Does Therapeutic Facilitation Add to Locomotor Outcome of Body Weight−Supported Treadmill Training in Nonambulatory Patients With Stroke? A Randomized Controlled Trial

Article Outline

• Abstract
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgment
• References
• Copyright

Abstract 

Yagura H, Hatakenaka M, Miyai I. Does therapeutic facilitation add to locomotor outcome of body weight−supported treadmill training in nonambulatory patients with stroke? A randomized controlled trial.

Objective

To assess benefit of the facilitation technique (FT) coupled with body weight–supported treadmill training (BWSTT) in nonambulatory patients with stroke.

Design

Randomized controlled trial.

Setting

Inpatient rehabilitation hospital.

Participants

Forty-nine patients with nonambulatory patients with stroke were randomly allocated to BWSTT coupled with the FT or mechanical assistance (control).

Interventions

Swinging and stance of the paretic leg were assisted using the FT or mechanically (control) during BWSTT.

Main Outcome Measures

The FIM instrument, Fugl-Meyer Assessment, gait speed, and cadence.

Results

Demographic and clinical features of the FT (n=22) and control (n=25) groups on admission were comparable after excluding 2 dropouts. There were no differences in the gains of the main outcome measures between the FT and control groups. Patients with severe impairment in the FT group had greater gains in arm function than those in the control group.

Conclusions

The FT did not add significantly to locomotor outcome of BWSTT in nonambulatory patients with stroke but it did require more therapists’ assistance.

Key Words:  Gait , Hemiplegia , Rehabilitation , Stroke , Treadmill test , Weight-bearing

WHILE IT IS NOT LIKELY that patients in the chronic phase of stroke with severe hemiparesis will achieve functional use of the paretic hand, considerable gains in gait function are expected after intensive rehabilitation in these patients.¹ This suggests that locomotor function is more likely to recover than hand function in patients who failed to recover within a few months after the onset of stroke. However, evidence-based strategies for locomotor training have not been established. For instance, it remains to be determined whether body weight−supported treadmill training (BWSTT) enhances locomotor recovery after stroke.², ³, ⁴, ⁵, ⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹² A recent review¹³ has also suggested that the effect of BWSTT on locomotor outcome of patients with stroke is not conclusive, although it is likely to be beneficial for patients with mild gait disturbance. The inconsistency of the results might be at least partially due to variations in locomotor status of patients, degree of body weight support (BWS), training speed, methods for assisting patients, and control interventions.⁵, ⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹² Although BWSTT enables even nonambulatory patients to be trained to walk, it requires additional assistance in stepping. There are at least 2 ways to provide this assistance. One is to assist the paretic foot mechanically by holding the paretic leg. The other is the facilitation technique, in which a therapist presses the hip of a patient forward and backward to ensure stable swing and stance of the paretic leg.¹⁴ The facilitation technique (FT) induced enhanced cortical activations in multiple motor areas, including the premotor cortex, during hemiparetic gait¹⁵ as well as improved gait performance, such as gait symmetry.¹⁶ The afferent input from the hip joint and adequate load appear to be essential to facilitate activities in the paretic leg.¹⁷ In healthy subjects, “unilateral walking,” where they only moved the unilateral leg on the treadmill with 70% of BWS, induced muscle activities on the contralateral leg that was not moving.¹⁷ Because such muscle activities were not seen during unilateral walking in patients with complete spinal cord injury,¹⁷ supraspinal control might play an important role in the interlimb coordination during gait.¹⁸, ¹⁹ Cortical areas, including the supplementary motor area, premotor cortex, and sensorimotor cortex, appear to be involved in the interlimb coordination.²⁰ An optical imaging study revealed that the FT enhanced activations in the medial sensorimotor cortex and the premotor cortex during hemiparetic gait.¹⁵ Increased activities in the premotor cortex are likely to be associated with improved locomotor outcome.²¹ We hypothesized that using FT might be more beneficial than simply assisting the paralyzed foot mechanically during BWSTT. We focused on severe patients who had failed to reach independent gait after 4 weeks of inpatient rehabilitation and investigated whether FT improved functional outcome of BWSTT in a randomized controlled trial.

