Archives of Physical Medicine and Rehabilitation

journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2013;94:1247-55

ORIGINAL ARTICLE

Cross-Education for Improving Strength and Mobility After

Distal Radius Fractures: A Randomized Controlled Trial
Charlene R.A. Magnus, PhD,a Cathy M. Arnold, PhD, PT,b
Geoffrey Johnston, MD, MBA, FRCSC,b Vanina Dal-Bello Haas, PhD, PT,c
Jenny Basran, MD, FRCPC,b Joel R. Krentz, MSc,a Jonathan P. Farthing, PhDa
From the aCollege of Kinesiology and bCollege of Medicine, University of Saskatchewan, Saskatoon, SK; and cSchool of Rehabilitation Science,
McMaster University, Hamilton, ON, Canada.

Abstract
Objective: To evaluate the effects of cross-education (contralateral effect of unilateral strength training) during recovery from unilateral distal
radius fractures on muscle strength, range of motion (ROM), and function.
Design: Randomized controlled trial (26-wk follow-up).
Setting: Hospital, orthopedic fracture clinic.
Participants: Women older than 50 years with a unilateral distal radius fracture. Fifty-one participants were randomized and 39 participants were
included in the final data analysis.
Interventions: Participants were randomized to standard rehabilitation (Control) or standard rehabilitation plus strength training (Train).
Standard rehabilitation included forearm casting for 40.4
6.2 days and hand exercises for the fractured extremity. Nonfractured hand strength
training for the training group began immediately postfracture and was conducted at home 3 times/week for 26 weeks.
Main Outcome Measures: The primary outcome measure was peak force (handgrip dynamometer). Secondary outcomes were ROM (flexion/
extension; supination/pronation) via goniometer and the Patient Rated Wrist Evaluation questionnaire score for the fractured arm.
Results: For the fractured hand, the training group (17.3
7.4kg) was significantly stronger than the control group (11.8
5.8kg) at 12 weeks
postfracture (P<.017). There were no significant strength differences between the training and control groups at 9 (12.5
8.2kg; 11.3
6.9kg) or 26
weeks (23.0
7.6kg; 19.6
5.5kg) postfracture, respectively. Fractured hand ROM showed that the training group had significantly improved wrist
flexion/extension (100.5
19.2 ) than the control group (80.2
18.7 ) at 12 weeks postfracture (P<.017). There were no significant differences
between the training and control groups for flexion/extension ROM at 9 (78.0
20.7 ; 81.7
25.7 ) or 26 weeks (104.4
15.5 ; 106.0
26.5 )
or supination/pronation ROM at 9 (153.9
23.9 ; 151.8
33.0 ), 12 (170.9
9.3 ; 156.7
20.8 ) or 26 weeks (169.4
11.9 ; 162.8
18.1 ),
respectively. There were no significant differences in Patient Rated Wrist Evaluation questionnaire scores between the training and control groups
at 9 (54.2
39.0; 65.2
28.9), 12 (36.4
37.2; 46.2
35.3), or 26 weeks (23.6
25.6; 19.4
16.5), respectively.
Conclusions: Strength training for the nonfractured limb after a distal radius fracture was associated with improved strength and ROM in the
fractured limb at 12 weeks postfracture. These results have important implications for rehabilitation strategies after unilateral injuries.
Archives of Physical Medicine and Rehabilitation 2013;94:1247-55
ª 2013 by the American Congress of Rehabilitation Medicine

Preliminary results of the study presented to the Canadian Society for Exercise Physiology, October
20, 2011, Quebec City, QC, Canada; the Canadian Orthopaedic Association, June 8, 2012, Ottawa, ON,
Canada; and the Canadian Physiotherapy Association, May 25, 2012, Saskatoon, SK, Canada.
Supported by the Royal University Hospital Foundation Grant, doctoral funding from the
Natural Sciences and Engineering Research Council of Canada, the Dean’s Scholarship
from the University of Saskatchewan, and a Graduate Scholarship from the University of
Saskatchewan.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit on the authors or on any organization with
which the authors are associated.
Cross-education is a neural adaptation defined as the increase in
strength or functional performance of the untrained contralateral
limb after unilateral training of the opposite homologous limb.1,2
The increase in strength in the untrained limb is related to the gain
in magnitude of the trained limb, and is on average 52% of the
strength gain observed in the trained muscle.2 Cross-education is
thought to be primarily controlled by neural mechanisms,2-6 but
the exact mechanisms are unknown.

0003-9993/13/$36 - see front matter ª 2013 by the American Congress of Rehabilitation Medicine
http://dx.doi.org/10.1016/j.apmr.2013.03.005
1248
A large gap in the literature remains in applying cross-
education to clinical rehabilitation settings. The potential benefit
of cross-education for rehabilitation from unilateral injuries (ie, a
fractured limb) is an obvious, clinically relevant extension
of the work; however, little research has been conducted in clin-
ical application of cross-education.7 Stromberg7 applied cross-
education after wrist/forearm surgeries, but several limitations
such as not including raw data, not accounting for baseline
differences, and not reporting details of the training program have
made it difficult to draw any conclusions from the results. Three
studies have applied cross-education to unilateral immobilization
in healthy (ie, nonfractured) limbs.6,8,9 Farthing et al6,8 found that
cross-education strength training on the nonimmobilized limb
provided a maintenance of strength in the immobilized healthy
limb after wearing a forearm cast for 3 weeks. Similarly, Magnus
et al9 found that strength training of the nonimmobilized arm
provided an increase in strength in the healthy immobilized arm
after wearing an arm sling for 4 weeks. These studies suggest that
cross-education can benefit a healthy immobilized limb. As yet,
there are no randomized controlled clinical trials that have
investigated these effects in real injuries that require limb immo-
bilization. More research in this area may help improve the
rehabilitation techniques clinicians use postinjury, and in turn may
improve function for those with unilateral injuries such as distal
radius fractures.
