Neurorehabilitation and Neural
Repair
http://nnr.sagepub.com/
Repetitive Mass Practice or Focused Precise Practice for Retraining Walking After Incomplete Spinal
Cord Injury? A Pilot Randomized Clinical Trial
Jaynie F. Yang, Kristin E. Musselman, Donna Livingstone, Kelly Brunton, Gregory Hendricks, Denise Hill and Monica
Gorassini
Neurorehabil Neural Repair published online 8 November 2013
DOI: 10.1177/1545968313508473
The online version of this article can be found at:
http://nnr.sagepub.com/content/early/2013/10/28/1545968313508473
Published by:
http://www.sagepublications.com
On behalf of:
American Society of Neurorehabilitation
Additional services and information for Neurorehabilitation and Neural Repair can be found at:
Email Alerts: http://nnr.sagepub.com/cgi/alerts
Subscriptions: http://nnr.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Nov 8, 2013
What is This?
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
508473
research-article2013
NNRXXX10.1177/1545968313508473Neurorehabilitation and Neural RepairYang et al
Original Article
Repetitive Mass Practice or Focused Precise
Practice for Retraining Walking After
Incomplete Spinal Cord Injury? A Pilot
Randomized Clinical Trial
Neurorehabilitation and
Neural Repair
XX(X) 1­–11
© The Author(s) 2013
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1545968313508473
nnr.sagepub.com
Jaynie F. Yang, PhD1, Kristin E. Musselman, PhD1,2, Donna Livingstone1,
Kelly Brunton, MScPT1, Gregory Hendricks1,3, Denise Hill, MD4,
and Monica Gorassini, PhD1
Abstract
Background. Retraining walking following spinal cord injury using visually guided tasks may be especially efficacious because
it engages the motor cortex, whose input may facilitate improvements in functional walking. Objectives. To contrast 2
methods of retraining, one emphasizing precise, visually guided walking over obstacles and on targets (Precision Training),
the other emphasizing mass practice of walking on a treadmill (Endurance Training). Methods. A randomized, single-blind,
crossover design was used. Twenty-two participants, ≥7 months postinjury, were randomly allocated to start with Precision
or Endurance Training. Each phase of training was 5 times per week for 2 months, followed by a 2-month rest. Measures of
walking speed, distance, skill, confidence, and depression were obtained before training, then monthly thereafter. Results.
Both forms of training led to significant improvements in walking, with Endurance Training inducing bigger improvements
in walking distance than Precision Training, especially for high-functioning walkers who had initial walking speeds >0.5 m/s.
The largest improvements in walking speed and distance occurred in the first month of Endurance Training, with minimal
changes in the second month of training. In contrast, improvements in walking skill occurred over both months during
both types of training. Retention of over ground walking speed, distance, and skill was excellent for both types of training.
Conclusions. Intensive walking training in the chronic phase after spinal cord injury is effective in improving over ground
walking. Visually guided tasks for training individuals with chronic spinal cord injury were not superior to mass practice on
a treadmill.
Keywords
treadmill, walking, physical therapy, rehabilitation, spinal cord injuries
Introduction
Retraining of walking after incomplete spinal cord injury
(SCI) has largely focused on mass practice with repetitive
stepping on a treadmill or over ground. Adjuncts during
such training have included partial body weight support,1-7
functional electrical stimulation of muscles,8,9 and/or
robots10-16 to assist or resist the walking movements (recent
reviews17-20). In our literature scan, only one study (a case
series) considered a very different form of training, called
“skill training,” which incorporated precise stepping over
ground that challenged balance.21 Large randomized controlled trials comparing over ground training with treadmill
training have shown no differences in outcome.5,6,22
In individuals with chronic incomplete SCI, improvements in walking were correlated with training-related
strengthening of the corticospinal tract.23 Even though
rhythmic stepping is controlled by the spinal cord and
brainstem,24 training that enhances the participation of the
motor cortex may be especially efficacious. One way to
engage the primary motor cortex is to use visually guided
walking, such as stepping precisely over obstacles and onto
targets. In cats, such precise stepping enhanced the firing of
neurons from the primary motor cortex, especially when the
paw stepped over obstacles.25 Training methods incorporating such precise stepping may better engage and strengthen
the corticospinal tract, thereby translating into better over
1
University of Alberta, Edmonton, Alberta, Canada
University of Saskatchewan, Saskatoon, Saskatchewan, Canada
3
Millard Health, Edmonton, Alberta, Canada
4
Foothills Hospital, Calgary, Alberta, Canada
2
Corresponding Author:
Jaynie F. Yang, Department of Physical Therapy, University of Alberta,
2-50 Corbett Hall, Edmonton, Alberta, T6G 2G4, Canada.
Email: jaynie.yang@ualberta.ca
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
2
Neurorehabilitation and Neural Repair XX(X)
ground walking. Alternatively, stepping on a treadmill may
already be sufficiently difficult to engage the motor cortex
in many individuals with SCI, so that making as many steps
as possible (ie, mass practice) may be more important in
training.
Here, we contrast 2 forms of training, based on the reasoning above. A randomized crossover study determined
how participants improved in over ground walking with
each form of training. Although many other electrophysiological measures were taken, this article will focus on the
outcome measures for walking.