Back to Article Outline

Methods 

In a hospital with 245 beds providing approximately 3 months of inpatient rehabilitation that is fully covered by the national insurance system,¹, ²² we selected patients within 3 months after the onset of stroke as possible candidates for this study. Patients were transferred to the hospital for multidisciplinary rehabilitation because they still needed assistance in activities of daily living (ADLs) after medical treatment coupled with a less intensive physical therapy (PT) 3 to 5 days a week in acute hospitals. Exclusion criteria included: (1) being more than 80 years old; (2) inability to understand the informed consent form because of impaired cognitive function; (3) previous stroke or dependence in ADLs prior to stroke; (4) history of myocardial infarction within 1 year; (5) uncontrolled hypertension; (6) symptomatic orthostatic hypotension; and (7) atrial fibrillation with uncontrolled rate. Next, only those who needed physical assistance for gait after 4 weeks of inpatient rehabilitation (40-min PT and occupational therapy [OT] 5d/wk and 40-min speech-language pathology [SLP] therapy 5d/wk as needed) were invited to the study. We expected that additional 1-point improvement in 7 items of the FIM instrument²³ related to the lower-extremity function (gait, stairs, transfer ×3, bathing, toileting) by using the FT. Because the average gains ± standard deviation (SD) in FIM total score in our institute were approximately 30±10, the standardized effect size was 7/10=0.7. Thus with α value of .05 and β value of .20, the minimal sample size estimated was 52.²⁴ Of 863 patients hospitalized from January 2002 to December 2002, 279 patients meeting the criteria were invited to the study. We obtained written informed consent from 49 patients (fig 1). The other patients either refused to participate in this study or could not make a decision within 5 days after the proposal.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Flow of patient participation.

BWSTT was started at the fifth week after admission in 49 patients. BWSTT was included in the ordinary PT sessions 3 days a week for 6 weeks (weeks 5–10). Approximately half of each 40-minute PT session was used for BWSTT. The patients also received additional PT sessions without BWSTT 2 days a week and OT and SLP therapy as before during the BWSTT period. By selecting from an envelope, the participants were randomly assigned to either type of BWSTT where therapists assisted the swing and stance of the paretic leg either mechanically (control group) or by using the FT (FT group). Patients could not be blinded to the assigned intervention type. In the FT group, experienced therapists assisted flexion of the knee for the initiation of the swing phase, and prevented the pelvis from being hitched up by handling the hip and pelvis.¹⁴ Patients with severe impairment in the FT group initially needed additional mechanical assistance both in the unaffected leg and affected leg. However, the FT group differed from the control group in that continuous sensorimotor stimulation was provided in the hip and pelvic regions during locomotor training on the treadmill. Degree of BWS provided by using the overhead harness with a pelvic belt and thigh strips attached to a suspension system¹⁵, ²⁵, ²⁶ was optimized for each patient to walk most comfortably (0%−50%) at the beginning of each session. Treadmill speed was increased progressively as patients and assisting physical therapists could tolerate (0.2−3.0km/h). Outcome measures included Fugl-Meyer Assessment (FMA) motor scale for the upper (UE) and lower extremity (LE),²⁷ total FIM score, FIM motor subscore, and FIM gait item. These measures were assessed at the baseline on admission, 4 weeks after admission before BWSTT was started, 10 weeks when 6 weeks of BWSTT was completed, and 16 weeks after admission. Gait measures including overground gait speed for 10m, stride, and cadence were evaluated at the baseline on admission, 4 weeks after admission before BWSTT was started, every 2 weeks during the BWSTT, and in follow-up periods up to 16 weeks by therapists who were not blinded to the allocation. Gait speed in nonambulatory patients even with assistance was rated as 0m/s. Stride and cadence could not be measured in such patients. This study was approved by the local ethics committee.

