Feedback-Controlled and Programmed Stretching of the Ankle Plantarflexors and Dorsiflexors in Stroke: Effects of a 4-Week Intervention Program

Article Outline

• Abstract
• Methods
  • Participants
  • Stretching Protocol
  • Outcome Evaluation
  • Statistical Analysis
• Results
• Discussion
• Conclusions
• Acknowledgment
• References
• Copyright

Abstract 

Selles RW, Li X, Lin F, Chung SG, Roth EJ, Zhang L-Q. Feedback-controlled and programmed stretching of the ankle plantarflexors and dorsiflexors in stroke: effects of a 4-week intervention program.

Objective

To investigate the effect of repeated feedback-controlled and programmed “intelligent” stretching of the ankle plantar- and dorsiflexors to treat subjects with ankle spasticity and/or contracture in stroke.

Design

Noncontrolled trial.

Setting

Institutional research center.

Participants

Subjects with spasticity and/or contracture after stroke.

Interventions

Stretching of the plantar- and dorsiflexors of the ankle 3 times a week for 45 minutes during a 4-week period by using a feedback-controlled and programmed stretching device.

Main Outcome Measures

Passive and active range of motion (ROM), muscle strength, joint stiffness, joint viscous damping, reflex excitability, comfortable walking speed, and subjective experiences of the subjects.

Results

Significant improvements were found in the passive ROM, maximum voluntary contraction, ankle stiffness, and comfortable walking speed. The visual analog scales indicated very positive subjective evaluation in terms of the comfort of stretching and the effect on their involved ankle.

Conclusions

Repeated feedback-controlled or intelligent stretching had a positive influence on the joint properties of the ankle with spasticity and/or contracture after stroke. The stretching device may be an effective and safe alternative to manual passive motion treatment by a therapist and has potential to be used to repeatedly and regularly stretch the ankle of subjects with spasticity and/or contracture without daily involvement of clinicians or physical therapists.

Key Words:  Biomechanics , Cerebrovascular accident , Muscle spasticity , Physical therapy , Rehabilitation

NEUROLOGIC IMPAIRMENTS SUCH AS stroke, spinal cord injury, multiple sclerosis, and cerebral palsy may lead to contractures and spasticity that can be severely disabling.¹, ², ³, ⁴ Spastic hypertonus and reflex hyperexcitability may disrupt the remaining functional use of muscles, impede motion, and cause severe pain. Lack of mobilization and prolonged spasticity may be accompanied by structural changes of muscle fibers and connective tissue, which may result in a reduction in joint range of motion (ROM) and a clinical contracture.¹, ³, ⁵, ⁶, ⁷, ⁸, ⁹

Presently, effective treatment options for contracture and/or spasticity are limited. One treatment option is passive movement or stretch of the joint. In physical therapy treatment,¹⁰, ¹¹, ¹², ¹³, ¹⁴ passive stretch can be applied by manually moving the joint through the ROM to reduce spasticity and contracture and restore movement function. However, effects may not be long lasting because of the limited and sometimes infrequent therapy, whereas manual stretching is laborious and the outcome may depend on the ability of the therapist to gauge the limits of the ROM or “end feel.”

It has been reported that continuous passive motion (ie, cyclic stretching) of the ankle joint might be effective in reducing passive ankle joint stiffness in healthy subjects.¹⁵, ¹⁶ In stroke, however, only a few studies have focused on repeated stretching of spastic joints. For example, Nuyens et al¹⁷ used an isokinetic dynamometer in 10 subjects with spasticity after stroke to perform repeated passive knee movements through the available ROM and found a decreased hypertonia. Harvey et al¹⁸ evaluated the effects of daily repeated stretching of the ankle during 4 weeks but reported no effects. Bressel and McNair¹⁶ examined the short-term effects of combined static and cyclic stretch of the calf muscles by using an isokinetic dynamometer in stroke subjects and reported decreased ankle stiffness.

To forcefully and safely stretch the ankle of subjects with spasticity and/or contracture to its extreme positions, we recently proposed an “intelligent” stretching device¹⁹ that stretches the joint with quantitative feedback control of the resistance torque and stretching velocity. In contrast to constant-speed continuous passive motion (CPM) machines and isokinetic dynamometers, in this device the stretching velocity decreases when resistance increases. In addition, although CPM machines and dynamometers stretch until a preset angle, this device forcefully stretches the joint into extreme position until a predefined resistance torque is reached. Once the predefined resistance torque is reached, the device will hold the ankle at the extreme position for a period of time to let stress relaxation occur. By using this control strategy, the stretching device moves quickly in the middle (nonspastic) ROM and slows down in the stiffer part of the ROM, while never exceeding preset stretching torques.

