Escolar Documentos
Profissional Documentos
Cultura Documentos
Robotic Gait Training For Indiviuals With Cerebral Palsy: A Systematic Review And
Meta-Analysis
Igor da Silveira Carvalho, Srgio Medeiros Pinto, Daniel das Virgens Chagas, Jomilto
Luiz Praxedes dos Santos, Tain de Sousa Oliveira, Luiz Alberto Batista
PII: S0003-9993(17)30474-4
DOI: 10.1016/j.apmr.2017.06.018
Reference: YAPMR 56958
Please cite this article as: da Silveira Carvalho I, Pinto SM, Chagas DdV, Praxedes dos Santos JL, de
Sousa Oliveira T, Batista LA, Robotic Gait Training For Indiviuals With Cerebral Palsy: A Systematic
Review And Meta-Analysis, ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION (2017), doi:
10.1016/j.apmr.2017.06.018.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Running Head: Robotic Gait Training
Title: Robotic Gait Training For Indiviuals With Cerebral Palsy: A Systematic Review
And Meta-Analysis
Authors: Igor da Silveira Carvalho1, Srgio Medeiros Pinto1, Daniel das Virgens
PT
Chagas2, Jomilto Luiz Praxedes dos Santos2, Tain de Sousa Oliveira3, Luiz Alberto
Batista4
RI
SC
1. MSc, Laboratory of Biomechanics and Motor Behavior, Graduate Program in
Medical Sciences - Rio de Janeiro State University (UERJ). Rio de Janeiro, Brazil.
U
2. PhD, Laboratory of Biomechanics and Motor Behavior, Institute of Physical
AN
Education and Sports - Rio de Janeiro State University (UERJ). Rio de Janeiro, Brazil.
and Sports - Rio de Janeiro State University (UERJ). Rio de Janeiro, Brazil.
D
Physical Education and Sports - Rio de Janeiro State University (UERJ). Rio de
Acknowledgements
The first author is supported by the National Counsel of Technological and Scientific
Development (CNPq).
ACCEPTED MANUSCRIPT
Corresponding Author: Igor da Silveira Carvalho, Laboratory of Biomechanics and
Motor Behavior, Institute of Physical Education and Sports - Rio de Janeiro State
PT
RI
U SC
AN
M
D
TE
C EP
AC
1
ACCEPTED MANUSCRIPT
1 Robotic Gait Training for Individuals with Cerebral Palsy: A Systematic Review
2 and Meta-Analysis
4 Abstract
PT
6 Objective: To identify the effects of robotic gait training practices in individuals with
7 cerebral palsy.
RI
8
SC
9 Data Sources: The search was performed in the following electronic databases:
11
U
database of systematic reviews), Web of Science, Scopus, Compendex, IEEE Xplore,
AN
12 ScienceDirect, Academic Search Premier, and PEDro.
13
M
14 Study Selection: Studies were included if they fulfilled the following criteria: (1) they
D
15 investigated the effects of robotic gait training, (2) they involved patients with cerebral
TE
16 palsy, and (3) they enrolled patients classified between levels I and IV using the Gross
18
19 Data Extraction: The information was extracted from the selected articles using the
C
20 descriptive-analytical method. The "Critical Review Form for Quantitative Studies" was
AC
21 used to quantitate the presence of critical components in the articles. To perform the
22 meta-analysis, the effects of the intervention were quantified by effect size (Cohen's
23 d).
24
25 Data Synthesis: Of the 133 identified studies, 10 met the inclusion criteria. The meta-
26 analysis showed positive effects on gait speed (0.21 [-0.09, 0.51]), endurance (0.21 [-
2
ACCEPTED MANUSCRIPT
27 0.06, 0.49]), and gross motor function in dimension D (0.18 [-0.10, 0.45]) and
29
30 Conclusion: The results obtained suggest that this training benefits people with
31 cerebral palsy, specifically by increasing walking speed and endurance and improving
PT
32 gross motor functions. For future studies, we suggest investigating device
RI
34 trials with larger sample sizes and individuals with homogeneous impairment.
SC
35
37
U
AN
38 Abbreviations:
51
52
3
ACCEPTED MANUSCRIPT
53 Introduction
54
55 Cerebral palsy (CP) is the most common cause of physical disability in childhood.1,2 It
PT
58 neurological nature and is caused by a non-progressive disorder that affects the brain
RI
60
SC
61 Motor impairments induced by CP are characterized, in part, by the reduced ability to
63 and fatigue.5,6
U
AN
64
65 For the patient, CP results in significant anatomical and functional changes such as a
M
67 changes in the posture of body segments. An example of this is equinus foot, which is
TE
68 a condition characterized by the heel being kept high during ambulation, which in turn
70
73
74 Gait is commonly affected by CP; therefore, people affected tend to have typical
77
4
ACCEPTED MANUSCRIPT
78 Changes such as these significantly impair the physical quality of the gait and are
80 during ambulation15 and pain,5,6 ultimately leading to a diminished quality of life. Thus,
81 the main goal of rehabilitation programs is to enable and/or restore motor function and
PT
83
RI
85 similar movements to the motor skills to be enabled or rehabilitated, are effective in