Distal radius fractures are one of the most common types of
fracture,10 especially in older women.11 Rehabilitation after a
distal radius fracture is quite slow, and it can often be difficult for
individuals to return to their normal level of functioning. Brogren
et al12 showed that 1 year postfracture, grip strength was 88% of
the nonfractured limb. Similarly, Trumble et al13 found that 2.4
years postfracture, grip strength was 69% of the nonfractured limb
and range of motion (ROM) was 75% of the nonfractured limb. A
Cochrane Review by Handoll et al11 examined the effects of
rehabilitation beginning both during and after immobilization in
adults with distal radius fractures. Fifteen randomized controlled
trials were included, whereby treatment was conservative and
involved plaster cast immobilization. The review found that there
was insufficient evidence to determine the best form of rehabili-
tation after distal radius fractures. New ways of improving reha-
bilitation to enhance recovery and to provide better functional
outcome are important to investigate.
One way of improving strength and functional gains in the
fractured hand may be to apply cross-education during recovery
from unilateral distal radius fractures. Unilateral distal radius
fractures represent an adequate clinical model to test the efficacy
of cross-education due to the standard immobilization intervention
of forearm casting for approximately 6 weeks. In our clinic, there
is no rigorous therapeutic intervention prescribed for individuals
beyond ROM exercises for the fractured limb, and potential
referral to physical therapy for more severe fractures. To our
knowledge, there are no rehabilitation protocols that incorporate
a formal strength training program of the nonfractured side as part
of the recovery for the fractured side after distal radius fractures.11
List of abbreviations:
ANOVA analysis of variance
MCAR missing completely at random
PRWE Patient Rated Wrist Evaluation
ROM range of motion
C.R.A. Magnus et al
The purpose of this study was to apply cross-education to
unilateral distal radius fractures in women 50 years of age and
older and to evaluate the effects on grip strength, ROM, and
function. The hypothesis was that strength training of the non-
fractured limb in addition to standard rehabilitation of the fractured
limb would provide better strength and functional outcome than
standard rehabilitation alone after a unilateral distal radius fracture.
Methods
Participants
Women aged 50 years and older with a unilateral distal radius
fracture were recruited for 1 year from the fracture clinic at Royal
University Hospital in Saskatoon, Saskatchewan, Canada, under
the direction of 1 orthopedic surgeon. Patients referred to the clinic
who met inclusion criteria were invited to participate in the study
before their first visit to the clinic. Exclusion criteria included
any prior upper body injury or joint problem interfering with daily
life, or any history of upper-extremity neurologic problems (eg,
stroke, multiple sclerosis, Parkinson’s disease, vestibular disor-
ders, reflex neuropathy). Participants were also excluded if the
fracture was >2 weeks old at the time of the first visit to the clinic
or if there were multiple fractures of the wrist and forearm. All
participants completed the Mini-Cognitive Assessment Instrument
for Dementia14 to screen for cognitive impairment. Those who
were unable to remember any words in the word recall and those
who scored an abnormal clock draw test and recalled only 1 or 2
words were not included in the study.
A sample size calculation was completed using G Power 3.115,a
for the primary outcome variable (ie, strength). On the basis of our
previous immobilization cross-education studies involving fore-
arm casting,6,8 we anticipated a 13% difference in affected limb
strength between training and control. Because we have no pre-
vious data on cross-education effects on injured participants, we
used a much smaller effect size estimate based on a 5% difference
between groups to achieve a more conservative sample size esti-
mate. Using alpha of .05 at 80% power, and an effect size of 0.2,
the total required sample size was 36 (ie, 18 per group). Before the
commencement of the study, all participants completed written
informed consent approved by the Biomedical Ethics Review
Board at the University of Saskatchewan with subsequent opera-
tional approval from the Saskatoon Health Region. Participants
completed the Waterloo Handedness Questionnaire16 at the first
clinic visit to determine handedness. The 10-item questionnaire is
scored from 20 to þ20, whereby negative scores indicate left-
handedness and positive scores indicate right-handedness. Partic-
ipant characteristics per group are shown in table 1.
Study design
Participants were randomly assigned to 1 of 2 groups using
a computer random number generator (see fig 1 for participant
enrollment flow diagram). Randomization was completed at the
first visit to the clinic by a researcher who did not conduct any of
the testing procedures. The orthopedic surgeon and all other testing
staff were blinded to the randomization of groups to limit any bias,
altered treatment, or encouragement during testing procedures.
Group 1 participants received the standard clinical rehabilitation
protocol after a distal radius fracture and strength trained their
nonfractured limb throughout the duration of the study (Train), and
www.archives-pmr.org
Cross-education after wrist fractures
Table 1 Descriptive characteristics for all randomized participants
Waterloo
Handedness
Group Age (y) Height (cm) Weight (kg) Score
Train (nZ27) 63.3
10.0 161.8
6.2 70.3
17.5 14.6
9.4
Control (nZ24) 62.7
10.2 159.6
6.8 70.1
19.7 16.4
8.1
NOTE. Values for age, height, weight, Waterloo Handedness Score, and casting period are mean
SD. Dominant/nondominant fracture, number of
surgeries, and number attended physiotherapy are frequencies per group. There are no significant differences between groups for any participant
characteristics.