The purpose of this study was to investigate the effect of repeated stretching of the ankle plantar- and dorsiflexors by using a feedback-controlled and programmed stretching device during a 4-week period in a group of 10 subjects with ankle spasticity and/or contracture after stroke. The stretching was applied 3 times a week for 45 minutes, whereas the effect of stretching was evaluated in terms of ROM, joint stiffness, joint viscous damping, reflex excitability, and comfortable walking speed. In addition, the subjective experiences of the subjects were scored by using visual analog scales (VASs).

Back to Article Outline

Methods 

Participants 

Ten subjects with limited ROM in the ankle caused by chronic spasticity and/or contracture after stroke were recruited. Each subject was initially examined by using the clinical Tendon Reflex Scale (range, 0–4)¹⁹ and the Modified Ashworth Scale (MAS; range, 0–4).²⁰, ²¹ Subjects were included when the scores on both scales were 1 or higher. The study was approved by the institutional review board of Northwestern University and all subjects signed a consent form.

Stretching Protocol 

The spastic ankle of the subjects was stretched 3 sessions a week during a 4-week period. The effective stretching time in each session was approximately 45 minutes. During the 4-week stretching period, subjects were instructed to maintain any regular therapy or exercise program they were involved in but to refrain from participating in new activities.

The stretching treatment was performed by using a custom-designed joint stretching device (fig 1) while the subject was comfortably seated. The leg was strapped to a leg support with the knee flexed 30° to stretch both the gastrocnemius and soleus muscles. The foot was clamped to a footplate at the dorsal side and at the heel. The footplate was fixed to the motor shaft, and a sensor measured the dorsiflexion-plantarflexion torque. The footplate could be adjusted in all directions to align the ankle axis with the motor shaft. When needed, a custom-designed pointer was used to align the ankle joint with the rotation axis of the device.¹⁹ The stretching device could be clamped to the chair of the subject to avoid relative movement between the device and the chair.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  The intelligent stretching device used to repeatedly stretch the ankle joint in subjects with spasticity and/or contracture after stroke. The left panel shows the main components of the device, as described in the Methods section. The right panel shows the position of the subject during the stretching. The foot of the subject was fixated on the footplate, and the lower leg was fixed to a leg support. The stretching device was fixed to the chair to prevent the device from moving relative to the subject. Abbreviation: LED, light-emitting diode.

The stretching device was driven by a servomotor controlled by a digital signal processor (DSP).²², ²³, ²⁴ The exact rules of the DSP controlled motion have been described elsewhere.¹⁹ In short, the stretching velocity was inversely proportional to the resistance torque, which was monitored at 2000Hz. As a result, stretching velocity was at a maximum when the joint provided little resistance, whereas the motor gradually slowed down when resistance increased near extreme positions to stretch the muscle-tendon complex slowly and safely (for typical force and angle profiles, fig 2). Once the predefined peak resistance torque was reached, the joint was held at the extreme position for a period of time to allow stress relaxation. In the middle ROM where the resistance was usually low, the motor stretched the slack muscles at higher speeds. The DSP controller would shut down the system if a signal was out of the predefined ranges to avoid excessive stretching of the joint.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Typical data from a stretching trial, indicating (A) the dorsiflexion (DF) angle as well as (B) the joint torque during the stretching trials. It can be seen from the curves that as the ankle joint moves into extreme positions, the resistance torque increases and the stretching velocity decreases gradually until the maximum resistance torque (10Nm in this trial for both directions) is reached. After a holding period, the movement direction is reversed.

In addition to torque limits, the operator also specified position limits that should not be exceeded by a prespecified amount of motion (typical value, 5°) when stretching the ankle. The additional amount of movement beyond the position limit was allowed for potential stretching-induced improvement. Two light-emitting diodes would indicate if the torque or the position limit was reached. For additional safety, mechanical and electric stops were added to restrict the motor ROM to a physiologic range and both the operator and the subject had stop switches.