Methods
2 months of rest. Measurements were taken at least 3 times
before the first training period (separated by at least 1 week)
to confirm baseline stability and then once monthly thereafter. By using a crossover design, we maximized the number
of participants receiving each type of training, and counterbalanced the order of training. This trial is registered in
clinicaltrials.gov (NCT01765153).
Variables Controlled
Participants maintained the same medication dosages and
exercise outside of our intervention throughout the study
period.
Participants
Clinical Outcome Measures
Inclusion Criteria
1. Spinal cord injury from trauma or other stable lesions
2. Injury occurred ≥7 months prior
3. Able to walk independently ≥5 m with a walking aid
and/or braces
4. Injury between C1 and L1
5. Able to attend training sessions 5 times per week
Recruitment started October 2007 and finished January
2012. The last participant completed follow-up in September
2012. The study was approved by the Health Research
Ethics Board, University of Alberta and Alberta Health
Services.
Physical therapists blinded to group assignment administered the measures. The primary outcome measure, the SCIFAP,27 measured walking skills in daily life (uninjured
score = 7). Secondary measures included the following: the
6-minute walk test (6MWT),28 the 10-meter walk test29 performed over 14 m (timing middle 10 m) at a self-selected
speed [10MWT(ss)] and a fast speed [10MWT(f)], and the
Walking Index for Spinal Cord Injury–Version II30,31 performed in 2 ways—self-selected [WISCI-II(ss)] and maximum [(WISCI-II(max)].32 Psychometric properties of these
tests have been established (6MWT,29,31,33 SCI-FAP,27,34
10MWT,29,31,33 WISCI-II29,33,35). Walking confidence was
measured with the Activities-specific Balance Confidence
scale,36 and depression was measured with the Center for
Epidemiologic Studies–Depression Scale.37 Manual muscle
strength of the lower extremities was assessed prior to any
treatment, including hip extension, flexion, abduction,
adduction, knee flexion, extension, ankle dorsiflexion, and
plantarflexion according to clinical standards,38 with modification for the plantarflexor group as previously reported.39
Study Design
Interventions
A randomized, single-blind, crossover design was used.
Sample size (n = 34) was estimated from previous data,21
using a difference score of 17% between groups for the
modified Emory Ambulation Profile (mEFAP)26 and a standard deviation of 25%, power of 0.8, and confidence level
of .05. We used the mEFAP because our primary outcome
measure, the Spinal Cord Injury–Functional Ambulation
Profile (SCI-FAP)27 was modeled after the mEFAP.
Participants were screened by a physical therapist. Once
suitability and written consent was obtained, participants
were block-randomized (block size 4) to start with either
Endurance or Precision Training by picking a label out of a
box. Each type of training was ~1 h/d, 5 times per week, for
2 months. After the first phase of training, participants
rested for 2 months with no training, then crossed over to
the other form of training for 2 months, followed by another
Precision Training. Participants had to step over obstacles of
different heights and onto targets of different sizes. Obstacles were 3-cm high Styrofoam blocks (stacked to increase
height), varying in width from 4 to 20 cm in increments of
4 cm. Targets were fabric circles, diameter from 7 to 10.5 cm.
The maximum ability for targets and obstacles was determined by trial-and-error prior to the beginning of Precision
Training. A different course was designed each day, consisting of a straight track (15 m) with ~15 obstacles and
targets (ie, ~1 m apart). Based on the error recorded each
day (see below), the course was altered for the next day to
aim for 80% success. The emphasis was on accuracy, not
speed. Extra steps between obstacles and targets were
allowed at the beginning of training, and rest breaks were
taken as needed. As participants improved, course difficulty
was increased by requiring that each obstacle/target was
Exclusion Criteria
1. Head injury
2. Cognitive or musculoskeletal impairments that limit
walking training
3. Seizures or head implants that preclude participation
in transcranial magnetic stimulation experiments
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
3
Yang et al
executed without taking extra steps in between. Reducing
gait aids also challenged walking balance.
Endurance Training. Participants walked on the treadmill,
with body weight support and manual assistance if needed.
The treadmill speed was initially set to be faster than their
over ground walking speed as measured by the 10MWT(ss).
Walking as far and as fast as possible with minimal rests
was encouraged. Progression involved reducing manual
assistance and body weight support, increasing walking
speed and distance, while reducing rest breaks.
Documentation of Training
Training parameters were recorded each session. The number of steps executed (StepWatch, Orthocare Innovations,
Oklahoma City, OK), total time, distance, and walking
speed for each walking bout were measured, along with
heart rate before and immediately after each bout of walking (Polar FT7, Polar, Lachine, Quebec, Canada). Errors
during Precision Training were defined as any touch to the
obstacle, and any target incompletely obscured by the foot.
A Success Index was calculated each day, defined as:
Success Index = [(∑ height × width of obstacles
cleared)/10] × [(∑ target diameters achieved)/10]
where units for height, width, and diameter were in meters,
summed (∑) over all successful obstacles and targets. The
division by 10 was arbitrary, to convert the score to a workable number. This new measure is used for descriptive purposes only.
Statistical Analysis
Baseline variables were compared between the two groups:
those starting with Endurance vs Precision Training. In all
comparisons between groups, a standard t test was used for
variables that were continuous, whereas the Mann–Whitney
U test was used for ordinal data (ie, WISCII(ss) and level
of injury). Paired t tests were used for within-participant
comparisons, and Wilcoxon signed rank test for ordinal
data. A P value of <.05 was considered significant for all
comparisons.