Group 2 received the standard clinical rehabilitation protocol after
a distal radius fracture (Control). The standard clinical rehabili-
tation protocol included forearm casting; 6 visits to the fracture
clinic at weeks 1, 3, 6, 9, 12, and 26 postfracture; and the adoption
of 3 paper-based exercise protocols designed by a panel of phys-
ical therapists targeting the fractured side (in cast, 6wk post-
fracture, and 9wk postfracture). The orthopedic surgeon coached
patients on each of the time-specific protocols at the appropriate
time. Standard rehabilitation began with active ROM exercises for
the neck, shoulder, elbow, fingers, and thumb while in the cast.
Once the cast was removed, exercises focused on improving active
and passive ROM of the fractured wrist and hand (ie, supination,
pronation, flexion, extension). Stretching continued at 9 weeks
postfracture, and strengthening exercises were integrated into the
exercise regimen. Strengthening exercises such as wrist curls and
gripping a soft ball/sponge/play dough were prescribed once per
day. Participants were instructed to complete the exercises 10 to 12
times per day. At 12 weeks postfracture, the patients were
encouraged to continue with their exercises, and no formal limi-
tations on their activity levels were imposed. The standard reha-
bilitation protocol encouraged patients to continue these exercises
throughout recovery; however, no training log or formalized
regimen was implemented to track adherence. All exercises were
to be completed at home, unsupervised on the patient’s own time,
with no prescribed exercises given to the nonfractured arm.
Standard rehabilitation did not require patients to see a physio-
therapist, but some were referred by the orthopedic surgeon or by
their own family physician and attended physiotherapy on their
own initiative (see table 1).
Training intervention
The training group participants (Train) strength trained their
nonfractured arm during the casting period and continued to
strength train their nonfractured arm throughout the follow-up
period (ie, 26wk total). The strength training intervention was
completed in addition to the standard clinical rehabilitation
protocol described above (Control). Strength training during the
casting period was progressive in nature, beginning with 2 sets of
8 repetitions and increasing up to a maximum of 5 sets of 8
repetitions of maximal voluntary effort handgrip contractions
as tolerated.
Strength training of the nonfractured side began immediately
after the first clinic visit. Handgrip training was performed using
standard handgrip trainersb to train finger, hand, and forearm
strength. The resistance levels in the handgrip trainers ranged
www.archives-pmr.org
1249
Number of
Dominant/ Fractures
Casting Nondominant Repaired Number Attended
Period (d) Fracture Surgically Physiotherapy
42.2
6.0 DominantZ12 9 7
NondominantZ15
38.9
6.0 DominantZ13 2 7
NondominantZ11
from extra light (0.7–2.3kg) to extra heavy (4.1–14.1kg). In the
event that the extra heavy handgrip trainersb were not strong
enough, participants used coil resistance handgrip trainersc to
progress their training. Each maximal handgrip contraction was
held for 3 seconds, and therefore was essentially isometric in
nature. Participants were instructed to increase resistance with the
coil resistance trainersc by beginning with the hand at the bottom
of the handles, and to move the hand up (closer to the coil) as the
exercises became less difficult. Strength was assessed for each
participant to determine which handgrip trainer would begin the
training program. Progression in resistance was individually
determined and monitored throughout the study by telephone
calls and at subsequent visits. Participants completed the exer-
cises 3 times per week, and recorded adherence in a training log
monitored by the researchers. Participants had to complete at
least 1 training session/week (on average) to be considered
trained and included in the final data analysis. The strength
training intervention was unsupervised and conducted individu-
ally at home. Training participants were contacted via telephone
biweekly to encourage adherence and to monitor training. To
ensure there was no effect of the phone calls on rehabilitation,
the control group participants were also called via telephone
biweekly and were asked how their wrist was feeling, and
whether there had been any changes in their wrist since the last
time they were contacted.
All participants were tested at regular visits to the clinic
(weeks 1, 3, 6, 9, 12, and 26); however, the present study displays
results only from 4 time points: weeks 1 (1e2wk postfracture), 9,
12, and 26. Weeks 3 and 6 were not included in the analysis
because only the nonfractured side could be measured at these
time points for practical reasons.
Strength
Isometric grip strength was assessed using a calibrated hand
dynamometer.d Testing was conducted with participants seated,
the shoulder completely adducted, elbow flexed at 90 , and the
wrist in neutral position (palm facing medially). The peak value
obtained from 3 maximal voluntary efforts was used for
comparison. The contractions were 3 seconds in duration with
each contraction separated by a 1-minute rest. The nonfractured
extremity was always tested first. At week 1, grip strength was
assessed on the nonfractured side only. Week 9 (ie, 3wk after cast
removal) was the first time point that participants were able to
complete a maximal contraction on the fractured side. Both sides
were tested for strength at weeks 9, 12, and 26. Participants were
1250 C.R.A. Magnus et al

Fig 1 Participant enrollment flow diagram. Final n for TrainZ18. Final n for ControlZ21. All dropouts left the study prior to the 9-week
follow-up testing point; therefore, no data were collected on the fractured arm and dropouts were not included in the final analysis.

instructed to squeeze the dynamometer as hard as they could for
the duration of the contraction and were given verbal encour-
agement at each trial. To minimize the learning effect, all
participants were familiarized with the dynamometer before the
contractions.
Range of motion
Active ROM was assessed manually using a goniometer for wrist
flexion, extension, supination, and pronation. Wrist flexion and
extension scores were added together to give a combined flexion/
extension range. Supination and pronation were also added
together for a combined supination/pronation range. All measures
were conducted with the participant seated, shoulder fully
adducted, and the elbow bent 90 . ROM was measured on the
fractured limb only at weeks 9, 12, and 26.