During the stretching sessions, the subjects were asked to relax and not to react to the stretch. In case a subject would react to a stretch by opposing the motion, it would reverse the stretching direction before reaching the extreme position. The maximum stretching velocity, peak resistance torque, and holding period at extreme position could be adjusted for each trial. Typical values were 30°/s peak stretching velocity, 10 to 25Nm peak resistance torque in dorsiflexion, 5 to 10Nm peak resistance torque in plantarflexion, and a 5-second holding period. These values were chosen based on estimated torques that are used by experienced physical therapists during manual stretching and on feedback from subjects during the stretching. To the best of our knowledge, no studies are available that suggest the optimal stretching torque for the ankle in stroke (see Discussion). When needed, stretching torque was adjusted for aspects such as the severity of the spasticity or contracture, the length and weight of a subject, and muscle tightness. It should be noted that the slow stretching velocity near the extreme position resulted in a total holding period in extreme and near-extreme positions of approximately 10 to 12 seconds (see fig 2).

Outcome Evaluation 

Outcome was evaluated before the first stretching trial and at the end of the 4-week period. Biomechanic and reflex properties of the spastic ankle were evaluated by using the intelligent stretching device (see fig 1). With this device, we were able to measure ankle position and torque about the dorsiflexion-plantarflexion axis, synchronized with electromyographic signals from the soleus, medial gastrocnemius, lateral gastrocnemius, and tibialis anterior muscles. All signals were sampled at 500Hz after low-pass filtering with a 230-Hz cutoff frequency. Delsys surface electrodes^a were used to record the electromyographic signals with electrode placement according to the Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles recommendations.²⁵, ²⁶

Passive ankle ROM during controlled torque and joint stiffness were evaluated by imposing the same movement pattern on the ankle as used during the stretching treatment. For each subject, at least 3 full stretching cycles were analyzed in which the ankle was stretched by using 10Nm of torque in both directions with a maximum stretching velocity of 5°/s (see fig 2). These angles and torques were averaged and plotted as a torque-angle curve (fig 3), from which the total passive ROM was determined. The quasistatic stiffness was calculated as the slope of this torque-angle curve.¹⁹, ²⁷, ²⁸ For both dorsiflexion and plantarflexion directions, we evaluated the stiffness at the neutral position, at 5-Nm resistance torque, and the minimum stiffness within the ROM. Finally, we calculated the energy loss in the hysteresis loop of the torque-angle curve as the area between the ascending and descending part of the curve to indicate joint viscosity. This energy loss was evaluated for the dorsiflexion side of the torque-angle curve (positive angles), the plantarflexion side (negative angles), and the total curve. The energy loss (in joules) was normalized by the corresponding ROM (in degrees).²⁹

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Torque-angle relation (hysteresis loop) obtained from the stretching data, indicating the relation between the ankle angle (positive indicates movement to dorsiflexion) and the resistance torque in the ankle. The curve is the average of 3 completed stretching cycles in a single subject. From the curve, the passive ROM at a controlled peak resistance torque (10Nm) was derived. The slope of the curve was used to estimate the quasistatic stiffness throughout the ROM; the joint viscosity was estimated through the energy loss in the hysteresis loop (the area enclosed).

Reflex properties of the spastic ankles were quantified by tapping the Achilles’ tendon with an instrumented hammer and measuring reflex-mediated electromyographic and joint torque responses.²³, ³⁰ Tendon taps were performed approximately 10 times at the location that elicited the maximal response with random intervals of about 3 seconds (fig 4). The ankle was fixed at the neutral position in the stretching device. Reflexes were measured isometrically to eliminate nonreflex-mediated contributions associated with limb movement. The subjects were asked to relax and not to react or anticipate the taps.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Typical example of tendon tapping data of the Achilles’ tendon from a single stroke subject. The subject was asked to relax during the tendon tapping. Average data (solid line) and SD (dotted line) from 10 taps are shown for all signals. (A) Tendon tapping force; (B) medial gastrocnemius electromyographic (EMG) response; (C) torque response in the ankle around the plantar- and dorsiflexion axis; and (D) impulse response obtained by scaling the torque response to the tendon tapping force, with the vertical lines indicating, from left to right, the start of the tendon tapping and the onset of the torque response.

Tendon tapping force and ankle plantarflexion torque were low-pass filtered digitally (cutoff, 90Hz), and the electromyographic signals were full-wave rectified and low-pass filtered to extract the linear envelope. Ankle plantarflexion torque and electromyographic signals were inspected to see whether there was any voluntary contraction. If so, the corresponding taps were excluded.