For statistical analyses of demographic data in the 2 groups (FT vs control), we used the chi-square test or Mann-Whitney U test. To compare the outcome measures of the 2 groups, we used 2-factor repeated-measures analysis of variance (ANOVA). Post hoc test was adjusted for multiple comparisons using a Bonferroni adjustment. We also used Cox proportional hazards analyses whether the group allocation might predict the possibility of reaching independent gait indoor and gait with supervision or light contact guard. Statistical significance was set at P less than .05.

Back to Article Outline

Results 

Of 49 patients, 23 patients were allocated to the FT group and 26 to the control group. BWSTT was discontinued at the first session in 2 patients (1 FT, 1 control). The main reason for the dropout was discomfort caused by a parachute harness during BWS. Otherwise there were no adverse effects associated with BWSTT. Further analyses were performed in 47 patients (FT=22, control=25). Demographic and clinical features of the patients on admission are comparable (table 1). Lesions were mainly located in the subcortical areas involving the internal capsule and basal ganglia. Three patients in the FT group had the premotor lesions. During 6 weeks of BWSTT, the degree of BWS and treadmill speed did not differ significantly between the groups (table 2). The FT group needed significantly more therapists than the control group (P<.003), although there was no interaction between group and time (F_6,44=1.3, P=.25). This might reflect that severely disabled patients tended to need mechanical assistance of the paretic leg in addition to the FT. The rate of recovery as measured by weekly gains in FIM total score was greater in the baseline period (2.9±2.6/wk) than in the BWSTT period (1.8±1.5/wk, P<.05), probably due to the common nature of recovery curve after stroke. Figure 2 shows serial changes of the outcome measures. Repeated-measures ANOVA for total FIM score revealed significant main effects for group (F_1,45=4.1, P<.05) and for time (F_3,135=105.4, P<.001), without significant interaction. This indicates that general disability in the control group was significantly less severe than in the FT group and that there was a relative failure of the randomization to balance the 2 groups on the main outcome measures. Mean changes ± SD of total FIM score per week in the FT and control groups were 1.7±0.9 and 1.9±1.2, respectively (not significant [NS]). Only main effect for time, but not for group, was significant for motor FIM score (F_3,135=49.8, P<.001) and gait speed (F_7,233=32.4, P<.001). Mean changes ± SD of motor FIM score per week in the FT and control group were 1.2±1.0 and 1.3±1.2, respectively (NS). Because stride and cadence were not measured in patients who remained nonambulatory, these parameters were not included in statistical analysis. Similarly, in FMA scores for the UE and LE, there were no group differences although both groups equally improved during the study period (UE: F_3,135=38.8, P<.001; LE: F_3,135=54.3, P<.001).

Table 1. Demographic and Clinical Features of Nonambulatory Patients With Stroke

NOTE. Values are mean ± SD unless otherwise indicated.

Abbreviations: CH, cerebral hemorrhage; CI, cerebral infarct; Com, combined cortical and subcortical lesion; Ctx, cortical lesion; L, left; MMSE, Mini-Mental State Examination; R, right; Sub, subcortical lesion.

Table 2. Interventions During the BWSTT Period in Each Group

Group	Week 5	Week 6	Week 7	Week 8	Week 9	Week 10
Degree of BWS (%)
FT	12.0±5.9	8.5±6.6	6.9±6.7	5.4±6.1	5.6±7.1	5.1±6.7
Control	10.2±10.5	8.6±9.6	7.5±10.8	5.0±8.5	4.0±8.7	3.0±8.4
Treadmill speed (km/h)
FT	0.61±0.33	0.73±0.34	0.85±0.30	0.96±0.36	1.0±0.35	1.1±0.36
Control	0.77±0.30	0.91±0.30	0.96±0.32	1.1±0.30	1.1±0.30	1.1±0.40
No. of therapists needed for assistance
FT	1.9±0.75	1.8±0.61	1.6±0.60	1.6±0.58	1.6±0.73	1.6±0.66
Control	1.1±0.68	1.2±0.87	1.1±0.66	1.0±0.66	1.0±0.66	0.8±0.6

NOTE. Values are mean ± SD.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Changes of the outcome measures in the FT and control groups (n=47). Improvement of (A) FMA UE scores, (B) FMA LE scores, (C) total FIM scores, (D) FIM gait scores, (E) FIM motor scores, and (F) gait speed were comparable. Values are mean ± standard error (SE) (error bars). Abbreviations: FIM-g, FIM gait score; FM, Fugl-Meyer Assessment.

