You are on page 1of 10

Neurorehabilitation and Neural

Repair
http://nnr.sagepub.com/

Validation of a New Biomechanical Model to Measure Muscle Tone in Spastic Muscles
Påvel G. Lindberg, Johan Gäverth, Mominul Islam, Anders Fagergren, Jörgen Borg and Hans Forssberg
Neurorehabil Neural Repair 2011 25: 617 originally published online 13 April 2011
DOI: 10.1177/1545968311403494
The online version of this article can be found at:
http://nnr.sagepub.com/content/25/7/617

Published by:
http://www.sagepublications.com

On behalf of:

American Society of Neurorehabilitation

Additional services and information for Neurorehabilitation and Neural Repair can be found at:
Email Alerts: http://nnr.sagepub.com/cgi/alerts
Subscriptions: http://nnr.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - Aug 4, 2011
OnlineFirst Version of Record - Apr 13, 2011
What is This?

Downloaded from nnr.sagepub.com at Karolinska Institutets Universitetsbibliotek on August 4, 2014

2. stroke. sagepub. MSc.1177/1545968311403494Lindb Reprints and permission: http://www. although it is a common and important symptom after a brain injury.sagepub. there is also a nonneural resistance caused by inertia. Jörgen Borg. A biomechanical model was created to estimate the components of the force resisting passive hand extension. PT. these clinical scales are still in practice since they are inexpensive and easy to use. Results. these approaches are difficult to apply in the clinic11 and do not discriminate between neural and mechanical (elasticity and 1 Karolinska Institute.7-10 Another disadvantage is that the Ashworth scale only measures the total force and does not allow separation of the neural and nonneural components. Stockholm. Keywords spasticity.nav Research Articles Validation of a New Biomechanical Model to Measure Muscle Tone in Spastic Muscles Neurorehabilitation and Neural Repair 25(7) 617­–625 © The Author(s) 2011 Reprints and permission: http://www.6 However. Sweden 3 Karolinska University Hospital. (b) elasticity (EC).com Påvel G. which is composed of the neural activation (stretch reflex) and the viscous resistance in the tissues being stretched12-15. Mominul Islam. elasticity. and (c) NC was velocity dependent. and it is routinely assessed to describe the neurological condition of the patient. (c) electromyography (EMG)17-19. PT. MSc1.403494 erg et alNeurorehabilitation and Neural Repair © The Author(s) 2011 NNR25710. the neural and nonneural components varied between patients.20 However. (b) sophisticated biomechanical models7. stretch reflex Introduction Spasticity is considered an important symptom by clinicians. PhD2. The model was validated in chronic stroke patients with varying degree of hand spasticity.3-5 Spasticity is usually measured by an examiner moving a joint and simultaneously estimating the resistance according to an ordinal scale. Methods. and Hans Forssberg.3.com/journalsPermissions. Electromyography (EMG) was recorded to measure the muscle activity induced by the passive stretch. such as the Tardieu or the Ashworth scales.” However.11 Previous approaches to separate out the neural component include using (a) the velocity-dependent resisting force. Astrid Lindgrens Barnsjukhus. Karolinska University Hospital. and because more sophisticated biomechanical methods are too complicated to be used in the clinical setting. MD. and (d) neural components (NC). Stockholm. biomechanical model.sagepub. and (d) ischemic nerve block. Sweden Danderyd University Hospital. but not in all. the total resisting force and NC correlated with the modified Ashworth score. Sweden Email: pavel.1 Although there is no proper definition of spasticity that is recognized by both clinicians and researchers. Conclusions.1 the most commonly used definition by Lance2 restricts spasticity to the “velocity dependent increase in the tonic stretch reflex with exaggerated tendon reflexes. Neuropediatrics (Q2:07). PhD1 Abstract Background. The results suggest that the model allows valid measurement of spasticity in the upper extremity of chronic stroke patients and that it can be used to separate the neural component induced by the stretch reflex from resistance caused by altered muscle properties. namely (a) inertia (IC). hand. There is no easy and reliable method to measure spasticity. Lindberg. PhD. The model was validated in 3 ways: (a) NC was reduced after an ischemic nerve block.com/journalsPermissions.se Downloaded from nnr. Sweden 2 Corresponding Author: Påvel G. and viscosity of the body part that is moved. the validity and reliability of these methods has been questioned in many studies.nav DOI: 10. MD1. Johan Gäverth. PT. In most of the patients. (c) viscosity (VC).1177/1545968311403494 http://nnr. PhD1. Objective. Lindberg. Anders Fagergren. the NC dominated. The aim of this study was to develop and validate a new method to measure spasticity that can be easily used in clinical practice. 2014 . According to the model. (b) NC correlated with the integrated EMG across subjects and in the same subject during the ischemic nerve block. PhD1. sagepub. In spite of the criticism.16. Stockholm.com at Karolinska Institutets Universitetsbibliotek on August 4. In addition. MD. Stockholm. and several studies have shown that muscle properties are altered after a cerebral lesion.lindberg@ki. Department of Women’s and Children’s Health and Stockholm Brain institute.

and by active muscle force. B. light blue) forces. A schematic illustration of the major components that contributed to the total passive resisting force trajectory. A force sensor was attached to the device under the hand in order to measure the total resisting force opposing the passive wrist movement. according to the following equation Fm(q) = Fp(q) + Fv(q) + Fr(q) + Fin(q). Materials and Methods The Biomechanical Model Our model is based on previous work by Koo and Mak. viscosity) components. A. fingers.At faster speeds (236°/s). and θ denotes a specific angle. Hand. Fr is the reflex forces.17-19 from the passive mechanical components. that allows estimation of the relative contribution of the neural and the non-neural components of spasticity. the elasticity (purple) and the end range stiffness (gray) constitute the major contribution to the total force (P3).sagepub. This is important since according to the definition of spasticity by Lance. A main feature of this model is that it separates active force produced by muscle contractions induced by stretch reflexes elicited by the muscle stretch2. where Fm is the total measured passive force. easy to apply clinically. In Figure 1. except a minimal acceleration peak in the beginning of the movement. ms.618 Neurorehabilitation and Neural Repair 25(7) Figure 1. elasticity and end range stiffness (EC). Fin is the inertia forces of both the limb and the moving parts of the measuring device.2 the velocity-dependent muscle resistance is a fundamental factor in the increased muscle tone. and inertial forces. Fv is the viscous force. Downloaded from nnr. viscosity. Surface electromyography (EMG) electrodes were placed above the belly of the flexor carpi radialis muscle (FCR). Original trajectories of the total resisting force. dark blue) and viscosity (VC. Therefore. millivolts. During the slow movement (5°/s). EMG recordings. Black line illustrates the total force at 5°/s and 236°/s. newtons. and the neural component (NC. yellow) from the stretch reflex (mV. and forearm were fastened to the platform of a custom built device that passively extended the wrist joint (50°) at controlled velocities (step motor). the first force peak (P1) occurring at about 15 ms after movement onset is composed of inertia (IC. Note that only the fast movement resulted in muscle activity. millisecond). C. 2014 . the purpose of this study was to develop and validate a new model.com at Karolinska Institutets Universitetsbibliotek on August 4. N. The later force peak at the stop of the movement (P2) consists of viscosity (VC).21 in which the resisting force produced during passive wrist extension can be regarded as a summation of passive elasticity. Fp is the passive elastic force. the variation of total force over time during the stretch is shown in conjunction with the passive and active components. and the position of the hand are displayed during slow (5°/s) and fast (236°/s) extension of the wrist.

Movements at a certain velocity were carried out in sets of 5 with 10-second intervals between subsequent movements and an 18-second pause between each set of different velocities. (b) inability to communicate and understand information of the study. Whereas the early viscosity could be calculated from the force trajectory.17 In the model. respectively). 3 force points were used to estimate the passive (inertia.23. see examples in Figure 1). Testing at multiple velocities allowed us to test which components were velocity dependent. 71°/s. and a is the acceleration. Force traces were averaged across 3 trials at each velocity (71°/s. Two specific force points were defined during fast passive stretch (71°/s. Subjects A total of 31 chronic stroke patients with right-sided hemiparesis were recruited (Table 1). 142°/s and (b) 142°/s. the length-dependent elasticity is recorded 1 second after the end of the slow stretching movement (5°/s. 5°/s. to be 20% of the early viscosity at P1: VC = (Total forceP1 . Karolinska Institutet. Inertia component (IC).8. Downloaded from nnr. the later viscosity had to be approximated. 11. and 17% at 236°/s.619 Lindberg et al In our model.IC.6% of the total body weight. 22% at 142°/s. Germany. Subjects were instructed to be relaxed throughout the testing session.com at Karolinska Institutets Universitetsbibliotek on August 4. In a previous report. viscosity. Clinical Assessment Clinical assessment of upper limb function included (a) upper limb subscale of the Motor Assessment Scale (UL-MAS). We therefore determined that our model should approximate the late viscosity at P2. The study was approved by the Regional Ethics Committee. suggesting that they had no or very little muscle activity induced by the stretch.(EC + VC). It includes both the linear elasticity and the nonlinear end range stiffness. for example. In these subjects. 236°/s. which consisted of passive extension of the wrist at 4 velocities ordered in 2 sequences to minimize order effects (a) 5°/s. the late peak during the stretch.23 There is also an exponential increase when the muscle is stretched close to its end range. A third force point (P3) was defined at fully stretched position after slow passive stretch (5°/s. All subjects gave written informed consent. 71°/s. and 21 m/s2 at 71°/s. Stockholm. the early viscosity component was defined as the resisting force that remained after the inertia component had been subtracted from the initial peak of the total resisting force at P1 (Figure 1): VCP1 = Total forceP1 . Acceleration profiles were consistent and increased with target velocity (9. the mean relationship between the late viscosity and the early viscosity calculated at P1 (see above) was 25% at 71°/s. The starting position of the hand was approximately at 20° flexion. according to the Declaration of Helsinki. Neural component (NC). followed by later stretch evoked responses adding to the first muscle contraction.5. in which the late viscosity is about 20% of the early viscosity. where IC is the inertia. The viscosity is the force produced by the viscous resistance. and 236°/s. Landsham. 142°/s. 142°/s. and end position was approximately 30° extension. sliding muscle fibers. Sweden. Figure 1A). m is the mass of hand and platform. and 236°/s) to give P1 and P2 values. Sweden) for later analysis. A total of 13 healthy subjects of comparable age without any history of neurological disorder participated as controls. 2014 . Elasticity is a length-dependent resisting force that increases the more the muscles and tendons are stretched (Figure 1C). 236°/s. P3 in Figure 1C). Umeå University.sagepub. and (d) present pharmacological treatment of spasticity. Inertia is the force resisting the acceleration of the hand and depends on the mass of the hand and movable platform (Figure 1C) and the acceleration according to IC = m × a. Viscosity component (VC). the initial peak during the stretch and P2.24 The viscosity depends on the velocity of the muscle stretch23 and is highest during the initial acceleration and continues at a lower level during the remaining muscle stretch. 142°/s.IC) × 0.24 In our model. elasticity) and active (neural) components. In the model.2. digitalized at 12-bit resolution and stored in SC/ZOOM (Department of Physiology. The mass of the hand was estimated to be 0.22 Elasticity component (EC). and 236°/s): P1. The muscle stretch can activate a spinal stretch reflex with a latency of about 40 ms. Experimental Setup and Protocol Passive extension of the wrist at controlled velocities was performed using a custom-built apparatus incorporating a computer-controlled step motor (Nanotec Electronic GmbH & Co KG. Force and position were sampled at 400 Hz. (c) less than 40° passive extension of the right wrist (to avoid fast stretches near the end of the movement range). the NC was estimated at P2 (maximal extension at the end of the passive movement) by subtracting the elasticity and viscosity components from the total force: NC = Total forceP2 . Halaki et al25 described that there is a rather stable relationship between the early and late viscosities. The P3 force was similarly achieved by averaging 3 trials at 5°/s. Exclusion criteria were (a) previous neurological disorder. We verified this estimation in 9 control subjects who had no visible EMG response.

2014 .7 28. IC I/NA I/frontoparietal.1 22.3 67.3 0. BMI.0 3.1 25.5 23.2 68.4 5.2 62.0 65.6 77. Crb. y BMI. F.3 75. insula. kg/m2 Lesion Type/Location M M M F F F M M M F F M M F F M M F F F M M M F F M M M F F M 58.6 20. BG.4 29.3 1.5 66. % Blocks 81 45 49 0 56 68 14 32 63 44 72 23 19 19 79 56 119 50 78 22 95 24 56 73 80 53 4 98 103 87 7 76 67 25 0 0 50 0 11 41 29 74 0 0 0 95 67 71 0 84 27 100 6 61 54 81 43 0 83 85 74 0   ULMAS. NA.6 21.2 7. internal capsule.2 1.9 2.sagepub.9 6.6 21.0 5. body mass index.4 68. Clinical Description of Patients   S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 S19 S20 S21 S22 S23 S24 S25 S26 S27 S28 S29 S30 S31 Gender Age.2 0.620 Neurorehabilitation and Neural Repair 25(7) Table 1. Rectified EMG was integrated during the whole passive Downloaded from nnr. Massachusetts) were placed on the muscle belly aligned with the muscle fibers of the flexor carpi radialis (FCR).2 28. cerebellum. Goteborg.1 1. An ischemic nerve block test26 was performed on a subgroup of the stroke subjects (n = 7) with high resisting total force and strong EMG responses.5 39.9 60. upper limb subscale of the Motor Assessment Scale.2 4.1 30. IC I/NA H/BG. IC.4 I/striatocapsular H/BG H/medulla oblongata H/BG.0 71.3 5. occipital I/striatocapsular H/frontoparietal IC H/frontoparietal I/IC I/parietal H/Parietal I/striatocapsular I/frontal.5 67.6 24. R.9 29.com at Karolinska Institutets Universitetsbibliotek on August 4. IC. insula H/BG H/Crb.7 4.6 3. female. IC.1 21.2 74. UL-MAS.9 60.6 R L R R R R R R R R R R R R R R L R R R R R L R R R R R R R R Modified Ashworth Scale Finger Flexors Wrist Flexors 2 3 0 3 3 2 1 3 3 1+ 1 2 2 0 0 0 0 3 0 3 0 3 1+ 1+ 0 1+ 1 0 0 2 1 1 1+ 0 3 4 2 4 2 1+ 1+ 1+ 4 3 0 0 0 0 3 0 1+ 0 3 1+ 1+ 0 1+ 1 0 0 2 1 Grip Box Strength.7 4.0 27.9 1. Ischemic nerve block. I.3 53. basal ganglia.0 63. 0-18 17 16 10 1 2 16 6 7 11 14 18 3 2 4 18 13 17 6 15 12 18 11 15 17 16 11 7 17 18 17 9 Abbreviations: M.6 10.3 21.7 56. insula H/striatocapsular I/IC I/striatocapsular. The efficacy of the nerve block was clinically assessed after each measurement (T0 − T35) by means of a light touch test according to the Fugl-Meyer scale and the ability to discriminate between sharp and dull.9 24. temporofrontal H/putamen I/insula.3 27.5 34.5 31. Hopkinton.8 37.8 23. Liberty Technology. left. Bipolar circular Ag-AgCl surface EMG electrodes with an interelectrode distance of 15 mm and an inbuilt amplification (MYO 115. right. and R/L. The resisting force and EMG were measured during passive muscle stretch before inflation (T0).5 54.3 29.5 66. H.8 1.3 3.6 (c) assessment of dull/sharp distinction and light touch according to the Fugl-Meyer Motor Assessment.2 7. infarction. BG Time Since Stroke.0 21.8 62.7 22.7 78.4 26.9 23.7 51. male. hemorrhage.3 26. insula I/corpus callosum I/pons I/thalamus I/thalamus.6 7. BG.5 21. every 5 minutes after the inflation started (T5-T25).8 55.4 10.7 6.3 28.3 31. L.6 1. not available.1 40. and finally at 5 and 10 minutes after the pressure in the cuff was released (T30-T35). Electromyography. (b) rating of muscle tone in finger and wrist flexors according to the Modified Ashworth Scale. (d) box and blocks test of manual dexterity.3 11.5 66.5 4.2 60.8 2. frontal H/frontoparietal I/striatocapsular I/caudate.2 24.9 28.3 52.8 43. Dominant y Hand 5.2 61.6 6. and (e) grip strength using the Grippit dynamometer (AB Detektor.5 51. A blood pressure cuff was placed around the subject’s upper arm and inflated to a pressure exceeding the systolic blood pressure with 30 mm Hg during 25 minutes. Sweden).

Composition of the resisting force during passive muscle stretch at 236°/s in 5 stroke patients with complete clinical ischemic nerve block 25 minutes after onset (T25).cos[anglehand × 3.3 N).5 N at the position 1. of the total force at the end of movement.05. Statistical tests were considered significant if P < . We also Figure 2.621 Lindberg et al movement from onset until fully extended wrist at P2.64. the NC varied depending on the position from +0. elasticity. After placement of the hand on the platform the force sensor measured slightly different forces across subjects (median controls 3. largest in the horizontal plane at zero degrees).62). and the area under the curve was quantified (mVs). Friedman analysis of variance (ANOVA) was used to assess velocity-dependent differences. P < . respectively. measured the consistency of the mechanical set-up by running 50 repetitive trials of the 500-g weight at the zero position. This difference of baseline force was not significant (P = . P2. 5 of the 7 patients had complete absence of both light touch sensation and capacity to discriminate between sharp and dull sensation. Downloaded from nnr.58. viscosity (VC). N.5 cm intervals from the center of the force sensor (0 cm) in both directions. and viscosity components. Nonparametric statistical tests were used throughout the study since some data had skewed distribution (median and interquartile range). A similar position-dependent variation was also found in both EC and VC with an amplitude of ±0. the elasticity (EC). Figure 3). Measurement Error We predicted measurement errors due to (a) variations in hand placement on the moving platform and (b) insufficient assumptions of the model.5 cm closer to the axis of rotation to −2.001 at 142°/s. To compensate for this angle-dependent gravitational variation. there was a significant reduction in NC at T25 compared with T0 (P < . The zero angle of the hand (anglehand = 0) was defined as the horizontal plane. that is.05) but no consistent change in EC or VC (P > . The first and most important way to validate the model was to investigate whether the neural component could be reduced by an ischemic nerve block. All 5 patients showed a large reduction in NC.5 cm further away from the axis of rotation. Wilcoxon matched pairs test was used. T0 indicates baseline level before the blood cuff pressure was increased. Figure 2). Mann– Whitney U test was used. P < . Force recordings included an angle-dependent variation of the gravitational force exerted to the hand (ie.001 at 236°/s. different hand size).5 cm. To reduce the influence of different baseline values. We estimated the measurement error due to hand position by placing a weight of 500 g (similar to that of the hand) at 0.14/180]). Spearman rank correlation test was used for correlations. the baseline force was subtracted from P1.81 m/s 2. The 2 patients with incomplete nerve block were excluded from group analysis of nerve block results. Correlation between the neural component and the integrated EMG of the FCR muscle.001 at 71°/s. see measuring errors in the Materials and Methods section). and VC represent the neural. 2014 . Results Validation of the Biomechanical Model Reduction of the neural component during ischemic nerve block. r = . and P3.14/180]) + mplatform × g(1 - cos[angleplatform × 3.5 N and patients 4. The zero angle of the platform (angleplatform = 0) was defined as the angle of maximum Ftotal without hand. The second way to validate the model was to correlate the variation of the NC with the variation of the muscle activity across subjects.5 N. At group level (n = 5). At T25 (ie. In relation to the 0 position. The NC showed a variation between +1 N and −1 N. and neural (NC) components are shown at baseline (T0) and T25 in the 5 patients with clinical signs of ischemic nerve block. This angle difference was compensated for in each subject by calculating the vertical component of the measured gravitational force during the slow 5°/s movement. the starting angle (measured by a goniometer) varied slightly between subjects even though the platform was in a standardized starting position. P < . In Figure 2.com at Karolinska Institutets Universitetsbibliotek on August 4. For post hoc analysis and for within-group analysis. the total force trace was corrected according to the following formula: Fcorrected = Ftotal + mhand × g(1 . For between-group analyses. The integrated EMG correlated positively with NC across the 3 velocities (r = . where g = 9.5 N at a position 1.62. Data Analysis and Statistics Because of anatomical variability (eg. r = . EC. newton. which indicated that the conduction of nerve signals in large afferents had been blocked.sagepub. (The asterisk in subject S27 indicates zero or negative values in NC.20. 0 to ±1. NC. after 25 minutes of ischemic nerve block).

the NC also correlated with the integrated EMG across time points (T0-T35) during the ischemic nerve block test (r = . 25% to 75% (N. In E. Figure 4C). Correlation between the neural component (NC. The VC increased with velocity in both patients (χ2 =58.01). The VC was lower in patients than in controls at all velocities (P < . millivolt second. df = 2. 142°/s. the integrated EMG is shown at the 3 velocities. df = 2. Figure 4. C.2 Figure 4A shows a clear velocity-dependent increase in the initial P1 peak and later during the ramp movement. S.86. P < . and D show the estimated force components.622 Neurorehabilitation and Neural Repair 25(7) Figure 3.001) and control subjects (χ2 =26.sagepub. at the 3 different velocities (elasticity only at 5°/s). C.3. Patients had significantly larger NC compared with controls at 142°/s and 236°/s (P < . B. Velocity-dependent modulation of the forces that resist the passive muscle stretch in stroke patients and control subjects. 2014 . and (C) 236°/s (mV s. control). The total force developed during stretch was larger in the patients compared with controls.001) but not in the control subjects.007). The mean total force trajectories during passive wrist extensions are shown in black for stroke patients and in gray for control subjects at 5°/s. Velocity dependence of the neural component. and neural components. stroke. The integrated EMG was also velocity-dependent Downloaded from nnr. elasticity. A. df = 2. respectively. Squares and whiskers show median and quartile ranges. viscosity. P < . P < . P = . In the patients who underwent ischemic nerve block.05). ordinate) and the integrated EMG (abscissa) in stroke patients (N = 31) at 3 different velocities of muscle stretch: (A) 71°/s. The third way to validate the model was to test whether the estimated NC would increase with increasing velocity of muscle stretch.com at Karolinska Institutets Universitetsbibliotek on August 4. newton).001. 71°/s. in particular during the end of stretch and at higher velocities (Figure 4A). (B) 142°/s. newton. The NC (Figure 4D) showed a significant velocitydependent increase in the stroke patients (χ2 =37. N. that is.4. and 236°/s.

Asterisks indicate zero or negative values in one of the components (see Materials and Methods section for details). time since stroke. P = .21 N [−0.53 N [0. Figure 4B).82 N].1). neither in the stroke patients nor in the control groups (P > . The most significant validation of the NC was the reduction of NC after an ischemic nerve block that prevented muscle contractions by inhibiting the stretch reflex. in which some had a relatively high EC (eg.19 that is. 2014 . and S12. and NC were not related to body mass index (BMI). elasticity (white).com at Karolinska Institutets Universitetsbibliotek on August 4. Arrows indicate subjects in which the neural component (NC) was relatively small in relation to the total force (N. these findings strongly support that the model of calculating NC is valid and that NC is produced by neural muscle activity. df = 2.623 Lindberg et al Figure 5. Inertia did not differ between groups at any velocity (P > . It varied substantially in patients and showed a trend toward increased elasticity and end range stiffness in patients compared with controls.1). Correlation between the resisting forces and modified Ashworth scores. There was a strong positive correlation between the Ashworth score and both NC and total force across velocities (r > .sagepub. We also investigated the correlations between the total force and various force components with the clinically scored muscle tone according to the modified Ashworth scale.001. and neural (black) components to the total force resisting the muscle stretch at 236°/s in the stroke patients. Nor did any of these 3 components correlate with each other.6. Variation of the Composition of Forces Resisting Muscle Stretch in Stroke Patients The model was used to estimate the partial contribution of the NC. The estimated NC is velocity dependent. Our main focus was to validate the neural component (NC).2). in the patients (ANOVA χ2 = 49.68 to 2.The patients are presented in the order of increasing total resistance. but the difference did not reach significance (patients. Discussion Validation of the Neural Component of the Model The purpose of the model was to determine the neural and nonneural force components that resist a passive stretch. EC. 0.17.001).05. EC. Figure 4E) but not in control subjects. The model is further supported by the strong correlation between NC and the integrated EMG both across subjects and in each of the subjects during the nerve block test. age. P < . NC Downloaded from nnr. whereas EC and VC did not show any consistent correlations. indicated by arrows in Figure 5). VC.2. In general. S18. Figure 4E). The distribution of the 3 components is shown for each stroke patient in Figure 5. The EC was only tested at 5°/s. control subjects. patients S28. However. it becomes larger with increasing speed of the passive stretch. which is fundamental in clinical measurements of spasticity. or gender (P > . 1. The contribution of the viscosity (gray).62 N]. making up the total force resisting the passive muscle stretch. Taken together. It was also higher in patients than in controls across all velocities (P < .38 to 2. and VC. there was also an individual variation among patients.1. newton). the NC was larger both in absolute and relative measures in subjects with larger total resistance. P < . In addition.

although the validity and reliability of this latter method has been questioned. not available in the clinical rating scales. The negative values were mainly present in stroke subjects and control subjects. The control of muscle tone. Inferior mechanical properties of spastic muscle bundles due to hypertrophic but compromised extracellular matrix material.28 It should be pointed out that patients with a limited range of movement. a few patients had small negative EC or VC values at 71°/s (<1 N). Errors can be introduced directly depending on the placement of the hand.17. the Swedish Research Council (5925). and the Department of Neurophysiology.28 and supports the view that the increased passive resistance is predominantly neural in origin.13 Interestingly. The model only requires monitoring of the force during slow and fast passive movements. disorganization of proteins such as collagen). This suggests that the present model is better Acknowledgments We thank the staff of the Department of Rehabilitation Medicine. Elbasiouny SM. Moroz D. which is in accordance with the expectations. Similar findings of reduced initial resistance peaks in spastic stroke patients have been reported by others.1 for example. However. The Robert Wartenburg lecture.sagepub. JG. Although further evidence is required. Neurology.4 However. less than 60° of wrist extension. Another advantage of this method. Errors will also be introduced by the individual discrepancies between the subjects and the general model. a financial relationship with the commercial organizations or products discussed in this article) as follows: The method described has been patented (WO/2008/121067) and the following authors own part of the commercial rights of the patent: PL. VINNOVA. The individual profile of components (Figure 5) may be of clinical value for targeting of treatment. 2010.   2. Bakr MM. Alternatively. Management of spasticity after spinal cord injury: current techniques and future directions. The results clearly show that the NC is sensitive to measurement errors and that several stroke patients ended up with negative values with values down to −1. patients with a large NC component should be considered for therapies aimed at reducing the exaggerated spinal stretch reflex. the finding that the VC was consistently smaller in patients than in controls across all velocities was not expected. if the center of mass of the hand does not fall directly above the force sensor or if there is slight misalignment of the wrist joint with the axis of rotation of the hand apparatus.19. for example. who had a low total resisting force. Increased elasticity may occur in some stroke patients. The negative values were within the range of measurement errors depending on the position of the hand on the platform (see Materials and Methods section) and can thus partly be explained by small variations of hand placement. EC might well be higher. whereas those with dominant mechanical components may benefit from other strategies such as exercise and stretch. In addition. Fridén J. Runesson E. were excluded from our study.28:464-471.624 Neurorehabilitation and Neural Repair 25(7) correlated well with the Ashworth score. the reduced muscle mass after stroke may also cause reduced viscosity. Nonneural Components of the Model Clinical Considerations The EC showed a trend to be increased in the patients compared with controls.com at Karolinska Institutets Universitetsbibliotek on August 4. Downloaded from nnr. and that in patients with such reduced range of movement. Einarsson F. JB. and HF. is the objective and quantitative character of the measurements and the potential to differentiate the neural and mechanical components of the total resisting force. Limitations The model presented is simple and should not be expected to give 100% accurate values.58:113-130. Lieber RL. 2014 . Muscle Nerve. the lack of a significant group difference in our study is consistent with other reports after stroke27. Lieber et al3 reported that although spastic muscles are stiffer and smaller. that is. The VC increased with increasing velocity (Figure 4C). no electrophysiological recording is required.5 N at 236°/s.7 suited for estimation of mechanical and neural components in patients with moderate to severe increase in muscle tone. They showed that the extracellular matrix (composed of fibrous proteins such as collagen) in spastic muscle occupies more of the relative space in the muscle bundle (since muscle size is reduced) and that the extracellular matrix is more disorganized. References   1. Mushahwar VK. Declaration of Conflicting Interests The author(s) declared a potential conflict of interest (eg. reflexes and movement. 2003. Karolinska University Hospital for assistance during data collection. Funding The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This study was supported by the Health Care Sciences Postgraduate School and the Strategic Neuroscience Program at Karolinska Institutet. AF.29 The method we have proposed may offer an alternative way for the clinician to objectively quantify “spasticity” in an easy and noninvasive way.24:23-33.   3. 1980. Danderyd University Hospital. Lance JW. Swedish Foundation for Strategic Research. it is intriguing to consider that viscosity is reduced secondary to such remodeling of the extracellular matrix (ie. Neurorehabil Neural Repair. and the Swedish Brain Foundation. the extracellular matrix exhibits inferior mechanical strength.

26. Lin CC. 13. Bohannon RW. 2000. Hwang IS. Stark L. 10. Smith GV. Brain Res. Harvey RL. Dewald JP. 2006. Lin TS.sagepub. Muscle Nerve. Dewald JP. Lee HM. and recommendations. Huang YZ. Meyer LH. J Electromyogr Kinesiol. Levin MF. Krüger M. 2006.22:399-414. Ada L. Mak AF. Del Conte C. 2006. 2014 . Fleuren JF. Clin Neurophysiol.5:18. Chen JJ.72:621-629. 1988.23:115-124. Disabil Rehabil. 1987. Miscio G. 16.141:446-459. Silver KH. 21. 19. J Neurol Neurosurg Psychiatry.39:2171-2182.28:60-69. Med Eng Phys. 2001. Quantitative features of the stretch response of extrinsic finger muscles in hemiparetic stroke. Mirbagheri MM. Bagni MA.482:391-400. 2010. Lee DG. Lee HM. Stretch-induced electromyographic activity and torque in spastic elbow muscles.81:46-52. Suresh S. Given JD. 23. 1995.59: 271-279. Cecchi G. Arch Phys Med Rehabil. Rymer WZ. Rymer WZ. Rymer WZ. 17.119:1737-1749. The relation between Ashworth scores and neuromechanical measurements of spasticity following stroke. Sarcomeric visco-elasticity of chemically skinned skeletal muscle fibres of the rabbit at rest. 2000. Olsson MC. O’Dwyer N. J Physiol.32:826-839. 2000. Rymer WZ. Dietz V. Thompson AJ. 2007. Burridge JH. Rymer WZ.111:1015-1022. Trippel M. Moammeri H. IEEE Trans Biomed Eng. Zajicek JP. 12. Kearney RE. 1993.181:579-593. 2001. 11. O’Dwyer NJO. Brain. 2003. Differential modulation of reflex activity in passive and active motor tasks.6:1094-1105. Colomo F. Musampa NK. 18. Van Wijck FM. Arch Phys Med Rehabil. 29. Ansari NN. Winters JM. 1995. Koo TK. Chen JJ. Stretch reflex adaptation in elbow flexors during repeated passive movements in unilateral brain-injured patients. Halaki M. J Neurol Neurosurg Psychiatry. Hayashi R. Relationship between stretch reflex thresholds and voluntary arm muscle activation in patients with spasticity. Dobrovolny CL. 1985.27:19-32. Neilson PD. Schmit BD. Kamper DG. Absence of mechanical evidence for attached weakly binding cross-bridges in frog relaxed muscle fibres.   9. Smith MB. Kamper DG. Hobart JC. Naghdi S. J Neuroeng Rehabil. Barbeau H. Wood DE. 2002. solutions. 2008. 14. Powers RK. Ann Neurol. 25. Analysis of fundamental human movement patterns through the use of in-depth antagonistic muscle models. Intrinsic and reflex stiffness in normal and spastic. Exp Brain Res.23:954-961. 2005. Rating scales as outcome measures for clinical trials in neurology: problems. Huang YZ. Effects of ischemic nerve block on the early and late components of the stretch reflex in the human forearm.10:1-14.   7. Voerman GE.   5. Relative contributions of neural mechanisms versus muscle mechanics in promoting finger extension deficits following stroke. Quantitative relations between hypertonia and stretch reflex threshold in spastic hemiparesis. Ranatunga KW.22:119-125. Cano SJ.   6. 24. Mirbagheri MM. Mathieu PA. Jalaie S. Quantitative measures of spasticity in post-stroke patients. Fibre type-specific increase in passive muscle tension in spinal cord-injured subjects with spasticity. Erren-Wolters CV.   8. Harvey RL.com at Karolinska Institutets Universitetsbibliotek on August 4. J Biomech. Ashworth Scales are unreliable for the assessment of muscle spasticity. 1987. Timecourse analysis of stretch reflexes in hemiparetic subjects using an on-line spasticity measurement system.403:341-344. 20.116:971-989. Joint dependent passive stiffness in paretic and contralateral limbs of spastic patients with hemiparetic stroke. Stop using the Ashworth Scale for the assessment of spasticity. 22. Becker WJ. Alibiglou L. Physiother Theory Pract. 2007. et al. Ibrahim IK. J Muscle Res Cell Motil. A neuromusculoskeletal model to simulate the constant angular velocity elbow extension test of spasticity. Ladouceur M. 2006. Phys Ther.577(pt 1):339-352.81: 269-278.67:206-207. Berger W. 15. Spasticity and muscle contracture post stroke. Hemiparetic muscle atrophy and increased intramuscular fat in stroke patients. Pianca D. 27. Ju MS. Ryan AS. Macko RF. Candeloro E. 28. Inter-rater reliability of a modified Ashworth scale of muscle spasticity. spinal cord injured subjects. J Neurol Neurosurg Psychiatry. Systematic nonlinear relations between displacement amplitude and joint mechanics at the human wrist. 2002. Exp Brain Res.83:1703-1707. 2000. Marder-Meyer J.625 Lindberg et al   4. Downloaded from nnr. et al. Brain. Muscle Nerve.28:309-318. J Physiol. Biomechanical approaches applied to the lower and upper limb for the measurement of spasticity: a systematic review of the literature. Quantitative analysis of the velocity related pathophysiology of spasticity and rigidity in the elbow flexors. Cathers I. Garzella P. Pisano F. 1996. Lancet Neurol. Colombo R. et al. Rymer WZ.