RESEARCH ARTICLE
Treadmill Training or Progressive Strength Training to Improve Walking in People with Multiple Sclerosis? A Randomized Parallel Group Trial Siri Merete Brændvik1,2*, Teija Koret2, Jorunn L. Helbostad1,2, Håvard Lorås3, Geir Bråthen1,4, Harald Olav Hovdal4 & Inger Lise Aamot2,5 1
Department of Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway
2
Clinical Services, St. Olavs University Hospital, Trondheim, Norway
3
Sør-Trøndelag University CollegeDepartment of Physical TherapyTrondheim, Norway
4
Department of Neurology, St. Olavs University Hospital, Trondheim, Norway
5
The K.G. Jebsen Center of Exercise in Medicine/Department of Circulation and Medical Imaging, Faculty of Medicine, Norwegian University
of Science and Technology, Trondheim, Norway
Abstract Background and purpose. The most effective treatment approach to improve walking in people with multiple sclerosis (MS) is not known. The aim of this trial was to assess the efficacy of treadmill training and progressive strength training on walking in people with MS. Methods. A single blinded randomized parallel group trial was carried out. Eligible participants were adults with MS with Expanded Disability Status Scale score ≤6. A total of 29 participants were randomized and 28 received the allocated exercise intervention, treadmill (n = 13) or strength training (n = 15). Both groups exercised 30 minutes, three times a week for 8 weeks. Primary outcome was The Functional Ambulation Profile evaluated by the GAITRite walkway. Secondary outcomes were walking work economy and balance control during walking, measured by a small lightweight accelerometer connected to the lower back. Testing was performed at baseline and the subsequent week after completion of training. Results. Two participants were lost to follow-up, and 11 (treadmill) and 15 (strength training) were left for analysis. The treadmill group increased their Functional Ambulation Profile score significantly compared with the strength training group (p = .037). A significant improvement in walking work economy (p = .024) and a reduction of root mean square of vertical acceleration (p = .047) also favoured the treadmill group. Discussion. The results indicate that task-specific training by treadmill walking is a favourable approach compared with strength training to improve walking in persons with mild and moderate MS. Implications for Physiotherapy practice, this study adds knowledge for the decision of optimal treatment approaches in people with MS. Copyright © 2015 John Wiley & Sons, Ltd. Received 27 January 2014; Revised 5 February 2015; Accepted 5 May 2015 Keywords multiple sclerosis; strength; treadmill; walking *Correspondence Siri Merete Brændvik, PT/PhD. Department of Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway. Email: [email protected]
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pri.1636
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
Multiple Sclerosis and Walking
Introduction Impaired walking is prominent in people with multiple sclerosis (MS) (Motl et al., 2010) and used as an indicator of disease progression (Kurtzke, 1983). The impairments have been demonstrated as reduced walking speed (Thoumie et al, 2005; Givon et al., 2009), increased variability in hip, knee and ankle kinematics compared with healthy controls (Crenshaw et al., 2006; Sosnoff et al., 2011) and impaired postural control (Martin et al., 2006). In addition, and presumably as a result, increased energy cost during walking has been reported in people with MS compared with healthy controls (Motl et al., 2011). MS is associated with a decrease in physical activity, which in turn can result in deconditioning in domains like aerobic capacity, muscle strength and balance. This process feeds back and leads to a vicious circle of inactivity (Motl et al., 2010). Importantly, this physiological deconditioning might influence the onset and progression of impaired walking in persons with MS (White and Castellano, 2008). From a clinical perspective, it is therefore essential to find interventions that aim to attenuate this process and maintain or improve walking. Exercise training is considered to be an important part of the rehabilitation in persons with MS. There is strong evidence in favour of exercise training compared with no exercise training for muscle power, exercise tolerance and mobility-related activities (Rietberg et al., 2004; Snook and Motl, 2009). However, no evidence is found to support the superiority of a particular approach on mobility in general or on walking in particular (Rietberg et al., 2004). Muscle weakness is one possible factor that may contribute to impair walking in MS (Schwid et al., 1999; Lambert et al., 2001; Cameron and Wagner, 2011). A strong correlation between strength and impaired walking has been reported (Thoumie et al., 2005). As such, progressive strength training (ST) might possibly prove to be efficient in improving walking. However, a recent systematic review reports that although there is strong evidence that progressive ST increases strength in persons with Expanded Disability Status Scale (EDSS) below 6.5, the transfer to improved walking is questionable (Kjolhede et al., 2012). Treadmill training is another approach used to improve walking. Walking on a treadmill allows for massed task-specific practice that is suggested to be critical in obtaining improvement of motor function following
stroke (Hornby et al., 2011). A recent systematic review (Swinnen et al., 2012) addressing the effect of different treadmill training (TT) modalities (unsupported, body weight supported and robot assisted) in people with MS concluded that TT improved walking speed and walking distance. However, it was not clear which type of TT was the most effective. Moreover, the included studies compared TT with either another TT modality to conventional gait training or no training. Therefore, the objective of this study was to compare the effect of two different exercise approaches, unsupported TT and progressive ST, on walking in people with mild to moderate MS. It was hypothesized that TT would be superior to progressive ST in improving walking.
Methods This was a single-blinded randomized trial, configured as a two-group pre-test–post-test design, incorporating two experimental groups and stratified for age and gender. The study took place at Trondheim University Hospital, Norway, and eligible participants were recruited in the period between January and November 2010. After the baseline testing, participants were randomly assigned to TT or progressive ST in a 1:1 ratio. Allocation of the participants was performed by the unit of Applied Clinical Research at the Norwegian University of Science and Technology, using a webbased randomization system. The study was conducted according to the Declaration of Helsinki and was approved by the Regional Committee for Medical and Health Research Ethics South East Norway. Written informed consent was obtained from the participants before inclusion. Sample size calculation was based on the primary outcome measure, the Functional Ambulation Profile (FAP) (Nelson, 1974). In order to detect a 10% difference in FAP score between the groups, with a two-sided 5% significance level and a power of 80%, a sample size of 13 participants per group was needed (Givon et al., 2009).
Participants Eligible participants were adults (aged 18 years and over) diagnosed with MS, having a maximum EDSS score of 6, no relapses of disease or new medications the last 6 months, no pronounced spasticity with need of Baclofen, sign of pyramidal affection examined by magnetic resonance imaging and with effect on gait Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
clinically examined by a neurologist, and no other conditions affecting their gait function. Eligible participants were identified through a systematic review of the outpatients’ medical record of patients under active follow-up at the Department of Neurology at the Trondheim University Hospital, where we aimed to identify all patients that fulfiled the eligibility criteria. All patients were clinically examined by an experienced physical therapist (T. K.) and a neurologist (H. H.).