SC
86 improving a patients motor function.18,19
87
88
U
Thus, different therapeutic resources emphasizing repetitive movements have been
AN
89 designed and incorporated into gait treatment programs for individuals with
92
TE
93 Similar to other therapeutic approaches, the availability of evidence that supports the
98
99 The aim of this study was to identify the effects of robotic gait training practices in
101
102
103
5
ACCEPTED MANUSCRIPT
104 Methods
105 This systematic review was conducted according to Preferred Reporting Items for
106 Systematic Reviews and Meta-Analyses (PRISMA) guidelines.22,23 The study did not
108
PT
109 Search Strategy
110
RI
111 The systematic literature search was restricted to full articles that were written in
SC
112 English, Spanish, or German and published from January 1980 to November 2016.
113
114
U
The search was performed in the following electronic databases: PubMed, EMBASE
AN
115 (Excerpta Medical), MEDLINE (OvidSP), CDSR (Cochrane database of systematic
117 Academic Search Premier, and PEDro. The terms used in the electronic search were
D
119
120 Two researchers (CI and PS or BLA) were responsible for reviewing the titles and
EP
121 abstracts of the articles found and selecting those that met the inclusion criteria. For
122 the items on which the primary analysts did not reach an agreement, we requested the
C
124
126
127 Studies were included if they fulfilled the following criteria: (1) they investigated the
128 effects of robotic gait training, (2) they involved patients with cerebral palsy, and (3)
6
ACCEPTED MANUSCRIPT
129 they enrolled patients classified between levels I and IV using the Gross Motor
131 Articles were excluded from this review if they were case reports and case series
132 studies. Investigations in which any concomitant interventions were used were also
133 excluded.
PT
134
RI
136
SC
137 The "Critical Review Form for Quantitative Studies"24 (Table I) was used to quantitate
139
U
AN
140 Table 1: Methodological quality of articles: Critical Review Form for Quantitative
141 Studies.
M
142
D
143 The information was extracted from the selected articles using the descriptive-
TE
144 analytical method.25 The collected data included author(s), year of publication,
146 age (range), intervention, parameters for intervention, outcome measures, dimensions
147 according to the International Classification of Functioning, Disability and Health (ICF),
C
149
151 To perform the meta-analysis, the effects of the intervention were quantified by effect
152 size (Cohen's d) and standardized with mean difference and 95% confidence intervals
153 (CI). Heterogeneity was obtained by q Cochran (Q < 1, absence of heterogeneity; Q >
156 result, the value is set to zero. The variables used in the meta-analysis were a 10-
157 meter walk test, six-minutes walk test, Gross Motor Function Measure Dimension D
158 (GMFM D), and Gross Motor Function Measure Dimension E (GMFM E). The
159 statistical analysis and graphic visualization were performed using R (http://www.r-
PT
160 project.org) with the Rcmdr packages and the RcmdrPlugin.MA plugin.
161
RI
162 Results
SC
163
164 This systematic review identified 300 articles. After removal of duplicate articles, two
165
U
reviewers independently evaluated 133 titles and found that 75 of these met the
AN
166 requirements to proceed to the abstract-reading stage. In the abstract-reading stage,
168 criteria, which qualified them for the full-text reading stage. The implementation of this
D
169 last procedure resulted in the selection of 10 articles that were classified as eligible for
TE
171
EP
173
C
174
AC
175 To confirm the presence of critical components of methodological quality, the selected
176 articles were evaluated using the "Critical Review Form" (Table 2).
177
178 Table 2: Results of methodological quality of articles: Critical Review Form for
180
8
ACCEPTED MANUSCRIPT
181 0, no; 1, yes; NA, Not applicable.
182
183 Concerning the study design, we identified one cohort study, one case-control study,
184 one prospective cohort study, two randomized clinical trials and five before and after
185 studies, which suggest a low level of evidence according to the Oxford Centre for
PT
186 Evidence-based Medicine (OCEBM).26
187 With regard to patient characteristics, we identified a wide range of age range, from 4
RI
188 to 22 years. We also found an extensive diversity of motor impairment in the
SC
189 participants, ranging from levels I to IV on the GMFCS.
190
191
U
Using the criteria of the ICF proposed by the World Health Organization (WHO), seven
AN
192 studies investigated individuals with limitations in activity, two studies investigated
193 individuals with limitations in function and activity, and one study investigated
M
195
TE
196 Two gait training devices were used in the studies. The Lokomat (Hocoma,
197 Switzerland) was used in nine studies and the GaitTrainer I (Reha-Stim, Germany)
EP
199
C
200 To identify the effects of training using robotic devices, researchers used different
AC
201 strategies. Functional tests were used in six studies, with the Gross Motor Function
203 Running, and Jumping). Five studies used the 10-meter Walk Test (10mWT), six
204 studies used the six-minute Walk Test (6minWT), and one study used the Functional
205 Ambulation Categories (FAC)27. Four studies analyzed gait kinematic parameters. The
9
ACCEPTED MANUSCRIPT
206 summaries of study characteristics are presented in two tables, Table 3 for Single
207 Subject Research Designs and Table 4 for Group Research Designs.
208
PT
211
RI
212 Meta-analysis
213 We performed a meta-analysis of four functional tests, 6minWT (Figure 2), 10mWT
SC
214 (Figure 3), the GMFM D (Figure 4), and the GMFM E (Figure 5). In the forest plot,
216
U
AN
217 The Q Cochran test and I metric showed low heterogeneity between the included
M
218 studies. For 6minWT, Q = 0.977 and I = 0, for 10mWT, Q = 0.964 and I = 0, for
219 GMFM D, Q = 1.0 and I = 0, and for GMFM E, Q = 0.998 and I = 0. However, we
D
220 chose to use the random-effect model for statistical analysis because the samples
TE
221 used in the studies included in this review presented important differences with
223
C
224 Discussion
AC
225 This systematic review aimed to identify the effects of robotic gait training practices in
227
229 In evaluating the OCEBM level of evidence of the selected articles, we identified five
230 before and after studies, one prospective cohort, and one case control; these study
10
ACCEPTED MANUSCRIPT
231 design types were classified as level IV. In addition, we found one study cohort design
232 that was classified as level III. Randomized clinical trials were conducted only in two
233 cases; however, because the sample size was less than 100, they only reached the
234 level II classification. This low level of evidence makes it difficult to generalize the
PT
236
RI
238 Participants in the selected studies had spastic diplegia, quadriplegia, hemiplegia,
SC
239 double hemiplegia or ataxia. In the study by Smania et al (2011)35, participants had
240 spastic diplegia and tetraplegia, with GMFCS classification levels between II and IV.