To determine if Precision and Endurance Training each
led to significant improvements, pre- and posttraining
scores for all outcomes were compared using the paired t
tests. Pretraining scores were either the average of the baseline scores (for phase 1) or the average of the 2 scores
obtained during the first rest period (for phase 2). Data were
collapsed across phase of training to counterbalance for a
potential order effect.
To determine if one type of training was superior to the
other, we compared the change scores (ie, posttraining
minus pretraining) from each type of training using the
standard t tests. To further determine if the changes
were greater than measurement error, the minimally detectable change (MDC) was obtained from the literature for the
10MWT,40,41 6MWT,40 and WISCI-II32 and SCI-FAP.34
Since individuals with high compared with low walking
function may respond differently to the training, we stratified individuals based on their walking speed (10 MWT(ss))
at the beginning of each type of training. The ability to walk
>0.5 m/s, that is, sufficient for community walking,42,43
separated individuals into high- (>0.5 m/s) versus low(<0.5 m/s) functioning walkers. The change scores from
each of these groups were compared for each type of training using the standard t tests.
The time course for improvement was determined by
comparing the change scores for the first month versus the
second month of training, using the paired t tests.
Retention of gains was determined by comparing the
outcome measures at the end of the 2-month rest with those
immediately after the training (paired t tests). A retention
index was calculated for each outcome measure as the ratio
of the score at the end of a 2-month rest divided by the score
immediately after training (ie, ratio of 1 = perfect retention).
The retention index was compared between the 2 types of
training, and between phases 1 and 2 using the standard t
tests. The correlation between retention index and final
walking speed indicated if achieving a higher final walking
speed led to better retention.
The P value for number of comparisons was not corrected. Because the statistical power is low with the small
number of participants, we did not want to risk missing a
difference if one existed. Nevertheless, we also determined
if using a single test on the primary outcome measure (SCIFAP) would change the results. A linear mixed model was
used, with factors treatment (Precision vs Endurance), carryover (no treatment preceding vs Precision or Endurance
preceding), and sequence of treatment (Precision first vs
Endurance first, participants nested within sequence).
Results
Participants
Seventy potential participants were screened (Figure 1).
Forty-eight were excluded because 13 were ineligible (ie,
lesion level too low, progressive spinal condition, and/or
other injury to the central nervous system), 18 were not able
to walk 5 m without an assistant, 9 lived too far away and
could not relocate, 4 were younger than 18 years, 2 had
other problems (pressure sores, substance abuse), and 2
were lost after initial contact. Twenty-two were randomized
and received the intervention. Twenty provided data for the
analyses. Eleven participants each started with Precision or
Endurance Training. One participant was excluded because
of low attendance (<3 times per week), and another was
excluded because wrist pain was aggravated during
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
4
Neurorehabilitation and Neural Repair XX(X)
very large number of participants (n = 360) would be needed
to obtain a statistical power of .8.
Baseline Measures From the Two Groups Were
Not Different
The groups starting with Endurance versus Precision
Training were not significantly different from each other
(Table 1).
Intensity of Training
The number of training sessions and the average walking
duration/session were the same for both training methods,
but the average walking speed and the average number of
steps (strides) executed per session were ≥3 times higher for
Endurance Training (Figure 2A). Peak change in heart rate
(ΔHR), averaged across all sessions, was also higher for
Endurance Training (Figure 2A).
Figure 1. CONSORT diagram summarizing the number of
participants in each arm of the study. Excluded individuals in the
Other category were mostly those who lived too far away and
could not relocate for personal reasons. Care provider codes:
G, K, and D are for the 3 physical therapists (PTs) who provided
the training; numbers that are fractions are because the PT
provided one phase of the training (ie, 0.5). See text for those
who discontinued or were excluded.
Precision Training with a walker. No other adverse events
resulted from the training. Of the 10 that started with and
completed Endurance Training, all completed Precision
Training. The first participant (P10) who started with
Endurance Training did the training over ground, thus her
data from the Endurance phase were not used. Endurance
Training was initially planned to be done over ground.
However, P10’s walking speed and heart rate were too limited by her over ground walking ability to induce a strong
training effect. Thus, all subsequent Endurance Training
was done on the treadmill, where walking speed can be
more easily increased by providing some body weight support. Of the 10 who started and completed Precision
Training, 1 dropped out after this phase (P11), and 1 started
chemotherapy during the Endurance phase of the training
(P19), so 8 in this group completed both phases of training.
Thus, when we collapsed the data across the order of training to compare Precision versus Endurance Training, there
were 20 participants for Precision Training and 17 participants for Endurance Training. The trial was stopped at this
point because an interim calculation of sample size, based
on the mean change scores in SCI-FAP, indicated that a
Progression During Training
Endurance Training was characterized by significant
improvements in distance, number of strides, and speed
(Figure 2B). The 5 participants who required body weight
support reduced the support by 60% ± 42% (not shown).
Precision Training was characterized by a significant
improvement in the Success Index, but no change in distance, speed, and number of strides. Thus, the original
premise that one method focused on mass practice of fast
and long walks, whereas the other focused on precise, slow
stepping, was achieved.