Multiple Sclerosis and Walking
for each exercise separately at the first training session. Two sets of six repetitions were performed for each exercise, and each leg was trained separately. Resistance was increased by a minimum of 0.25 kg once the participants were able to perform two sets of six repetitions. There was a 2-minute recovery period between each set and between different exercises. Total duration of the training programme was 30 minutes. Outcome measures
Interventions Participants took part in group exercise three times a week for 8 weeks (maximal 24 training sessions) in and outpatient clinical setting. Exercise sessions were supervised by experienced physical therapists working at the hospital and who were not involved in the assessments. The treadmill training was carried out without body weight support, but the participants were allowed to use the handrails for balance support. The training consisted of three different walking bouts, each lasting for 7 minutes: 1) preferred walking speed at an increased slope; 2) walking with verbal guidance on gait pattern, focusing on heel strike, loading and toe-off and knee control during weight-bearing; and 3) fast gait speed defined as a 10% increase in walking speed relative to preferred speed. During the 8-week intervention, walking speed in bout 3 was gradually increased with 10% to 40% of preferred speed. Preferred walking speed was used as departure point for each training session to calculate the intensity on increased velocity and inclined walking. The participants wore a heart rate monitor and were instructed to keep their exercise intensity below 70% of maximal heart rate that was obtained during the treadmill test, to keep focus on walking pattern and avoid exhaustion during gait. Each session lasted for 30 minutes, including 2-minute break between the bouts. The ST was conducted according to current guidelines from the American College of Sports Medicine position stand (2009). The programme consisted of five exercises, comprising both concentric and eccentric components: 1) knee extensor strength performed in a leg press device (lying position); 2) plantar flexion strength in leg press device (extended knees); 3) standing hip abductions with low pulley; 4) dorsal flexion strength in sitting position with the use of pulley; and 5) core and back muscles in sitting position using pull-down device. Work load during training was 80% of one repetition maximum that was set Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
The primary outcome measure was the FAP score. The FAP provides a single score for gait, based on the principles that walking ability attains several sub-skills, like a stable base of support, weight transfer in the mediolateral direction and forward progression (Nelson, 1974). The score ranges from 0–100, where 100 is the best score. In healthy adult populations, the mean FAP is reported to be 95 with a standard deviation of 1 (Givon et al., 2009). The FAP score is reported to be a valid measure of gait impairments in persons with MS classified with an EDSS level between 4 and 6 (Sosnoff et al., 2011), also in those with relatively short duration of the disease (Givon et al., 2009). Reliability of the FAP score in the MS population has to our knowledge not been reported, neither has sensitivity to change. Secondary outcome measures were balance control during walking, measured as trunk acceleration amplitudes, and walking work economy. Trunk accelerations have previously been tested for precision and accuracy (Moe-Nilssen, 1998a), and acceleration amplitudes have shown to be sensitive measures in discriminating between MS patients and healthy controls with respect to balance during walking and quiet standing (Spain et al., 2012). Trunk accelerations have been found to be reliable in healthy persons (Henriksen et al., 2004), as well as in children with cerebral palsy (Saether et al., 2014). We are not aware of published reliability measures on trunk accelerations in the MS population, but unpublished data from our own patient database show test–retest reliability (1 week apart) assessed by an intraclass correlation coefficient (ICC 2.1) for acceleration amplitude in anteroposterior, mediolateral and vertical direction, ranging from 0.87 to 0.94. Procedure and equipment The participants were tested, prior to randomization, the week before the training started (Pre) and within a week after the last training session (Post).
S. M. Brændvik et al.
Multiple Sclerosis and Walking
The GAITRite walkway system (CIR Systems, Inc., Sparta, NJ, USA) was used to obtain the FAP score. The GAITRite is an electronic walkway with sensors arranged in a grid-like pattern to identify footfall contacts, and it has been found to be a valid method for measuring temporal and spatial gait parameters (Webster et al., 2005). The active length of the walkway was 6.1 m. An additional 0.90 m at each end of the walkway was added to give room for acceleration and deceleration. The participants walked according to the following instruction: ‘Walk as you usually do, at your preferable speed’. All participants completed two walks (back and forth along the walkway once). They wore good walking shoes, and the same shoes were used at all test sessions. One test trial was performed before measurement. Concurrently, balance control during walking was measured using a small lightweight inertial sensor (MTx, XSens, Enschede, Netherlands), attached to the lower back (L3 region) by an elastic belt. This sensor contains tri-axial units of accelerometers, gyroscopes and magnetometers and was connected to a batteryoperated communication unit also worn by the subjects (Aaslund et al., 2013). Data were sampled at a frequency of 100 Hz and transferred wireless to a laptop in real time for processing. Only acceleration data were used in the current study and were reported along anteroposterior, mediolateral and vertical axes. Photoelectric timing units were synchronized to the sensor system and placed at each end of the active area of the electronic walkway to register time sequences for the acceleration data for each walk. Energy expenditure during walking was measured during a treadmill exercise test, and respiratory gas analysis was performed using a Metamax II (Cortex Biophysics, Leipzig, Germany). Metamax II has shown to be valid and reliable for gas measurements (Medbø et al., 2002). After 5 minutes of familiarization to treadmill walking, oxygen consumption (VO2) was measured during 2 minutes walking at a pre-set speed (3 km hour 1). The participants were encouraged to walk without handrail support. Heart rate was measured continuously during the test. In addition, peak oxygen consumption (VO2peak) was measured using an individually ramp protocol (Fletcher et al., 2013) in order to calculate the intended exercise intensity during treadmill training. To obtain a stability estimate of the FAP score, the walking protocol was repeated in a retest 5 days after Pre, just prior to the start of intervention. Intraclass
correlation coefficient of the FAP score was found to be 0.91 (ICC 2.1), with a 95% confidence interval of .79–.96, reflecting high level of agreement between test occasions. Measurement error (within-subject standard deviation) was 3.6 FAP scores. Data analysis The GAITRite software system (Version 3.8) calculates spatial and temporal parameters based on processed footfalls. Walking speed is calculated by dividing the distance walked by ambulation time, and the normalization is carried out by dividing it to leg length (GAITRite manual). The FAP score is calculated by the software integrating values of normalized preferred speed, step and leg length ratio, step time, right–left asymmetry of step length and dynamic base of support. See Gouelle et al. (2011) for a more detailed description. Mean FAP score for the two walks was used in the analysis. Accelerometer data were processed using custom-made software run in MATLAB 7.3.0. (Mathworks Inc., Natick, MA, USA). Root mean square (RMS) of trunk acceleration amplitudes in AP (APaccRMS), ML (MLaccRMS) and V (VaccRMS) directions were calculated. Statistics for Windows, version 19 (IBM SPSS Statistics, Armonk, NY, USA: IBM Corp.) was used for statistical analysis. Primary and secondary outcome measures were analysed for differences between groups at Post using a general linear model (ANCOVA) with postintervention values as a dependent variable and with intervention as a factor. Baseline values were used as covariates to control for differences between groups at baseline. Change in absolute speed from Pre to Post was included as covariate in order to control for velocity dependency in trunk acceleration amplitudes (Kavanagh, 2009). Effect size is presented as mean difference between groups at Post with 95% confidence interval. Within-group effects on change were tested with paired samples t-tests or Wilcoxon Signed Ranked Test (strength measures).
SPSS
Results The flow of participants throughout the study is shown in Figure 1. All except one participant in the TT group completed the training period. This person experienced Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
Multiple Sclerosis and Walking
Figure 1. Flow of participants through enrollment, intervention allocation, follow-up and data analysis
a fall after one of the training sessions that resulted in a fracture of the hip. One person in the TT group did not perform the post-test because of influenza. Because of these missing data, considered to be missing completely at random, an intention to treat approach was not feasible. Therefore, listwise deletion was used, resulting in a ‘complete case analysis’, including only those with known outcome. This resulted in 11 participants in the TT group and 15 in the ST group. Baseline demographics and clinical characteristics of the participants are shown in Table 1. The participants completed on average 22 out of 24 training sessions (identical in both groups), with a standard deviation of 1.7 for the TT group and 1.8 for the ST group. No adverse effects of the training were reported, but one participant had to reduce her work employment slightly during the training period. Results for all outcome measures are shown in Table 2. There were no significant differences between groups at baseline in the outcome measures except for higher APaccRMS and MLaccRMS values in the Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
TT group compared with the ST group (t-test p = .029 and .041, respectively). A significant difference of 3.9 scores between groups at post-test was found for the FAP score in favour of the TT group. The TT group had a close to significant improvement in their FAP score (p = .051) following training, while the ST group remained unchanged (p = . 844). Moreover, a significant difference between groups at post-test was also found for two of the secondary outcome measures. The walking work economy improved significantly more in the TT group, with a mean difference between groups of 1.03 mL kg 1 minute 1 following training. The within-group results showed that both groups improved significantly or close to significance (p = . 025 for the TT group and p = .061 for the ST group). There was no significance within-group or between-group differences in VO2peak. Also, a significant reduction in the VaccRMS was found in the TT group compared with the ST group, with a mean difference of 0.352 m second2 (p = .047). The within-group result was however not significant.