Next, we divided the patients into 2 subgroups based on the locomotor status as measured by the FIM gait item (FIM gait) on admission. In patients with severe gait disorder (FIM gait score, 1; n=22), improvement of FIM, FIM gait, gait speed, and FMA score for the LE were comparable between the FT and control group. However, improvement of FMA score for the UE was greater in the FT group (n=13) than the control group (n=9), as indicated by significant interaction between group and time (F_3,60=3.3, P<.05) as well as a significant main effect for time (F_3,135=24.8, P<.001) but not for group (fig 3). In patients with mild to moderate gait disorder (FIM gait range, 2–4; n=25), improvement of all these parameters was comparable.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Changes of the outcome measures in patients with severe gait disorder (FIM gait score, 1 on admission, n=22). ANOVA indicated that improvement of (A) FMA UE scores was significantly better in the FT group than in the control group. (B) Improvement of FMA LE scores, (C) total FIM scores, (D) FIM gait scores, (E) FIM motor scores, and (F) gait speed were comparable. Values are mean ± SE (error bars).

Finally, to compare real-world outcome of locomotor function, we performed Cox proportional hazards analyses with adjustment of age and using the baseline locomotor status (FIM gait score on admission and week 4) as covariates. Reaching independent gait indoor (FIM gait score, 6 or 7) was not associated with group (hazard ratio [HR] = .53; 95% confidence interval [CI], 0.12−2.25) or the baseline locomotor status. Reaching gait with supervision (FIM gait score, 5) was not associated with group (HR=.83; 95% CI, 0.37−1.85), but was associated with the locomotor status (FIM gait) at 4 weeks before BWSTT was started (HR=1.73; 95% CI, 1.07−2.79; P<.05).

Back to Article Outline

Discussion 

We hypothesized that using the FT might be more beneficial than simply assisting the paretic leg during BWSTT to nonambulatory patients with stroke. However, we failed to show the benefit of the FT coupled with BWSTT to patients with stroke presenting with severe gait disorder. There might be several reasons for the failure. Number of studied patients might be small. However, this is unlikely because number of recruited patients was close enough to the estimated sample size. The plotted figures also support that increasing sample size would not change our statistical findings. Second, the control group showed less severe impairment and disability on admission and greater improvement during the baseline period before BWSTT was started compared with the FT group. Third, the damage in the descending motor pathways and the basal ganglia causing severe hemiparesis could have diminished the effect of the FT in our patients.²⁸ Fourth, the settings of treadmill speed in our study appear to be slower than those in the previous literature.⁴, ⁵, ⁹, ¹² However, this is unlikely because we studied patients with most severe gait disorder who were nonambulatory even after 4-week inpatient rehabilitation and indeed we tried the highest speed that patients could tolerate. Finally, more patients in the FT group had the premotor lesion that might have altered locomotor outcome.²⁹

Interestingly, patients with the most severe gait disorder in the FT group showed significantly greater improvement of the UE function than those in the control group. One of the adverse effects of BWS is inducing associated movements of the affected UE in an effort to adapt oneself to the unusual condition of BWS.³⁰ We indeed observed that associated movements were less frequent in the FT group than in the control group, although this phenomenon was not evaluated quantitatively. Because the interlimb coordination of the legs and arms, such as arm swing during gait, is mediated by the propriospinal neurons,³¹ it might be possible that afferent input induced by the hip manipulation might reduce associated movements of the affected UE via these neurons. Enhanced cortical activation in the motor related areas induced by the FT¹⁵ might also be associated with greater improvement of the affected UE. To clarify whether such improvement of the UE function in patients with severe gait disorder is clinically relevant, further studies are needed.