Endurance Versus Precision Training
Most of the outcome measures were significantly improved
from pre- to posttraining, shown as change scores with a
white asterisk (*) in Figure 3. The estimated MDCs (see
Statistical Analysis in Methods section) are shown as
dashed horizontal lines. Endurance Training led to
improvements that were ≥MDCs for more measures than
Precision Training. Improvement in the 6MWT was significantly greater for Endurance compared with Precision
Training (Figure 3, black *, P = .045, standard t test); no
other comparisons were significant (P values for standard t
tests: SCI-FAP = .35, 10 MWT both (ss) and (f) = .37,
Activities-specific Balance Confidence scale = .52, and for
the Mann–Whitney U test: WISCI(ss) = .42 and
WISCI(max) = .82). The linear mixed model analysis
showed the same results for the SCI-FAP, as described
above. Furthermore, since most participants retained their
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
5
Yang et al
Table 1. Participant Characteristics and Initial Walking Scores.
Participants
Endurance 1st Age (Years) Injury Type
P1 (F)
P2 (M)
P3 (M)
P4 (M)
P5 (M)
P6 (M)
P7 (M)
P8 (M)
P9 (F)
P10 (F)
Mean (SD)
Participants
Precision 1st
P11 (M)
P12 (M)
P13 (F)
P14 (M)
P15 (M)
P16 (M)
P17 (M)
P18 (F)
P19 (F)
P20 (M)
Mean (SD)
49
24
25
57
48
65
60
46
63
45
48 (14)
MVA
MVA
MVA
MVA
Fall
MVA
Bull attack
MVA
MVA
Surgical clot
Age (Years) Injury Type
43
63
21
61
44
34
32
41
50
52
Time Since
Injury Level Injury (Years)
C6
T6
T4
C5
T12
C3
C3
C6
C4
T2
10MWT
(m/s)
6MWT (m)
0.05
0.37
0.13
0.90
0.23
0.05
0.47
0.14
1.13
0.33
0.38 (0.37)
17
114
48
298
119
19
172
43
278
115
122(101)
10MWT
(m/s)
6MWT (m)
SCI-FAP
17.5
20.0
1.0
2.4
0.9
1.2
0.8
3.5
1.8
1.4
1.27
0.82
0.25
0.90
0.59
1.10
0.16
0.17
0.20
0.37
392
254
67
292
222
337
53
38
67
120
10
31
262
14
65
820
231
602
165
117
5.0 (7.3)
0.58 (0.42)
184 (131)
2.5
1.0
1.1
34.9
0.7
1.2
0.6
7.3
6.0
2.3
5.7 (10.5)
Time Since
Injury Level Injury (Years)
Sport
Sport
MVA
Sport
MVA
Gunshot
Fall
Infection
Tumor
Surgical
bleeding
44 (13)
T12
C4
C6
C5
L1
C4
T2
T12
T6
T10
SCI-FAP
MMT
WISCI (Maximum 80)
649
13
119
12
235
9
59
9
140
16
864
13
286
13
308
13
10
20
133
12
280 (273) 13 (3)
51.5
40
48
43
43.5
39
65.5
36
51.5
39.5
45.8 (8.7)
MMT
WISCI (Maximum 80)
20
15
9
20
9
56
13
9
13
13
151 (184) 14 (5)
62.5
36
36.5
45
41
53.5
48.5
50
48
47.7 (8.4)
Abbreviations: F, female; M, male; MVA, motor vehicle accident; 10MWT, 10-meter walk test; 6MWT, 6-minute walk test; SCI-FAP, Spinal Cord
Injury–Functional Ambulation Profile; WISCI, Walking Index for Spinal Cord Injury; MMT, Manual Muscle Test.
gains during the rest period after phase 1 training, we also
compared the effects of Endurance and Precision Training
from phase 1 separately (see Supplemental Data available online at http://nnr.sagepub.com/supplemental). The
change scores were still very similar between Precision
and Endurance, none of them being statistically different.
More of the change scores from phase 1 exceeded the MDC
than shown in Figure 3, because the changes were greater
in phase 1 compared with phase 2.
Low Versus High Functioning Walkers
High- (n = 8 Endurance, n = 9 Precision) and low- (n = 9
Endurance, n = 11 Precision) functioning walkers tended to
respond differently to the 2 methods of training (Figure 4).
High-functioning walkers improved more with Endurance
Training, as reflected by significant differences in the 6MWT
(P = .03). In contrast, there were no significant differences
between the 2 training methods for low-functioning walkers.
Time Course of Improvement During Training
In Precision Training, the improvements in walking speed,
distance, and skill (SCI-FAP) occurred approximately
equally over the 2 months (compare hatched vs dotted
gray bars, Figure 5). Conversely, in Endurance Training,
improvements in walking speed and distance (ie,
10MWT(ss), 10MWT(f), 6 MWT) occurred mostly during
the first month of training, while improvements in SCI-FAP
occurred over both months.
Characteristics of Those Who Did Not Improve
Three individuals (P2, P4, P16; 18%) did not improve by
>MDC in any walking outcome during Endurance Training,
2 of the 3 (P2, P4), together with 4 others (P1, P10, P11,
P14; ie, 6 total; 30%) did not improve during Precision
Training. There were no differences between time since
injury, age at onset of training, initial walking ability, MMT
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
6
Neurorehabilitation and Neural Repair XX(X)
Figure 2. Intensity and progression in training. (A) Measures of training intensity are shown for Precision and Endurance Training
(mean and standard error of the mean). Significant differences (standard t test; P < .05) indicated with *. (B) Progression of training
measured during Precision and Endurance phases from initial (average of first 3 training sessions; hatched) to final (average of last 3
training sessions; cross hatched) training. Significant improvements (paired t test; P < .05) indicated with *.
