Received: 16 January 2019 Revised: 19 September 2019 Accepted: 21 September 2019

DOI: 10.1002/mus.26727

CLINICAL RESEARCH ARTICLE

Ankle bracing practices in ambulatory, corticosteroid-naive

boys with Duchenne muscular dystrophy

Victoria Kern MPT1 | Matthew Wicklund MD2 | Anne Haulman MPH3 |

Michael P. McDermott PhD4,5 | William B. Martens BA5 | Robert C. Griggs MD5 |
Ashutosh Kumar MD6 The Muscle Study Group and TREAT-NMD
1
Rehabilitative Medicine, Penn State Milton
S. Hershey Medical Center, Hershey, Abstract
Pennsylvania, Introduction: Loss of ambulation in Duchenne muscular dystrophy presages scoliosis,
2
Department of Neurology, University
respiratory failure, and death. Strategies to maintain ankle range of motion are
of Colorado School of Medicine, Aurora,
Colorado employed, but little evidence exists to support these approaches and limited informa-
3
Department of Neurology, Penn State Milton tion is available concerning current practice.
S. Hershey Medical Center, Hershey,
Pennsylvania, Methods: In this study we assessed baseline bracing data from 187 boys participating
4
Department of Biostatistics and in a multicenter, international clinical trial.
Computational Biology, University of
Results: Ankle-foot orthoses (AFOs) were recommended for 54% of the boys, with
Rochester Medical Center, Rochester,
New York, nighttime static AFOs and nighttime dynamic AFOs utilized in 94% and 6% of these
5
Department of Neurology, University of boys, respectively. Daytime static AFOs were recommended for 3 boys. Compliance
Rochester Medical Center, Rochester,
New York,
with bracing recommendations was 54% for nighttime static braces and 67% for
6
Division of Pediatric Neurology, Penn State nighttime dynamic braces.
Milton S. Hershey Medical Center, Hershey, Discussion: The basis for the variation in recommended AFO use is unknown and
Pennsylvania,
requires further study. Long-term follow-up of boys may permit assessment of the
Correspondence effects of AFO use.
Ashutosh Kumar, Division of Pediatric
Neurology, Penn State Milton S. Hershey
Medical Center, 90 Hope Drive, Hershey, KEYWORDS
PA 17036. bracing, corticosteroids, Duchenne muscular dystrophy, dynamic ankle-foot orthoses,
Email: ashukmc@gmail.com pediatrics, static ankle-foot orthoses

Funding information
National Institute of Neurological Disorders
and Stroke of the National Institutes of Health,
Grant/Award Numbers: U01NS061795,
U01NS061799; PTC Therapeutics; Telethon
Italy; Muscular Dystrophy Association; Parent
Project Muscular Dystrophy

1 | I N T RO D UC T I O N fibrosis and tissue damage. As strength declines, boys routinely use

postural compensations to maintain the ability to stand and walk,
In Duchenne muscular dystrophy (DMD), loss of ankle range of including widening the base of support, avoidance of hip-extension
motion (ROM), specifically dorsiflexion, is well documented.1,2 As excursion, retraction of the shoulder girdle, and accentuation of lum-
weakness progresses, extensibility of muscle tissue is lost due to bar lordosis, all of which shift the center of mass forward and reduce
the strain on weakened muscle groups.3 This forward weight shift
Abbreviations: ADF, ankle dorsiflexion; AFO, ankle-foot orthoses; CE, clinical evaluator; CRF, reinforces toe walking, further shortening the gastrocnemius/soleus
case report form; DMD, Duchenne muscular dystrophy; FOR-DMD trial, Finding the Optimal
complex with reduced excursion into ankle dorsiflexion (ADF) in the
Steroid Regimen in Duchenne Muscular Dystrophy trial; FVC, forced vital capacity; NSAA,
North Star Ambulatory Assessment; PT, physical therapist; ROM, range of motion. gait cycle. These modifications result in a loss of efficient, safe,