(mediolateral) direction. Significant at 0.05 level.
Within-group and between-group summery results, presented as mean (standard deviation) of the FAP score, the WWE and AccRMS (acceleration root mean square) in AP (anteroposterior), V (vertical) and ML
mean square in vertical direction; MLaccRMS = acceleration root mean square in mediolateral direction.
3.9 (.25, 7.5) 1.3 ( 1.9, .150) .033 ( 2.15, 2.81) .352 ( .649, .055) .107 ( .293, .080) .844 .061 .455 .242 .170 (6.2) (1.57) (.32) (.57) (.23) 90.3 12.66 1.55 2.19 1.44 (10.9) (1.91) (.28) (.59) (.23) 91.7 13.34 1.49 2.06 1.37 .051 .025 .877 .117 .062 (3.7) (1.33) (.38) (.44) (.42) FAP (0–100) 1 1 WWE (mL kg minute ) 2 APaccRMS (m second ) 2 VaccRMS (m second ) 2 MLaccRMS (m second )
Pre
Treadmill training (n = 11) Table 2. Summary results
The main result in our study was that TT was superior to progressive ST in improving walking in people with MS. There were significant improvements in walking work economy and reduced amplitude in vertical trunk acceleration in favour of TT compared with ST. We are not aware of any other studies comparing TT and progressive ST on walking in the MS population. Both interventions may be defined as exercise training with a potential beneficial effect on walking (Rietberg et al., 2004; Snook and Motl, 2009). The two training modalities were comparable in duration and frequency (30 minutes, three times a week over a period of 8 weeks), setting (outpatient clinical setting with equal therapeutic supervision) and number of completed training sessions. The between-group difference in FAP score in favour of the treadmill group was 3.9 scores that is above the calculated measurement error (3.6 scores). A meta-analysis of exercise in patients with MS reports an overall standardized effect size of 0.19 on walking mobility (Snook and Motl, 2009), and the authors conclude that although small, this is most likely clinically significant. Our absolute effect size of 3.9 scores is equivalent to a Cohen’s d of 0.59. Taken together, this gives strength to our hypothesis that TT is preferable to progressive ST when the aim is to improve walking in persons with mild to moderate MS. A beneficial effect of TT on walking is in line with, and adds further evidence, to the review by Swinnen et al. (2012), reporting that TT improves walking in MS.
Pre
Discussion
94.2 11.38 1.79 2.12 1.60
Strength training (n = 15)
Strength in the ST group increased with 25% to 137% (p-values ≤ .011) for the four different strength tests.
91.0 (6.6) 12.81 (2.47) 1.78 (.37) 2.32 (.59) 1.85 (.66)
PP = primary progressive; SP = secondary progressive.
Paired samples t-test
EDSS = Expanded Disability Status Scale; RR = relapsing remitting;
Post
49.1 (7.4) 5/10 26 (6.0) 3.2 (1.4) 9/5/1 6.2 (6.6) 11/4
Paired samples t-test
46.6 (6.2) 4/7 28 (3.7) 3.1 (1.6) 10/0/1 8.3 (6.4) 6/5
.037 .024 .622 .047 .125
ANCOVA
Strength training (n = 15)
Post
Age mean (SD) Gender: males/females Body mass index mean (SD) EDSS mean (SD) Classification: RR/PP/SP Time since onset mean (SD) Working: Yes/no
Treadmill training (n = 11)
Mean between group difference at post-test (95%CI)
Table 1. Subject characteristics
ANCOVA = analysis of covariance; FAP = Functional Ambulation Profile; WWE = Walking Work Economy; APaccRMS = acceleration root mean square in anteroposterior direction; VaccRMS = acceleration root
S. M. Brændvik et al.
Multiple Sclerosis and Walking
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
A significant difference between groups in favour of the TT was also detected in two of the secondary outcome measures. First, walking work economy improved significantly more in the treadmill group than in the ST group. The improvement in the treadmill group was above 10%, which is regarded a clinically relevant change (Wezenberg et al., 2011). Second, the vertical acceleration amplitude decreased significantly in the treadmill group compared with the ST group. Trunk acceleration amplitudes during gait are suggested as a surface measure of center of mass displacement (Moe-Nilssen, 1998b) and is thus a reflection of balance control during gait (Kavanagh and Menz, 2008). Increased trunk accelerations during walking are reported in MS (Spain et al., 2012), as well as in other patient populations experiencing balance problems (Iosa et al., 2012) and may be a result of excessive reactive adjustments, possibly explained by inadequate anticipatory balance control (Moe-Nilssen, 1998b). Moreover, altered balance control is suggested to increase metabolic cost during walking (Wezenberg et al., 2011), which is compatible with a close relation between trunk accelerations and energy expenditure during walking (Bouten et al., 1997). Eventually, our results indicate that walking has become both easier and more stable in those who performed TT compared with those who performed the ST. Despite randomization, the treadmill group had significantly higher acceleration amplitudes in both mediolateral and anteroposterior directions at baseline compared with the ST group. From a functional perspective, this might indicate that balance control during walking was poorer in the treadmill group, and accordingly, they had more potential for improvement. This could explain why the treadmill group showed a trend towards improvement (p = .062) in mediolateral accelerations following training, while the ST group remained unchanged. In order to control for imbalance in baseline scores in the statistical analysis, we used an ANCOVA when testing differences between the groups at Post because this statistics adjust each patient’s follow-up score for his or her baseline score. However, because sample size and power calculation in the current study were based on the primary outcome measure, the FAP score, and not trunk accelerations, the acceleration results need to be verified in a larger sample. Despite a significant increase in lower limb muscle strength following ST, and previously reported Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Multiple Sclerosis and Walking
correlation between strength and walking (Thoumie et al., 2005), less improvements in walking were obtained in the ST group. There could be several reasons for this. First, it could be that muscle weakness was not the primary limitation for walking in this group of relatively well-functioning patients with MS. Walking is a submaximal task, and the muscular strength of the study participants could be sufficient to fulfil the task requirements. Second, we assessed gait function over a relatively short distance. The use of alternative tests assessing gait over longer distances that put greater demands on muscle endurance could have given a different result. There are a few considerations regarding this study. First, we do not know whether there was a lasting effect on walking in the treadmill group. Moreover, it could be speculated whether the difference in walking work economy reflects a learning effect because the treadmill group were both tested and trained on a treadmill. However, both the FAP score and the trunk accelerations were measured during over ground walking. Further, it would have been interesting to know whether the improvement in walking was reflected in an increase in daily activities and participation. Our study sample comprised persons with mild to moderate disease progression. Accordingly, the results may not be generalized to more severely affected persons. Also, the treadmill programme was fairly rigorous, and although it was well tolerated in the current study sample, it might not be feasible for more severely affected patients.
Conclusion The results from this study indicate that unsupported TT is a more favourable approach than progressive ST when the aim is to improve walking in persons with MS classified with an EDSS level of 6 or better. These results are clinically important as they bring forth knowledge about the efficiency of different training modalities for this group of patients.
Conflict of interest The authors declare no conflict of interest.
