Medical Engineering and Physics 38 (2016) 275–279

Contents lists available at ScienceDirect

Medical Engineering and Physics

journal homepage: www.elsevier.com/locate/medengphy

The influence of a polymer damper on swing-through crutch

gait biomechanics
Megan K. MacGillivray a,b, Ranita H.K. Manocha c, Bonita Sawatzky b,d,∗
a
Rehabilitation Sciences, University of British Columbia, Vancouver, British Columbia, Canada
b
International Collaboration on Repair Discoveries, Vancouver, British Columbia, Canada
c
Department of Physical Medicine & Rehabilitation, Western University, London, Ontario, Canada
d
Department of Orthopaedics, University of British Columbia, , Vancouver, British Columbia, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Forearm crutch technology has evolved slowly compared to other assistive mobility devices, despite the
Received 15 July 2015 highly repetitive nature of forearm crutch gait and the high incidence of overuse injuries. Using 13 able-
Revised 20 December 2015
bodied volunteers between the ages of 19 and 27, we compared the ground reaction forces of a novel
Accepted 31 December 2015
crutch design featuring an elastomeric polymer situated below the handle to an identical design without
a damper system and to a commercially available generic rigid forearm crutch model. There were no dif-
Keywords: ferences in peak vertical force or impulse between crutches. The crutch with the damper system demon-
Forearm crutch strated a significantly smaller peak braking force and impulse compared to the generic forearm crutch
Biomechanics model. However, the crutch with the damper system demonstrated a significantly larger peak propulsive
Swing-through crutch gait
force and impulse compared to both crutch models. This finding indicates that a forearm crutch with
Ground reaction forces
a damper system may help to propel the crutch forward when walking on level surfaces, which could
Damper
impact forward momentum.
© 2016 Published by Elsevier Ltd on behalf of IPEM.

1. Introduction Crutch-assisted gait requires significantly more energy com-

Although the forearm crutch has existed for nearly 5000 years
[1], its design and technology has advanced slowly compared to
other forms of assistive mobility devices. Many assistive walk-
ing devices have historically been characterized as a ‘rigid sup-
port with an underarm crosspiece’ in reference to the early
axillary crutch [2]. Problems associated with this primitive crutch
design were identified as early as the 1900s and have included
compression neuropathy of the radial nerve, brachial plexus, and
axillary artery [3–5]. The forearm crutch, also known as the Lof-
strand or Canadian crutch, is thought to enhance control during
gait [6]. However, long-term use of the forearm crutch has been
associated with overuse injuries of the upper extremities such
as ulnar neuropathy at the wrist and elbow and ulnar fractures
[7–11].
Abbreviations: % BW, percent body weight; s, seconds; sd, standard deviation;
ηp 2 , partial eta squared.
∗
Corresponding author at: 818 West 10th Avenue, Vancouver, British Columbia
V5Z 1M9, Canada. Tel.: +1 604 675 8806.
E-mail address: bonitas@mail.ubc.ca, bonita.sawatzky@ubc.ca (B. Sawatzky).
pared to unassisted gait in able-bodied individuals and conse-
quently many crutch users may also experience fatigue [12–14].
Swing-through gait is the most energy-demanding form of crutch
gait and has been reported to require 78% more energy compared
to unassisted gait [13]. An additional study found that ambulating
using swing-through gait with axillary crutches expended twice as
much energy compared to unassisted gait [12]. Swing-through gait
consists of the individual simultaneously advancing both crutches
forward, swinging the torso past the crutches, and then bearing
weight on the foot or feet (Fig. 1) [15], thus resulting in repetitive
loading of the upper limbs and leg(s).
The reliance on crutches for mobility requires repetitive joint
loading of the upper extremities, therefore any reductions in verti-
cal ground reaction forces and impulses may help to reduce impact
on the body. Many forearm crutches are made of rigid materials
such as steel, aluminum, and hard plastics, which are not designed
to absorb impulse (i.e. dissipate kinetic energy). Although these
rigid materials are less expensive, they likely do little to reduce
joint impact. Shock absorption systems have been used in a vari-
ety of devices such as bicycles, prosthetics, wheelchair forks, and
footwear for many years, but have only been implemented into
commercially available crutch designs over the past few decades