Muscle & Nerve. 2019;1–6. wileyonlinelibrary.com/journal/mus © 2019 Wiley Periodicals, Inc. 1

2 KERN ET AL.

propulsive gait. Infrequent excursion into dorsiflexion in the gait cycle, The clinical evaluators (CEs) at the different sites underwent
coupled with the cellular changes in the muscles themselves, cause hands-on training with the lead physical therapists (PTs) of the FOR-
muscle stiffness and gradual loss of ankle ROM. DMD trial. Reliability of the assessments was verified via review by
Maintenance of functional ankle ROM remains important for the lead PTs of videotapes of the assessments of the first two patients
standing, ambulation, and functional mobility. Early recognition of the enrolled at each site. Feedback for consistency and accuracy was pro-
importance of ankle ROM is assumed to be a key to a functional pro- vided back to the CE by phone call or meeting.
gram and is considered best practice in physical therapy.1,4,5 To In this study, because interest was in the association between gas-
achieve prolonged positional stretching of the Achilles tendons, cal- trocnemius shortening and ankle dorsiflexion, measurement of ankle
ves, and ankle soft tissues, a variety of positional stretching devices dorsiflexion was undertaken in the supine position with the knee in full
may be applied during the day or overnight. Options include bi-valved extension. The calcaneus was held in neutral alignment while pressure
casts, static ankle-foot orthoses (AFOs) (fixed at one angle, typically was applied over the midsection of the foot to dorsiflex the ankle as
 much as possible, preventing inversion. The axis of the goniometer was
neutral or 0 ), and dynamic AFOs, which include a hinged ankle and
one of several different types mechanisms to produce low load stress over the lateral malleolus, with the stationary arm aligned with the fibu-
into dorsiflexion. These braces are designed to be worn for long lar head along the lateral aspect of the lower leg. The “moving arm” was
periods, typically overnight, and are not for use while walking as they held parallel to the lateral aspect of the fifth metatarsal, and aligned
provide continual low load-stretching stress, which would interfere with the posterior third of the foot to ensure the gastrocnemius range
with the ankle actions during walking. was being monitored and not that of the plantar structures of the foot.
Currently, there is no reliable evidence indicating which of many The range of dorsiflexion past plantargrade was noted as +X and range
different types of interventions for ankle ROM management is most lacking from plantargrade as −X . Normal range is 20 dorsiflexion to
effective or best tolerated by patients and families, nor whether pre- 50 plantarflexion.9 The CE used an appropriately sized goniometer to
serving ADF ROM is associated with or supports prolonged ambula- measure ankle ROM to the nearest degree. Passive range was measured
tion. No evidence yet exists to indicate at what time, age, or joint
flexibility level intervention should be initiated. The Finding the Opti-
TABLE 1 Characteristics of enrolled participants
mal Steroid Regimen in Duchenne Muscular Dystrophy (FOR-DMD)
Mean (SD)
trial gathered data on a large group of boys with DMD that included
Category or number (%)
information on ankle passive range of dorsiflexion motion, the type of
Age (years) 5.9 (1.0)
AFOs recommended, and the recommended and actual wearing times.
Degrees of ROM of ankle dorsiflexion
We examined baseline data from the FOR-DMD trial to describe cur-
rent international practices of prescribing different AFOs in a cohort Left 5.6 (9.1)

of DMD patients from five countries. We expected to find differences Right 5.7 (9.4)

between countries with regard to AFO prescription, given different Bracing recommendation
health-care systems, and expected to find similar compliance across None 86 (46.0%)
different styles of bracing given that the recommendations were being Static daytime 2 (1.1%)
made by perceived experts in DMD care. Static nighttime 92 (49.2%)
Static daytime and nighttime 1 (0.5%)

2 | METHODS Static and dynamic nighttime 2 (1.1%)

Dynamic nighttime 4 (2.1%)