Patient Rated Wrist Evaluation
The Patient Rated Wrist Evaluation (PRWE)17 is a 15-item
questionnaire designed to assess wrist pain and function with
activities of daily living. Respondents self-reported levels of wrist
pain and function using a scale ranging from 0 to 10 (0Zno pain/
no difficulty; 10Zworst pain/unable to do activity). A total score
was calculated by adding the responses for each question (best
scoreZ0; worst scoreZ150). Results for the PRWE questionnaire
are shown for weeks 1, 9, 12, and 26. Week 1 was answered
as a retrospective prefracture score. Weeks 9, 12, and 26 were
completed for the corresponding time point postfracture.

www.archives-pmr.org
Cross-education after wrist fractures 1251

Data analysis differences between groups for strength at week 1. There was
a significant GroupTime interaction (F3,37Z4.01, PZ.009,
All data were analyzed using SPSS 20.0e and were checked for partial h2Z.098), and a significant TimeArm interaction
normality using skewness and kurtosis tests. The primary outcome (F3,37Z108.38, P<.001, partial h2Z.745). Post hoc analysis for
for the study was strength for the fractured arm, and secondary the fractured arm showed that there was a significant difference
outcomes were ROM and the PRWE questionnaire score for the between the training and control groups at 12 weeks postfracture
fractured arm. As described in the Consolidated Standards of (17.3
7.4kg and 11.8
5.8kg, respectively) (Bonferroni adjusted,
Reporting Trials statement, a modified intention-to-treat anal- P<.05/3Z.017) (fig 2). There were no significant strength
ysis18,19 was used to determine whether the trial would work in differences in the fractured arm between the training and control
a group of adhering participants. This was the first clinical trial groups at 9 (12.5
8.2kg; 11.3
6.9kg) or 26 weeks (23.0
7.6kg;
that attempted to apply cross-education in a wrist fracture setting, 19.6
5.5kg) postfracture, respectively. Week 1 was significantly
and future trials should use intention-to-treat analysis to determine different than all other time points for both the training group and
the outcome in both adhering and nonadhering participants once the control group in the fractured arm (Bonferroni adjusted for
the preliminary trial has been conducted. Group series mean was multiple comparisons, P<.05). For the fractured arm of the
used to replace missing data if they were determined to be missing training group, week 9 (12.5
8.2kg) was significantly different
completely at random (MCAR).20 Strength was analyzed using than week 12 (17.3
7.4kg) and weeks 9 and 12 were significantly
a Group (Train, Control)Time (Week 1, 9, 12, and 26)Arm different than week 26 (23.0
7.6kg) (Bonferroni adjusted for
(Fractured, Nonfractured) repeated-measures analysis of variance multiple comparisons, P<.05). For the fractured arm of the control
(ANOVA). Week 1 strength values for the nonfractured arm were group, weeks 9 (11.3
6.9kg) and 12 (11.8
5.8kg) were signifi-
used as week 1 strength values for the fractured arm. The non- cantly different than week 26 (19.6
5.5kg) (Bonferroni adjusted
fractured arm baseline measurement was also assumed as the for multiple comparisons, P<.05).
fractured arm baseline measurement because it was impossible to The nonfractured arm of the training group significantly
get a strength measure on the fractured arm at week 1. ROM increased strength from week 1 (28.1
6.0kg) to weeks 9
(fractured arm only) was analyzed using a Group (Train, Con- (30.8
6.9kg), 12 (30.7
6.5kg), and 26 (31.0
6.9kg) (Bonferroni
trol)Time (Week 9, 12, and 26) repeated-measures ANOVA. The adjusted for multiple comparisons, P<.05) (fig 3). The non-
PRWE questionnaire was analyzed using a Group (Train, Con- fractured arm of the control group significantly decreased strength
trol)Time (Week 1, 9, 12, and 26) repeated-measures ANOVA. from week 9 (26.9
4.4kg) to week 12 (24.9
4.4kg) (Bonferroni
Week 1 for the PRWE questionnaire was a retrospective pre- adjusted for multiple comparisons, P<.05). There was a significant
fracture score. If significant main effects or interactions were difference between groups at 9, 12, and 26 weeks for nonfractured
detected, simple main effects analysis continued using 1-way arm strength (P<.05), and when Bonferroni adjusted, week 12
ANOVA and Bonferroni adjustments. Bonferroni adjustments remained significantly different (P<.05/3ZP<.017). Raw strength
were made using SPSS programming where possible (identified by data can be viewed in table 3.
stating Bonferroni adjustments were made for multiple compari-
sons) or were adjusted for manually by dividing by the number of Range of motion
tests (ie, P<.05/3Z.017). Significance was accepted at P<.05.
Little’s MCAR test was used to determine that missing data for
Results ROM (c212Z11.75, PZ.466) were MCAR. There were a total of
24 of 234 missing data points for ROM (see table 2 for
Fifty-one women with an average age of 63.0
10.0 years, height a description of missing data). Therefore, group series means were
of 160.7
6.5cm, and weight of 70.2
18.4kg were randomized in used to replace all missing data.20 For ROM, there was a signifi-
the study (see fig 1 for participant enrollment flow diagram and cant GroupTime interaction for flexion/extension (F2,37Z8.20,
table 1 for participant characteristics). Of the 40 women who PZ.001, partial h2Z0.181) and a significant time main effect for
initiated the intervention or control period and completed the study,
there was one nonadherent participant in the training group who Table 2 Number of missing data points per group
did not complete the minimum requirement of strength training Week Week Week Week Total
sessions (at least 1 training session/week) and therefore was not Variable Group 1 9 12 26 Missing
included in the data analysis following a modified intention-to- Strength Train 0 1 1 0 21
treat18,19 approach. Dropouts were not included in the final data Control 0 5 8 6
analysis because all left the study before the 9-week follow-up Flexion/ Train NT 2 1 0 24
testing point; therefore, no data were collected on the fractured extension Control NT 2 4 3
arm (ie, no group comparisons could be made for the primary ROM
outcome). The final number of participants included in the analysis Supination/ Train NT 2 1 0
was 18 in the training group and 21 in the control group. pronation Control NT 2 4 3
ROM
Strength PRWE Train 7 1 0 2 28
questionnaire Control 8 3 2 5
Little’s MCAR test was used to determine that missing data for
NOTE. ROM was not tested at week 1. Train nZ18, Control nZ21. Table
strength (c232Z39.89, PZ.159) were MCAR. There were a total of
includes all participants in the final analysis. The total number of data
21 of 312 missing data points for strength (see table 2 for points for strength was 312, ROM 234, and PRWE questionnaire 156.