241
U
Meyer-Heim et al (2009)29 included subjects with GMFCS levels between II and IV. In
AN
242 the study by Borggraefe et al (2010, b)30 individuals showed the highest variability in
244
D
245 The range of motor impairment limits this systematic review because the difference
TE
246 between the categories of GMFCS classification is significant.39 Individuals who are
247 classified as level I are able to walk without the aid of devices but with reduced gait
EP
248 speed and impaired balance and coordination. Individuals classified at level IV can
250
AC
252
253 In this review, we identified the use of two devices for gait training. The Lokomat was
254 used in nine studies, and the Gait Trainer I (GTI) was used in the remaining one.
255 Although the objective of this study was not to compare the effects of equipment on
256 gait, the fact that there were only two devices necessitates a brief comment.
11
ACCEPTED MANUSCRIPT
257 According to the data provided, the study that used the GTI35 identified statistically
258 significant positive effects by using the device. There were six studies using the
259 Lokomat that identified statistically significant positive results and three studies that
261
PT
262 Unfortunately, these results do not achieve value-establishment about the advantages
263 of one device compared with the other because the difference in equipment training
RI
264 frequency prevents this type of conclusion.
SC
265
267
U
There are some important parameters to be considered in the design of an exercise
AN
268 regime and/or motor technique, some of which are inherent to the intervention
269 process.
M
270
D
271 Eight articles stated the total number of sessions, weekly frequency, and duration of
TE
272 each session. The studies of Drubicki et al (2013),37 Shroeder et al (2014),33 and
273 Arellano-Martinez (2013)32 did not report the weekly frequency, thus making it difficult
EP
275
C
276 It can be observed in the included studies that the total number of sessions varied
AC
277 between 10 and 40, the weekly frequency varied between two and five and the
278 training duration varied between 20 and 60 minutes. The optimal level of intensity in
279 intervention programs for cerebral palsy is still being debated,40 and the results of the
281
12
ACCEPTED MANUSCRIPT
282 The weekly frequency analysis of intervention programs used in the studies could not
283 be used to establish a consensus. The data suggest that in general, a weekly
284 frequency of training greater than or equal to four days per week (observed in four
285 studies), results in significantly improved gait speed and endurance and step length.
286 Only one study showed improvements in gait with a training frequency less than four.
PT
287 This frequency intervention can be classified as intensive because it is greater than
288 three times per week.43 Thus, this finding is in agreement with Tsorlakis et al (2004),42
RI
289 who found that therapeutic intervention programs with intensive training achieve
SC
290 significantly higher motor gains than non-intensive training.
291
292
U
Concerning the duration of training, only two studies proposed training for 20 minutes,
AN
293 and they did not show satisfactory results. Studies that presented significant positive
295 (2007)44 in their systematic review studying the effect of robotic training for stroke
D
296 patients did not conclude what duration would be the most effective and suggested
TE
298
EP
300
C
301 The device configuration is defined as the possibility of adjusting the equipment
AC
302 parameters for gait training, such as body weight support, step length, cadence,
303 walking speed, and adjustments in the angular variation of the hip, knee and ankle
304 joints.
305
306 Regarding device configuration for gait training, only studies by Arellano-Martinez et al
310
PT
313 accounted for the anthropometric measurements of the subjects for the device
314 configuration. Anthropometric variables are related to gait motor behavior and interfere
RI
315 with the dimensions of the step and stride.46
SC
316
317 Another study that described device configuration was that by Smania et al (2011).35
318
U
The equipment was adjusted according to the gait parameters collected from
AN
319 individuals in the pretest. However, in other studies included in this review, it was not
320 possible to identify the criteria that were used to select the device configuration
M
321 parameters, thus making it difficult to establish valid standards for scientifically
D
323
325 In the 10 articles selected for this review, seven studies showed statistically significant
326 positive effects on gait speed and endurance in patients with cerebral palsy.
C
327
AC
329 improvements in the gait speed and the step size by videography. Smania et al
330 (2011)35 showed more appropriate timing in the angular variation of the hip joint during
331 the initial contact, midstance, and initial swing, and they also found an improvement in
332 gait speed and the step size. Corroborating this study, Patritti et al (2009) 47 also
335 The gait speed was also measured in the 10mWT.28,29,31,34,35 Significant improvement
336 was observed in five studies. Mayer-Heim et al (2007)48 showed an increase in gait
337 speed in patients with neurological damage, including cerebral palsy. This
338 improvement may be related to the increased muscle strength from training.20,49
PT
339 Although muscle strength was not measured in these studies, in previous studies,
340 subjects reported feeling muscle fatigue after treatment; this suggests that training
RI
341 with a robotic device produces an active response in lower limb muscles.20 The results
SC
342 obtained by Gizzi et al (2012)50 reinforce this conclusion because the
343 electromyographic signals obtained in their examination showed that training with a
344
U
robotic device induces muscle activity similar to that of walking.