Because the tapping force was rather brief, it could be approximated as a pulse, and the reflex torque impulse response h_Mf(t) could be approximated as the reflex torque response M(t) scaled by the area of the force pulse f(t)³¹:

In addition to quantifying the reflex-mediated electromyographic and torque responses, the reflex gain was quantified by the peak impulse response and the reflex-mediated torque delay was calculated as the interval between the onsets of the tapping force and the reflex torque (see fig 3).³¹

To study the contractile properties of the muscles, we measured active ROM and maximum voluntary contraction (MVC) of the ankle at the neutral position in the stretching device. For the active ROM, the subjects were asked to repeatedly move the ankle into maximal dorsiflexion and plantarflexion in the stretching device, whereas the motor did not provide any significant resistance. For the MVC measurements, the subjects were asked to maximally push the ankle 3 times into dorsiflexion and 3 times in plantarflexion with the ankle fixed in neutral position. The maximum torque during these trials was selected.

To evaluate comfortable walking speed, we used the 10-m walking test³² on a level floor. VASs were used to score the subjects’ subjective experiences with the stretching sessions.

Statistical Analysis 

For all outcome variables, we calculated the group mean and standard deviation (SD) at baseline and follow-up. Paired t tests were used to test whether the change between baseline and follow-up was statistically significant, with a significance level of .05.

Back to Article Outline

Results 

Characteristics of the subjects are described in table 1. Spasticity of most subjects was relatively severe (median MAS score, 3), and clonus was found in most subjects when manually forcing the ankle into dorsiflexion. All subjects completed the full 4-week stretching treatment.

Table 1. Baseline Characteristics of the Subjects

The passive plantar- and dorsiflexion ROM of the ankle increased significantly with 8.6° on average (P=.001). In addition, significant decreases were found in 5 of the 6 stiffness parameters (table 2). No significant changes in the energy loss variables of the hysteresis curve were found, indicating no significant change in the viscous damping of the ankle.

Table 2. Passive ROM, Stiffness, and Energy Loss of the Spastic Ankles Measured at Baseline and Follow-Up

NOTE. Values are mean ± SD. P values indicate the results of the paired t test comparing baseline and follow-up.

Abbreviations: DF, dorsiflexion; PF, plantarflexion.

The mean total active ROM in the plantarflexor and dorsiflexor directions ± SD was 29.7°±27.8° at baseline and did not change significantly at follow-up (29.6°±30.6°, P=.989). The MVC of the subjects significantly increased in plantarflexion direction (from 8.2±6.7Nm to 9.6±6.8Nm, P=.028). The P value of the MVC in dorsiflexion direction just exceeded the significance level (from 8.6±7.6Nm to 10.6±9.7Nm, P=.053).

Comparing the reflex excitability of the spastic ankle at baseline and follow-up, we found no significant differences in the input (ie, the force of the tendon tapping [P=.988; table 3]). In addition, we did not find any significant differences in the reflex output variables.

Table 3. Reflex Excitability of the Spastic Ankle Measured From the Tendon Tapping Experiments

NOTE. Values are mean ± SD. P values indicate the results of the paired t test comparing baseline and follow-up.

Abbreviation: EMG, electromyographic.

Mean comfortable walking speed of the subjects was relatively low at baseline (mean, .52±.21m/s) but was significantly increased at follow-up to .60±.28m/s. The VAS scores indicated that the general feeling of the ankle (score range, 0 [very bad] to 100 [very good]) showed a large and statistically significant improvement (from 35.4±30.0 to 75.1±19.8, P=.021). The subjective evaluation of the stiffness of the ankle (score range, 0 [very stiff] to 100 [not stiff]) significantly changed (P=.002) from 36.9±28.4 to 73.3±4.2. Subjects reported only limited pain (score range, 0 [not painful] to 100 [very painful]) at baseline (8.0±10.6), and this pain was not significantly different at follow-up (10.3±8.8, P=.518). After the last stretching session, subjects scored their general experiences with the stretching (score range, 0 [very unpleasant] to 100 [very pleasant]) on average as 91.1±10.6.

