft-lb, respectively (r.99). At the 2 posttest assessments, supramaximal contraction torques were 43.9 and 45.1ft-lb (r.99). Maximal Voluntary Isometric Torque At the pretreatment assessment, patients in the study and control groups had a similar mean MVIT, at 13.5 and 16.1ft-lb, respectively. After the 3-week intervention, patients in the study group increased their mean MVIT by 77% to 23.9ft-lb (fig 2). Nearly half this increase was obtained at the week 1 interim assessment (mean MVIT score, 18.1ftlb). For patients in the control group, mean MVIT also increased after 3 weeks, however, to a lesser extent. Mean MVIT increased by just 31% to 21.0ft-lb at the posttreatment assessment. Gains in MVIT for the control group were more gradual than those observed in the study group (wk 1, 18.0ft-lb; wk 2, 19.1ft-lb). Repeated-measures ANOVA determined that there was a statistically significant effect for assessment time; however, there was no group effect or group-by-time interaction. Supramaximal Contraction Torque At the posttest, supramaximal contraction torque increased in 7 of the study group subjects (range, ftlb) and decreased in the remaining 3 subjects (range, ft-lb). For the control group, 4 subjects had increased posttest supramaximal contraction torque (range, ft-lb), 4 subjects had a decrease (range, ft-lb), and 1 subject was unchanged. The average supramaximal contraction torque for the study group increased from 37.3 to 45.7ft-lb during the 3-week intervention, with a median increase of 10.75ft-lb (interquartile range [IQR], 17.2). For the control group, the average supramaximal torque increased from 41.6 to 46.5ft-lb, and the median change score was 0.0ft-lb (IQR, 5.10). Nonparametric statistical testing determined that the change in supramaximal torque for the study group was significantly greater than that recorded in the control group (P.05). Motor Unit Recruitment Interpolated twitch testing indicated that the ability to activate motor units during maximal voluntary effort was initially similar between the study and control groups (pretreatment mean recruitment, 34.7% and 29.5%, respectively). For patients in the study group, improvements in recruitment ability were documented at each of the subsequent assessments (fig 3). Mean motor unit recruitment increased to 45.2% by week 1 and to 50.9% by week 2. At the end of the 3-week intervention Fig 2. Mean knee extension MVIT for study and control groups at 4 assessment times: pretreatment, white; week 1, stripped; week 2, gray; and posttreatment, black. Error bars represent SEM.

4 EFFECT OF ELECTRIC STIMULATION ON MOTOR UNIT RECRUITMENT, Newsam 2043 Fig 3. Mean motor unit recruitment ability for study and control groups at 4 assessment times: pretreatment, white; week 1, stripped; week 2, gray; and posttreatment, black. Error bars represent SEM. *Significantly different from pretreatment; **significantly different from pretreatment and from control group at same assessment time. period, mean recruitment of the study group had increased to 52.8% of complete motor unit activation. For patients in the control group, motor unit recruitment was unchanged at the interim and posttreatment assessments (28.8%, 26.0%, 28.6%). Repeated-measures ANOVA determined that there was a statistically significant effect for assessment time and group-bytime interaction. Simple main effect for group did not differ significantly. Post hoc testing revealed that significant differences were related to improvements in recruitment observed in the study group. At the week 2 and posttreatment assessments, the study group showed significant improvement in recruitment compared with pretreatment assessment. In addition, the study group had significantly greater recruitment ability than that of the control group at both interim assessments and at the posttreatment assessment. DISCUSSION After CVA, residual limb paresis can greatly affect function Accordingly, rehabilitative efforts are targeted at increasing muscle force production. Reduced stays in the inpatient rehabilitation setting, however, limit the ability to produce significant muscle hypertrophy. Consequently, interventions that can rapidly enhance motor recruitment are essential for the recovery of muscle force production. In our study, the addition of an electric stimulation facilitation program to a standard rehabilitation intervention significantly improved quadriceps motor unit recruitment after stroke. Initially, participants were able to volitionally activate approximately 33% of the total motor units available, as determined by interpolated twitch testing. After the 3-week intervention, however, individuals treated with electric stimulation were able to recruit more than 50% of the total motor units available. Most of this improvement in recruitment ability occurred after only 1 week of treatment, which suggests that clinical benefits can be realized with a brief electric stimulation facilitation program. The addition of an electric stimulation facilitation program also yielded considerable gains in voluntary force generating capacity. Over the course of the 3-week intervention period, mean MVIT increased by 77% in the study group but by only 31% in the control group. This more than 2-fold differential in MVIT change was likely functionally beneficial for the patients in the study group 10 ; however, no functional measures (eg, gait velocity or balance assessment) were performed in our investigation. Similar to the changes in recruitment ability, most of the improvement in MVIT occurred after only 1 week of the electric stimulation facilitation program. Strength gains associated with