Abbreviations: SCI-FAP, Spinal Cord Injury–Functional Ambulation Profile; 10MWT, 10-meter walk test; WISCI, Walking Index for Spinal Cord Injury;
ss, self-selected; f, fast; CES-D, Center for Epidemiologic Studies–Depression Scale; 6MWT, 6-minute walk test; ABC, Activities-specific Balance
Confidence.
prior to training, and training intensity between those who
did and did not improve.
Retention of Improvements
We lost 2 individuals to follow-up after the second phase of
training, both cases being after Precision Training. One did
not attend the final assessment, while the other joined another
intensive treatment program and was ineligible for followup. We compared retention after each type of training (average follow-up time 1.9 ± 0.4 months for Endurance, and
1.9 ± 0.5 months for Precision Training), and after phase 1
versus phase 2 of training. In addition, individuals who
improved by >MDC were analyzed separately, to avoid
biasing the results by individuals who did not improve during
training, but maintained their function nevertheless.
Regardless of the training type, or the phase of training, participants retained their gains whether those who did not
respond to treatment were included or not. Furthermore,
retention of walking ability was the same for the 2 training
methods (ie, average Retention Index for walking measures:
Endurance=1.04 ± 0.05 (mean ± SD), Precision = 1.05 ±
0.21). Correlations between the final walking speed and the
Retention Index were small to moderate (r = 0.05-0.38),
indicating high- and low-functioning walkers retained their
gains equally well. Correlation between Center for
Epidemiologic Studies–Depression Scale with the Retention
Index was also low (ranging from r = 0.03 to 0.08), indicating retention was unrelated to depression scores in our
cohort.
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
7
Yang et al
Figure 3. Change scores for all outcome measures with each type of training. Mean change is shown with one standard error of
the mean. Statistically significant improvement within each method is shown with a white * in the bar (paired t test comparing scores
pre- and posttraining; P < .05). Statistically significant difference between the 2 methods is illustrated with a black * above the bars
(standard t test; P < .05). The minimally detectable change (dashed line) is shown for walking measures. The number of participants in
each group was 20 for Precision Training and 17 for Endurance Training.
Abbreviations: SCI-FAP, Spinal Cord Injury–Functional Ambulation Profile; 10MWT, 10-meter walk test; ss, self-selected; f, fast; 6MWT, 6-minute walk test.
Figure 4. Response of high versus low functioning individuals to Precision versus Endurance Training. Individuals were considered
high or low functioning based on their walking speed pre-training, for each phase of the training, using the 10MWT(ss) (ie, high >0.5
m/s, low <0.5 m/s). Change scores are shown with one standard error of the mean. Statistically significant difference between the two
methods is shown with * (standard t test; P < .05). Horizontal lines indicate minimally detectable difference (see text).
Abbreviations: SCI-FAP, Spinal Cord Injury–Functional Ambulation Profile; 10MWT, 10-meter walk test; ss, self-selected; f, fast; 6MWT, 6-minute walk test.
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
8
Neurorehabilitation and Neural Repair XX(X)
Figure 5. Time course of improvements in each type of training. Improvements in walking speed (10MWT) and distance (6MWT)
tended to occur in the first month of training, especially with Endurance Training (black bars). Improvements in walking skill (SCI-FAP)
were seen over both months, for both types of training. Statistically significant difference change scores in the 2 months is shown with
*(paired t test; P < .05).
Abbreviations: SCI-FAP, Spinal Cord Injury–Functional Ambulation Profile; 10MWT, 10-meter walk test; ss, self-selected; f, fast; 6MWT, 6-minute
walk test.
Discussion
Is Mass Practice Better Than Precise Practice?
Endurance Training (mass practice) was superior to
Precision Training (skill practice) for inducing large
improvements in walking distance. Surprisingly, it also
improved walking skill just as much as Precision Training
(Figure 3), opposite to our original hypothesis that Precision
Training might be superior to Endurance Training, because
it engages the motor cortex25 (see Introduction section).
The findings may mean that challenging individuals with
SCI to walk faster and longer on a treadmill is sufficient to
induce corticospinal tract plasticity. Indeed, increases in
motor-evoked potentials with single-pulse transcranial
magnetic stimulation recorded before and after endurance
training supports this.44 Furthermore, Endurance Training
involved 3 times as many steps at more than 3 times the
speed in each training session (Figure 2), so the number of
steps and/or the speed of walking may also be very important to induce plasticity. The improvements could also
have been due to cardiovascular conditioning, since the
peak ΔHR during Endurance Training was 56 ± 14 beats/
min compared with Precision Training, which was 39 ± 12
beats/min, although cardiovascular conditioning should
not have affected measures of walking skill. To identify the
key ingredient(s) responsible for the differences between
Endurance and Precision Training, such as walking speed,
number of steps per session, or ΔHR during training will
require control of these variables in future studies.
Another possible explanation why Precision Training
was not superior to Endurance Training is that our outcome
measures could not gauge the high-level improvements
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
9
Yang et al
made during Precision Training in the high-functioning
group. The types of improvements expected from Precision
Training are dynamic balance and skill during walking.