AN
345
346 Another possible cause for increased gait speed is modification of the intersegmental
M
347 pattern coordination of the lower limbs, as suggested by the data obtained by
D
348 Wagenaar and van Emmerik (2000),51 who identified a positive association between
TE
349 gait speed and intersegmental coordination in healthy individuals, and Meyns et al
350 (2012),52 who undertook a similar study in individuals with cerebral palsy. Krishnan et
EP
351 al (2012)53 reported improved muscle coordination in stroke patients who underwent
353
AC
354 Individuals affected by cerebral palsy have a higher energy expenditure when walking
355 than non-affected individuals.54-56 Cardiovascular and respiratory ability while walking
356 were measured using the 6minWT, which has been validated for this purpose.57 The
357 variables associated with these skills were measured in six studies; the studies of
361 with neurological injury, including cerebral palsy. This evidence was reinforced by
362 findings by Turiel et al (2007),58 who suggested that cardiovascular function improves
363 in individuals with spinal cord injury after training, and by Chang et al (2012),59 who
PT
365
366 Studies show a positive association between the best performance, characterized by
RI
367 the greatest distance, and a decrease in energy expenditure to perform the task.60 The
SC
368 reduction in energy cost may be related to decreased co-contraction of the spastic
369 muscle and a more efficient gait pattern.61,62 This reduction allows greater mobility,
370
U
activity, and participation because the decrease in energy expenditure during walking
AN
371 is strongly associated with increased GMFM.63
372
M
373 Among the observed effects, changes in gross motor skills were measured by GMFM-
D
374 66 for dimensions D and E. In six studies, four showed significant improvements in
TE
375 both dimensions. The study of Mayer-Heim et al (2009)29 showed improvements only
377 improvements.
378
C
379 A positive correlation was found between the improvement in the score of dimension E
AC
380 of the GMFM-66, the distance traveled and the time during gait training, which
382
383 It should be noted that in studies in which the results are stratified by participant
384 GMFCS function, a greater improvement in gross motor function was observed in
385 patients classified as GMFCS I or II than in patients classified as GMFCS III or IV.
16
ACCEPTED MANUSCRIPT
386 These findings are consistent with those of other studies that indicate a lower potential
388
390 (2013)32 identified changes before and after training. Specifically, patients tended to
PT
391 improve by one level after training. Therefore, we suggest that in these studies, the
RI
393
SC
394 The minimum clinically important difference (MCID) is the smallest change that an
395 instrument can detect, making it possible to interpret whether this observed change
396
U
results in improvement or worsening of the individual's symptoms.65,66 Oeffinger et al
AN
397 (2008)67 identified the MCID in the dimensions D and E of GMFM-66 for individuals
398 with cerebral palsy. Their data point to a large effect size in studies that demonstrate
M
399 an increase equal to or greater than 1.8 in dimension D. The articles selected by this
D
400 review showed an amplitude in the improvement of gross motor function between 2.7
TE
401 and 6.3, demonstrating a large effect size in dimension D. With regard to dimension E,
402 Oeffinger et al (2008)67 indicate significant clinical effects; an increase of 1.6 for
EP
403 medium-effect size and 2.6 for large-effect size. The data obtained in our study shows
404 that all articles selected reached a large effect size when evaluating the E dimension
C
405 of GMFM-66.
AC
406
407 With regard to the tests 10mWT and 6minWT, we identified studies that established
408 the MCID for stroke 68,69 and spinal cord injury70; however, we did not find studies with
409 this information involving individuals with cerebral palsy. This made it infeasible to
410 analyze the results obtained by this review. Therefore, we suggest future studies be
17
ACCEPTED MANUSCRIPT
411 performed to determine the MCID of the respective tests for people with cerebral
412 palsy.
413
414 Regarding the meta-analysis, the forest-plot of the variables is shown in different-
415 sized boxes plotted for each of the individual studies. The area of the box representing
PT
416 the weight that the study takes in the analysis provides a visual representation of the
417 relative contribution that each study makes to the overall effect.71 The area of the
RI
418 black squares reflects the weight of the study in the meta-analysis. Methods used for
SC
419 meta-analysis employ a weighted average of the results in which the larger trials
420 generally have more influence than the smaller ones. In practice, the weights are often
421
U
the inverse of the variance of the treatment effect, which relates closely to sample
AN
422 size.72 The representative diamond shows a positive effect of gait training using
423 robotic devices. However, the confidence intervals indicated a low significance. This
M
424 may have occurred because of the small sample sizes. Therefore, future studies with
D
425 larger sample sizes and individuals with homogeneous levels of motor impairment
TE
427
EP
429 The evidence in this review suggests that, despite the heterogeneity and the small
C
430 sample size, robotic gait training is effective in individuals with cerebral palsy. The
AC
431 main findings show improvements in gait speed, gait endurance, and gross motor
432 function. Furthermore, the data suggest that this improvement is best observed in
434
435
436
18
ACCEPTED MANUSCRIPT
437 Study limitations
438 This systematic review is limited by the low level of evidence observed and the wide
439 range of GMFCS levels. Furthermore, none of the reviewed articles reported a priori
441
PT
442 Conclusion
RI
444 between robotic gait training and improvements in gait function. However, the results
SC
445 obtained by the meta-analysis suggest that the training benefits people with cerebral
446 palsy, specifically by increasing walking speed and endurance and improving gross
447
U
motor function. These benefits were observed in studies with a weekly frequency of
AN
448 training greater than or equal to four days per week and with a duration of training
450
D
451 Despite the variability in the frequency and duration of training regimens, it is possible
TE
452 to identify significant positive results in studies that have proposed more intense
453 training.