http://dx.doi.org/10.1016/j.medengphy.2015.12.010
1350-4533/© 2016 Published by Elsevier Ltd on behalf of IPEM.
276 M.K. MacGillivray et al. / Medical Engineering and Physics 38 (2016) 275–279

Fig. 1. Schematic of swing-through crutch gait. The schematic of swing-through crutch gait identifies the main phases of this gait pattern including crutch stance and swing
phases.

[16–21]. The use of damping materials in crutch designs may help

to reduce overall impact and consequently repetitive overuse in-
juries.
Minimal research has evaluated the influence of a shock ab-
sorption system on swing-through forearm crutch gait kinetics (i.e.
peak forces and impulses). Kinetic characteristics of a conventional
aluminum axillary crutch were compared to a crutch model with
a helical compression spring (spring constant 22.4 kN/m, preload
10 N) at the distal end of the shaft in ten able-bodied individ-
uals [22]. The authors discovered that while the spring-loaded
crutch decreased vertical crutch impulse, it also slowed walk-
ing velocity and increased peak vertical crutch ground reaction
force. The authors hypothesized that their results might be a re-
sult of a ‘bottoming-out’ effect of the spring although the kine-
matic data did not confirm this [22]. An additional study found
that spring-loaded axillary crutches (spring constant 12.95 kN/m,
preload 220 N) increased peak forward velocity by 5% but did not
change preferred ambulation speed, compared to standard axillary
crutches [18]. An extensive spring study conducted by Shortell et
al. found a spring constant of 21.9 kN/m to be suitable for individu-
als between 53–90 kg based on interviewing participants regarding
their preference following trials with 30 different linear compres-
sion springs [19].
The focus of this research was to evaluate a new forearm
crutch, designed to decrease overall impact to improve joint health
(see Patent # 20110240077) [23]. Among other design features (e.g.
rotating footpad, carbon tube in lower shaft, and ergonomic de-
sign), this crutch model incorporated a centrally positioned elas- Fig. 2. Schematic of the CarbonDamp crutch model. The specific geometry of the
tomeric damper-system (Fig. 2). The system is inserted below the Carbon and CarbonDamp crutches is indicated in the diagram. The damper sys-
handle to (1) reduce the moment of inertia caused by the weight tem is located below the handle and the elastomeric polymer is situated within
of the damper system by keeping it close to the body (reducing the shaft of the crutch (white cylinder).
the lever arm); (2) reduce environmental contaminants from in-
terfering with the damper system; and (3) allow for easier height
vertical force and impulse compared to a similar crutch model
adjustment. The damper includes interchangeable polyurethane
without a damper system and a generic rigid forearm crutch.
elastomers (polymers) with varying spring constants for different
weight ranges which were selected based on the work by Shortell
et al. [19]. 2. Methods
It is postulated that the elastomeric damper would dissipate
kinetic energy upon crutch loading and thus reduce peak verti- 2.1. Participant recruitment
cal ground reaction forces and impulse. The purpose of this study
was to determine the influence of the damper system on swing- Thirteen healthy able-bodied individuals (age range 19–27
through crutch gait ground reaction forces. We hypothesized that years; mean height (sd) 174 (9.6) cm; mean body mass (sd)
the crutch with the elastomeric damper system would reduce peak 66.3 (11.6) kg) were recruited. Ten participants reported being
 M.K. MacGillivray et al. / Medical Engineering and Physics 38 (2016) 275–279
right-hand dominant, two were left-hand dominant, and one was
ambidextrous. Individuals were excluded from participating if they
had any musculoskeletal conditions, had previously used crutches
for mobility impairments, did not speak English, or exceeded a
weight of 113.4 kg (maximum recommended weight for the alu-
minum crutch model). The crutch model with the elastomeric
damper integrated different density elastomeric polymers to ac-
commodate four ranges of mass. The four different elastomeric
polymers have durometer readings of 70 A (up to 54.4 kg), 80 A
(54.4–90.7 kg), 83.5 A (90.7–127 kg), and 85 A (over 127 kg). Twelve
participants used the mid-range polymer (80 A) and one partici-
pant used the higher range polymer (83.5 A) to ensure the spring
constant fell within the recommended range based on mass [19].
2.2. Devices tested
Three crutch models were tested (Table 1). The “Generic” fore-
arm crutch was fabricated from aluminum and reflected standard
over-the-counter forearm crutch models. Two SideStixTM crutch
models were also examined. The upper shafts of the SideStixTM
crutch models were fabricated from 6061 aluminum alloy and the
lower shafts were composed of carbon fiber to minimize weight.
The handles were orientated at 100° from the upper shaft and 100°
from the lower shaft, in contrast to the typical 90° used in ba-
sic models (Fig. 2) [23]. A commercially available ergonomic hand-
grip was also used (Ergon, GP1, Koblenz, Germany). One SideStixTM
model (“CarbonDamp”) contained a damper. The foot of this crutch
contained a ball-and-socket rotating joint with a Vibram® rubber
footpad (Vibram S.P.A, Varese, Italy), allowing for full contact with