Indeed, all participants could walk over much higher and
wider obstacles, as documented by the Success Index
(Figure 2B). Our primary measure of walking skill was the
SCI-FAP, which was designed to measure daily walking
skills, so the ceiling may be too low to measure these higher
level skills. A normal score on the SCI-FAP is 7; 3 participants had scores just above 7 prior to Precision Training.
The other measure that may reflect walking balance is
the WISCI, but the ceiling was too low for some (ie,
WISCI(ss) = 20 for 4 participants before Precision
Training). Thus, better outcome measures to gauge highlevel walking skill are needed.
The improvements in walking outcomes with both types
of training resembled those reported by others. For example, improvement in self-selected walking speed in randomized controlled trials of people with chronic injury, with n >
10 per group7,16,22 were mean change scores of: 0 or 0.01
m/s,16 0.01 m/s to 0.09 m/s,22 and 0.02 m/s and 0.4 m/s,7
while we report 0.07 m/s for Endurance and 0.04 m/s for
Precision Training. Our change scores could have been
reduced because we collapsed the data across 2 phases. As
most of the gains were made in phase 1, collapsing the data
will under estimating the change scores overall. The change
scores for phase 1 only are 0.07 m/s for Endurance and 0.06
m/s for Precision (Supplementary Data). Our participants
were not unique compared with those in other studies with
respect to chronicity of the injury or walking ability prior to
training. Our training parameters (number of sessions, duration of sessions) are also close to the median values reported
by others (reviewed in Yang and Musselman19). Finally, the
number of steps executed in Precision Training is low, but
this parameter is rarely reported in the literature. In the case
where the target number of steps had been reported (8001250),10 Endurance Training is comparable (1189 ± 557),
but Precision Training is lower (369 ± 180). Thus, we suggest that this training parameter may be very important to
follow in the future.
High- and Low-Functioning Walkers Benefitted
Differently From the Two Types of Training
Endurance Training induced greater change than Precision
Training in the 6MWT for high-functioning walkers,
whereas there were no differences for any of the walking
outcomes for the low-functioning walkers. The difference
in response could not be explained by a relative difference
in the way they trained, since the number of strides executed in training was approximately 3 times higher for
Endurance Training compared with Precision Training for
both groups. Thus, we suggest that the method of training is
less critical for individuals with low walking function,
while higher functioning individuals may benefit more
from Endurance Training. Of course, these conclusions
cannot be generalized to very low functioning individuals,
since they were excluded from participating.
Improvements in Walking Speed and Distance
Are Acquired Faster Than Walking Skill
The largest changes in walking speed and distance were
induced by Endurance Training in the first month (Figure
5). This suggests that the fastest way to boost walking speed
and endurance is to use mass practice as we did in Endurance
Training. In contrast, walking skill (ie, SCI-FAP) improved
more slowly, extending over the 2 months with both types
of training. This agrees with our previous case report, in
which we trained individuals with a program of alternating
treadmill and over ground walking training,21 in which
improvements were found over at least 6 months. Thus, it is
important that treatment not be terminated too early.
Furthermore, it is insufficient to only measure walking
speed and distance in walking outcome; skill is an important and different aspect of walking, which has a different
time course for improvement.
Retention of Walking Ability After Training
Retention of walking speed, distance, and skill was good
after both forms of training. Since the relationship between
final walking speed and retention was weak to modest, it
suggests that other factors unrelated to walking speed may
have been more important to retention, such as the motivation of the individual. In the future, it would be useful to
quantify the activities participants engaged in during the
rest period, to better understand the factors affecting
retention.
Limitations
First, the participants were a sample of convenience. As
the study required heavy time commitment from participants, they were among the highly motivated, a selection
bias. Second, the sample size was small, resulting in moderate statistical power for detecting small differences
(Cohen’s d45 between 0.21 and 0.69 for walking outcome
measures). Third, 4 participants (P5, P11, P15, P18) had
lesions at T12-L1, whose response to training could be
different from participants with higher levels of injury.
However, no differences were seen between the improvements made by these individuals compared to the rest of
the cohort (ie, Δ10MWT(ss) during Endurance = 0.040.06 m/s for T12-L1 lesions, 0.07 ± 0.12 m/s for higher
lesions, and during Precision = 0.0-0.15 m/s for T12-L1
lesions, 0.04 ± 0.06 m/s for higher lesions). Fourth, 4 participants (P5, P7, P15, P17) were injured less than a year
ago, so it is possible that spontaneous recovery influenced
their improvement, even though their baseline scores were
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
10
Neurorehabilitation and Neural Repair XX(X)
stable. Indeed, their improvements in 10MWT (ss) were
higher than the mean of the other participants, although
still within the range of improvements observed in the
other individuals (0.04-0.16 m/s for <1 year postinjury vs
0.06 ± 0.12 m/s >1 year postinjury for Endurance, and
0-0.15 m/s for <1 year postinjury vs 0.03 ± 0.06 m/s for >1
year postinjury for Precision). Finally, it was difficult to
enforce the same level of physical activity in our participants, especially during the rest periods.
Conclusions
The 2 forms of training, although very different, both led to
improvements in over ground walking. Mass practice
(Endurance Training) resulted in greater changes in walking distance. Both led to improvements in walking skills
needed in daily life. Finally, better outcome measures to
gauge improvements, especially in walking balance and
skill, are still needed as the ceiling in many measures
remains low.