Back to Article Outline

Conclusions 

The FT did not add significantly to locomotor outcome of BWSTT in nonambulatory patients with stroke, while requiring more therapists’ assistance. Further studies are needed to determine whether the greater improvement in FMA scores of the UE in the FT group is clinically relevant.

Back to Article Outline

Acknowledgment 

We thank all the therapists at Bobath Memorial Hospital for their role in patient selection and evaluation.

Back to Article Outline

References 

1. Yagura H , Miyai I , Seike Y , Suzuki T , Yanagihara T . Benefit of inpatient multidisciplinary rehabilitation up to 1 year after stroke . Arch Phys Med Rehabil . 2003;84:1687–1691
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (227 KB)
   • CrossRef
2. Hesse S , Bertelt C , Schaffrin A , Malezic M , Mauritz KH . Restoration of gait in nonambulatory hemiparetic patients by treadmill training with partial body-weight support . Arch Phys Med Rehabil . 1994;75:1087–1093
   • View In Article
   • MEDLINE
   • CrossRef
3. Hesse S , Bertelt C , Jahnke MT , et al.   Treadmill training with partial body weight support compared with physiotherapy in nonambulatory hemiparetic patients . Stroke . 1995;26:976–981
   • View In Article
   • MEDLINE
   • CrossRef
4. Hesse S , Helm B , Krajnik J , Gregoric M , Mauritz KH . Treadmill training with partial body weight support (influence of body weight release on the gait of hemiparetic patients) . J Neuro Rehabil . 1997;11:15–20
   • View In Article
5. Hesse S , Konrad M , Uhlenbrock D . Treadmill walking with partial body weight support versus floor walking in hemiparetic subjects . Arch Phys Med Rehabil . 1999;80:1–7
   • View In Article
   • Full-Text PDF (1694 KB)
   • CrossRef
6. 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;29:1122–1128
   • View In Article
   • MEDLINE
   • CrossRef
7. Kosak MC , Reding MJ . Comparison of partial body weight-supported treadmill gait training versus aggressive bracing assisted walking post stroke . Neurorehabil Neural Repair . 2000;14:13–19
   • View In Article
   • MEDLINE
8. Nilsson L , Carlsson J , Danielsson A , et al.   Walking training of patients with hemiparesis at an early stage after stroke (a comparison of walking training on a treadmill with body weight support and walking training on the ground) . Clin Rehabil . 2001;15:515–527
   • View In Article
   • MEDLINE
   • CrossRef
9. Pohl M , Mehrholz J , Ritschel C , Ruckriem S . Speed-dependent treadmill training in ambulatory hemiparetic stroke patients (a randomized controlled trial) . Stroke . 2002;33:553–558
   • View In Article
   • CrossRef
10. da Cunha IT , Lim PA , Qureshy H , Henson H , Monga T , Protas EJ . Gait outcomes after acute stroke rehabilitation with supported treadmill ambulation training (a randomized controlled pilot study) . Arch Phys Med Rehabil . 2002;83:1258–1265
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (97 KB)
    • CrossRef
11. Sullivan KJ , Knowlton BJ , Dobkin BH . Step training with body weight support (effect of treadmill speed and practice paradigms on poststroke locomotor recovery) . Arch Phys Med Rehabil . 2002;83:683–691
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (152 KB)
    • CrossRef
12. Miller EW , Quinn ME , Seddon PG . Body weight support treadmill and overground ambulation training for two patients with chronic disability secondary to stroke . Phys Ther . 2002;82:53–61
    • View In Article
    • MEDLINE
13. Moseley AM , Stark A , Cameron ID , Pollock A . Treadmill training and body weight support for walking after stroke . Cochrane Database Syst Rev . 2005;(4):CD002840
    • View In Article
14. Davis PM . Steps to follow . 2nd ed.. New York: Springer; 2000;