The FOR-DMD trial is an international, multicenter, randomized, Bracing use

double-blind, parallel-group trial comparing 3-year outcomes for the Static daytime (n = 4*)
three most commonly used corticosteroid regimens. The trial is being Days/week 6.0 (1.2)
performed at 32 sites in five countries (United States, Canada, United Hours/week 17.8 (21.5)
Kingdom, Italy, and Germany) in genetically confirmed DMD boys from Static nighttime
age 4 to 7 years 11 months (at randomization), although two boys age Days/week (n = 48†) 6.4 (1.4)
8.1 years were ultimately enrolled. Boys were required to be steroid- Hours/week (n = 46†) 42.4 (28.0)
naive and able to walk at least 6 minutes to enter the study. Details of
Dynamic nighttime (n = 4)
this study have been published.6 Physical therapy assessments include
Days/week 7.0 (0.0)
measurements of passive ROM, muscle extensibility, posture, strength,
Hours/week 45.3 (17.2)
and function using the North Star Ambulatory Assessment (NSAA) and
timed function studies.7,8 The study was approved by the local institu- Abbreviations: ROM, range of motion; SD, standard deviation.
*Daytime static ankle-foot orthoses were used by 5 boys; data on
tional review board/ethics committee at each participating institution.
frequency/hours of use were missing for 1 boy.
Written informed consent was obtained from the parents/legal guard- †
Nighttime static AFOs were used by 51 boys; data on frequency of use
ians of the study participants. were missing for 3 boys and data on hours of use were missing for 5 boys.
KERN ET AL. 3

so that the subject needed to relax, whereas the CE ranged the joint. this study focused on the brace interventions, data on other
Measurements were taken with a “moderate” degree of stretch applied stretching strategies are not included in the analyses.
and with the child trying to dorsiflex the foot at the same time. The In addition to the specific style of brace recommended by the CE,
child was asked to count to ten while measuring the range. the prescribed frequency and duration and the reported compliance
Data from the trial regarding the use of brace interventions for (documented as hours used) were collected. Compliance was deter-
ankle range of motion specifically were entered by the CE at each site mined through documentation of the parental reports on actual use.
under the following headings: Static AFO; Contracture Correction Pairwise comparisons among groups defined by type of AFO used
Device; Serial Casting; and Stretching Exercises. With the input of were compared with respect to age and ADF ROM, using Wilcoxon
several lead PTs in each participating country, descriptions of the rank sum tests. All statistical analyses were performed using SAS ver-
braces they documented under each heading were gathered. “Static sion 9.4 (SAS Institute, Cary, North Carolina). A significance level of
AFO” was universally applied to mean a rigid plastic AFO with non- 5% was used for hypothesis testing.
elastic straps across the proximal tibia and ankle mortise, possibly also
across the forefoot. These are generally custom molded to each 3 | RESULTS
patient and are set at neutral ankle dorsiflexion (0 ) with a neutral
hindfoot (0 of calcaneal varus/valgus) and allow no motion of the There were 196 boys randomized in the FOR-DMD trial. Baseline data
ankle joint. Under “Contracture Correction Device,” most PTs docu- on bracing were available in 187 (95.4%) of these boys (Table 1). Fifty-
mented any type of jointed or hinged brace that provided some low four percent of boys received a recommendation for bracing of any
load stress into ADF by either nonelastic straps or a mechanical kind (ie, 101 of 187), mostly for nighttime static AFOs (N = 95). The fre-
spring-loaded hinge. These are generally designed to provide and quency of recommendation of nighttime use of a static AFO varied
allow gradual motion into greater degrees of ADF over several hours across countries as follows: 46 of 78 (59%) in the United States; 11 of
of use as the soft tissue relaxes and extends in response to the load- 22 (50%) in Italy; 28 of 57 (49%) in the United Kingdom; 5 of 12 (42%)
ing stress. In Canada, a process of serial splinting was occasionally in Canada; and 5 of 18 (28%) in Germany. Fifty-one occurrences of
used whereby trained occupational therapists mold new static splints actual use of a static AFO at night were documented; therefore, 44 fam-
periodically in response to increasing range of motion. By report of ilies (46%) did not comply with this recommendation.
one of the Canadian PTs, this process was not documented on the Daytime use of static AFOs was rarely recommended at any site.
case report forms (CRFs) as it did not fall clearly into any category. As It was recommended for three boys (1.6%; 1 each in the United