exercise have historically been attributed to 2 factors: enhanced neural recruitment and muscle fiber hypertrophy. In their classic study, Moritani and devries 11 described a linear relationship between improvements in voluntary force production and electromyographic intensity during the first 2 weeks of a voluntary exercise program. This linear relationship did not persist, however, because voluntary force-generating capacity continued to improve without a concomitant increase in electromyographic intensity. Moritani and devries concluded that the immediate strength gains were related to enhanced neural recruitment, whereas gains made later in the exercise program were attributed to muscle fiber hypertrophy. Our investigation, however, also documented a significant increase in the supramaximal contraction torque at the posttest, which suggests that muscle hypertrophy occurred after only 3 weeks. Additionally, the group receiving electric stimulation showed a significantly greater change in supramaximal contraction torque than that recorded in the control group. These findings, along with evidence for increased activity of metabolic and genetic markers for muscle hypertrophy after a single bout of exercise, 12,13 suggest that the timeline for muscle hypertrophy may differ for persons with considerable disuse atrophy. Use of electric stimulation for the purpose of increasing voluntary muscle force output has been well substantiated. Delitto and Snyder-Mackler 14 have proposed 2 theories to explain the mechanism by which electric stimulation increases muscle strength. The first proposed mechanism is related to the intensity of the muscle contraction produced during the stimulation training program (ie, overload principle). In healthy subjects, Currier and Mann 15 established a minimal stimulation training intensity of 60% MVIT for yielding significant strength gains. Similarly, Snyder-Mackler et al 16 identified a linear relationship between stimulation training intensity (% MVIT of uninvolved limb) and quadriceps strength recovery after anterior cruciate ligament reconstruction. The strength gains reported in our investigation were likely not related to overload of the muscle because training intensities were very low. An alternative mechanism by which electric stimulation may have improved MVIT relates to physiologic differences between voluntary muscle contractions and those produced by electric stimulation. 14 Externally applied electric stimulation preferentially activates large-diameter motoneurons, innervating fast-twitch muscle fibers and effectively reversing recruitment order. Consequently, motor units that are typically activated with only the most aggressive voluntary exercise programs are activated first with stimulated contractions. Targeted activation of fast-conducting motor units has been proposed as a means of enhancing voluntary force production at lower stimulation training intensities. 17 A final consideration of how electric stimulation may have increased motor unit recruitment and MVIT was the effect of stimulation on the sensory system. As previously described, stimulation intensity was set to the minimal amplitude necessary, to provide control of knee extension, and most patients required only sensory cues by the end of the 3-week intervention. Trimble and Enoka 17 reported that submotor electric stimulation did not increase the peak twitch force during H-

5 2044 EFFECT OF ELECTRIC STIMULATION ON MOTOR UNIT RECRUITMENT, Newsam reflex testing, which suggests that segmental-level sensory drive did not augment muscle output. Electric stimulation, however, provides a multimodal bombardment of the CNS, 18 which results in increased cortical activity 19 and improved sensory awareness. 20 For individuals who have sustained a stroke, the nonspecific nature of the sensory drive potentially increased the likelihood of successfully activating cortical awareness. The increased muscle performance shown in the study group could directly benefit patients receiving rehabilitation for stroke. Increased MVIT and motor unit recruitment allows for greater possibilities for volitional exercise or functional training opportunities. 21 Patients in the study group increased from 11% of normal knee extension torque to 19% of normal. 22 This would translate roughly to an increase in manual muscle test score from 3/5 to 3 /5 for people with selective control. 23 Continued improvement in recruitment also could be expected once the initial gains were realized. For example, the control subject dropped from the statistical analysis showed considerable improvement without use of electric stimulation. This likely reflects the greater initial ability. Logically, a goal of an electric stimulation program would be to provide immediate gains in muscle performance, followed by a transition to a program more reliant on volitional exercise and functional training. Another area of potential benefit to patients receiving rehabilitation for stroke would be functional gains. Although no direct measurement of function was recorded in our study, other investigations have established a relationship between knee extensor strength and function. Suzuki et al 24 reported that knee extensor strength was the best predictor of maximal walking speed in patients with hemiparesis after 8 weeks of gait training. Additionally, knee extensor strength gains have been associated with improvements in the limits of postural stability in middle-aged and elderly persons. 