    • View In Article
15. Miyai I , Yagura H , Oda I , et al.   Premotor cortex is involved in restoration of gait in stroke . Ann Neurol . 2002;52:188–194
    • View In Article
    • MEDLINE
    • CrossRef
16. Hesse S , Jahnke MT , Schaffrin A , Lucke D , Reiter F , Konrad M . Immediate effects of therapeutic facilitation on the gait of hemiparetic patients as compared with walking with and without a cane . Electroencephalogr Clin Neurophysiol . 1998;109:515–522
    • View In Article
    • MEDLINE
17. Dietz V , Muller R , Colombo G . Locomotor activity in spinal man (significance of afferent input from joint and load receptors) . Brain . 2002;125:2626–2634
    • View In Article
    • MEDLINE
    • CrossRef
18. Armstrong DM . The supraspinal control of mammalian locomotion . J Physiol . 1988;405:1–37
    • View In Article
    • MEDLINE
19. Drew T , Prentice S , Schepens B . Cortical and brainstem control of locomotion . Prog Brain Res . 2004;143:251–261
    • View In Article
    • MEDLINE
    • CrossRef
20. Debaere F , Swinnen SP , Beatse E , Sunaert S , Van Hecke P , Duysens J . Brain areas involved in interlimb coordination (a distributed network) . Neuroimage . 2001;14:947–958
    • View In Article
    • MEDLINE
    • CrossRef
21. Miyai I , Yagura H , Hatakenaka M , Oda I , Konishi I , Kubota K . Longitudinal optical imaging study for locomotor recovery after stroke . Stroke . 2003;34:2866–2870
    • View In Article
    • CrossRef
22. Miyai I , Suzuki T , Kii K , Kang J , Kajiura I . Functional outcome of multidisciplinary rehabilitation in chronic stroke . J Neuro Rehabil . 1998;12:95–99
    • View In Article
23. Keith RA , Granger CV , Hamilton BB , Sherwin FS . The Functional Independence Measure (a new tool for rehabilitation) . In:  Eisenberg MG ,  Grzesiak RC editor. Advances in clinical rehabilitation . Vol 2: New York: Springer; 1987;p. 6–18
    • View In Article
24. Hulley SB , Cummings SR , Browner WS , Grady D , Hearst N , Norman TB . Designing clinical research (an epidemiologic approach) . 2nd ed.. Philadelphia: Lippincott Williams & Wilkins; 2001;
    • View In Article
25. Miyai I , Fujimoto Y , Ueda Y , et al.   Treadmill training with body weight support (its effect on Parkinson’s disease) . Arch Phys Med Rehabil . 2000;81:849–852
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (46 KB)
    • CrossRef
26. Miyai I , Fujimoto Y , Yamamoto H , et al.   Long-term effect of body weight-supported treadmill training in Parkinson’s disease (a randomized controlled trial) . Arch Phys Med Rehabil . 2002;83:1370–1373
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (76 KB)
    • CrossRef
27. Fugl-Meyer AR , Jaasko L , Leyman I , Olsson S , Steglind S . The post-stroke hemiparetic patient I. A method for evaluation of physical performance . Scand J Rehabil Med . 1975;7: 13-3
    • View In Article
28. Miyai I , Blau AD , Reding MJ , Volpe BT . Patients with stroke confined to basal ganglia have diminished response to rehabilitation efforts . Neurology . 1997;48:95–101
    • View In Article
    • MEDLINE
29. Miyai I , Suzuki T , Kang J , Kubota K , Volpe BT . Middle cerebral artery stroke that includes the premotor cortex reduces mobility outcome . Stroke . 1999;30:1380–1383
    • View In Article
    • MEDLINE
    • CrossRef
30. Davies PM . Weight-supported treadmill training . Neurorehabil Neural Repair . 1999;13:167–169
    • View In Article
31. Dietz V . Spinal networks involved in interlimb co-ordination and reflex regulation of locomotion . In:  Swinnen SP ,  Duysens J editor. Neurobehavioral determinants of interlimb coordination . Norwell: Kluwer Academic Publishers; 2004;p. 93–108
    • View In Article