Back to Article Outline

Discussion 

In this study, we found that repeated feedback-controlled or intelligent stretching of the plantarflexors and dorsiflexors of the ankles of subjects with spasticity and/or contracture after stroke positively affected the passive ROM, MVC, ankle stiffness, and comfortable walking speed, whereas no significant changes were found in the active ROM, the energy loss, and the reflex excitability of the spastic ankle. The VAS scores indicated very positive scores of the subjects in terms of the comfort of the stretching and the subjective evaluation of their involved ankle.

Several studies have reported positive effects of stretching subjects with joint spasticity and/or contracture after stroke. Prolonged static stretching of the plantarflexor muscles has been reported to reduce passive ankle joint resistance, increase ankle joint ROM, and improve gait characteristics in spastic ankles.³³, ³⁴, ³⁵, ³⁶ Bressel and McNair¹⁶ investigated 30 minutes of static stretching and 30 minutes of cyclic stretching of the calf muscles in stroke subjects and found a significant decrease in joint stiffness. Nuyens et al¹⁷ reported that a single repeated passive movement session in stroke subjects decreased the spastic hypertonia in the ankle through a combination of reflexive and mechanical factors. To our knowledge, this study is the first to report on multiple sessions of repeated feedback-controlled or intelligent stretching in subjects with chronic spasticity and/or contracture. Generally, the changes we found in passive ROM and joint stiffness for this multiple stretching are in line with the previously mentioned studies reporting effects for stretching on ankle stiffness, passive ROM, and comfortable walking speed.

Although most of the present literature has described the beneficial effects of stretching in joints with spasticity and/or contracture in terms of joint mechanical properties (eg, passive ROM, stiffness), not many studies studied the effects on reflex excitability. Nuyens¹⁷ evaluated electromyographic responses during the repeated passive movement and reported a decrease of spastic hypertonia in stroke subjects through a combination of reflexive and mechanical factors. Although our study confirmed a decreased stiffness after repeated stretching, we did not find changes in the reflex electromyographic response or in the reflex-mediated joint torque. It should be noted, however, that the reflex response showed a relatively large variation between subjects. In addition, electromyographic signals from different measurement sessions are known to be highly variable. More research is therefore needed to study the effect of repeated stretching on the reflex excitability of subjects with spasticity and/or contracture.

Stretching technique (position-controlled vs intelligent stretching), stretching duration, and stretching force may be important factors to consider when using muscle stretching to prevent or treat joint spasticity and/or contracture. In our study, an intelligent stretching approach was used, in which stretching velocity was adjusted based on the measured resistance and position of the joint. This control strategy assures a forceful and safe stretching treatment that can be adjusted to the needs of individual subjects. We did not compare the stretching of the plantarflexors and dorsiflexors of the ankles by using the present device with alternatives such as position-controlled passive continuous motion, manual stretching, or splinting. Therefore, further research is needed to study the relative effectiveness of these interventions. In addition, long-term follow-up measurements should be included in future studies to determine whether improvements are retained in these subjects.

From the present literature, it is not clear what stretching torque should be applied to safely obtain optimal results. Bressel and McNair,¹⁶ Harvey et al,¹⁸ and Nuyens et al¹⁷ stretched the ankle to a maximum of 7.5Nm resistance torque, whereas Yeh et al³⁶ applied a 12Nm constant stretching torque to the plantarflexors. The intelligent stretching device allowed us to differentiate between the stretching torque in dorsiflexion and plantarflexion. Peak resistance torques ranged between 10 and 25Nm in dorsiflexion direction (stretching the plantarflexors) and between 5 and 10Nm in plantarflexion direction. For the plantarflexion direction, we did not exceed 10Nm because dorsiflexors are relatively weak and less affected by spasticity. In a number of subjects with severe spasticity in the plantarflexors, based on positive subject responses, we increased the stretching torque to a maximum of 30Nm in dorsiflexion without finding any negative effects.

Despite the small number of subjects in this study, we found statistically significant changes in a relatively large number of variables. Because of the long time since the first stroke in the subjects and the chronic nature of spasticity, it can be assumed that the condition of the ankle was relatively stable and that the obtained improvements are not because of natural recovery.

Back to Article Outline

Conclusions 

Based on our results, we believe a larger and controlled study is warranted to further evaluate intelligent stretching of the spastic ankle in stroke subjects. We believe that the intelligent stretching device that was used may be an effective and safe alternative to manual passive motion treatment by a therapist and has the potential to be used to repeatedly and regularly stretch the ankle of subjects with spasticity and/or contracture without daily involvement of clinicians or physical therapists.