25 The stimulation program described in our study was both brief and dynamic. Daily treatment sessions lasted no more than 40 minutes and were incorporated into a standard therapy session. The greatest improvements in muscle performance were measured after just 1 week of treatment. The intent of this stimulation facilitation program was to continually challenge the patient s motor control capacity. As motor control improved, stimulation amplitude required to adequately stabilize the knee during weight bearing was decreased to a sensory level. It is unclear, however, whether maintaining stimulation amplitude at the initial, higher setting would have translated to continued enhancement of muscle performance. CONCLUSIONS An electric stimulation facilitation program integrated into weight-bearing activities can significantly increase quadriceps recruitment in persons recovering from stroke. An increase in the supramaximal contraction torque at the posttest suggests that the timeline for muscle hypertrophy may differ for persons with disuse atrophy. Acknowledgments: We thank Dr. Jans Axelgaard and Axelgaard Manufacturing for the donation of the stimulation electrodes used in this investigation. References 1. Munsat T, McNeal D. Effects of nerve stimulation on human muscle. Arch Neurol 1976;33: Glanz M, Klawansky S, Stason W, Berkey C, Chalmers TC. Functional electrostimulation in poststroke rehabilitation: a metaanalysis of the randomized controlled trials. Arch Phys Med Rehabil 1996;77: Levin MF, Hui-Chan CW. Relief of hemiparetic spasticity by TENS is associated with improvement in reflex and voluntary motor functions. Electroencephalogr Clin Neurophysiol 1992;85: Merletti R, Zelaaschi F, Latella D, Galli M, Angeli S, Sessa MB. A control study of muscle force recovery in hemiparetic patients during treatment with functional electrical stimulation. Scand J Rehabil Med 1978;10: Winchester P, Montgomery J, Bowman B, Hislop H. Effects of feedback stimulation training and cyclical electrical stimulation on knee extension in hemiparetic patients. Phys Ther 1983;63: Carnstam B, Larsson LE, Prevec TS. Improvement of gait following functional electrical stimulation. I. Investigations on changes in voluntary strength and proprioceptive reflexes. Scand J Rehabil Med 1977;9: Sharp SA, Brouwer BJ. Isokinetic strength training of the hemiparetic knee: effects on function and spasticity. Arch Phys Med Rehabil 1997;78: Bohannon RW, Andrews AW. Relationship between impairments in strength of limb muscle action following stroke. Percept Mot Skills 1998;87: Weiss A, Suzuki T, Bean J, Fielding RA. High intensity strength training improves strength and functional performance after stroke. Am J Phys Med Rehabil 2000;79: Kim CM, Eng JJ. The relationship of lower extremity muscle torque to locomotor performance in people with stroke. Phys Ther 2003;83: Moritani T, devries HA. Neural factors versus hypertrophy in the time course of muscle strength gain. Am J Phys Med 1979;58: Booth FW, Tseng BS, Fluck M, Carson JA. Molecular and cellular adaptation of muscle in response to physical training. Acta Physiol Scand 1998;162: Goldspink G. Changes in muscle mass and phenotype and the expression of autocrine and systemic growth factors by muscle in response to stretch and overload. J Anat 1999;194: Delitto A, Snyder-Mackler L. Two theories of muscle strength augmentation using percutaneous electrical stimulation. Phys Ther 1990;70: Currier DP, Mann R. Muscular strength development by electrical stimulation in healthy individuals. Phys Ther 1983;63: Snyder-Mackler L, Delitto A, Stralka SW, Bailey SL. Use of electrical stimulation to enhance recovery of quadriceps femoris muscle force production in patients following anterior cruciate ligament reconstruction. Phys Ther 1994;74: Trimble MH, Enoka RM. Mechanisms underlying the training effects associated with neuromuscular electrical stimulation. Phys Ther 1991;71: Baker LL, Wederich C, McNeal D, Newsam CJ, Waters RL. Neuromuscular electrical stimulation: a practical guide. 4th ed. Downey (CA): Los Amigos Research and Educational Institute; Golaszewski S, Kremser C, Wagner M, Felber S, Aichner F, Dimitrijevic MM. Functional magnetic resonance imaging of the human motor cortex before and after whole-hand afferent electrical stimulation. Scand J Rehabil Med 1998;31: Perennou DA, Leblond C, Amblard B, Micallef JP, Herisson C, Pelissier JY. Transcutaneous electric nerve stimulation reduces neglect-related postural instability after stroke. Arch Phys Med Rehabil 2001;82: Alon G, Dar A, Katz-Behiri D, Wingarden H, Nathan R. Efficacy of a hybrid upper limb neuromuscular electrical stimulation system in lessening selected impairments and dysfunctions consequent to cerebral damage. J Neurol Rehabil 1998;12:73-80.

6 EFFECT OF ELECTRIC STIMULATION ON MOTOR UNIT RECRUITMENT, Newsam Murray MP, Gardner GM, Mollinger LA, Sepic SB. Strength of isometric and isokinetic contractions. Phys Ther 1980;60: Beasley WC. Quantitative muscle testing: principles and applications to research and clinical services. Arch Phys Med Rehabil 1961;42: Suzuki K, Imada G, Iwaya T, Handa T, Kurogo H. Determinants and predictors of the maximum walking speed during computerassisted gait training in hemiparetic stroke patients. Arch Phys Med Rehabil 1999;80: Ryushi T, Kumagai K, Hayase H, Abe T, Shibuya K, Ono A. Effect of resistive knee extension training on postural control measures in middle aged and elderly persons. J Physiol Anthropol Appl Human Sci 2000;19: Suppliers a. PALS neurostimulation electrodes; Axelgaard Manufacturing Co, Fallbrook, CA b. Focus Electrical Stimulator; EMPI Inc, 599 Cardigan Rd, St. Paul, MN c. Lido Active; Loredan Inc, Davis, CA d. UltraStim; Neuromedics Inc, Clute, TX e. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL