Doi:10.1016/j.neulet.2006.07.06

Neuroscience Letters 406 (2006) 289–292 Effect of acute leg cycling on the soleus H-reflex and modified Ashworth scale scores in individuals with multiple sclerosis Robert W. Motl , Erin M. Snook, Marcus L. Hinkle, Edward McAuley Department of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, 350 Freer Hall, Urbana, IL 61801, United States Received 21 April 2006; received in revised form 8 July 2006; accepted 27 July 2006 Abstract
This study examined the effect of a single bout of unloaded leg cycling on the soleus H-reflex and modified Ashworth scale (MAS) in 27 individuals with multiple sclerosis (MS) who had spasticity of the leg muscles, but were not currently taking anti-spastic medications. The soleusH-reflex and MAS data were collected before and 10, 30, and 60 min after 20 min of unloaded leg cycling and a control condition. The acutebout of unloaded leg cycling resulted in concomitant and prolonged reductions in the soleus H-reflex and MAS scores compared with the controlcondition. This provides converging evidence for the anti-spastic potential of acute unloaded leg cycling in individuals with MS.
2006 Elsevier Ireland Ltd. All rights reserved.
Keywords: Hoffmann reflex; Exercise; Neurological disease; Spasticity Spasticity is a motor disorder characterized by a velocity- The present study examined the effect of a single bout of dependent increase in tonic stretch reflexes and exaggerated unloaded leg cycling on the H-reflex in individuals with MS.
tendon jerks resulting from hyper-excitability of the stretch Based on previous research we expected that a single reflex it is a prevalent symptom associated with multiple bout of unloaded leg cycling would reduce the amplitude of the sclerosis (MS) Indeed, the amplitude of the H-reflex, soleus H-reflex in individuals with MS. As a secondary pur- an electrically evoked analogue of the monosynaptic stretch pose, this study examined the effect of a single bout of unloaded reflex, has been demonstrated to be elevated in individuals with leg cycling on modified Ashworth scale (MAS) scores We MS when compared with healthy control participants expected that a single bout of unloaded leg cycling would reduce spastic agents such as baclofen have been effective in reducing MAS scores in individuals with MS, consistent with the effects spasticity in individuals with MS as demonstrated by an attenua- tion of the H-reflex amplitude owever, anti-spastic medi- The procedure was approved by an Institutional Review cations require a prescription, are costly, and side effects include Board, and participants provided informed consent. Participants fatigue, weakness, and functional limitations. Preliminary work (N = 27) were individuals with relapsing-remitting, primary pro- has demonstrated that acute bouts of loaded, unloaded (i.e., no gressive, or secondary progressive MS with expanded disability resistance), and passive leg cycling have reduced the amplitude status scale (EDSS) between 0.5 and 4.5 and MAS of the H-reflex recorded in the soleus muscle Such scores between 1 and 3. Those with EDSS scores between 0.5 findings suggest that acute leg cycling, particularly unloaded and 4.5 have minimal disability should be fully capable of leg cycling, may be a possible therapeutic modality for treating completing a bout of cycling performed against no resistance.
MS-induced spasticity. Unloaded leg cycling might reduce exag- We note that individuals with MS and EDSS scores between gerated stretch reflex activity and muscle tone, while avoiding 2 and 4.5 have completed 30 min bouts of moderate intensity complications of increased body temperature that are possible cycling exercise in a previous study with MAS scores with intense loaded cycling exercise in individuals with MS between 1 and 3 have slight to moderate increases in muscle tone and relative ease of joint movement are likely to benefitfrom acute exercise. We excluded all individuals currently usingoral or intrathecal anti-spastic medications, and none of the par- ticipants reported engaging in a cycling exercise program within Corresponding author. Tel.: +1 217 265 0886; fax: +1 217 333 3124.
their home or a gym. The mean age of the sample was 44.9 0304-3940/$ – see front matter 2006 Elsevier Ireland Ltd. All rights reserved.
doi: R.W. Motl et al. / Neuroscience Letters 406 (2006) 289–292 (S.D. = 8.3) years and 25 of the 27 participants were females.
4 (fixed muscle contracture). The reliability of the MAS scores Of the 27 participants, 25 had relapsing-remitting MS and two across time within the control condition was adequate based on participants had primary progressive MS. The average duration an intraclass correlation coefficient of 0.90. The reliability of since the initial MS diagnosis was 8.3 (S.D. = 5.7) years.
the pre-condition MAS scores across conditions was adequate The H-reflex was measured in the soleus muscle of the right based on an intraclass correlation coefficient of 0.76.
leg with the participant in a comfortable semi-reclined posi- Participants undertook three experimental sessions. During tion. The H-reflex was evoked by stimulating the tibial nerve in the 1st session, participants provided informed consent and the popliteal fossa through a monopolar stimulating electrode demographic information, underwent EDSS, MAS, and H-reflex with an anode placed superior to the patella. The stimulus was a assessments, and rode a cycle ergometer for 5 min. During the single, 1-ms rectangular pulse delivered every 10 s (Grass S88; 2nd and 3rd sessions, participants undertook exercise or control Astro-med Inc., West Warwick, RI). The H-reflex was measured conditions, and the order of conditions was counter-balanced; using bipolar electrodes placed 2 cm apart along the ipsilat- the clinician who collected the MAS data was not blinded to eral soleus muscle and standard electromyography (EMG). The the order of conditions. The exercise condition involved 20 min EMG signal was band-pass filtered (3 Hz–1 kHz) and amplified of unloaded leg cycle ergometry (i.e., pedaling against no resis- by 1000 (Grass P511; Astro-med Inc., West Warwick, RI), and tance), and was selected because it should not be excessively sampled at 2500 Hz (Grass PolyVIEW; Astro-med Inc., West fatiguing for individuals with intact mobility and minimal dis- Warwick, RI). An initial H-reflex recruitment curve was made ability. We collected perceived exertion data using Borg’s 6–20 by progressively increasing the intensity of the stimulation in an scale very 5 min during the exercise condition and the aver- effort to locate the largest obtainable H-wave and M-wave mea- age value was 10.8 (S.D. = 2.5) and corresponded with a verbal sured as peak-to-peak amplitude of the non-rectified wave. The anchor of “fairly light” exertion. The control condition involved maximal H-wave (Hmax) and maximal M-wave (Mmax) were sitting in a comfortable chair for 20 min and served as a control then measured as an average of five subsequent recordings of for passage of time and instrumentation effects. The H-reflex the largest obtainable H-wave and M-wave, respectively. Impor- and MAS data were collected before and 10, 30, and 60 min tantly, the Hmax corresponded with an M-wave of approximately 10% of Mmax thereby maintaining the constancy of nerve stimu- The data were analyzed using a 2 (condition: exercise and lation and the Mmax was evoked with supramaximal stimulation control) × 4 (time: pre-condition, 10-min post-condition, 30 min of the tibial nerve. The Hmax was expressed as the Hmax/Mmax post-condition, and 60 min post-condition) repeated measures ratio. The reliability of the Hmax/Mmax ratio across time within ANOVA based on the multivariate F statistic (Pillai’s Trace) the control condition was adequate based on an intraclass cor- Effect sizes associated with F statistics were expressed as eta- relation coefficient of 0.99 (two-way mixed effects model with squared (η2). Effect sizes based on a difference in mean scores people as a random effects and measure as a fixed effect). The were expressed as Cohen’s d The bivariate relationships reliability of the pre-condition Hmax/Mmax ratio across condi- between the Hmax/Mmax ratio and MAS scores across the two tions was adequate based on an intraclass correlation coefficient conditions and four time points were examined using Pearson product-moment correlation coefficients.
Spasticity of the right calf muscle was assessed by a trained The mean scores for the Hmax/Mmax ratio are provided in clinician using the MAS the participant lying in the There was a statistically significant condition × time same comfortable semi-reclined position used for recording the interaction on Hmax/Mmax ratio values [F (3, 24) = 42.13, H-reflex. The clinician underwent 3 months of training and p < 0.0001, η2 = 0.84]. The Hmax/Mmax ratio was significantly practicum by a neurologist and two physical therapists in using reduced 10-min (d = −0.38), 30-min (d = −0.48), and 60-min the MAS. The MAS provides a measure of muscle hypertonia on (d = −0.54) after unloaded leg cycling. There was not a sta- a five-point scale, ranging from 0 (normal muscle tone) through tistically significant change in the Hmax/Mmax ratio 10-min Table 1Soleus Hmax/Mmax ratio, Hmax, Mmax, and modified Ashworth scale (MAS) values before and 10, 30, and 60 min after 20 min of unloaded cycling exercise or acontrol condition of quite seated rest in 27 individuals with multiple sclerosis Note: Pre, pre-condition; Post 10, 10 min post-condition; Post 30, 30 min post-condition; Post 60, 60 min post-condition.