the ground during crutch stance phase to reduce slippage and ac-
commodate normal shoulder rotation [23]. The other SideStixTM
crutch did not contain a damper system (“Carbon”).
2.3. Data collection
Participants were fitted to each of the three crutch models de-
scribed in Table 1 based on standard guidelines [24,25]. Partici-
pants were given instructions on how to perform swing-through
gait and were provided with approximately 15 min to become ac-
quainted with this style of gait while using each of the three crutch
models. Participants were instructed to practice walking across the
6-m testing space with the two disguised force plates flush with
the floor within the walkway. Each individual’s starting point was
adjusted so that they landed on the force plates with the crutch
footpads only.
Force plates (Bertec Corporation, Columbus, Ohio, USA), which
collected at a sampling frequency of 1000 Hz, were used to mea-
sure ground reaction forces under the left and right crutch foot-
pads during crutch stance phase. For each trial, participants walked
approximately 6 m at a consistent self-selected speed. Data collec-
tion continued until there were 5 successful trials for each of the
three crutch conditions. Trials were considered successful only if
the left and right crutches contacted their respective force plates.
The testing order of the three crutch models were randomized,
however all trials with a particular crutch type were completed
successively.
2.4. Data processing
Data analysis was performed using a custom MATLAB program
(Mathworks, Natwick, MA, USA). Force plate data were filtered
(4th-order low-pass Butterworth filter with a cut-off frequency of
50 Hz) to remove noise. For each trial, the data was parsed to
crutch stance phase, starting when the crutch footpad began load-
ing the force plate and ending when the crutch tip was no longer
277
in contact with the force plate. The anterior-posterior and medial-
lateral forces were cut based on crutch stance phase established
by the vertical force profile. Force data were normalized in time to
100% of crutch stance phase, normalized to the participants’ body
weight, and averaged across five trials for each crutch type. The
peak ground reaction forces, impulses, and time spent in crutch
stance phase were calculated. Velocities were determined based on
the time it took the participant to walk the 6 m to complete a trial,
and were averaged across the 5 trials for each crutch type.
2.5. Statistical analysis
Data from the left and right sides were averaged for statis-
tics involving peak force, impulse, and time spent in crutch stance
phase. This allowed us to remove any effect of hand dominance
if present. Statistics were performed on the averaged data from
all 13 able-bodied participants. All statistical computations were
completed with PASW Statistics 18.0 (SPSS Inc. Chicago, IL, USA).
A repeated-measures analysis of variance (ANOVA) was used to as-
sess differences in kinetic variables between the three crutch types.
Pairwise comparisons were used to determine which crutches dif-
fered when crutch type was significant. The significance level was
set to α = 0.05 for all tests. Holm’s Sequential Bonferroni adjust-
ments were made for the three pairwise comparisons.
3. Results
3.1. Ground reaction forces
Peak force: There were no differences in peak vertical or peak
medial-lateral forces between the three crutch types. There was
a significant difference in peak values for braking and propulsive
forces respectively between crutch models (Table 2). The Carbon-
Damp crutch demonstrated a smaller peak braking force compared
to the Aluminum model (p = 0.005), but not the Carbon model
(p = 0.036). The CarbonDamp crutch demonstrated a larger peak
propulsive force than the Aluminum (p = 0.008) and Carbon mod-
els (p = 0.023).
Impulse: There were no differences in vertical impulse between
the CarbonDamp and Aluminum models (p = 0.031). There were
no differences in medial-lateral impulses between the three crutch
types. The CarbonDamp crutch demonstrated a smaller braking im-
pulse compared to the Aluminum model (p = 0.010), but not the
Carbon model (p = 0.096). The CarbonDamp model demonstrated a
larger propulsive impulse than the Aluminum (p = 0.010) and Car-
bon models (p = 0.019).
The CarbonDamp crutch exhibited a significantly larger peak
propulsive force and impulse compared to both the Carbon and
Aluminum crutch models. Additionally, the CarbonDamp crutch ex-
hibited smaller peak braking force and impulse compared to the
Aluminum crutch model.
4. Discussion
We did not observe any differences in peak vertical force or im-
pulse between the crutch models, refuting our hypothesis. Segura
et al. found that an axillary crutch with a spring located on the
distal end of the shaft resulted in increased peak vertical forces
which they postulated could be because of a bottoming out effect
or by variations in gait technique used with the different crutches
[22]. Similar to our study, the authors also observed that the verti-
cal impulse was smaller with their spring-loaded crutch compared
to their standard axillary crutch [22]. It is possible that the differ-
ent footpads that came with the CarbonDamp and Carbon crutch
models may have affected the peak vertical force or impulse.
278 M.K. MacGillivray et al. / Medical Engineering and Physics 38 (2016) 275–279