EP
454
455 Regarding device configuration, it was not possible to identify the criteria for
C
458
459 For future studies, we suggest investigating device configuration parameters and
460 conducting a large number of randomized controlled trials with larger sample sizes
462
19
ACCEPTED MANUSCRIPT
463 Disclosure of interests
464 The authors have stated that they had no interests which might be perceived as
466
467
PT
468
469
RI
470
SC
471
472
U
473
AN
474
475
M
476
D
477
TE
478
479
EP
480
481
C
482
AC
483
484
485
486
487
488
20
ACCEPTED MANUSCRIPT
489 REFERENCES
490
491 1. Pakula AT, Braun K, Yeargin-Allsopp M. Cerebral Palsy: Classification and
492 Epidemiology. Phys Med Rehabil Clin N Am 2009;20:425-52.
493
494 2. Bax MC, Flodmark O, Tydeman C. Definition and classification of cerebral
495 palsy. From syndrome toward disease. Dev Med Child Neurol Suppl 2007; 109:
496 39-41.
497
PT
498 3. Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M, Damiano D, Bernard
499 D, Jacobsson B. A report: the definition and classification of cerebral palsy April
500 2006. Dev Med Child Neurol Suppl 2007;109(suppl 109), 8-14.
501
RI
502 4. Gage JR. Gait analysis in cerebral palsy. New York: Mac Keith; 1991.
503
504 5. Houlihan CM. Walking function, pain, and fatigue in adults with cerebral palsy.
SC
505 Dev Med Child Neurol 2009;51(5):338-339.
506
507 6. Opheim A, Jahnsen R, Olsson E, Stanghelle JK. Walking function, pain, and
U
508 fatigue in adults with cerebral palsy: A 7-year follow-up study. Dev Med Child
509 Neurol 2009;51(5):381-388.
AN
510
511 7. Bell KJ, Ounpuu S, DeLuca PA, Romness MJ. Natural progression of gait in
512 children with cerebral palsy. J Pediatr Orthop 2002; 22(5): 677-682.
513
M
517
518 9. Novacheck TF, Gage, JR. Orthopedic management of spasticity in cerebral
TE
522 in young people with cerebral palsy. Developmental Medicine & Child
523 Neurology. 2003; 45(10):652-7.
524
525 11. Goldstein M, Harper DC. Management of cerebral palsy: equinus gait. Dev Med
C
PT
549
550 18. Bonnyaud R, Zory R, Boudarham J, Pradon D, Bensmail D, Roche, N. Effect of
551 a robotic restraint gait training versus robotic conventional gait training on gait
552 parameters in stroke patients. Exp Brain Res 2014; 232: 3142.
RI
553
554 19. Borggraefe I, Meyer-Heim A, Kumar A, Schaefer J.S, Berweck S, Heinen F.
555 Improved gait parameters after robotic-assisted locomotor treadmill therapy in a
SC
556 6-year-old child with cerebral palsy. Movement Disorders 2008; 23(2): 280
557 283.
558
U
559 20. Hesse S, Uhlenbrock D. A mechanized gait trainer for restoration of gait.
560 Journal of Rehabilitation Research & Development 2000; 37(6): 710-708.
AN
561
562 21. Borggraefe I, Schaefer JS, Klaiber M, Dabrowski E, Ammann-Reiffer C, Knecht
563 B. Robotic-assisted treadmill therapy improves walking and standing
564 performance in children and adolescents with cerebral palsy. European Journal
M
568 Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA
569 Statement. Ann Intern Med. 2009; 151: 264-269.
TE
570
571 23. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gtzsche PC, Ioannidis JP, et al.
572 The PRISMA Statement for Reporting Systematic Reviews and Meta-Analyses
EP
PT
600
601 30. Borggrfe I, Schaefer JS, Klaiber M, Dabrowski E, Ammann-Reiffer C, Knecht
602 B, Berweck S, Heinen F, Meyer-Heim A. Robotic-assisted treadmill therapy
603 improves walking and standing performance in children and adolescents with
RI
604 cerebral palsy. European Journal of Paediatric Neurology 2010; 14(6): 496-502.
605 (34)
606
SC
607 31. Borggrfe I, Kiwull L, Schaefer JS, Koerte I, Blaschek A, Meyer-Heim A,
608 Heinen, F. Sustainability of motor performance after robotic-assisted treadmill
609 therapy in children: an open, non-randomized baseline-treatment
U
610 study. European Journal of Physical and Rehabilitation Medicine 2010; 46(2):
611 125-131.
AN
612
613 32. Arellano-Martnez IT, Rodrguez-Reyes G, Quiones-Uriostegui I, Arellano-
614 Saldaa ME. Spatial-temporal analysis and clinical gait findings: comparison of
615 two treatment modalities in children with cerebral palsyspastic hemiplegia.
M
628
629 35. Smania N, Bonetti P, Gandolfi M, Cosentino A, Waldner A, Hesse S, Werner C,
AC
630 Bisoffi G, Geroin C, Munari D. Improved gait after repetitive locomotor training
631 in children with cerebral palsy. American Journal of Physical Medicine &
632 Rehabilitation 2011; 90(2): 137-149.
633
634 36. Romei M, Montinaro A, Piccinini L, Maghini C, Germiniasi C, Bo I, Molteni A,
635 Turconi AC. Efficacy of robotic-assisted gait training compared with intensive
636 task-oriented physiotherapy for children with Cerebral Palsy. In 2012 4th IEEE
637 RAS & EMBS International Conference on Biomedical Robotics and
638 Biomechatronics (BioRob) 2012; 1890-1894.
639
640 37. Druzbicki M, Rusek, W, Snela S, Dudek J, Szczepanik M, Zak E, Durmala J,
641 Czernuszenko A, Bonikowski M, Sobota, G. Functional effects of robotic-
23
ACCEPTED MANUSCRIPT
642 assisted locomotor treadmill therapy in children with cerebral palsy. Journal of
643 rehabilitation medicine 2013; 45(4): 358-363.