Acknowledgements We gratefully thank the participants for their involvement and physical therapists and staff at the Department of Neurology, St. Olavs Hospital for their help. Special
Multiple Sclerosis and Walking
thanks to Kari Espeset for her valuable help in recruiting participants. We also thank Mona Lyngstad for providing training supervision. This study was founded by the Liaison Committee between the Central Norway Regional Health Authority and the Sør-Trøndelag University College.
REFERENCES American College of Sports Medicine position stand. Progression models in resistance training for healthy adults. Medicine and Science in Sports and Exercise 2009; 41: 687–708. Aaslund MK, Helbostad JL & Moe-Nilssen R. Walking during body-weight-supported treadmill training and acute responses to varying walking speed and bodyweight support in ambulatory patients post-stroke. Physiotherapy Theory and Practice 2013; 29: 278–289. Bouten CV, Sauren AA, Verduin M & Janssen JD. Effects of placement and orientation of body-fixed accelerometers on the assessment of energy expenditure during walking. Medical and Biological Engineering and Computing 1997; 35: 50–56. Cameron MH & Wagner JM. Gait abnormalities in multiple sclerosis: Pathogenesis, evaluation, and advances in treatment. Current Neurology and Neuroscience Reports 2011; 11: 507–515. Crenshaw SJ, Royer TD, Richards JG & Hudson DJ. Gait variability in people with multiple sclerosis. Multiple Sclerosis 2006; 12: 613–619. Fletcher GF, Ades PA, Kligfield P, Arena R, Balady GJ, Bittner VA, Coke LA, Fleg JL, Forman DE, Gerber TC, Gulati M, Madan K, Rhodes J, Thompson PD & Williams MA. Exercise standards for testing and training: A scientific statement from the American Heart Association. Circulation 2013; 128: 873–934. Givon U, Zeilig G & Achiron A. Gait analysis in multiple sclerosis: Characterization of temporal-spatial parameters using GAITRite functional ambulation system. Gait & Posture 2009; 29: 138–142. Gouelle A, Megrot F, Presedo A, Pennecot GF & Yelnik A. Validity of Functional Ambulation Performance Score for the evaluation of spatiotemporal parameters ofchildren’s gait. Journal of Motor Behavior 2011; 43: 95–100. Henriksen M, Lund H, Moe-Nilssen R, Bliddal H, Danneskiod-Samsøe B. Test–retest reliability of trunk accelerometric gait analysis. Gait & Posture 2004; 19(3): 288–297. Hornby TG, Straube DS, Kinnaird CR, Holleran CL, Echauz AJ, Rodriguez KS, Wagner EJ, Narducci EA. Importance of specificity, amount, and intensity of
S. M. Brændvik et al.
locomotor training to improve ambulatory function in patients poststroke. Topics in Stroke Rehabilitation 2011; 18(4): 293–307. Iosa M, Marro T, Paolucci S & Morelli D. Stability and harmony of gait in children with cerebral palsy. Research in Developmental Disabilities 2012; 33: 129–135. Kavanagh JJ. Lower trunk motion and speed-dependence during walking. Journal of Neuroengineering and Rehabilitation 2009; 6: 9. Kavanagh JJ & Menz HB. Accelerometry: A technique for quantifying movement patterns during walking. Gait & Posture 2008; 28: 1–15. Kjolhede T, Vissing K, Dalgas U. Multiple sclerosis and progressive resistance training: A systematic review. Multiple Sclerosis 2012; 18(9): 1215–1228. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: An Expanded Disability Status Scale (EDSS). Neurology 1983; 33: 1444–1452. Lambert CP, Archer RL & Evans WJ. Muscle strength and fatigue during isokinetic exercise in individuals with multiple sclerosis. Medicine and Science in Sports and Exercise 2001; 33: 1613–1619. Martin CL, Phillips BA, Kilpatrick TJ, Butzkueven H, Tubridy N, Mcdonald E & Galea MP. Gait and balance impairment in early multiple sclerosis in the absence of clinical disability. Multiple Sclerosis 2006; 12: 620–628. Medbø JI, Mamen A, Welde B, von Heimburg E, Stokke R. Examination of the Metamax I and II oxygen analysers during exercise studies in the laboratory. Scandinavian Journal of Clinical and Laboratory Investigation 2002; 62: 585–598. Moe-Nilssen R. A new method for evaluating motor control in gait under real-life environmental conditions. Part 1: The instrument. Clinical Biomechanics (Bristol, Avon) 1998a; 13: 320–327. Moe-Nilssen R. A new method for evaluating motor control in gait under real-life environmental conditions. Part 2: Gait analysis. Clinical Biomechanics (Bristol, Avon) 1998b; 13: 328–335. Motl RW, Goldman MD, Benedict RH. Walking impairment in patients with multiple sclerosis: exercise training as a treatment option. Neuropsychiatric Disease and Treatment 2010; 6: 767–774. Motl RW, Suh Y, Dlugonski D, Weikert M, Agiovlasitis S, Fernhall B & Goldman M. Oxygen cost of treadmill and over-ground walking in mildly disabled persons with multiple sclerosis. Neurological Sciences 2011; 32: 255–262. Nelson AJ. Functional Ambulation Profile. Physical Therapy 1974; 54: 1059–1065. Rietberg MB, Brooks D, Uitdehaag BMJ & Kwakkel G. Exercise therapy for multiple sclerosis. Cochrane Database of Systematic Reviews 2004. DOI: 10.1002/14651858. CD003980.pub2. Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
Saether R, Helbostad JL, Adde L, Brændvik S, Lydersen S, Vik T. Gait characteristics in children and adolescents with cerebral palsy assessed with a trunk-worn accelerometer. Research in Developmental Disabilities 2014; 35(7): 1773–1781. Epub 2014 Mar 26. Schwid SR, Thornton CA, Pandya S, Manzur KL, Sanjak M, Petrie MD, Mcdermott MP & Goodman AD. Quantitative assessment of motor fatigue and strength in MS. Neurology 1999; 53: 743–750. Snook EM & Motl RW. Effect of exercise training on walking mobility in multiple sclerosis: A meta-analysis. Neurorehabilitation and Neural Repair 2009; 23: 108–116. Sosnoff JJ, Weikert M, Dlugonski D, Smith DC & Motl RW. Quantifying gait impairment in multiple sclerosis using GAITRite technology. Gait & Posture 2011; 34: 145–147. Spain RI, St George RJ, Salarian A, Mancini M, Wagner JM, Horak FB & Bourdette D. Body-worn motion sensors detect balance and gait deficits in people with multiple sclerosis who have normal walking speed. Gait & Posture 2012; 35: 573–578.
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Multiple Sclerosis and Walking
Swinnen E, Beckwee D, Pinte D, Meeusen R, Baeyens JP & Kerckhofs E. Treadmill training in multiple sclerosis: Can body weight support or robot assistance provide added value? A systematic review. Mult Scler Int 2012; 2012: 240–274. Thoumie P, Lamotte D, Cantalloube S, Faucher M & Amarenco G. Motor determinants of gait in 100 ambulatory patients with multiple sclerosis. Multiple Sclerosis 2005; 11: 485–491. Webster KE, Wittwer JE & Feller JA. Validity of the GAITRite walkway system for the measurement of averaged and individual step parameters of gait. Gait & Posture 2005; 22: 317–321. Wezenberg D, De Haan A, Van Bennekom CA & Houdijk H. Mind your step: metabolic energy cost while walking an enforced gait pattern. Gait & Posture 2011; 33: 544–549. White LJ & Castellano V. Exercise and brain health— implications for multiple sclerosis: Part 1—neuronal growth factors. Sports Medicine2008; 38: 91–100.