F I G U R E 1 Distribution of age by use of

AFOs. Each dot indicates an individual subject.
Abbreviation: AFOs, ankle-foot orthoses
4 KERN ET AL.

F I G U R E 2 Distribution of ADF ROM, measured as degrees from plantargrade, by use of AFOs. Data are shown for the left (A) and right
(B) ankles. Abbreviations: ADF ROM, ankle dorsiflexion range of motion; AFOs, ankle-foot orthoses

States, Canada, and Italy), whereas it was documented as used by five based on the published standards requires each PT to determine the
boys (2.7%; 4 in the United States and 1 in Canada). “best” options.5
The use of “contracture correction devices” or dynamic AFOs, was Bracing was used in just over 50% of the patients for prevention
recommended for two boys (10%) in Germany and four boys (5%) in of ankle plantarflexion contractures in the five countries participating
the United States. Neither of the two German boys complied with the in the FOR-DMD trial. As all participants entered the study with the
recommendation, whereas all four of boys in the United States ability to walk functional distances (as measured by the 6-minute walk
reported compliance. The remaining countries did not recommend nor test), the lack of use of static AFOs during the day seems to confirm
use dynamic AFOs. No one documented the use of serial casting. that permitting ambulatory compensations is practiced in the five
No statistically significant differences were noted between type countries in FOR-DMD. Wearing a static AFO during the day would
of AFO employed and age (Figure 1) or ADF ROM (Figure 2). The limit the participant's ability to walk by preventing functional compen-
small sample sizes for all types of AFO used, other than nighttime use sations for weakness or would require seated time during the day for
of static AFOs, precluded meaningful comparisons. We noted that prolonged stretching. Enforced sitting in an otherwise ambulatory boy
three of the four boys using dynamic AFOs were under the age of with DMD is generally believed to be counterproductive to prolonga-
5.5 years, whereas the ages of boys using static AFOs tended to be tion of the ambulatory stage of the disease.10,11 This is the likely rea-
somewhat higher (Figure 1). A smaller ADF ROM tended to be associ- son serial casting appears to no longer be used in the five countries in
ated with use of the dynamic style of brace (at or just below 0 or the study, as even fairly lightweight fiberglass casts limit ambulation in
neutral), whereas static braces tended to be more often used with ini- boys with progressive weakness. Serial casting would increase resis-
tial passive dorsiflexion greater than 0 (Figure 2), although compari- tance to leg motions during walking, thereby overstressing fragile
sons are limited by the small sample sizes. There was little relationship muscle tissue and possibly accelerating disease progression.12 The
between ankle ROM and age, although there were some boys who fact that more than half of the boys received a prescription for some
exhibited more impairment after the age of 7 years (see Figure S1
kind of bracing indicates that many PTs are applying strategies of pro-
online).
longed positional stretching in their practice. However, there was no
uniform practice with regard to bracing prescriptions. The lack of a
4 | DISCUSSION mechanism to verify that the brace prescription came only from the
CE may also have impact on the nonuniform prescription patterns
The standards of care for DMD include broad recommendations for a identified. This variability may be linked to several issues with regard
spectrum of therapeutic interventions for ROM maintenance, includ- to treating boys with DMD, such as lack of availability of certain types
ing stretching, bracing, and use of equipment to achieve supported of braces, lack of training, reliance on other strategies to address ADF
standing position. Few details have been presented, and prescription ROM, and the different theoretical paradigms.
KERN ET AL. 5