644
645 38. Sackett, DL. Rules of evidence and clinical recommendations on the use
646 of antithrombotic agents. Chest 1989; 95(2): 2S-4S.
647
648 39. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B.
649 Development and reliability of a system to classify gross motor function in
650 children with cerebral palsy. Developmental Medicine & Child Neurology. 1997;
PT
651 39(4):214-23.
652
653 40. Gagliardi C, Maghini C, Germiniasi C, Stefanoni G, Molteni F, Burt DM Turconi
654 AC. The effect of frequency of cerebral palsy treatment: a matched-pair pilot
RI
655 study. Pediatric neurology 2008; 39(5): 335-340.
656
657 41. Bower E, Michell D, Burnett M, Campbell MJ, McLellan DL. Randomized
SC
658 controlled trial of physiotherapy in 56 children with cerebral palsy followed for
659 18 months. Developmental Medicine & Child Neurology. 2001; 43(1):4-15.
660
U
661 42. Tsorlakis N, Evaggelinou C, Grouios G, Tsorbatzoudis C. Effect of intensive
662 neurodevelopmental treatment in gross motor function of children with cerebral
AN
663 palsy. Developmental Medicine & Child Neurology, 2004; 46(11), 740-745.
664
665 43. Arpino C, Vescio MF, De Luca A, Curatolo P. Efficacy of intensive versus
666 nonintensive physiotherapy in children with cerebral palsy: a meta-analysis.
M
672
673 45. Vaughan CL, Davis BL, Oconnor JC. Dynamics of Human Gait. 2. ed. Cape
674 Town, South Africa: Kiboho Publishers, 1999.
EP
675
676 46. Perry J. Gait analysis: normal and pathological function: SLACK Incorporated;
677 1992.
678
C
681 palsy after robotic gait training. Gait & Posture. 2009; 30:S9-10.
682
683 48. MeyerHeim A, Borggraefe I, AmmannReiffer C, Berweck S, Sennhauser FH,
684 Colombo G, Knecht B, Heinen F. Feasibility of roboticassisted locomotor
685 training in children with central gait impairment. Developmental Medicine &
686 Child Neurology. 2007; 49(12):900-6.
687
688 49. Bottos M, Gericke C. Ambulatory capacity in cerebral palsy: prognostic criteria
689 and consequences for intervention. Dev Med Child Neurol 2003; 45: 786-90
690
691 50. Gizzi L, Nielsen JF, Felici F, Moreno JC, Pons JL, Farina D: Motor modules in
692 robot-aided walking. J Neuroeng Rehabil 2012; 9(1): 1-12.
24
ACCEPTED MANUSCRIPT
693
694 51. Wagenaar RC, Van Emmerik REA. Resonant frequencies of arms and legs
695 identify different walking patterns. J Biomech 2000; 33(7): 853-861.
696
697 52. Meyns P, Van Gestel L, Bruijn SM, Desloovere K, Swinnen SP, Duysens J. Is
698 interlimb coordination during walking preserved in children with cerebral palsy?
699 Research in developmental disabilities 2012; 33(5): 1418-1428.
700
701 53. Krishnan C, Ranganathan R, Kantak SS, Dhaher YY, Rymer WZ. Active robotic
PT
702 training improves locomotor function in a stroke survivor. Journal of
703 Neuroengineering and Rehabilitation, 2012: 9(1); 1-13.
704
705 54. Rose J, Gamble JG, Burgos A, Medeiros J, Haskell WL. Energy expenditure
RI
706 index of walking for normal children and for children with cerebral palsy. Dev
707 Med Child Neurol 1990; 32: 33340.
708
SC
709 55. Rose J, Haskell WL, Gamble JG. A comparison of oxygen pulse and respiratory
710 exchange ratio in cerebral palsied and nondisabled children. Arch Phys Med
711 Rehabil 1993; 74: 70205.
U
712
713 56. Thomas SS, Buckon CE, Schwartz MH, Russman BS, Sussman MD, Aiona
AN
714 MD. Variability and minimum detectable change for walking energy efficiency
715 variables in children with cerebral palsy. Developmental medicine and child
716 neurology 2009; 51(8): 615-621.
717
M
718 57. Maher CA, Williams MT, Olds TS. The six-minute walk test for children with
719 cerebral palsy. International Journal of Rehabilitation Research. 2008;
720 31(2):185-8.
D
721
722 58. Turiel M, Sitia S, Cicala S, Magagnin V, Bo I, Porta A, Caiani E, Ricci C, Licari
TE
726
727 59. Chang WH, Kim MS, Huh JP, Lee PK, Kim YH. Effects of Robot-Assisted Gait
728 Training on Cardiopulmonary Fitness in Subacute Stroke Patients A
729 Randomized Controlled Study. Neurorehabilitation and neural
C
732 60. Piccinini L, Cimolin V, Galli M, Berti M, Crivellini M, Turconi AC. Quantification
733 of energy expenditure during gait in children affected by cerebral palsy. Eura
734 Medicophys 2007; 43: 7-12.
735
736 61. Waters RL, Mulroy S: The energy expenditure of normal and pathologic gait.