Acknowledgments
We thank Drs Blair Calancie and Hubertus van Hedel for helpful
comments on earlier versions of this manuscript. We thank the
physical therapists, Sarah Pletch, Tina Carter, and Michelle
Krappala, for serving as blinded assessors, the physical therapy
assistants, Rachelle Lohlun and Katelyn Brown, for assisting with
the training and organization, and the participants for devoting
their time and effort to the study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article: This
project was supported by the Canadian Institutes of Health
Research, Christopher and Dana Reeve Paralysis Foundation,
Alberta Paraplegic Foundation, Rick Hansen Institute, and the
Rehabilitation Medicine Students Association.
References
1.Hicks AL, Adams MM, Martin Ginis K, et al. Long-term
body-weight-supported treadmill training and subsequent
follow-up in persons with chronic SCI: effects on functional
walking ability and measures of subjective well-being. Spinal
Cord. 2005;43:291-298.
2. Behrman AL, Harkema SJ. Locomotor training after human
spinal cord injury: a series of case studies. Phys Ther.
2000;80:688-700.
3. Fritz SL, Merlo-Rains AM, Rivers ED, et al. An intensive
intervention for improving gait, balance, and mobility in
individuals with chronic incomplete spinal cord injury: a
pilot study of activity tolerance and benefits. Arch Phys Med
Rehabil. 2011;92:1776-1784.
4.Harkema SJ, Schmidt-Read M, Lorenz DJ, Edgerton VR,
Behrman AL. Balance and ambulation improvements in individuals with chronic incomplete spinal cord injury using locomotor training-based rehabilitation. Arch Phys Med Rehabil.
2012;93:1508-1517.
5.Dobkin B, Apple D, Barbeau H, et al. Weight-supported
treadmill vs over-ground training for walking after acute
incomplete SCI. Neurology. 2006;66:484-493.
6. Alexeeva N, Sames C, Jacobs PL, et al. Comparison of training methods to improve walking in persons with chronic
spinal cord injury: a randomized clinical trial. J Spinal Cord
Med. 2011;34:362-379.
7. Lucareli PR, Lima MO, Lima FP, de Almeida JG, Brech GC,
D’Andrea Greve JM. Gait analysis following treadmill training with body weight support versus conventional physical
therapy: a prospective randomized controlled single blind
study. Spinal Cord. 2011;49:1001-1007.
8. Field-Fote EC. Combined use of body weight support, functional electric stimulation, and treadmill training to improve
walking ability in individuals with chronic incomplete spinal
cord injury. Arch Phys Med Rehabil. 2001;82:818-824.
9. Postans NJ, Hasler JP, Granat MH, Maxwell DJ. Functional
electric stimulation to augment partial weight-bearing supported treadmill training for patients with acute incomplete
spinal cord injury: a pilot study. Arch Phys Med Rehabil.
2004;85:604-610.
10. Hesse S, Werner C, Bardeleben A. Electromechanical gait
training with functional electrical stimulation: case studies in
spinal cord injury. Spinal Cord. 2004;42:346-352.
11. Wirz M, Zemon DH, Rupp R, et al. Effectiveness of automated locomotor training in patients with chronic incomplete
spinal cord injury: a multicenter trial. Arch Phys Med Rehabil.
2005;86:672-680.
12. Winchester P, McColl R, Querry R, et al. Changes in supraspinal activation patterns following robotic locomotor therapy
in motor-incomplete spinal cord injury. Neurorehabil Neural
Repair. 2005;19:313-324.
13. Wu M, Landry JM, Schmit BD, Hornby TG, Yen SC. Robotic
resistance treadmill training improves locomotor function
in human spinal cord injury: a pilot study. Arch Phys Med
Rehabil. 2012;93:782-789.
14. Schwartz I, Sajina A, Neeb M, Fisher I, Katz-Luerer M, Meiner
Z. Locomotor training using a robotic device in patients with
subacute spinal cord injury. Spinal Cord. 2011;49:1062-1067.
15. Benito-Penalva J, Edwards DJ, Opisso E, et al. Gait training
in human spinal cord injury using electromechanical systems:
effect of device type and patient characteristics. Arch Phys
Med Rehabil. 2012;93:404-412.
16.Alcobendas-Maestro M, Esclarin-Ruz A, Casado-Lopez
RM, et al. Lokomat robotic-assisted versus overground training within 3 to 6 months of incomplete spinal cord lesion:
randomized controlled trial. Neurorehabil Neural Repair.
2012;26:1058-1063.
17. Swinnen E, Duerinck S, Baeyens JP, Meeusen R, Kerckhofs
E. Effectiveness of robot-assisted gait training in persons
with spinal cord injury: a systematic review. J Rehabil Med.
2010;42:520-526.
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013
11
Yang et al
18. Morawietz C, Moffat F. The effects of locomotor training after
incomplete spinal cord injury: a systematic review [published
online July 9, 2013]. Arch Phys Med Rehabil. doi:10.1016/j.
apmr.2013.06.023.
19. Yang JF, Musselman KE. Training to achieve over ground
walking after spinal cord injury: a review of who, what, when,
and how. J Spinal Cord Med. 2012;35:293-304.
