1370

Long-Term Effect of Body WeightSupported Treadmill Training in Parkinsons Disease: A Randomized Controlled Trial
Ichiro Miyai, MD, PhD, Yasuyuki Fujimoto, RPT, Hiroshi Yamamoto, RPT, Yoshishige Ueda, RPT, Toshio Saito, MD, Sonoko Nozaki, MD, PhD, Jin Kang, MD, PhD
ABSTRACT. Miyai I, Fujimoto Y, Yamamoto H, Ueda Y, Saito T, Nozaki S, Kang J. Long-term effect of body weight supported treadmill training in Parkinsons disease: a randomized controlled trial. Arch Phys Med Rehabil 2002;83: 1370-3. Objective: To investigate whether body weightsupported treadmill training (BWSTT) is of long-term benet for patients with Parkinsons disease (PD). Design: Randomized controlled trial. Setting: Inpatient rehabilitation unit for neurologic diseases in Japan. Participants: Twenty-four patients (Hoehn and Yahr stages 2.5 or 3) who were not demented (Mini-Mental State Examination score, 27). Interventions: Patients were randomized to receive either a 45-minute session of BWSTT (up to 20% of body weight supported) or conventional physical therapy (PT) for 3 days a week for 1 month. Main Outcome Measures: Outcome measures were evaluated at baseline and at 1, 2, 3, and 6 months. Measures included the Unied Parkinsons Disease Rating Scale (UPDRS), ambulation speed (s/10m), and number of steps taken for a 10-m walk as a parameter for stride length. Results: Four patients needed modication of medications in the follow-up period. Twenty patients (BWSTT, n 11; PT, n 9) without modied medications were analyzed for functional outcome. Age, duration of PD, gender, and doses of medications were comparable. There was no difference in the baseline UPDRS (BWSTT 33.3; PT 32.6), speed (BWSTT 10.8; PT 11.5), and steps (BWSTT 23.4; PT 22.8). The BWSTT group had signicantly greater improvement than the PT group (Mann-Whitney U test, Bonferroni adjustment for multiple comparison) in ambulation speed at 1 month (BWSTT 8.5; PT 10.8; P .005); and in the number of steps at 1 (BWSTT 20.0; PT 22.7; P .005), 2 (BWSTT 19.5; PT 22.4; P .005), 3 (BWSTT 20.1; PT 23.1; P .005), and 4 months (BWSTT 21.0; PT 23.0; P .006). Conclusions: BWSTT has a lasting effect specically on short-step gait in PD. Key Words: Body weight; Parkinson disease; Rehabilitation; Treadmill test. 2002 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation LTHOUGH AN prois in improving activities of daily A gramand benecialINTENSIVE physical rehabilitationliving (ADLs) mobility in patients with Parkinsons disease (PD),1 it remains unclear how long its effect is sustained. Other strategies of neurorehabilitation for PD, such as visual, auditory, or somatosensory external cues, have been reported to improve their motor performance only temporally.2-4 In stroke patients, however, several trials5-9 have shown multidisciplinary rehabilitation to be effective in improving mortality, ADLs, mobility, medical costs, and disposition. However, there is no specic strategy or technique that is superior to others.10 This is partially because those rehabilitative interventions are highly variable among therapists, even among those with the same approaches, such as neurodevelopmental technique and proprioceptive neuromuscular facilitation. Locomotor training on a treadmill with partial body weight supported with an overhead harness, a pelvic belt, and thigh straps (body weightsupported treadmill training [BWSTT]) has an advantage in that it is easy to replicate in any rehabilitation facility.11 Recent studies have shown that BWSTT is effective in improving the mobility outcome of patients with spinal cord injury (SCI),12-15 stroke,16-18 and cerebral palsy.19 Favorable shortterm effects of BWSTT on gait disturbance in PD has also been reported.20 To investigate whether BWSTT has a lasting benet for PD patients, we compared functional outcomes of BWSTT and conventional physical therapy (PT) in a randomized controlled trial. We hypothesized that BWSTT has a longer effect on gait disorder in PD than does PT. METHODS Twenty-four PD patients (12 men, 12 women) with Hoehn and Yahr stage 2.5 or 3 who were not demented (Mini-Mental State Examination [MMSE] score, 27) were enrolled in this study. Diagnosis of PD was based on the presence of rest tremor, bradykinesia, rigidity, positive response to levodopa, and no evidence of vascular lesions on magnetic resonance imaging.21 No patients exhibited on-off phenomenon, but all patients experienced freezing phenomenon during gait. Written informed consent was obtained from each patient, who was then randomized to receive either 45-minute sessions of BWSTT or 45-minute sessions of conventional PT 3 days a week for 1 month, for a total of 12 sessions. At the start of the BWSTT session, patients were observed by a therapist to optimize percentage unweighing by using the overhead harness with a pelvic belt and thigh strips attached to a suspension system. Patients were asked to walk with 0%, 10%, 20%, and 30% of their body weight supported. In each percentage of body weight support (BWS), patients walked at