737 Gait & Posture 1999; 9: 207-31
738
739 62. Unnithan VB, Dowling JJ, Frost G, Bar-Or O. Role of cocontraction in the O2
740 cost of walking in children with cerebral palsy. Medicine and science in sports
741 and exercise 1996; 28(12): 1498-1504.
742
25
ACCEPTED MANUSCRIPT
743 63. Kamp FA, Lennon N, Holmes L, Dallmeijer AJ, Henley J, Miller F. Energy cost
744 of walking in children with spastic cerebral palsy: relationship with age, body
745 composition and mobility capacity. Gait & posture 2014; 40(1): 209-214.
746
747 64. Hanna SE, Bartlett DJ, Rivard LM, Russell DJ. Reference curves for the Gross
748 Motor Function Measure: percentiles for clinical description and tracking over
749 time among children with cerebral palsy. Physical therapy 2008; 88(5):596-607.
750
751 65. Beaton DE, Bombardier C, Katz JN, Wright JG. A taxonomy for
PT
752 responsiveness. J Clin Epidemiol. 2001;54(12):1204-17. 14.
753
754 66. Revicki DA, Hays RD, Cella D, Sloan J. Recommended methods for
755 determining responsiveness and minimally important differences for patient-
RI
756 reported outcomes. J Clin Epidemiol. 2008;61(2):102-9.
757
758 67. Oeffinger D, Bagley A, Rogers S, Gorton G, Kryscio R, Abel M, Damiano D,
SC
759 Barnes D, Tylkowski C. Outcome tools used for ambulatory children with
760 cerebral palsy: responsiveness and minimum clinically important differences.
761 Developmental medicine & child neurology. 2008;50(12):918-25.
U
762
763 68. Tang A, Eng J, Rand D. Relationship between perceived and measured
AN
764 changes in walking after stroke. Journal of neurologic physical therapy: JNPT.
765 2012 Sep;36(3):115.
766
767 69. Perera S, Mody SH, Woodman RC, Studenski SA. Meaningful change and
M
771 70. Forrest GF, Hutchinson K, Lorenz DJ, Buehner JJ, VanHiel LR, Sisto SA,
772 Basso DM. Are the 10 meter and 6 minute walk tests redundant in patients with
TE
776 for undertaking reviews in health care. Centre for Reviews and Dissemination;
777 2009.
778
779 72. Egger M, Davey-Smith G, Altman D, editors. Systematic reviews in health care:
C
781
782
783
784
785
786
787
788
789
790
791
792
793
26
ACCEPTED MANUSCRIPT
794 List of Figures
795
796 Figure1: PRISMA flow diagram.
PT
800 Figure 5: Forest Plot of GMFM E.
801
RI
802
SC
803
804
805
U
AN
806
807
M
808
D
809
TE
810
811
EP
812
813
C
814
AC
815
816
817
818
819
820
27
ACCEPTED MANUSCRIPT
821 List of Tables
822 Table 1: Methodological quality of articles: Critical Review Form for Quantitative
823 Studies.
PT
826 Table 3: Summary of study characteristics: Single Subject Research Designs
RI
828
SC
829
U
AN
M
D
TE
C EP
AC
Table 1
ACCEPTED MANUSCRIPT
Table 1
Study purpose
Literature
PT
2- Was relevant background literature reviewed?
RI
Design
SC
Sample
Intervention
TE
Results
C
results?
PT
RI
U SC
AN
M
D
TE
C EP
AC
Table 2
ACCEPTED MANUSCRIPT
Table 2
Studies Questions
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Total
Borggrfe et al.27 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 12
28
Meyer-Heim et al. 1 1 1 1 0 1 1 0 1 1 1 1 1 1 1 13
Borggrfe et al.29 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 14
30
Borggrfe et al.(b) 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 14
PT
Arellano-Martnez et al.31 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 14
Verazaluce-Rodrguez et al.32 1 1 1 1 0 1 1 0 1 1 1 1 NA 1 1 12
RI
33
Schroeder et al. 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 12
Smania et al.29 1 1 1 1 0 1 1 0 1 1 1 1 1 1 1 13
SC
35
Romei et al. 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 12
Drubicki et al.36 1 1 1 1 0 1 1 0 1 1 1 1 1 1 1 13
U
AN
M
D
TE
C EP
AC
Table 3
ACCEPTED MANUSCRIPT
Table 3
PT
RI
U SC
AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
et al., design spastic 4 sessions per week Dimension D and Dimension D Dimension D
PT
GMFCS I-IV session.
RI
p = 0.004 d = 0.1740771
SC
3. 6minWT
2. 10mWT 2. 10mWT
U
p = 0.005 d = 0.2714188
AN
3. 6minWT 3. 6minWT
M
p = 0.005 d = 0.2392387
D
Meyer- Before After IV 22 Bilateral 4 - 12 Lokomat 20 sessions, 3 to 5 No reports. 1. GMFM-66 Activity 1. GMFM-66 1. GMFM 66
TE
Heim et al., Design spastic sessions per week, 45 Dimension D. Dimension D. Dimension D
p = 0.169 d = 0.0705761
AC
3. 6minWT
2. 10MWT 2. 10mWT
3. 6minWT 3. 6minWT
p = 0.093 d = 0.154707
ACCEPTED MANUSCRIPT
4. FAC 4. FAC
p = 0.063 d = 0.242
Borggraefe Before - After IV 20 Bilateral 4,5 20,7 Lokomat 12 sessions, 4 sessions 1. BWS was 1. GMFM - 66 Activity 1. GMFM 66 1. GMFM 66
et al., Design Spastic per week, 50 minutes started at 100% Dimension D Dimension D Dimension D
PT
201030 GMFCS I-II per session and then reduced Dimension E p = 0.001 d = 0.1973145
(10) as much as
RI
GMFCS III- possible. Dimension E Dimension E.