Contrary to our hypothesis, the differences we observed are not impacted the walking outcomes, which will be clarified in future longi-
statistically significant, possibly due to the small cohorts in each brac- tudinal data from this study. Further studies looking at other
ing category. In this study we have provided a description of the cur- stretching interventions and bracing would shed additional light and
rent bracing practices as observed in the FOR-DMD study and, provide guidance on brace prescription and best practices.
therefore, we are unable to make strong recommendations at this PTs rely on their experience with braces to recommend styles.
time. We also cannot draw any conclusions concerning differences in When prescribing for the DMD population, they would be best served
compliance between the types of braces prescribed due to the small by having a discussion about the child's current sleep schedule and
number of boys who received a prescription for the dynamic type. any sleeping problems, as nighttime splinting must not interfere with
Although static AFOs at night were the most frequently rec- sleep hygiene. In addition, if the PT is unfamiliar with dynamic brace
ommended, there appeared to be a trend toward recommending styles and options available in their area, working with an experienced
dynamic AFOs when the initial ADF ROM was more reduced. How- certified orthotist may prove fruitful to their practice. Our study rep-
ever, the conclusions one can draw are limited by the small number of resents a starting point for subsequent work to standardize care and
boys using dynamic AFOs (n = 4). It is not clear why boys who pres- institute best practices for the care of boys with DMD. The longitudi-
ented with impaired ADF ROM did not receive a brace prescription. nal data from the FOR-DMD study may also provide information con-
There are multiple factors involved in prescribing, choosing, and cerning whether consistent, early use of ankle bracing impacts
obtaining different styles of brace. These factors include availability, subsequent disability, as defined by walking measures. As we await
cost, reimbursement, ease of use, family compliance, and therapist the conclusion of the FOR-DMD trial, we recommend that PTs pre-
and physician training and experience. scribe bracing that they believe would be best tolerated by the patient
Prolonged positional stretching is one strategy to address ankle and additional stretching interventions to maintain ankle range of
ROM management. Outpatient or school-based skilled physical ther- motion and support prolonged ambulation.
apy, home programs of manual or exercise-based stretching, and use
of standing frames (devices that maintain a patient in static standing
ACKNOWLEDG MENTS
for a set period of time) are also used. Reporting on these strategies
was included in the FOR-DMD assessments but was not included in The authors acknowledge the patient and family organizations, includ-
this analysis as there was no standardization of skilled physical ther- ing Action Duchenne, Muscular Dystrophy UK, Muscular Dystrophy
apy treatments and no details on the selection and style of standing Canada, and the Benni & Co/Parent Project, for their promotion of
frames. Therefore, meaningful comparisons were not be possible. our study. The FOR-DMD Steering Committee and the study site
Although these initial data have revealed variability in ROM inter- investigators are members of the Muscle Study Group. The content is
vention use in a large sample of young boys in the ambulatory stages solely the responsibility of the authors and does not necessarily repre-
of DMD, many more questions arisesuch as: (1) What are the main sent the official views of the National Institutes of Health.
factors underlying the choice to recommend or not recommend brac-
ing in the ambulant DMD boy? (2) Does bracing preserve ADF ROM?
CONFLICTS OF INTEREST
(3) Does preserving ADF ROM prolong ambulation? Evidence is still
lacking for the effectiveness of prolonged positional stretching of the The authors declare no potential conflicts of interest.
Achilles tendon and posterior calf structures in prolonging ambulation
for boys with DMD. Evidence is also lacking for determination of the
ET HICAL PU BLICAT ION ST AT E ME NT
most efficacious interventions.
The present findings represent a starting point from which further We confirm that we have read the Journal's position on issues
studies of preservation of ADF ROM can proceed. As longitudinal involved in ethical publication and affirm that this report is consistent
data become available from the FOR-DMD study, we will assess the with those guidelines.
impact of different interventions on long-term ankle flexibility and
walking. This information can guide best practice for PTs and physi-
OR CID
cians toward the most efficient strategies to support prolongation of
ambulation. FOR-DMD is an interventional study with respect to cor- Ashutosh Kumar https://orcid.org/0000-0002-4949-1395
ticosteroids, but is only observational with respect to ankle bracing,
thus limiting our ability to draw conclusions regarding causal relation-
ships involving the latter type of therapy. There may be variation due RE FE RE NCE S