From the Neurorehabilitation Research Institute, Bobath Memorial Hospital (Miyai); and Departments of Rehabilitation (Fujimoto, Yamamoto, Ueda) and Neurology (Saito, Nozaki, Kang), Toneyama National Hospital, Osaka, Japan. Supported by Funds for Comprehensive Research on Aging and Health from the Ministry of Health and Welfare, Japan. Presented in part at the American Society of Neurorehabilitations 7th annual meeting, April 28, 2000, San Diego. No commercial party having a direct nancial interest in the results of the research supporting this article has or will confer a benet upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint request to Ichiro Miyai, MD, PhD, Neurorehabilitation Research Institute, Bobath Memorial Hospital, 1-6-5, Higashinakahama, Joto-ku, Osaka, 536-0023, Japan, e-mail: webeo@ga2.so-net.ne.jp. 0003-9993/02/8310-6992$35.00/0 doi:10.1053/apmr.2002.34603

Arch Phys Med Rehabil Vol 83, October 2002

BODY-WEIGHT SUPPORT IN PARKINSONS DISEASE, Miyai Table 1: Clinical Characteristics of PD Patients

BWSTT PT

1371

Patients (n) Sex (male/female) Age (y) Hoehn and Yahr stage Duration of PD (y) MMSE Medications Levodopa (mg) Trihexyphenidyl (mg) Amantadine (mg) Bromocriptine (mg) Pergolide ( g) DOPS (mg)

11 5/6 69.5 1.9 2.9 0.1 4.1 0.8 28.3 0.5 241.0 1.5 45.0 3.2 100.0 45.5 29.5 0.8 18.4 1.3 50.0 31.2

9 5/4 69.8 1.5 2.8 0.1 4.5 0.7 28.7 0.6 255.6 0.9 68.8 3.1 11.1 55.6 22.8 0.7 21.0 1.7 41.5 37.7

NS NS NS NS NS NS NS NS NS NS NS

NOTE. Data are mean SEM unless otherwise indicated. Abbreviations: NS, not signicant; DOPS, L-threo-3,4-dihydroxyphenylserine (a norepinephrine precursor).

There were no differences in the baseline values of UPDRS (BWSTT 33.3 2.9; PT 32.6 2.8), gait speed (BWSTT 10.8 0.9s/10m; PT 11.5 1.8s/10m), and number of steps taken for the 10-m walk (BWSTT 23.4 2.3; PT 22.8 2.2). In age-matched unimpaired subjects (n 5: 2 men, 3 women; mean age, 68.0 1.5y), the average gait speed standard error of the mean (SEM) was 7.9 0.6, and the number of steps was 14.3 1.1. In patients with PD, the BWSTT group had signicantly greater improvement over the PT group in gait speed at 1 month (BWSTT 8.5 0.7; PT 10.8 1.8; P .005) and in the number of steps for the 10-m walk (g 1) at 1 (BWSTT 20.0 2.1; PT 22.7 2.0; P .005), 2 (BWSTT 19.5 1.7; PT 22.4 1.8; P .005), 3 (BWSTT 20.1 1.9; PT 23.1 2.1; P .005), and 4 months (BWSTT 20.1 1.9; PT 23.1 2.1; P .006). There were no signicant differences in UPDRS and its subscores between the groups after adjustments for multiple comparisons (table 2). DISCUSSION In a previous study,20 we showed that BWSTT was superior to conventional PT in improving short-term mobility outcome in PD. In this study, we not only conrmed our previous results, but we also found that the effects lasted for 4 months. BWSTT was specically effective in improving short-step gait of PD patients. The favorable outcome in BWSTT might be attributed to BWS rather than to treadmill walking because all patients initially could tolerate a higher treadmill speed when walking with BWS than without. A controlled trial that compared the effect of treadmill training both with and without BWS in stroke also showed a greater efcacy with the BWS.18 However, the mechanisms responsible for the functional improvement remain unclear. Clinical improvement of gait disturbance through the use of external cues is observed only when these cues are given2-4; it is unlikely that these external cues induce implicit motor learning. Normal stride length can be explicitly elicited in PD by using attentional strategies as well as external cues.23 A lasting improvement in gait after BWSTT occurred without using such attentional strategies. Training of arm movements, however, resulted in task-specic motor learning.24 Neuroimaging studies have revealed reduced activity in the medial frontal motor areas during hypokinetic gait and