Treadmill Training or Progressive Strength Training to Improve Walking in People with Multiple Sclerosis? A Randomized Parallel Group Trial Siri Merete Brændvik1,2*, Teija Koret2, Jorunn L. Helbostad1,2, Håvard Lorås3, Geir Bråthen1,4, Harald Olav Hovdal4 & Inger Lise Aamot2,5 1
Department of Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway
2
Clinical Services, St. Olavs University Hospital, Trondheim, Norway
3
Sør-Trøndelag University CollegeDepartment of Physical TherapyTrondheim, Norway
4
Department of Neurology, St. Olavs University Hospital, Trondheim, Norway
5
The K.G. Jebsen Center of Exercise in Medicine/Department of Circulation and Medical Imaging, Faculty of Medicine, Norwegian University
of Science and Technology, Trondheim, Norway
Abstract Background and purpose. The most effective treatment approach to improve walking in people with multiple sclerosis (MS) is not known. The aim of this trial was to assess the efficacy of treadmill training and progressive strength training on walking in people with MS. Methods. A single blinded randomized parallel group trial was carried out. Eligible participants were adults with MS with Expanded Disability Status Scale score ≤6. A total of 29 participants were randomized and 28 received the allocated exercise intervention, treadmill (n = 13) or strength training (n = 15). Both groups exercised 30 minutes, three times a week for 8 weeks. Primary outcome was The Functional Ambulation Profile evaluated by the GAITRite walkway. Secondary outcomes were walking work economy and balance control during walking, measured by a small lightweight accelerometer connected to the lower back. Testing was performed at baseline and the subsequent week after completion of training. Results. Two participants were lost to follow-up, and 11 (treadmill) and 15 (strength training) were left for analysis. The treadmill group increased their Functional Ambulation Profile score significantly compared with the strength training group (p = .037). A significant improvement in walking work economy (p = .024) and a reduction of root mean square of vertical acceleration (p = .047) also favoured the treadmill group. Discussion. The results indicate that task-specific training by treadmill walking is a favourable approach compared with strength training to improve walking in persons with mild and moderate MS. Implications for Physiotherapy practice, this study adds knowledge for the decision of optimal treatment approaches in people with MS. Copyright © 2015 John Wiley & Sons, Ltd. Received 27 January 2014; Revised 5 February 2015; Accepted 5 May 2015 Keywords multiple sclerosis; strength; treadmill; walking *Correspondence Siri Merete Brændvik, PT/PhD. Department of Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway. Email: [email protected]
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pri.1636
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
Multiple Sclerosis and Walking
Introduction Impaired walking is prominent in people with multiple sclerosis (MS) (Motl et al., 2010) and used as an indicator of disease progression (Kurtzke, 1983). The impairments have been demonstrated as reduced walking speed (Thoumie et al, 2005; Givon et al., 2009), increased variability in hip, knee and ankle kinematics compared with healthy controls (Crenshaw et al., 2006; Sosnoff et al., 2011) and impaired postural control (Martin et al., 2006). In addition, and presumably as a result, increased energy cost during walking has been reported in people with MS compared with healthy controls (Motl et al., 2011). MS is associated with a decrease in physical activity, which in turn can result in deconditioning in domains like aerobic capacity, muscle strength and balance. This process feeds back and leads to a vicious circle of inactivity (Motl et al., 2010). Importantly, this physiological deconditioning might influence the onset and progression of impaired walking in persons with MS (White and Castellano, 2008). From a clinical perspective, it is therefore essential to find interventions that aim to attenuate this process and maintain or improve walking. Exercise training is considered to be an important part of the rehabilitation in persons with MS. There is strong evidence in favour of exercise training compared with no exercise training for muscle power, exercise tolerance and mobility-related activities (Rietberg et al., 2004; Snook and Motl, 2009). However, no evidence is found to support the superiority of a particular approach on mobility in general or on walking in particular (Rietberg et al., 2004). Muscle weakness is one possible factor that may contribute to impair walking in MS (Schwid et al., 1999; Lambert et al., 2001; Cameron and Wagner, 2011). A strong correlation between strength and impaired walking has been reported (Thoumie et al., 2005). As such, progressive strength training (ST) might possibly prove to be efficient in improving walking. However, a recent systematic review reports that although there is strong evidence that progressive ST increases strength in persons with Expanded Disability Status Scale (EDSS) below 6.5, the transfer to improved walking is questionable (Kjolhede et al., 2012). Treadmill training is another approach used to improve walking. Walking on a treadmill allows for massed task-specific practice that is suggested to be critical in obtaining improvement of motor function following
stroke (Hornby et al., 2011). A recent systematic review (Swinnen et al., 2012) addressing the effect of different treadmill training (TT) modalities (unsupported, body weight supported and robot assisted) in people with MS concluded that TT improved walking speed and walking distance. However, it was not clear which type of TT was the most effective. Moreover, the included studies compared TT with either another TT modality to conventional gait training or no training. Therefore, the objective of this study was to compare the effect of two different exercise approaches, unsupported TT and progressive ST, on walking in people with mild to moderate MS. It was hypothesized that TT would be superior to progressive ST in improving walking.
Methods This was a single-blinded randomized trial, configured as a two-group pre-test–post-test design, incorporating two experimental groups and stratified for age and gender. The study took place at Trondheim University Hospital, Norway, and eligible participants were recruited in the period between January and November 2010. After the baseline testing, participants were randomly assigned to TT or progressive ST in a 1:1 ratio. Allocation of the participants was performed by the unit of Applied Clinical Research at the Norwegian University of Science and Technology, using a webbased randomization system. The study was conducted according to the Declaration of Helsinki and was approved by the Regional Committee for Medical and Health Research Ethics South East Norway. Written informed consent was obtained from the participants before inclusion. Sample size calculation was based on the primary outcome measure, the Functional Ambulation Profile (FAP) (Nelson, 1974). In order to detect a 10% difference in FAP score between the groups, with a two-sided 5% significance level and a power of 80%, a sample size of 13 participants per group was needed (Givon et al., 2009).
Participants Eligible participants were adults (aged 18 years and over) diagnosed with MS, having a maximum EDSS score of 6, no relapses of disease or new medications the last 6 months, no pronounced spasticity with need of Baclofen, sign of pyramidal affection examined by magnetic resonance imaging and with effect on gait Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
clinically examined by a neurologist, and no other conditions affecting their gait function. Eligible participants were identified through a systematic review of the outpatients’ medical record of patients under active follow-up at the Department of Neurology at the Trondheim University Hospital, where we aimed to identify all patients that fulfiled the eligibility criteria. All patients were clinically examined by an experienced physical therapist (T. K.) and a neurologist (H. H.).