a description of missing data). Therefore, group series means were Abbreviation: NT, not tested.
used to replace all missing data.20 There were no significant

www.archives-pmr.org
1252
Fig 2 Fractured limb handgrip strength (mean
SE). There was
a significant GroupTime interaction, and a significant TimeArm
interaction for strength (P<.05). NOTE. Dotted line is week 1 non-
fractured limb strength. *Significantly different than all other time
points. **Significantly different than week 9. ***Significantly
different than weeks 9 and 12 (adjusted for multiple comparisons,
P<.05). #Significant difference between groups (Bonferroni-adjusted,
P<.05/3Z.017).
flexion/extension (F2,37Z30.09, P<.001, partial h2Z0.449) and
supination/pronation (F2,37Z8.13, PZ.001, partial h2Z.180).
Post hoc analyses revealed that for flexion/extension ROM, there
was a significant difference between the training group
(100.5
19.2 ) and the control group (80.2
28.7 ) at 12 weeks
postfracture (Bonferroni adjusted, P<.05/3Z.017) (fig 4). There
were no significant differences between the training and control
groups for flexion/extension ROM at 9 (78.0
20.7 ;
81.7
25.7 ) or 26 weeks (104.4
15.5 ; 106.0
26.5 ) (see fig
4) or for supination/pronation ROM at 9 (153.9
23.9 ;
151.8
33.0 ), 12 (170.9
9.3 ; 156.7
20.8 ), or 26 weeks
(169.4
11.9 ; 162.8
18.1 ), respectively (fig 5). For flexion/
extension in the training group, week 9 (78.1
20.7 ) was
significantly different than weeks 12 (100.5
19.2 ) and 26
(104.4
15.5 ) (see fig 4) (Bonferroni adjusted for multiple
comparisons, P<.05). For the control group, weeks 9
(81.7
25.7 ) and 12 (80.2
28.7 ) were significantly different
than week 26 (106.0
26.5 ) (Bonferroni adjusted for multiple
Fig 3 Nonfractured limb handgrip strength (mean
SE). There was
a significant GroupTime interaction, and a significant TimeArm
interaction for strength (P<.05). *Significantly different than week 1.
**Significantly different than week 9 (adjusted for multiple compar-
isons, P<.05). #Significant difference between groups (unadjusted).
##
Significant difference between groups (Bonferroni-adjusted, P<.05/
3Z.017).
C.R.A. Magnus et al
comparisons, P<.05). There were no other significant differences
for supination/pronation ROM (see fig 5). Raw ROM data can be
viewed in table 4.
Patient Rated Wrist Evaluation
Little’s MCAR test was used to determine that missing data for the
PRWE questionnaire (c220Z22.87, PZ.295) were MCAR. There
were a total of 28 of 156 missing data points for the PRWE
questionnaire (see table 2 for a description of missing data).
Therefore, group series means were used to replace all missing
data.20 There were no significant differences between groups at
week 1 for the PRWE questionnaire. There was a time main effect
pooled across group (F3,37Z48.93, P<.001, partial h2Z.569);
however, there were no significant differences between the
training and control groups at 9 (54.2
39.0; 65.2
28.9), 12
(36.4
37.2; 46.2
35.3), or 26 weeks (23.6
25.6; 19.4
16.5),
respectively (table 5). No other significant differences were found.
Discussion
The main finding of this study was that strength training the
nonfractured arm after a distal radius fracture improved strength
and ROM at 12 weeks postfracture in the fractured arm. These
results demonstrate that cross-education strength training may be
beneficial to older women in recovery after a distal radial fracture.
To our knowledge, this is the first randomized controlled trial to
demonstrate the efficacy of cross-education of strength in a clin-
ical setting involving limb fractures. These results have potential
for changing current rehabilitation protocols in the early recovery
stages after distal radius fracture, following further investigation.
The study showed that the training group had a quicker
recovery in both strength and ROM on the fractured limb than the
control group. The control group had a 4.4% increase in strength
from week 9 to week 12, whereas the training group had a 38.4%
increase in strength from week 9 to week 12 (see fig 2). The
difference between the training and control groups is approxi-
mately 34%; therefore, the effect of the training could be per-
ceived as a 34% difference in recovery of strength between 9 and
12 weeks. The average transfer of strength in healthy, non-
immobilization cross-education studies is 52% of the strength in
the trained limb.2 Carroll et al2 suggested that if cross-education
were applied in a rehabilitation setting, the effect may be small
and not show significant improvements in activities of daily living.
Although a strength training effect was found, this did not transfer
to significant differences between the groups for the PRWE
questionnaire score. Results for flexion/extension ROM showed
that the control group had a slight decline in range of 1.8% from
9 to 12 weeks postfracture and the training group showed an
increase in range of 28.7% (see fig 4). This indicates that the
control group had no improvement in flexion/extension ROM from
9 to 12 weeks whereas the training group was almost fully
recovered by 12 weeks postfracture.