a pace of 0.5, 1.0, 1.5, 2.0, 2.5, and 3.0km/h, as tolerated. All patients were most comfortable when walking with 20% of BWS and most uncomfortable walking with 30% of BWS. (Therefore, patients were not trained with 30% of BWS.) The most comfortable treadmill speed differed for each patients. Consequently, all patients were trained at 20% of BWS for 10 minutes, 10% of BWS for 10 minutes, and then 0% of BWS for 10 minutes, with a 15-minute rest in 1 session. Treadmill speed was initiated at 0.5km/h and increased to 3.0km/h by increments of 0.5km/h as tolerated. The PT program included general conditioning, range-of-motion exercise, ADL training, and gait training. The time spent training in walking in PT was equal to that spent in BWSTT. Patients in both groups also received 45-minute sessions of occupational therapy for training of ADLs and transfers 3 days a week during the study. Medications for PD were not modied throughout the study. Outcome measures included the Unied Parkinsons Disease Rating Scale22 (UPDRS), gait speed, and the number of steps taken for a 10-m walk at the baseline, 1, 2, 3, 4, 5 and 6 months. For evaluation of gait speed, we measured the time required for the 10-m walk. We assessed the number of steps as a parameter for stride length because mean length of stride was equal to 10m 2/number of steps per 10m. Thus, fewer steps per 10m resulted in longer strides. To measure these gait parameters, each patient walked 10m 5 times, and we measured time and number of steps simultaneously. Median values of the 5 walks were used as data to ensure the reliability of the data. We also evaluated changes in ADLs, motor, mental, and complication subscores of the UPDRS. Patients were evaluated for functional outcome at the same time of the day (10 AM11 AM). The Mann-Whitney U test was used for statistical analysis. Statistical signicance was accepted at P less than .05. To adjust for multiple comparison (at 7 points) using the Bonferroni correction, the signicance level was set at P less than .007. RESULTS Four patients (1 in the BWSTT group, 3 in the PT group) had to modify their medications for PD during the follow-up period. Thus, 20 patients (BWSTT, n 11; PT, n 9) were analyzed for functional outcome. Age, duration of PD, gender, and doses of PD medications were comparable between the groups (table 1). At the initial session, all patients could walk without freezing phenomenon at higher treadmill speeds with 10% and 20% of BWS (1.0 2.0km/h) than with no BWS (0.0 1.0 km/h).

Fig 1. Changes of number of steps taken for the 10-m walk. The lower, middle, and upper horizontal lines of the boxes represent 25th, 50th, and 75th percentiles, respectively. The vertical lines extend from the 10th to the 90th percentiles. *P<.005 (signicantly greater improvement from the baseline in the BWSTT group than in the PT group); P .006, Mann-Whitney U test (signicantly greater improvement from the baseline in the BWSTT group than in the PT group).

Arch Phys Med Rehabil Vol 83, October 2002

1372

BODY-WEIGHT SUPPORT IN PARKINSONS DISEASE, Miyai Table 2: Comparison of Functional Outcome Between the BWSTT and PT Groups
Baseline 1 Month 2 Months 3 Months 4 Months 5 Months 6 Months

UPDRS (total) BWSTT PT UPDRS (ADLs) BWSTT PT UPDRS (motor) BWSTT PT UPDRS (mental) BWSTT PT UPDRS (complications) BWSTT PT Speed (s/10m) BWSTT PT Steps (/10m) BWSTT PT