Multiple Sclerosis and Walking
for each exercise separately at the first training session. Two sets of six repetitions were performed for each exercise, and each leg was trained separately. Resistance was increased by a minimum of 0.25 kg once the participants were able to perform two sets of six repetitions. There was a 2-minute recovery period between each set and between different exercises. Total duration of the training programme was 30 minutes. Outcome measures
Interventions Participants took part in group exercise three times a week for 8 weeks (maximal 24 training sessions) in and outpatient clinical setting. Exercise sessions were supervised by experienced physical therapists working at the hospital and who were not involved in the assessments. The treadmill training was carried out without body weight support, but the participants were allowed to use the handrails for balance support. The training consisted of three different walking bouts, each lasting for 7 minutes: 1) preferred walking speed at an increased slope; 2) walking with verbal guidance on gait pattern, focusing on heel strike, loading and toe-off and knee control during weight-bearing; and 3) fast gait speed defined as a 10% increase in walking speed relative to preferred speed. During the 8-week intervention, walking speed in bout 3 was gradually increased with 10% to 40% of preferred speed. Preferred walking speed was used as departure point for each training session to calculate the intensity on increased velocity and inclined walking. The participants wore a heart rate monitor and were instructed to keep their exercise intensity below 70% of maximal heart rate that was obtained during the treadmill test, to keep focus on walking pattern and avoid exhaustion during gait. Each session lasted for 30 minutes, including 2-minute break between the bouts. The ST was conducted according to current guidelines from the American College of Sports Medicine position stand (2009). The programme consisted of five exercises, comprising both concentric and eccentric components: 1) knee extensor strength performed in a leg press device (lying position); 2) plantar flexion strength in leg press device (extended knees); 3) standing hip abductions with low pulley; 4) dorsal flexion strength in sitting position with the use of pulley; and 5) core and back muscles in sitting position using pull-down device. Work load during training was 80% of one repetition maximum that was set Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
The primary outcome measure was the FAP score. The FAP provides a single score for gait, based on the principles that walking ability attains several sub-skills, like a stable base of support, weight transfer in the mediolateral direction and forward progression (Nelson, 1974). The score ranges from 0–100, where 100 is the best score. In healthy adult populations, the mean FAP is reported to be 95 with a standard deviation of 1 (Givon et al., 2009). The FAP score is reported to be a valid measure of gait impairments in persons with MS classified with an EDSS level between 4 and 6 (Sosnoff et al., 2011), also in those with relatively short duration of the disease (Givon et al., 2009). Reliability of the FAP score in the MS population has to our knowledge not been reported, neither has sensitivity to change. Secondary outcome measures were balance control during walking, measured as trunk acceleration amplitudes, and walking work economy. Trunk accelerations have previously been tested for precision and accuracy (Moe-Nilssen, 1998a), and acceleration amplitudes have shown to be sensitive measures in discriminating between MS patients and healthy controls with respect to balance during walking and quiet standing (Spain et al., 2012). Trunk accelerations have been found to be reliable in healthy persons (Henriksen et al., 2004), as well as in children with cerebral palsy (Saether et al., 2014). We are not aware of published reliability measures on trunk accelerations in the MS population, but unpublished data from our own patient database show test–retest reliability (1 week apart) assessed by an intraclass correlation coefficient (ICC 2.1) for acceleration amplitude in anteroposterior, mediolateral and vertical direction, ranging from 0.87 to 0.94. Procedure and equipment The participants were tested, prior to randomization, the week before the training started (Pre) and within a week after the last training session (Post).
S. M. Brændvik et al.
Multiple Sclerosis and Walking
The GAITRite walkway system (CIR Systems, Inc., Sparta, NJ, USA) was used to obtain the FAP score. The GAITRite is an electronic walkway with sensors arranged in a grid-like pattern to identify footfall contacts, and it has been found to be a valid method for measuring temporal and spatial gait parameters (Webster et al., 2005). The active length of the walkway was 6.1 m. An additional 0.90 m at each end of the walkway was added to give room for acceleration and deceleration. The participants walked according to the following instruction: ‘Walk as you usually do, at your preferable speed’. All participants completed two walks (back and forth along the walkway once). They wore good walking shoes, and the same shoes were used at all test sessions. One test trial was performed before measurement. Concurrently, balance control during walking was measured using a small lightweight inertial sensor (MTx, XSens, Enschede, Netherlands), attached to the lower back (L3 region) by an elastic belt. This sensor contains tri-axial units of accelerometers, gyroscopes and magnetometers and was connected to a batteryoperated communication unit also worn by the subjects (Aaslund et al., 2013). Data were sampled at a frequency of 100 Hz and transferred wireless to a laptop in real time for processing. Only acceleration data were used in the current study and were reported along anteroposterior, mediolateral and vertical axes. Photoelectric timing units were synchronized to the sensor system and placed at each end of the active area of the electronic walkway to register time sequences for the acceleration data for each walk. Energy expenditure during walking was measured during a treadmill exercise test, and respiratory gas analysis was performed using a Metamax II (Cortex Biophysics, Leipzig, Germany). Metamax II has shown to be valid and reliable for gas measurements (Medbø et al., 2002). After 5 minutes of familiarization to treadmill walking, oxygen consumption (VO2) was measured during 2 minutes walking at a pre-set speed (3 km hour 1). The participants were encouraged to walk without handrail support. Heart rate was measured continuously during the test. In addition, peak oxygen consumption (VO2peak) was measured using an individually ramp protocol (Fletcher et al., 2013) in order to calculate the intended exercise intensity during treadmill training. To obtain a stability estimate of the FAP score, the walking protocol was repeated in a retest 5 days after Pre, just prior to the start of intervention. Intraclass
correlation coefficient of the FAP score was found to be 0.91 (ICC 2.1), with a 95% confidence interval of .79–.96, reflecting high level of agreement between test occasions. Measurement error (within-subject standard deviation) was 3.6 FAP scores. Data analysis The GAITRite software system (Version 3.8) calculates spatial and temporal parameters based on processed footfalls. Walking speed is calculated by dividing the distance walked by ambulation time, and the normalization is carried out by dividing it to leg length (GAITRite manual). The FAP score is calculated by the software integrating values of normalized preferred speed, step and leg length ratio, step time, right–left asymmetry of step length and dynamic base of support. See Gouelle et al. (2011) for a more detailed description. Mean FAP score for the two walks was used in the analysis. Accelerometer data were processed using custom-made software run in MATLAB 7.3.0. (Mathworks Inc., Natick, MA, USA). Root mean square (RMS) of trunk acceleration amplitudes in AP (APaccRMS), ML (MLaccRMS) and V (VaccRMS) directions were calculated. Statistics for Windows, version 19 (IBM SPSS Statistics, Armonk, NY, USA: IBM Corp.) was used for statistical analysis. Primary and secondary outcome measures were analysed for differences between groups at Post using a general linear model (ANCOVA) with postintervention values as a dependent variable and with intervention as a factor. Baseline values were used as covariates to control for differences between groups at baseline. Change in absolute speed from Pre to Post was included as covariate in order to control for velocity dependency in trunk acceleration amplitudes (Kavanagh, 2009). Effect size is presented as mean difference between groups at Post with 95% confidence interval. Within-group effects on change were tested with paired samples t-tests or Wilcoxon Signed Ranked Test (strength measures).
SPSS
Results The flow of participants throughout the study is shown in Figure 1. All except one participant in the TT group completed the training period. This person experienced Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
Multiple Sclerosis and Walking
Figure 1. Flow of participants through enrollment, intervention allocation, follow-up and data analysis
a fall after one of the training sessions that resulted in a fracture of the hip. One person in the TT group did not perform the post-test because of influenza. Because of these missing data, considered to be missing completely at random, an intention to treat approach was not feasible. Therefore, listwise deletion was used, resulting in a ‘complete case analysis’, including only those with known outcome. This resulted in 11 participants in the TT group and 15 in the ST group. Baseline demographics and clinical characteristics of the participants are shown in Table 1. The participants completed on average 22 out of 24 training sessions (identical in both groups), with a standard deviation of 1.7 for the TT group and 1.8 for the ST group. No adverse effects of the training were reported, but one participant had to reduce her work employment slightly during the training period. Results for all outcome measures are shown in Table 2. There were no significant differences between groups at baseline in the outcome measures except for higher APaccRMS and MLaccRMS values in the Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
TT group compared with the ST group (t-test p = .029 and .041, respectively). A significant difference of 3.9 scores between groups at post-test was found for the FAP score in favour of the TT group. The TT group had a close to significant improvement in their FAP score (p = .051) following training, while the ST group remained unchanged (p = . 844). Moreover, a significant difference between groups at post-test was also found for two of the secondary outcome measures. The walking work economy improved significantly more in the TT group, with a mean difference between groups of 1.03 mL kg 1 minute 1 following training. The within-group results showed that both groups improved significantly or close to significance (p = . 025 for the TT group and p = .061 for the ST group). There was no significance within-group or between-group differences in VO2peak. Also, a significant reduction in the VaccRMS was found in the TT group compared with the ST group, with a mean difference of 0.352 m second2 (p = .047). The within-group result was however not significant.