20.Wessels M, Lucas C, Eriks I, de Groot S. Body weightsupported gait training for restoration of walking in people
with an incomplete spinal cord injury: a systematic review. J
Rehabil Med. 2010;42:513-519.
21. Musselman KE, Fouad K, Misiaszek JE, Yang JF. Training
of walking skills overground and on the treadmill: case series
on individuals with incomplete spinal cord injury. Phys Ther.
2009;89:601-611.
22. Field-Fote EC, Roach KE. Influence of a locomotor training approach on walking speed and distance in people with
chronic spinal cord injury: a randomized clinical trial. Phys
Ther. 2011;91:48-60.
23. Thomas SL, Gorassini MA. Increases in corticospinal tract
function by treadmill training after incomplete spinal cord
injury. J Neurophysiol. 2005;94:2844-2855.
24. Rossignol S, Frigon A. Recovery of locomotion after spinal
cord injury: some facts and mechanisms. Annu Rev Neurosci.
2011;34:413-440.
25. Drew T, Andujar JE, Lajoie K, Yakovenko S. Cortical mechanisms involved in visuomotor coordination during precision
walking. Brain Res Rev. 2008;57:199-211.
26. Wolf SL, Catlin PA, Gage K, Gurucharri K, Robertson R,
Stephen K. Establishing the reliability and validity of measurements of walking time using the Emory Functional
Ambulation Profile. Phys Ther. 1999;79:1122-1133.
27. Musselman K, Brunton K, Lam T, Yang J. Spinal cord injury
functional ambulation profile: a new measure of walking ability. Neurorehabil Neural Repair. 2011;25:285-293.
28. Guyatt GH, Sullivan MJ, Thompson PJ, et al. The 6-minute
walk: a new measure of exercise capacity in patients with
chronic heart failure. Can Med Assoc J. 1985;132:919-923.
29. van Hedel HJ, Wirz M, Dietz V. Assessing walking ability in
subjects with spinal cord injury: validity and reliability of 3
walking tests. Arch Phys Med Rehabil. 2005;86:190-196.
30. Dittuno PL, Ditunno JF Jr. Walking index for spinal cord injury
(WISCI II): scale revision. Spinal Cord. 2001;39:654-656.
31. Ditunno JF Jr, Barbeau H, Dobkin BH, et al. Validity of the
walking scale for spinal cord injury and other domains of
function in a multicenter clinical trial. Neurorehabil Neural
Repair. 2007;21:539-550.
32. Burns AS, Delparte JJ, Patrick M, Marino RJ, Ditunno JF.
The reproducibility and convergent validity of the walking
index for spinal cord injury (WISCI) in chronic spinal cord
injury. Neurorehabil Neural Repair. 2011;25:149-157.
33. van Hedel HJ, Wirz M, Curt A. Improving walking assessment in subjects with an incomplete spinal cord injury:
responsiveness. Spinal Cord. 2006;44:352-356.
34. Musselman K, Yang JF. The Spinal Cord Injury Functional
Ambulation Profile (SCI-FAP): a preliminary look at responsiveness [published online October 10, 2013]. Phys Ther.
doi:10.2522/ptj.20130071.
35. Ditunno JF Jr, Ditunno PL, Graziani V, et al. Walking index
for spinal cord injury (WISCI): an international multicenter
validity and reliability study. Spinal Cord. 2000;38:234-243.
36.Powell LE, Myers AM. The Activities-specific Balance
Confidence (ABC) Scale. J Gerontol A Biol Sci Med Sci.
1995;50:M28-M34.
37.Weissman MM, Sholomskas D, Pottenger M, Prusoff BA,
Locke BZ. Assessing depressive symptoms in five psychiatric populations: a validation study. Am J Epidemiol.
1977;106:203-214.
38.Kendall FP, McCreary EK, Provance PG, Rodgers MM,
Romani WA. Muscles Testing and Function with Posture
and Pain. Philadelphia, PA: Lippincott Williams & Wilkins;
2005.
39. Yang JF, Norton J, Nevett-Duchcherer J, Roy FD, Gross DP,
Gorassini MA. Volitional muscle strength in the legs predicts changes in walking speed following locomotor training in people with chronic spinal cord injury. Phys Ther.
2011;91:931-943.
40. Lam T, Noonan VK, Eng JJ, Team SR. A systematic review
of functional ambulation outcome measures in spinal cord
injury. Spinal Cord. 2008;46:246-254.
41. Musselman K. Clinical significance testing in rehablitation
research: what, why and how? Phys Ther Rev. 2007;12:287296.
42. Robinett CS, Vondran MA. Functional ambulation velocity
and distance requirements in rural and urban communities. A
clinical report. Phys Ther. 1988;68:1371-1373.
43. Andrews AW, Chinworth SA, Bourassa M, Garvin M, Benton
D, Tanner S. Update on distance and velocity requirements for
community ambulation. J Geriatr Phys Ther. 2010;33:128134.
44. Zwedie E, Roy F, Yang J, Gorassini MA. Changes in the
spared corticospinal tract induced by precision and endurance training after incomplete spinal cord injury. Paper
presented at: Neuroscience 2012: Annual Meeting of the
Society for Neuroscience; October 13-17, 2012; New
Orleans, LA.
45. Cohen J. Statistical Power Analysis for the Behavioral
Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum; 1988.
Downloaded from nnr.sagepub.com at UNIV OF DELAWARE LIB on November 18, 2013