33.3 2.9 32.6 2.8 13.0 1.6 13.2 1.4 18.5 1.2 18.6 1.4 1.1 0.5 0.3 0.2 0.7 0.3 0.2 0.1 10.8 0.9 11.5 1.8 23.4 2.3 22.8 2.2

27.8 3.2 30.1 2.5 10.1 1.6 12.1 1.1 15.5 1.3 17.3 1.4 0.8 0.3 0.1 0.1 0.7 0.3 0.4 0.2 8.5 0.7* 10.8 1.8 20.0 2.1* 22.7 2.0

27.7 3.1 30.4 2.3 11.0 1.5 12.7 1.2 15.6 1.3 17.9 1.3 0.6 0.2 0.2 0.1 0.5 0.3 0.3 0.2 9.0 0.9 10.7 1.5 19.5 1.7* 22.4 1.8

28.8 3.0 31.7 2.6 11.5 1.4 13.0 1.3 16.2 1.4 18.4 1.4 0.6 0.2 0.1 0.1 0.5 0.3 0.1 0.1 9.2 0.9 11.1 1.6 20.1 1.9* 23.1 2.1

28.8 3.0 31.9 2.6 11.8 1.5 13.1 1.1 15.8 1.3 18.0 1.8 0.6 0.2 0.1 0.1 0.5 0.3 0.1 0.1 9.4 1.0 11.4 1.6 21.0 2.4 23.0 2.7

29.2 2.9 32.1 2.6 12.0 1.4 13.3 1.2 16.1 1.3 18.6 1.5 0.6 0.2 0.1 0.1 0.5 0.3 0.1 0.1 9.7 1.1 11.5 1.7 20.5 2.0 23.6 2.5

29.7 3.1 32.6 2.7 12.4 1.5 13.6 1.1 16.3 1.4 18.9 1.7 0.6 0.2 0.1 0.1 0.5 0.3 0.1 0.1 9.8 1.1 11.4 1.7 21.2 2.2 23.8 2.7

NOTE. Data are mean SEM. The Mann-Whitney U test was performed for the baseline values and the gain from the baseline of each outcome measure. To adjust for multiple comparison using the Bonferroni correction, signicant level was set at P .007. * P .005. P .006.

enhanced activity of the lateral premotor cortex when gait was improved through visual cues in PD patients.25,26 Thus, it is possible that BWSTT induces implicit motor learning by enhancing such alternative brain networks. Further studies are necessary to determine whether such reorganization of motorrelated areas also occurs consistently after BWSTT. Enhancement of a central pattern generator, which has been postulated as a mechanism for the efcacy of BWSTT for SCI11 and stroke,27 might be also associated with functional improvements seen in PD patients. CONCLUSION Our results suggest that BWSTT has a lasting effect on gait disturbance in PD patients. BWSTT specically improved short-step gait, which is a clinical characteristic of PD, beyond the time of the intervention. Further studies are needed to determine if the effects of BWSTT can be realized in outpatient clinics. If so, it might be possible to delay increasing PD medications doses, thus reducing for a longer time the risk of unfavorable side effects such as on-off phenomenon and dyskinesia. Further studies are also needed regarding the mechanisms that result in clinical improvement of gait disturbance after BWSTT.
References 1. Comella CL, Stebbins GT, Brown-Toms N, Goetz CG. Physical therapy and Parkinsons disease: a controlled clinical trial. Neurology 1994;44:376-8. 2. Dietz MA, Goetz CG, Stebbins GT. Evaluation of a modied inverted walking stick as a treatment for parkinsonians freezing episodes. Mov Dis 1990;5:243-7. 3. Dam M, Tonin P, Casson S, et al. Effects of conventional and sensory-enhanced physiotherapy on disability of Parkinsons disease patients. Adv Neurol 1996;69:551-5. 4. Thaut MH, McIntosh GC, Rice RR, Miller RA, Rathbun J, Brault JM. Rhythmic auditory stimulation in gait training for Parkinsons disease patients. Mov Dis 1996;11:193-200.
Arch Phys Med Rehabil Vol 83, October 2002