(mediolateral) direction. Significant at 0.05 level.
Within-group and between-group summery results, presented as mean (standard deviation) of the FAP score, the WWE and AccRMS (acceleration root mean square) in AP (anteroposterior), V (vertical) and ML
mean square in vertical direction; MLaccRMS = acceleration root mean square in mediolateral direction.
3.9 (.25, 7.5) 1.3 ( 1.9, .150) .033 ( 2.15, 2.81) .352 ( .649, .055) .107 ( .293, .080) .844 .061 .455 .242 .170 (6.2) (1.57) (.32) (.57) (.23) 90.3 12.66 1.55 2.19 1.44 (10.9) (1.91) (.28) (.59) (.23) 91.7 13.34 1.49 2.06 1.37 .051 .025 .877 .117 .062 (3.7) (1.33) (.38) (.44) (.42) FAP (0–100) 1 1 WWE (mL kg minute ) 2 APaccRMS (m second ) 2 VaccRMS (m second ) 2 MLaccRMS (m second )
Pre
Treadmill training (n = 11) Table 2. Summary results
The main result in our study was that TT was superior to progressive ST in improving walking in people with MS. There were significant improvements in walking work economy and reduced amplitude in vertical trunk acceleration in favour of TT compared with ST. We are not aware of any other studies comparing TT and progressive ST on walking in the MS population. Both interventions may be defined as exercise training with a potential beneficial effect on walking (Rietberg et al., 2004; Snook and Motl, 2009). The two training modalities were comparable in duration and frequency (30 minutes, three times a week over a period of 8 weeks), setting (outpatient clinical setting with equal therapeutic supervision) and number of completed training sessions. The between-group difference in FAP score in favour of the treadmill group was 3.9 scores that is above the calculated measurement error (3.6 scores). A meta-analysis of exercise in patients with MS reports an overall standardized effect size of 0.19 on walking mobility (Snook and Motl, 2009), and the authors conclude that although small, this is most likely clinically significant. Our absolute effect size of 3.9 scores is equivalent to a Cohen’s d of 0.59. Taken together, this gives strength to our hypothesis that TT is preferable to progressive ST when the aim is to improve walking in persons with mild to moderate MS. A beneficial effect of TT on walking is in line with, and adds further evidence, to the review by Swinnen et al. (2012), reporting that TT improves walking in MS.
Pre
Discussion
94.2 11.38 1.79 2.12 1.60
Strength training (n = 15)
Strength in the ST group increased with 25% to 137% (p-values ≤ .011) for the four different strength tests.
91.0 (6.6) 12.81 (2.47) 1.78 (.37) 2.32 (.59) 1.85 (.66)
PP = primary progressive; SP = secondary progressive.
Paired samples t-test
EDSS = Expanded Disability Status Scale; RR = relapsing remitting;
Post
49.1 (7.4) 5/10 26 (6.0) 3.2 (1.4) 9/5/1 6.2 (6.6) 11/4
Paired samples t-test
46.6 (6.2) 4/7 28 (3.7) 3.1 (1.6) 10/0/1 8.3 (6.4) 6/5
.037 .024 .622 .047 .125
ANCOVA
Strength training (n = 15)
Post
Age mean (SD) Gender: males/females Body mass index mean (SD) EDSS mean (SD) Classification: RR/PP/SP Time since onset mean (SD) Working: Yes/no
Treadmill training (n = 11)
Mean between group difference at post-test (95%CI)
Table 1. Subject characteristics
ANCOVA = analysis of covariance; FAP = Functional Ambulation Profile; WWE = Walking Work Economy; APaccRMS = acceleration root mean square in anteroposterior direction; VaccRMS = acceleration root
S. M. Brændvik et al.
Multiple Sclerosis and Walking
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
A significant difference between groups in favour of the TT was also detected in two of the secondary outcome measures. First, walking work economy improved significantly more in the treadmill group than in the ST group. The improvement in the treadmill group was above 10%, which is regarded a clinically relevant change (Wezenberg et al., 2011). Second, the vertical acceleration amplitude decreased significantly in the treadmill group compared with the ST group. Trunk acceleration amplitudes during gait are suggested as a surface measure of center of mass displacement (Moe-Nilssen, 1998b) and is thus a reflection of balance control during gait (Kavanagh and Menz, 2008). Increased trunk accelerations during walking are reported in MS (Spain et al., 2012), as well as in other patient populations experiencing balance problems (Iosa et al., 2012) and may be a result of excessive reactive adjustments, possibly explained by inadequate anticipatory balance control (Moe-Nilssen, 1998b). Moreover, altered balance control is suggested to increase metabolic cost during walking (Wezenberg et al., 2011), which is compatible with a close relation between trunk accelerations and energy expenditure during walking (Bouten et al., 1997). Eventually, our results indicate that walking has become both easier and more stable in those who performed TT compared with those who performed the ST. Despite randomization, the treadmill group had significantly higher acceleration amplitudes in both mediolateral and anteroposterior directions at baseline compared with the ST group. From a functional perspective, this might indicate that balance control during walking was poorer in the treadmill group, and accordingly, they had more potential for improvement. This could explain why the treadmill group showed a trend towards improvement (p = .062) in mediolateral accelerations following training, while the ST group remained unchanged. In order to control for imbalance in baseline scores in the statistical analysis, we used an ANCOVA when testing differences between the groups at Post because this statistics adjust each patient’s follow-up score for his or her baseline score. However, because sample size and power calculation in the current study were based on the primary outcome measure, the FAP score, and not trunk accelerations, the acceleration results need to be verified in a larger sample. Despite a significant increase in lower limb muscle strength following ST, and previously reported Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Multiple Sclerosis and Walking
correlation between strength and walking (Thoumie et al., 2005), less improvements in walking were obtained in the ST group. There could be several reasons for this. First, it could be that muscle weakness was not the primary limitation for walking in this group of relatively well-functioning patients with MS. Walking is a submaximal task, and the muscular strength of the study participants could be sufficient to fulfil the task requirements. Second, we assessed gait function over a relatively short distance. The use of alternative tests assessing gait over longer distances that put greater demands on muscle endurance could have given a different result. There are a few considerations regarding this study. First, we do not know whether there was a lasting effect on walking in the treadmill group. Moreover, it could be speculated whether the difference in walking work economy reflects a learning effect because the treadmill group were both tested and trained on a treadmill. However, both the FAP score and the trunk accelerations were measured during over ground walking. Further, it would have been interesting to know whether the improvement in walking was reflected in an increase in daily activities and participation. Our study sample comprised persons with mild to moderate disease progression. Accordingly, the results may not be generalized to more severely affected persons. Also, the treadmill programme was fairly rigorous, and although it was well tolerated in the current study sample, it might not be feasible for more severely affected patients.
Conclusion The results from this study indicate that unsupported TT is a more favourable approach than progressive ST when the aim is to improve walking in persons with MS classified with an EDSS level of 6 or better. These results are clinically important as they bring forth knowledge about the efficiency of different training modalities for this group of patients.
Conflict of interest The authors declare no conflict of interest.