The decline in strength after wrist fracture is comparable with
other literature investigating grip strength after wrist fractures.21
Földhazy et al21 found that grip strength after 12 weeks of distal
radius fracture was approximately 65% of the nonfractured limb,
whereas our results showed that grip strength was 62% of the
nonfractured limb for the training group and 45% for the control
group. At 26 weeks postfracture, Földhazy21 showed that strength
was 76% of the nonfractured limb, and the training and control
www.archives-pmr.org
Cross-education after wrist fractures
Table 3 Raw handgrip strength data (kg)
Nonfractured Arm
Group Week 1 Week 9 Week 12
y y
Train (nZ18) 28.1
6.0 30.8
6.9* 30.7
6.5
Control (nZ21) 26.4
5.0 26.9
4.4 24.9
4.4
NOTE. Values are mean
SD. Only significant differences between groups are displayed in the table. Week 1 strength for the fractured arm was not
measured; therefore, values are borrowed from the nonfractured arm at week 1.
* Significant difference between groups (unadjusted, P<.05).
y
Significant difference between groups (Bonferroni-adjusted, P<.05/3Z.017).
groups values were 82% and 74%, respectively. Although the
studies have similar changes in strength, it is difficult to directly
compare the 2 studies considering Földhazy et al’s21 study
included participants with only nonsurgical fractures and the
present study included participants with both surgical and
nonsurgical fractures.
Significant differences in strength between the training and
control groups were evident at 12 weeks postfracture, roughly 6
weeks after the immobilization period ended. Why significant
differences were shown at 12 weeks and not 9 weeks is unknown.
It is possible that at 9 weeks postfracture, participants were still
very sore, and potentially so sore that the pain during a handgrip
contraction prevented comfortable and maximal strength testing.
At 12 weeks postfracture, the participants may have been much
more comfortable completing a maximal handgrip test, which may
have accounted for the significant difference at 12 weeks.
Cross-education literature from noninjury settings has shown
the effect occurs in training programs varying from 3 to 8 weeks in
duration1,22-24; therefore, it may be expected that significant
effects would be shown before 12 weeks postfracture (ie, at 9
weeks). Perhaps if participants were tested weekly for strength
between weeks 9 and 12, significant differences may have been
shown before 12 weeks postfracture. The significant effect at 12
weeks may be due to the time course of the injury itself, which
could have altered the neurologic transfer in strength that would
normally be shown in cross-education training without an injury.
More research is needed to further investigate the time course and
mechanisms behind these effects in a clinical setting.
Importantly, the cross-education home-based strength training
program effectively increased strength in the nonfractured hand
of the training group from week 1 to weeks 9, 12, and 26. The
Fig 4 Flexion/extension ROM for fractured hand only (mean
SE).
*Significantly different than week 9. **Significantly different than
weeks 9 and 12 (adjusted for multiple comparisons, P<.05). #Significant
difference between groups (Bonferroni-adjusted, P<.05/3Z.017).
www.archives-pmr.org
1253
Fractured Arm
Week 26 Week 1 Week 9 Week 12 Week 26
31.0
6.9* 28.1
6.0 12.5
8.2 17.3
7.4 23.0
7.6
27.0
5.0 26.4
5.0 11.3
6.9 11.8
5.8 19.6
5.5
training group showed an average increase of 9.6% in strength
(average from weeks 9, 12, and 26). This increase in strength is
comparable to that in Farthing et al’s6 study, which showed
a 10.7% increase in handgrip strength in the trained limb using
a supervised laboratory-based training program. The present study
is also novel because of the unsupervised, at-home strength
training program. Cross-education strength training studies have
typically been completed in supervised controlled laboratory
environments. Therefore, a 9.6% increase in strength from an
at-home grip strength program demonstrates that this type of
training is quite feasible in a clinical environment where super-
vised training is more difficult or impossible.
Cross-education is known to produce contralateral limb
strength adaptations following unilateral training; however, there
is no apparent evidence to suggest that cross-education strength
training can produce increases in ROM of an opposite limb. The
present study showed that the training group had significantly
improved wrist flexion/extension ROM at 12 weeks postfracture
compared with the control group. Evidence is limited in exam-
ining the effects of cross-education on ROM. Although Nelson
et al25 investigated the effect of unilateral stretching on strength in
the opposite limb, to our knowledge there are no studies that have
investigated the effect of unilateral strength training on ROM in
the opposite limb. More research is needed to determine whether
cross-education strength training can produce increased ROM in
the opposite limb.
There were no significant differences between groups for
supination/pronation ROM. This is likely due to the quick
recovery of supination/pronation for both groups, and a ceiling
effect for the total ROM. The training group already had a range
of 153.9 at 9 weeks postfracture and improved to 169.4 at
26 weeks postfracture. Similarly, the control group had a range of
151.8 at 9 weeks and 162.8 at 26 weeks postfracture.
Fig 5 Supination/pronation ROM for fractured hand only (mean
SE).
No significant differences.
1254
Table 4 Raw ROM data (degrees)
Flexion/Extension
Group Week 9 Week 12
Train (nZ18) 78.0
20.7 100.5
19.2*
Control (nZ21) 81.7
25.7 80.2
28.7
NOTE. Values are mean
SD. Only significant differences between groups are displayed in the table.
* Significant difference between groups (Bonferroni-adjusted, P<.05/3Z.017).
There were no significant differences between the training and
control groups for the PRWE questionnaire scores (see table 5).