SC
IV (10) p < 0.001 d = 0.1884934
2. The guidance
U
force was
individually
AN
adapted according
to clinical
M
judgment.
D
TE
3. Walking speed
gradually
C
increased to 1.8
AC
km/h.
Borggraefe Before IV 14 Bilateral 4,5 19,2 Lokomat 12 sessions 1. Walking speed 1. GMFM - 66 Function 1. GMFM 66 1. GMFM 66
et al., 2010 After Design Spastic 3 weeks initially set at 1.1 Dimension D and Dimension D Dimension D
(b)31 GMFCS I-IV 4 sessions per week km/h increased to Dimension E Activity p = 0.003 d = 0.1457511
PT
stance.
3. 6minWT 3. 6minWT
RI
3. Force guidance p = 0.005 d = 0.3303943
SC
was set at 5%
U
the DGO.
Arellano- Case Control IV 14 Hemiplegic 4 14 Lokomat 10 sessions 1. DGO group 1. GMFCS Activity GMFCS from II No raw data
AN
Martnez et GMFCS II 30 minutes Based on weight to I provided for
M
anthropometrics Step length
D
meausure Stride length Gait speed
TE
EP Gait speed p = 0.025
Lesft stride
p = 0.025
C
AC
Veraluce- Before After IV 33 Tetraplegic Mean age Lokomat 40 sessions, Blocks of 1. Gait velocity 1. GMFCS Activity 1. GMFCS No raw data
Rodriguez Design and 7,2 years 10 sesions, 2 sessions was initially set at 8 pacients provided for
et al., Double per week, 20 minutes 0.7-0.8 km/h and 2. Body Weight requalified as II calculation.
201433 Hemiplegia per session. the leading force Support (BWS) (24%).
GMFCS II at 100%.
p = 0.442
4. Guiding force
3. Speed
p = 0.081
PT
4. Guiding force
p = 0.662
RI
Schroeder Prospective IV 18 Bilateral 5 21,8 Lokomat 12 sessions, 3 weeks 1. BWS was 1. GMFM-66 Activity 1.GMFM 66 1. GMFM 66
SC
et al., Cohort Spastic training, 30 to 60 started at 100% Dimension D p < 0.001 Dimension D
201434 GMFCS I to minutes per session. and then reduced Dimension E d = 0.1270163
U
IV as much as Dimension D:
AN
Dimension E
d = 0.1012577
M
force was p < 0.01 2. 10mWT
D
individually 4. COPM d=0
TE
adapted according 2. 10mWT
3. 6minWT:
C
p = 0.076
AC
4. COPM
Performance
p = 0.010
Satisfaction
ACCEPTED MANUSCRIPT
p = 0.046
PT
RI
U SC
AN
M
D
TE
EP
C
AC
Table 4
ACCEPTED MANUSCRIPT
Table 4
PT
RI
U SC
AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Smania et Randomized IIb 9 9 Tetraplegic Trainer I per weeks, 40 minutes gait speed were Function p = 0.008 d = 0.558
PT
al., 201135 Controlled Trial. GMFCS II -IV (GTI) per session. individually set 2. 6minWT
RI
Experimental Group parameters recorded 3. Gait p = 0.008 d = 0.438
SC
30 min + 10 passive at the Analysis:
U
analysis. The experimental
AN
Control Group Spatiotemporal group showed
M
PT was gradually (speed, changes in gait
D
course of the 2 step length). length
TE
weeks.
progressively
C
decreased from 30%
AC
to 0% over the
duration of the
sessions.
ACCEPTED MANUSCRIPT
Romei et Cohort design III 9 10 Bilateral 4-16 Lokomat 40 sessions, 10 weeks, 1.body-weight 1. GMFM-66 Activity 1. GMFM-66: 1. . GMFM-88.
al., 201236 Spastic CP for 30 minutes per support fixed at 50% p > 0.05 Dimension D
PT
20 sessions of RAGT leading force at Dimension E d = 0.170
RI
20 of task-oriented 2. 6minWT
SC
physiotherapy 2. Gait velocity was 4. Gait 3. 6minWT: d = 0.084
U
km/h for all the (3DGA).
AN
40 sessions gradually increased step length, p > 0.05
M
oriented physiotherapy km/h for the and cadence. 5. Gillette Gait
D
(4 sessions/week). youngest children Index:
TE
(10% every 5 5. Gillette Gait p > 0.05
Druzbicki Randomized IIb 26 9 Spastic 6 - 13 Lokomat 20 sessions, 4 sessions 1. Based on the 1.Gait Activity 1.Gait Analysis. No raw data
et al., Controlled Trial. Diplegia per weeks, 45 minutes measurement Analysis. Temporospatial provided for
Mean step
width
Mean gait
PT
speed
Mean step
RI
length
SC
Gait Kinematic
U
parameters:
Range of pelvic
AN
motion.
M
D
TE
EP
C
AC
ACCEPTED MANUSCRIPT
PT
RI
U SC
AN
M
D
TE
EP
C
AC
AC
C EP
TE
D
M
AN
U SC
RI
PT
AC
C EP
TE
D
M
AN
U SC
RI
PT
AC
C EP
TE
D
M
AN
U SC
RI
PT
AC
C EP
TE
D
M
AN
U SC
RI
PT
Highlights
ACCEPTED MANUSCRIPT
Highlights
established.
PT
RI
U SC
AN
M
D
TE
C EP
AC