Acknowledgements We gratefully thank the participants for their involvement and physical therapists and staff at the Department of Neurology, St. Olavs Hospital for their help. Special
Multiple Sclerosis and Walking
thanks to Kari Espeset for her valuable help in recruiting participants. We also thank Mona Lyngstad for providing training supervision. This study was founded by the Liaison Committee between the Central Norway Regional Health Authority and the Sør-Trøndelag University College.
REFERENCES American College of Sports Medicine position stand. Progression models in resistance training for healthy adults. Medicine and Science in Sports and Exercise 2009; 41: 687–708. Aaslund MK, Helbostad JL & Moe-Nilssen R. Walking during body-weight-supported treadmill training and acute responses to varying walking speed and bodyweight support in ambulatory patients post-stroke. Physiotherapy Theory and Practice 2013; 29: 278–289. Bouten CV, Sauren AA, Verduin M & Janssen JD. Effects of placement and orientation of body-fixed accelerometers on the assessment of energy expenditure during walking. Medical and Biological Engineering and Computing 1997; 35: 50–56. Cameron MH & Wagner JM. Gait abnormalities in multiple sclerosis: Pathogenesis, evaluation, and advances in treatment. Current Neurology and Neuroscience Reports 2011; 11: 507–515. Crenshaw SJ, Royer TD, Richards JG & Hudson DJ. Gait variability in people with multiple sclerosis. Multiple Sclerosis 2006; 12: 613–619. Fletcher GF, Ades PA, Kligfield P, Arena R, Balady GJ, Bittner VA, Coke LA, Fleg JL, Forman DE, Gerber TC, Gulati M, Madan K, Rhodes J, Thompson PD & Williams MA. Exercise standards for testing and training: A scientific statement from the American Heart Association. Circulation 2013; 128: 873–934. Givon U, Zeilig G & Achiron A. Gait analysis in multiple sclerosis: Characterization of temporal-spatial parameters using GAITRite functional ambulation system. Gait & Posture 2009; 29: 138–142. Gouelle A, Megrot F, Presedo A, Pennecot GF & Yelnik A. Validity of Functional Ambulation Performance Score for the evaluation of spatiotemporal parameters ofchildren’s gait. Journal of Motor Behavior 2011; 43: 95–100. Henriksen M, Lund H, Moe-Nilssen R, Bliddal H, Danneskiod-Samsøe B. Test–retest reliability of trunk accelerometric gait analysis. Gait & Posture 2004; 19(3): 288–297. Hornby TG, Straube DS, Kinnaird CR, Holleran CL, Echauz AJ, Rodriguez KS, Wagner EJ, Narducci EA. Importance of specificity, amount, and intensity of
S. M. Brændvik et al.
locomotor training to improve ambulatory function in patients poststroke. Topics in Stroke Rehabilitation 2011; 18(4): 293–307. Iosa M, Marro T, Paolucci S & Morelli D. Stability and harmony of gait in children with cerebral palsy. Research in Developmental Disabilities 2012; 33: 129–135. Kavanagh JJ. Lower trunk motion and speed-dependence during walking. Journal of Neuroengineering and Rehabilitation 2009; 6: 9. Kavanagh JJ & Menz HB. Accelerometry: A technique for quantifying movement patterns during walking. Gait & Posture 2008; 28: 1–15. Kjolhede T, Vissing K, Dalgas U. Multiple sclerosis and progressive resistance training: A systematic review. Multiple Sclerosis 2012; 18(9): 1215–1228. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: An Expanded Disability Status Scale (EDSS). Neurology 1983; 33: 1444–1452. Lambert CP, Archer RL & Evans WJ. Muscle strength and fatigue during isokinetic exercise in individuals with multiple sclerosis. Medicine and Science in Sports and Exercise 2001; 33: 1613–1619. Martin CL, Phillips BA, Kilpatrick TJ, Butzkueven H, Tubridy N, Mcdonald E & Galea MP. Gait and balance impairment in early multiple sclerosis in the absence of clinical disability. Multiple Sclerosis 2006; 12: 620–628. Medbø JI, Mamen A, Welde B, von Heimburg E, Stokke R. Examination of the Metamax I and II oxygen analysers during exercise studies in the laboratory. Scandinavian Journal of Clinical and Laboratory Investigation 2002; 62: 585–598. Moe-Nilssen R. A new method for evaluating motor control in gait under real-life environmental conditions. Part 1: The instrument. Clinical Biomechanics (Bristol, Avon) 1998a; 13: 320–327. Moe-Nilssen R. A new method for evaluating motor control in gait under real-life environmental conditions. Part 2: Gait analysis. Clinical Biomechanics (Bristol, Avon) 1998b; 13: 328–335. Motl RW, Goldman MD, Benedict RH. Walking impairment in patients with multiple sclerosis: exercise training as a treatment option. Neuropsychiatric Disease and Treatment 2010; 6: 767–774. Motl RW, Suh Y, Dlugonski D, Weikert M, Agiovlasitis S, Fernhall B & Goldman M. Oxygen cost of treadmill and over-ground walking in mildly disabled persons with multiple sclerosis. Neurological Sciences 2011; 32: 255–262. Nelson AJ. Functional Ambulation Profile. Physical Therapy 1974; 54: 1059–1065. Rietberg MB, Brooks D, Uitdehaag BMJ & Kwakkel G. Exercise therapy for multiple sclerosis. Cochrane Database of Systematic Reviews 2004. DOI: 10.1002/14651858. CD003980.pub2. Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. M. Brændvik et al.
Saether R, Helbostad JL, Adde L, Brændvik S, Lydersen S, Vik T. Gait characteristics in children and adolescents with cerebral palsy assessed with a trunk-worn accelerometer. Research in Developmental Disabilities 2014; 35(7): 1773–1781. Epub 2014 Mar 26. Schwid SR, Thornton CA, Pandya S, Manzur KL, Sanjak M, Petrie MD, Mcdermott MP & Goodman AD. Quantitative assessment of motor fatigue and strength in MS. Neurology 1999; 53: 743–750. Snook EM & Motl RW. Effect of exercise training on walking mobility in multiple sclerosis: A meta-analysis. Neurorehabilitation and Neural Repair 2009; 23: 108–116. Sosnoff JJ, Weikert M, Dlugonski D, Smith DC & Motl RW. Quantifying gait impairment in multiple sclerosis using GAITRite technology. Gait & Posture 2011; 34: 145–147. Spain RI, St George RJ, Salarian A, Mancini M, Wagner JM, Horak FB & Bourdette D. Body-worn motion sensors detect balance and gait deficits in people with multiple sclerosis who have normal walking speed. Gait & Posture 2012; 35: 573–578.
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Multiple Sclerosis and Walking
Swinnen E, Beckwee D, Pinte D, Meeusen R, Baeyens JP & Kerckhofs E. Treadmill training in multiple sclerosis: Can body weight support or robot assistance provide added value? A systematic review. Mult Scler Int 2012; 2012: 240–274. Thoumie P, Lamotte D, Cantalloube S, Faucher M & Amarenco G. Motor determinants of gait in 100 ambulatory patients with multiple sclerosis. Multiple Sclerosis 2005; 11: 485–491. Webster KE, Wittwer JE & Feller JA. Validity of the GAITRite walkway system for the measurement of averaged and individual step parameters of gait. Gait & Posture 2005; 22: 317–321. Wezenberg D, De Haan A, Van Bennekom CA & Houdijk H. Mind your step: metabolic energy cost while walking an enforced gait pattern. Gait & Posture 2011; 33: 544–549. White LJ & Castellano V. Exercise and brain health— implications for multiple sclerosis: Part 1—neuronal growth factors. Sports Medicine2008; 38: 91–100.