Suppliers

Back to Article Outline

Acknowledgment 

We thank Chris Bolt for her assistance in the data collection.

Back to Article Outline

References 

1. Bell KR , Vandenborne K . Contracture and limb deformities . In:  Lazar RB editors. Principles of neurologic rehabilitation . New York: McGraw-Hill; 1998;p. 309–328
   • View In Article
2. Lance JW . Symposium synopsis . In:  Feldman RG ,  Young RR ,  Koella WP editor. Spasticity (disordered motor control) . Chicago: Year Book Medical Publishers; 1980;p. 485–494
   • View In Article
3. Rymer WZ , Katz RT . Mechanisms of spastic hypertonia . Phys Med Rehabil (State Art Rev) . 1994;8:441–454
   • View In Article
4. O’Dwyer NJ , Ada L , Neilson PD . Spasticity and muscle contracture following stroke . Brain . 1996;119(Pt 5):1737–1749
   • View In Article
   • CrossRef
5. Tabary JC , Tabary C , Tardieu C , Tardieu G , Goldspink G . Physiological and structural changes in the cat’s soleus muscle due to immobilization at different lengths by plaster cast . J Physiol . 1972;224:231–244
   • View In Article
   • MEDLINE
6. Tabary JC , Tabary C , Tardieu C , Tardieu G , Goldspink G . Experimental rapid sarcomere loss with concomitant hypoextensibility . Muscle Nerve . 1981;4:198–203
   • View In Article
   • MEDLINE
   • CrossRef
7. Tardieu C , Tabary JC , Tabary C , Tardieu G . Adaptation of connective tissue length to immobilization in the lengthened and shortened positions in cat soleus muscle . J Physiol (Paris) . 1982;78:214–220
   • View In Article
8. Tardieu C , Lespargot A , Tabary C , Bret MD . For how long must the soleus muscle be stretched each day to prevent contracture? . Dev Med Child Neurol . 1988;30:3–10
   • View In Article
   • MEDLINE
   • CrossRef
9. Williams PE , Goldspink G . Connective tissue changes in immobilised muscle . J Anat . 1984;138:343–350
   • View In Article
10. Stokes M . Neurological physiotherapy . London: Mosby Int; 1998;
    • View In Article
11. Carr J , Shepherd R . Neurological rehabilitation (optimizing motor performance) . Oxford: Butterworth-Heinemann; 1998;
    • View In Article
12. Bohannon RW , Larkin PA . Passive ankle dorsiflexion increases in patients after a regimen of tilt table-wedge board standing. A clinical report . Phys Ther . 1985;65:1676–1678
    • View In Article
    • MEDLINE
13. Dombovy ML , Sandok BA , Basford JR . Rehabilitation for stroke (a review) . Stroke . 1986;17:363–369
    • View In Article
    • MEDLINE
    • CrossRef
14. Moseley A . The effect of casting combined with stretching on passive ankle dorsiflexion in adults with traumatic head injuries . Phys Ther . 1997;77:248–259
    • View In Article
15. McNair P , Dombroski E , Hewson D , Stanley S . Stretching at the ankle joint (viscoelastic responses to holds and continuous passive motion) . Med Sci Sports Exerc . 2001;33:354–358
    • View In Article
    • MEDLINE
    • CrossRef
16. Bressel E , McNair PJ . The effect of prolonged static and cyclic stretching on ankle joint stiffness, torque relaxation, and gait in people with stroke . Phys Ther . 2002;82:880–887
    • View In Article
    • MEDLINE
17. Nuyens GE , De Weerdt WJ , Spaepen AJ , Kiekens C , Feys HM . Reduction of spastic hypertonia during repeated passive knee movements in stroke patients . Arch Phys Med Rehabil . 2002;83:930–935
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (78 KB)
    • CrossRef
18. Harvey L , Batty J , Crosbie J , Poulter S , Herbert R . A randomized trial assessing the effects of 4 weeks of daily stretching on ankle mobility in patients with spinal cord injuries . Arch Phys Med Rehabil . 2000;81:1340–1347
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (149 KB)
    • CrossRef
19. Zhang LQ , Chung SG , Bai Z , et al.   Intelligent stretching of ankle joints with contracture/spasticity . IEEE Trans Neural Syst Rehabil Eng . 2002;10:149–157