Significant differences were likely not detected because of the
high variability of the measure. Based on current data, there is no
evidence to suggest that the cross-education intervention had
a significant impact on self-reported pain and function during
activities of daily living of the fractured limb, despite evidence for
improved strength and ROM. The PRWE questionnaire is the most
commonly used instrument for evaluating outcome in patients
with distal radius fractures26; however, it may not be sensitive
enough to detect small changes between groups. A more direct
measure of function may be necessary to accurately assess the
effects on recovery. The minimal clinically important difference,
defined as the minimum change in a score that indicates a mean-
ingful difference for the patient,27 may be more relevant for
evaluating changes in function. The minimal clinically important
difference for the PRWE questionnaire is a change of 24 points
(16%).28 The training group had a change of 30.6 points from 9 to
26 weeks, and the control group had a change of 45.8 points from
9 to 26 weeks, indicating that both groups had a clinical
improvement in self-reported function by 26 weeks.
Study limitations
One limitation of the present study is we cannot be absolutely
certain that the training group completed the exercises as pre-
scribed. The training program was taught to the intervention group
at the initial visit to the clinic and was completed unsupervised at
the participants’ homes. Training logs were given out to track
adherence and monitor progression. This self-monitored at-home
program was chosen because it could be implemented in such
a manner that would decrease participant travel burden and
decrease clinical visits. Despite the unsupervised nature of the
program and some uncertainty regarding adherence (ie, self-report
training logs), it was effective for increasing strength. Arguably
the strength increase for the training group participants was partly
due to using their nonfractured arm more than normal for daily
activities. However, the control group participants would have also
used their nonfractured arm the same amount as the training group
participants, and the control group participants showed a signifi-
cant decrease in strength from week 9 to week 12. This is
Table 5 PRWE questionnaire scores
Group Week 1 Week 9 Week 12 Week 26
Train (nZ18) 3.1
4.2 54.2
39.0 36.4
37.2 23.6
25.6
Control (nZ21) 6.4
6.0 65.2
28.9 46.2
35.3 19.4
16.5
NOTE. Values are mean
SD. There are no significant differences.
A high score is more symptomatic. Week 1 is retrospective prefracture
score. Weeks 9, 12, and 26 are postfracture scores for corresponding
time point.
C.R.A. Magnus et al
Supination/Pronation
Week 26 Week 9 Week 12 Week 26
104.4
15.5 153.9
23.9 170.9
9.3 169.4
11.9
106.0
26.5 151.8
33.0 156.7
20.8 162.8
18.1
important because it points toward the possibility of a global
strength decline in both arms after a unilateral distal radius frac-
ture when attempting to implement the current clinical practice.
This may suggest that patients may not be adhering to the standard
rehabilitation program. Compliance to the standard rehabilitation
program was not recorded and is not part of the standard clinical
practice; therefore, there was no indication of adherence to the
standard rehabilitation program for either group, which is an
additional limitation to the study.
Another potential limitation is that we did not account for the
effect of physical therapy or surgery in the analysis. Conveniently,
of the participants included in the final data analysis, the number
who attended physical therapy was very similar between groups
(Train 6; Control 7); therefore, this even distribution should not
have affected the results if the physical therapy treatments re-
ceived in each group were similar. The number of surgeries for
participants included in the final data analysis was 6 in the training
group and 2 in the control group. The small sample size limited
the ability to separate out the number of participants who received
surgery in the data analysis, which is a limitation of the study. We
were unable to do any subgroup analyses because of the small
sample size. Further analysis could have consisted of dividing by
age, and surgical/nonsurgical fractures. Future studies may
investigate these factors and may also look at a longer term of
follow-up to determine the effects on overall function.
Conclusions
This intervention study found that strength training the non-
fractured limb was associated with significantly improved strength
and ROM in the fractured limb via cross-education in the early
stages of rehabilitation. This study marks a crucial advancement in
the field because, to the best of our knowledge, it is the first
randomized controlled trial to demonstrate that training of
a homologous noninjured limb may benefit an immobilized,
injured limb. These findings may have potential implications for
altering the current clinical rehabilitation protocols after wrist
fractures. More investigations are warranted before changes to
clinical practice can be recommended. Future research may
investigate cross-education effects in other types of injuries, the
effects over longer follow-up, and the mechanisms behind the
effect. This work adds to the recent translational study by Dragert
and Zehr29 where strength training of the less-affected limb was
effective for inducing bilateral neuromuscular plasticity in chronic
stroke patients.
Suppliers
a. G Power 3.1, Department of Experimental Psychology, Hein-
rich-Heine-University, 40225 Düsseldorf, Germany. Software
available as a free download.
www.archives-pmr.org
Cross-education after wrist fractures
b. DIGI-FLEX handgrip trainers; CANDO, PO Box 1174, White
Plains, NY 10602.
c. ZoN Fitness Resistance Hand Grips, 1950 Stanley St, North-
brook, IL 60065.
d. Baseline Hydraulic Hand Dynamometer, PO Box 1500, White
Plains, NY 10602.
e. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
Keywords
Colles’ fracture; Forearm; Hand strength; Rehabilitation; Strength
training; Wrist joint
Corresponding author
Jonathan P. Farthing, PhD, College of Kinesiology, University of
Saskatchewan, 87 Campus Dr, Saskatoon, SK, Canada S7N 5B2.
E-mail address: jon.farthing@usask.ca.
Acknowledgments
We thank the Royal University Hospital staff at the Outpatients
Orthopaedic clinic for their contributions, and Karla Koehn,
administrative assistant to Dr Johnston. We also thank the
Saskatoon Health Region for the operational approval to conduct
the study. We thank our research assistants, Nolan Carrier, BSc,
Shane Schwanbeck, MSc, Mike Smith, MSc, Dominika Pindus,
MSc, Kellie Boychuk, BSc, and Samantha Mitchell, MSc, for
their work.