R.W. Motl et al. / Neuroscience Letters 406 (2006) 289–292 (d = 0.05), 30-min (d = −0.05), and 60-min (d = 0.09) after the a control group of healthy individuals, and therefore are unable to directly compare baseline values in those with MS versus The interaction on Hmax/Mmax ratio values was explained by changes in the Hmax. There was a statistically significant We observed an average reduction in the Hmax/Mmax ratio of condition × time interaction on Hmax values [F (3, 24) = 11.90, nearly 1 S.D. or approximately 15% for the assessments taken p < 0.0001, η2 = .60], but not on Mmax values [F (3, 24) = 0.60, 10, 30, and 60 min after the bout of unloaded leg cycling in this p = 0.62, η2 = 0.07]. The Hmax was significantly decreased 10- sample of individuals with MS. Previous research has reported min (d = −0.16), 30-min (d = −0.20), and 60-min (d = −0.22) a reduction in the Hmax/Mmax ratio of nearly 1 S.D. or approx- after unloaded leg cycling, whereas there was not a statisti- imately 25% for assessments taken 10 and 30 min after a bout cally significant change in the Hmax 10-min (d = 0.03), 30-min of unloaded leg cycling in a sample of healthy participants (d = 0.07), and 60-min (d = 0.12) after the control condition con- This provides further evidence of a consistent reduction in the trol. The mean Hmax and Mmax values are provided in H-reflex after unloaded leg cycling, and this effect occurs in indi- The mean scores for the MAS are provided in viduals with MS and healthy controls. One novel feature of our was a statistically significant condition × time interaction on study is that the effect of unloaded leg cycling lasted for 60 min, MAS scores [F (3, 24) = 8.58, p < 0.0001, η2 = .52]. The MAS and this is 30 min longer than examined in previous research score was significantly decreased 10-min (d = −0.53), 30-min using healthy controls Hence, the present results suggest (d = −0.43), and 60-min (d = −0.37) after unloaded leg cycling.
that the effect of unloaded leg cycling on the H-reflex is not There was a statistically significant increase in the MAS score a transient response, but rather a response that lasts for a pro- 30-min (d = 0.28) and 60-min (d = 0.41) after the control condi- longed period of time. The exact duration of the post-exercise attenuation of the H-reflex before returning to pre-condition or With the control condition, there were no statistically sig- baseline levels was not examined in the present study, but serves nificant correlations between the Hmax/Mmax ratio and MAS as an important direction for subsequent research.
scores for the pre-condition (r = 0.01) or the 10 (r = 0.03), 30 The control condition was associated with an increase in (r = 0.16), and 60 min (r = 0.05) post-condition assessments.
mean values of the Hmax and Mmax across time, but the size of Similarly, there were no statistically significant correlations the Hmax as a percentage of the Mmax remained constant across between the Hmax/Mmax ratio and MAS scores for the pre- time in the control condition. The observed pattern of change condition (r = −0.14) or the 10 (r = −0.04), 30 (r = 0.09), and in mean values of the Hmax and Mmax across time is, in part, 60 min (r = 0.02) post-condition assessments with the exercise inconsistent with previous research That previous research measured the Hmax and Mmax every 3 min with a stimulus inter- Spasticity is a common MS-related symptom that is fre- val of 4 s during resting conditions across a 1–3 h period of time quently managed with pharmacological agents Those in 13 healthy controls and seven patients. The amplitudes of the anti-spastic medications require a prescription, are costly, and Hmax and Mmax generally decreased in parallel over time, and have unwanted side effects. The present study provides evidence the excitability of the monosynaptic pathway remained fairly that a single bout of unloaded leg cycling might serve as an constant given that the size of the Hmax as a percentage of the effective non-pharmacological strategy for managing spastic- Mmax was stable across time. Hence, we observed a different ity in individuals with MS. Indeed, the acute bout of unloaded direction of change in the amplitudes of the Hmax and Mmax, leg cycling resulted in concomitant and prolonged reductions but observed a constant Hmax/Mmax ratio across time in the con- in the H-reflex and MAS scores compared with a control con- trol condition. Although the exact cause of this different pattern dition. This provides converging evidence for the anti-spastic of change in the amplitudes of the Hmax and Mmax across time potential of acute unloaded leg cycling in individuals with MS.
is not clear, one possibility may be the frequency and rate of Hence, there is now a stronger basis for the recommendation of recording the H-reflex. That is, we recorded the H-reflex only incorporating exercise into a plan for spasticity management in four times using a 10 s stimulus interval intermittently across the course of an 80 min period of time, whereas previous research Previous research has demonstrated that the amplitude of the the H-reflex every 3 min with a 4 s stimulus interval H-reflex is elevated in individuals with MS when compared with consistently across a 1–3 h period of time.
healthy control participants The value of the Hmax/Mmax Herein, we examined the effect of unloaded leg cycling on ratio in the samples of individuals with MS and the controls the amplitude of the H-reflex recorded in the soleus muscle of were 0.57 and 0.44, respectively studies have reported the leg in a sample of individuals with MS. Hence, in this study similar Hmax/Mmax ratio values of approximately 0.60 to 0.64 we did not examine the possibility that the effect was localized in individuals with MS who have spasticity The val- or generalized. Importantly, our previous research has exam- ues of the Hmax/Mmax ratio in the current study were 0.65 and ined the effect of moderate intensity leg cycling on the H-reflex 0.66 before the control and exercise sessions, respectively. The recorded in soleus muscle of the leg and the flexor carpi radi- elevated values for the Hmax/Mmax ratio that we reported in this alis muscle of the arm We reported that the bout of leg study provide evidence that our sample of individuals with MS cycling attenuated the H-reflex in the soleus muscle, but not did have an elevated H-reflex, consistent with having spasticity, the flexor carpi radialis muscle. This would suggest that the compared with the expected value of the H-reflex for healthy effect of an acute bout of leg cycling on the H-reflex is local- control samples. This is important because we did not include ized within the activated muscle group rather than an effect that R.W. Motl et al. / Neuroscience Letters 406 (2006) 289–292 is generalized beyond those muscles used during the bout of unloaded, loaded, and graded exercise sessions on spasticity.
This is important for identifying the training parameters that This study examined the effect of unloaded leg cycling exer- might maximize any anti-spastic effect of acute and chronic cise on the H-reflex and MAS scores in individuals with MS.
Unloaded leg cycling involves pedaling against no resistanceand has minimal metabolic demand, whereas loaded leg cycling Acknowledgement
involves pedaling against a resistance that can vary in mag-nitude and metabolic demand. We selected an unloaded leg This investigation was supported by a Pilot Research Award cycling exercise stimulus because many individuals with MS from the National Multiple Sclerosis Society (PP1021).
experience a profound detraining effect across the course of thedisease and have problems with muscle weakness and fatigue References
an unloaded leg cycling stimulus is associated withminimal demands and can be quite easily performed by indi- [1] A.M.O. Bakheit, V.A. Maynard, J. Curnow, N. Hudson, S. Kodapala, The relation between Ashworth scale scores and the excitability of the ␣ motor viduals with MS who are detrained or who experience muscle neurons in patients with post-stroke muscle spasticity, J. Neurol. Neurosurg.
weakness and fatigue. Indeed, the average rating of perceived exertion collected with the unloaded leg cycling exercise was [2] R.W. Bohannon, M.B. Smith, Interrater reliability of a modified Ashworth 10.8 (S.D. = 2.5) and corresponded with a verbal anchor of scale of muscle spasticity, Phys. Ther. 67 (1987) 206–207.
“fairly light” exertion the participants in this study.
[3] G. Borg, Borg’s Perceived Exertion and Pain Scales, Human Kinetics, Importantly, there were not statistically significant correla- [4] J. Cohen, Statistical Power Analysis for the Behavioral Sciences, third ed., tions between the Hmax/Mmax ratio and MAS score across each of the four time points for both conditions. This lack of a sig- [5] C. Crone, L.L. Johnsen, H. Hultborn, G.B. Ørsnes, Amplitude of the maxi- nificant association is consistent with previous research mum motor response (Mmax) in human muscles typically decreases during Although the MAS is widely used as a measure of spasticity the course of an experiment, Exp. Brain Res. 124 (1999) 265–270.
[6] H.J. Keselman, Testing treatment effects in repeated measures designs: an in research and clinical practice, the measure has several limi- update for psychophysiological researchers, Psychophysiology 35 (1998) tations (e.g., reliance upon subjective judgment of clinician and lack of specificity for spasticity) and has questionable validity [7] J.F. Kurtzke, Rating neurological impairment in multiple sclerosis: an and reliability problems question the meaningfulness expanded disability status scale (EDSS), Neurology 33 (1983) 1444–1452.
of the observed reduction in the MAS after exercise in the cur- [8] J.W. Lance, The control of muscle tone, reflexes, and movement: Robert Wartenberg Lecture, Neurology 30 (1980) 1303–1313.
rent study and the use of the MAS in future studies of exercise [9] H. Morita, C. Crone, D. Christenhuis, N.T. Petersen, J.B. Nielsen, Modula- and spasticity in individuals with MS.
tion of presynaptic inhibition and disynaptic reciprocal Ia inhibition during There are several directions for future research involving voluntary movement in spasticity, Brain 124 (2001) 826–837.
exercise and spasticity in individuals with MS. One direc- [10] R.W. Motl, R.K. Dishman, Acute leg-cycling exercise attenuates the H- tion involves examining the effects of unloaded leg cycling on reflex recorded in soleus but not flexor carpi radialis, Muscle Nerve 27(2003) 609–613.
spasticity in individuals who are currently using pharmacolog- [11] R.W. Motl, B.D. Knowles, R.K. Dishman, Acute bouts of active and pas- ical agents such as baclofen or tizanidine. This is important sive leg cycling attenuate the amplitude of the soleus H-reflex in humans, as exercise is often recommended along with pharmacologi- cal agents for reducing spasticity but there is, as of yet, [12] J.F. Nielsen, T. Sinkjær, Peripheral and central effect of baclofen on ankle no evidence documenting that exercise still has an effect when joint stiffness in multiple sclerosis, Muscle Nerve 23 (2000) 98–105.
[13] G. Ørsnes, C. Crone, C. Krarup, N. Petersen, J. Nielsen, The effect of combined with pharmacological agents such as baclofen and baclofen on the transmission in spinal pathways in spastic multiple sclerosis tizanidine. One related future research direction involves com- patients, Clin. Neurophysiol. 111 (2000) 1372–1379.
paring the independent and additive effects of acute unloaded [14] A.D. Pandyan, F.M. Van Wijck, S. Stark, P. Vuadens, G.R. Johnson, M.P.
leg cycling and medications on spasticity in individuals with Barnes, The construct validity of a spasticity measurement device for clin- MS. This would provide direct evidence for unloaded cycling ical practice: an alternative to the Ashworth scales, Disabil. Rehabil. 15(2006) 579–585.
as a viable alternative to medications for managing spastic- [15] M.A. Rizzo, O.C. Hadjimichael, J. Preiningerova, T.L. Vollmer, Prevalence ity. Another future direction involves testing the efficacy of a and treatment of spasticity reported by multiple sclerosis patients, Mult.
chronic exercise stimulus for managing spasticity. That is, we only tested the effect of a single bout of unloaded leg cycling, and [16] R.T. Schapiro, Managing symptoms of multiple sclerosis, Neurol. Clin. 23 future research might consider the effects of chronic unloaded [17] A.T. White, T.E. Wilson, S.L. Davis, J.H. Petejan, Effect of precool- leg cycling on the H-reflex in spasticity management. One last ing on physical performance in multiple sclerosis, Mult. Scler. 6 (2000) future direction involves an evaluation of the effects of passive,

Source: http://www.epl.illinois.edu/files/pubs/2006/%5BMotl%5D%20Effect%20of%20acute%20leg%20cycling%20on%20the%20soleus%20H-reflex%20and%20modified%20Ashworth%20scale%20scores%20in%20individuals%20with%20multiple%20sclerosis.pdf