to different prescribing practices in each country, whether the care of 1. Kinali M, Main M, Eliahoo J, et al. Predictive factors for the develop-
the DMD patient was overseen by different types of physicians, and ment of scoliosis in Duchenne muscular dystrophy. Eur J Paediatr Neu-
the level of independence of the PT in prescribing the intervention. rol. 2007;11:160-166.
2. Zanotelli E. Predicting the loss of ambulation in Duchenne muscular
Statistical methods for causal inference in observational studies may
dystrophy. Arq Neuropsiquiatr. 2014;72:1-2.
be useful in addressing these issues.13 In addition, the fact that the 3. Sutherland DH, Olsen R, Cooper L, et al. The pathomechanics of gait in
boys were on one of three possible steroid regimens may have Duchenne muscular dystrophy. Dev Med Child Neurol. 1981;23:3-22.
6 KERN ET AL.

4. Birnkrant D, Bushby K, Bann CM, et al. Diagnosis and management of 11. Case LE, Apkon SD, Eagle M, et al. Rehabilitation management of the
Duchenne muscular dystrophy, part 1: diagnosis, and neuromuscular, patient with Duchenne muscular dystrophy. Pediatrics. 2018;142
rehabilitation, endocrine, and gastrointestinal and nutritional manage- (suppl 2):S17-S33.
ment. Lancet Neurol. 2018;17:251-267. 12. Glanzman AM, Flickinger JM, Dholakia KH, Bonnemann CG,
5. Bushby K, Finkel R, Birnkrant DJ, et al. Diagnosis and management of Finkel RS. Serial casting for the management of ankle contracture
Duchenne muscular dystrophy, part 2: Implementation of multi- in Duchenne muscular dystrophy. Pediatr Phys Ther. 2011;23:
disciplinary care. Lancet Neurol. 2010;9:177-189. 275-279.
6. Guglieri M, Bushby K, McDermott MP, et al. Developing standardized 13. Imbens GW, Rubin DB. Causal Inference for Statistics, Social, and Bio-
corticosteroid treatment for Duchenne muscular dystrophy. Contemp medical Sciences: An Introduction. New York, NY: Cambridge Univer-
Clin Trials. 2017;58:34-39. sity Press; 2015.
7. Mayhew AG, Cano SJ, Scott E, et al. and the North Star Clinical Net-
work for Neuromuscular Disease. Detecting meaningful change using
the North Star Ambulatory Assessment in Duchenne muscular dystro- SUPPORTING INF ORMATION
phy. Dev Med Child Neurol. 2013;55:1046-1052.
8. McDonald CM, Henricson EK, Abresch RT, et al. The 6-minute walk
Additional supporting information may be found online in the
test and other endpoints in Duchenne muscular dystrophy: longitudi- Supporting Information section at the end of this article.
nal natural history observations over 48 weeks from a multicenter
study. Muscle Nerve. 2013;48:343-356.
9. Soucie JM, Wang C, Forsyth A, et al. Range of motion measurements: How to cite this article: Kern V, Wicklund M, Haulman A,
reference values and a database for comparison studies. Haemophilia. et al. Ankle bracing practices in ambulatory, corticosteroid-
2011;17:500-507.
naive boys with Duchenne muscular dystrophy. Muscle Nerve.
10. Jansen M, de Groot LJ, van Alfen N, Geurts A. Physical training in
boys with Duchenne muscular dystrophy: the protocol of the No Use 2019;1–6. https://doi.org/10.1002/mus.26727
is Disuse study. BMC Pediatr. 2010;10:55.