5. Feigenson JS, Gitlow HS, Greenberg SD. The disability oriented rehabilitation unita major factor inuencing stroke outcome. Stroke 1979;10:5-7. 6. Smith DS, Goldenberg E, Ashburn A, et al. Remedial therapy after stroke: a randomized controlled trial. BMJ 1981;282:517-20. 7. Strand T, Asplund K, Eriksson S, Hugg E, Lithner F, Webster PO. A non-intensive stroke unit reduces functional disability and the need for long-term hospitalization. Stroke 1985;16:29-34. 8. Kaste M, Palomaki H, Sarna S. Where and how should elderly stroke patients be treated? A randomized trial. Stroke 1995;26: 249-53. 9. Jrgensen HS, Nakayama H, Raaschou HO, Larsen K, Hubbe P, Olsen TS. The effect of a stroke unit: reductions in mortality, discharge rate to nursing home, length of hospital stay, and cost. A community-based study. Stroke 1995;26:1178-82. 10. Price JR, Reding MJ. Physical therapy philosophies and strategies. In: Good DC, Couch JR Jr, editors. Handbook of neurorehabilitation. New York: Marcel Dekker; 1994. p 181-96. 11. Dobkin BH. An overview of treadmill locomotor training with partial body weight support: a neurophysiologically sound approach whose time has come for randomized clinical trials. Neurorehabil Neural Repair 1999;13:157-65. 12. Visintin M, Barbeau H. The effects of body weight support on the locomotor pattern of spastic paretic patients. Can J Neurol Sci 1989;16:315-25. 13. Dobkin BH, Edgerton VR, Fowler E, Hodgson J. Training induces rhythmic locomotor EMG patterns in a subject with complete spinal cord injury. Neurology 1992;42(Suppl):207-8. 14. Wernig A, Muller S. Laufband locomotion with body weight support improved walking in persons with severe spinal cord injuries. Paraplegia 1992;30:229-38. 15. Barbeau H, Danakas M, Arsenault B. The effect of locomotor training in spinal cord injured subjects: a preliminary study. Restor Neurol Neurosci 1993;5:81-4. 16. Hesse S, Bertelt C, Schaffrin A, Malezic M, Mauritz KH. Restoration of gait in nonambulatory hemiplegic patients by treadmill training with partial body-weight support. Arch Phys Med Rehabil 1994;75:1087-93.

BODY-WEIGHT SUPPORT IN PARKINSONS DISEASE, Miyai

1373

17. Hesse S, Bertelt C, Jahnke MT, et al. Treadmill training with partial body-weight support as compared to physiotherapy in non-ambulatory hemiparetic patients. Stroke 1995;26:976-81. 18. Visintin M, Barbeau H, Korner-Bitensky N, Mayo NE. A new approach to retrain gait in stroke patients through body weight support and treadmill stimulation. Stroke 1998;29:1122-8. 19. Schindl MR, Forstner C, Kern H, Hesse S. Treadmill training with partial body weight support in nonambulatory patients with cerebral palsy. Arch Phys Med Rehabil 2000;81:301-6. 20. Miyai I, Fujimoto Y, Yamamoto H, et al. Treadmill training with partial body weight support: its effect on Parkinsons disease. Arch Phys Med Rehabil 2000;81:849-52. 21. Brooks DJ. The early diagnosis of Parkinsons disease. Ann Neurol 1998;44 Suppl 1:S10-8. 22. Fahn S, Elton RL. Unied Parkinsons Disease Rating Scale. In: Fahn S, Marsden CD, Calne D, Goldstein M, editors. Recent

23. 24. 25. 26. 27.

developments in Parkinsons disease. Vol 2. Florham Park (NJ): Macmillan Healthcare Information; 1987. p 153-63. Morris ME, Iansek R, Matyas TA, Summers JJ. Stride length regulation in Parkinsons disease. Normalization strategies and underlying mechanisms. Brain 1996;119:551-68. Platz T, Brown RG, Marsden CD. Training improves the speed of aimed movements in Parkinsons disease. Brain 1998;121:505-14. Hanakawa T, Katsumi Y, Fukuyama H, et al. Mechanisms underlying gait disturbance in Parkinsons disease. A single photon emission computed tomography study. Brain 1999;122:1271-82. Hanakawa T, Fukuyama H, Katsumi Y, Honda M, Shibasaki H. Enhanced lateral premotor activity during paradoxical gait in Parkinsons disease. Ann Neurol 1999;45:329-36. Hesse S. Treadmill training with partial body weight support in hemiparetic patientsfurther research needed. Neurorehabil Neural Repair 1999;13:179-81.

Arch Phys Med Rehabil Vol 83, October 2002