    • View In Article
    • MEDLINE
    • CrossRef
20. Pandyan AD , Johnson GR , Price CI , Curless RH , Barnes MP , Rodgers H . A review of the properties and limitations of the Ashworth and modified Ashworth Scales as measures of spasticity . Clin Rehabil . 1999;13:373–383
    • View In Article
    • MEDLINE
    • CrossRef
21. Bohannon RW , Smith MB . Interrater reliability of a modified Ashworth scale of muscle spasticity . Phys Ther . 1987;67:206–207
    • View In Article
    • MEDLINE
22. Ahmed I . Digital control applications with the TMS320 family . Dallas: Texas Instruments; 1991;
    • View In Article
23. Zhang LQ , Rymer WZ . Simultaneous and nonlinear identification of mechanical and reflex properties of human elbow joint muscles . IEEE Trans Biomed Eng . 1997;44:1192–1209
    • View In Article
    • MEDLINE
    • CrossRef
24. Zhang LQ , Nuber GW , Butler JP , Bowen MK , Rymer WZ . In vivo human knee joint dynamic properties as functions of muscle contraction and joint position . J Biomech . 1998;31:71–76
    • View In Article
    • Full-Text PDF (112 KB)
    • CrossRef
25. Hermens HJ , Freriks B , Disselhorst-Klug C , Rau G . Development of recommendations for SEMG sensors and sensor placement procedures . J Electromyogr Kinesiol . 2000;10:361–374
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (437 KB)
    • CrossRef
26. Hermens HJ , Freriks B , Merletti R , et al.   European recommendations for surface ElectroMyoGraphy. Results of the SENIAM project . Enschede: Roessingh Res Dev; 1999;
    • View In Article
27. Olmstead TG , Wevers HW , Bryant JT , Gouw GJ . Effect of muscular activity on valgus/varus laxity and stiffness of the knee . J Biomech . 1986;19:565–577
    • View In Article
    • MEDLINE
    • CrossRef
28. Zhang LQ , Wang G . Dynamic and static control of the human knee joint in abduction-adduction . J Biomech . 2001;34:1107–1115
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (405 KB)
    • CrossRef
29. Chung SG , Van Rey E , Bai Z , Roth EJ , Zhang LQ . Biomechanic changes in passive properties of hemiplegic ankles with spastic hypertonia . Arch Phys Med Rehabil . 2004;85:1638–1646
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (371 KB)
    • CrossRef
30. Zhang LQ , Wang G , Nishida T , Xu D , Sliwa JA , Rymer WZ . Hyperactive tendon reflexes in spastic multiple sclerosis (measures and mechanisms of action) . Arch Phys Med Rehabil . 2000;81:901–909
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (134 KB)
    • CrossRef
31. Zhang LQ , Huang H , Sliwa JA , Rymer WZ . System identification of tendon reflex dynamics . IEEE Trans Rehabil Eng . 1999;7:193–203
    • View In Article
    • MEDLINE
    • CrossRef
32. Salbach NM , Mayo NE , Higgins J , Ahmed S , Finch LE , Richards CL . Responsiveness and predictability of gait speed and other disability measures in acute stroke . Arch Phys Med Rehabil . 2001;82:1204–1212
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (162 KB)
    • CrossRef
33. Odeen I , Knutsson E . Evaluation of the effects of muscle stretch and weight load in patients with spastic paraplegia . Scand J Rehabil Med . 1981;13:117–121
    • View In Article
    • MEDLINE
34. Tsai KH , Yeh CY , Chang HY , Chen JJ . Effects of a single session of prolonged muscle stretch on spastic muscle of stroke patients . Proc Natl Sci Counc Repub China B . 2001;25:76–81
    • View In Article
    • MEDLINE
35. Brouwer B , Davidson LK , Olney SJ . Serial casting in idiopathic toe-walkers and children with spastic cerebral palsy . J Pediatr Orthop . 2000;20:221–225
    • View In Article
    • MEDLINE
    • CrossRef
36. Yeh CY , Chen JJ , Tsai KH . Quantitative analysis of ankle hypertonia after prolonged stretch in subjects with stroke . J Neurosci Methods . 2004;137:305–314
    • View In Article
    • MEDLINE
    • CrossRef

• a DelSys Inc, 650 Beacon St, 6th Fl, Boston, MA 02215.