Table 1
Crutch models used for testing.

Crutch Lower shaft Damper system Foot Handle grip (orientation) Crutch weight (pair)

© TM ©
SideStix (CarbonDamp) Carbon fiber Present Rotating foot, proprietary SideStix , Vibram® pad Ergon GP1 (100°) 1.7 kg
SideStix© (Carbon) Carbon fiber Absent Non-rotating Fetterman© Tornado gel tip Ergon© GP1 (100°) 1.6 kg
Generic (Aluminum) Aluminum Absent Generic, non-rotating Generic (90°) 1.9 kg

Table 2
Differences in spatial-temporal and kinetic variables based on crutch model.

Mean ± SD Aluminum CarbonDamp (damper) Carbon (no damper) Sig. Effect size ηp 2

Walking velocity (m/s) 0.9 ± 0.1 1.0 ± 0.1 0.9 ± 0.1 p = 0.23 0.11
Time in crutch stance phase (s) 0.8 ± 0.1 0.8 ± 0.1 0.8 ± 0.1 p = 0.39 0.08
Peak force (% BW)
Vertical 51.7 ± 2.1 52.4 ± 2.5 51.8 ± 2.2 p = 0.068 0.20
Lateral 2.9 ± 1.0 2.9 ± 1.4 2.9 ± 1.3 p = 0.981 0.00
Posterior (braking) 4.1 ± 1.6 3.3 ± 1.2 3.9 ± 1.5 p = 0.004 0.37
Anterior (propulsive) 10.0 ± 1.4 10.8 ± 1.7 10.3 ± 1.8 p = 0.011 0.31
Impulse (% BW s)
Vertical 30.7 ± 2.8 29.8 ± 2.8 30.4 ± 2.9 p = 0.050 0.22
Lateral 1.7 ± 0.6 1.5 ± 0.7 1.6 ± 0.6 p = 0.341 0.09
Posterior (braking) 0.8 ± 0.5 0.7 ± 0.4 0.8 ± 0.5 p = 0.012 0.31
Anterior (propulsive) 2.6 ± 0.6 2.9 ± 0.6 2.7 ± 0.6 p = 0.010 0.32