References
1. Farthing JP, Chilibeck PD, Binsted G. Cross-education of arm
muscular strength is unidirectional in right-handed individuals. Med
Sci Sports Exerc 2005;37:1594-600.
2. Carroll TJ, Herbert RD, Munn J, Lee M, Gandevia SC. Contralateral
effects of strength training: evidence and possible mechanisms. J Appl
Physiol 2006;101:1514-22.
3. Lagerquist O, Zehr EP, Docherty D. Increased spinal reflex excitability
is not associated with neural plasticity underlying the cross-education
effect. J Appl Physiol 2006;100:83-90.
4. Farthing JP, Borowsky R, Chilibeck PD, Binsted G, Sarty GE. Neuro-
physiological adaptions associated with cross-education of strength.
Brain Topogr 2007;20:77-88.
5. Fimland MS, Helegerud J, Solstad GM, Iversen VM, Leivseth G,
Hoff J. Neural adaptations underlying cross-education after unilateral
strength training. Eur J Appl Physiol 2009;107:723-30.
6. Farthing JP, Krentz JR, Magnus CRA, et al. Changes in functional
magnetic resonance imaging cortical activation with cross education to
an immobilized limb. Med Sci Sports Exerc 2011;43:1394-405.
7. Stromberg BV. Contralateral therapy in upper extremity rehabilitation.
Am J Phys Med 1986;65:135-43.
8. Farthing JP, Krentz JR, Magnus CRA. Strength training the free limb
attenuates strength loss during unilateral limb immobilization. J Appl
Physiol 2009;106:830-6.
9. Magnus CRA, Barss TS, Lanovaz JL, Farthing JP. The effects of
cross-education on the muscle after a period of unilateral limb
www.archives-pmr.org
1255
immobilization using a shoulder sling and swathe. J Appl Physiol
2010;109:1887-94.
10. Larsen CF, Lauritsen J. Epidemiology of acute wrist trauma. Int J
Epidemiol 1993;22:911-6.
11. Handoll HHG, Madhok R, Howe TE. Rehabilitation for distal radial
fractures in adults. Cochrane Database of Systematic Reviews
2006;(3):Art. No. CD003324. CD003324.pub2.
12. Brogren E, Hofer M, Petranek M, Dahlin LB, Atroshi I. Fractures
of the distal radius in women aged 50 to 75 years: natural course
of patient-reported outcome, wrist motion and grip strength
between 1 year and 2-4 years after fracture. J Hand Surg (Eur)
2011;36:568-76.
13. Trumble TE, Schmitt SR, Vedder NB. Factors affecting functional
outcome of displaced intra-articular distal radius fractures. J Hand
Surg 1994;19A:325-40.
14. Borson S, Scanlan J, Brush M, Vitaliano P, Dokmak A. The mini-cog:
a cognitive “vital signs” measure for dementia screening in multi-
lingual elderly. Int J Geriatr Psychiatry 2000;15:1021-7.
15. Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses
using G*Power 3.1: tests for correlation and regression analyses.
Behav Res Methods 2009:1149-60.
16. Bryden MP. Measuring handedness with questionnaires. Neuro-
psychologia 1977;15:617-24.
17. MacDermid JC, Turgeon T, Richards RS, Beadle M, Roth JH. Patient
rating of wrist pain and disability: a reliable and valid measurement
tool. J Orthop Trauma 1998;12:577-86.
18. Abraha I, Montedori A, Romagnoli C. Modified intention to treat:
frequency, definition and implication for clinical trials [abstract]. Sao
Paulo, Brazil: XV Cochrane Colloquium; 2007. p 86-7.
19. Abraha I, Montedori A. Modified intention to treat reporting in
randomized controlled trials: systematic review. BMJ 2010;340:
c2697.
20. Tabachnick BG, Fidell LS. Using multivariate statistics. 5th ed.
Boston: Pearson/Allyn and Bacon; 2007.
21. Földhazy Z, Törnkvist H, Elmstedt E, Andersson G, Hagsten B,
Ahrengart L. Long-term outcome of nonsurgically treated distal radius
fractures. J Hand Surg 2007;32A:1374-84.
22. Kannus P, Alosa D, Cook L, et al. Effect of one-legged exercise on the
strength, power and endurance of the contralateral leg: a randomized,
controlled study using isometric and concentric isokinetic training.
Eur J Appl Physiol 1992;64:117-26.
23. Farthing JP, Chilibeck PD. The effects of eccentric training at different
velocities on cross-education. Eur J Appl Physiol 2003;89:570-7.
24. Munn J, Herbert RD, Hancock MJ, Gandevia SC. Training with
unilateral resistance exercise increases contralateral strength. J Appl
Physiol 2005;99:1880-4.
25. Nelson AG, Kokkonen J, Winchester JB, et al. A 10-week stretching
program increases strength in the contralateral muscle. J Strength
Cond Res 2012;26:832-6.
26. Changulani M, Okonkwo U, Keswani T, Kalairajah Y. Outcome
evaluation measures for wrist and hand e which one to choose? Int
Orthop 2008;32:1-6.
27. Smith MV, Calfee RP, Baumgarten KM, Brophy RH, Wright RW.
Upper extremity-specific measures of disability and outcomes in
orthopaedic surgery. J Bone Joint Surg Am 2012;94:277-85.
28. Schmitt JS, Di Fabio RP. Reliable change and minimum important
difference (MID) proportions facilitated group responsiveness
comparisons using individual threshold criteria. J Clin Epidemiol
2004;57:1008-18.
29. Dragert K, Zehr EP. High-intensity unilateral dorsiflexor resistance
training results in bilateral neuromuscular plasticity after stroke. Exp
Brain Res 2013;225:93-104.