Sig. = significance; % BW = percent body weight; s = seconds; SD = standard deviation; ηp 2 = partial eta squared.

The CarbonDamp crutch demonstrated potential benefits over elastomeric polymer may help to conserve energy while walking
both the Carbon and Aluminum crutches. The CarbonDamp crutch on flat surfaces.
exhibited a larger peak propulsive force and impulse compared
to the other two crutch models. Additionally, the CarbonDamp
5. Study limitations
crutch exhibited a smaller peak braking force and impulse than
the Aluminum crutch. These findings indicate that using a crutch
The sample size of this study was small, but comparable to
with an elastomeric damper could help conserve forward momen-
other studies in the field [17,18,22,26]. The effect size for most vari-
tum by increasing propulsive peak forces and impulses on level
ables was moderate to large suggesting a larger population size
surfaces.
would not change the results. The able-bodied participants in this
The increase in peak propulsive force and impulse observed in
study had minimal experience with crutch gait and therefore may
the CarbonDamp crutch may be related to the findings of Seeley
have been more variable with their swing-through gait pattern
et al., who observed a 5% increase in peak forward velocity when
than a population of chronic crutch users because they had just
walking with a spring-loaded axillary crutch compared to a tra-
learned a new motor skill. The use of able-bodied participants re-
ditional axillary crutch [18]. This increased peak forward velocity
duces the generalizability of the findings to chronic crutch users
may be a result of increased peak propulsive force and impulse
however we elected to examine inexperienced crutch users so that
due to return of kinetic energy as the polymer decompresses. The
participants would not be biased towards a specific crutch model.
larger peak forward velocity did not translate into increased over-
Additionally, participants only walked 6 m in this study, however
all velocity in that study [18]. Similarly, in our study we did not
the most frequent bouts of gait are short in number of steps and
observe any differences in velocity between the three crutches.
duration [27].
The type and location of a spring can greatly change the prop-
Although we have specifically looked at the damper system
erties of the crutch. Traditionally, metallic helical springs have
in this study, other design characteristics may be influencing the
been implemented into the lower shaft or foot of the crutch
results. The Carbon and CarbonDamp crutches weighed 300 and
[17,18,22]. Springs located near the ground, however, are likely to
200 g less, respectively, than the Aluminum crutches. The carbon-
change the weight distribution of the crutch (creating greater an-
fiber shaft may be advantageous in damping vibration that occurs
gular momentum because of a longer lever arm), and may also
when the footpad contacts the ground. The different angles of the
be affected by contaminants including dirt and water. In the Car-
handles used might also have affected the movement pattern used
bonDamp crutches, an elastomeric polymer rather than a helical
with these crutches.
spring was used because of increased durability with exposure
to contaminants and theoretically greater damping effect with
smaller vertical displacement when compressed (i.e. users would 6. Conclusions
feel shock absorption without a ‘falling’ sensation). The elastomeric
polymers are also lightweight compared to some metallic helical The crutch with a damper system did not demonstrate differ-
springs. In addition, the polymer was situated more proximally, ences in peak vertical force or impulse. Additionally, the crutch
below the crutch handle, to help shorten the lever arm of the with the damper system exhibited decreased peak braking force
device. This may provide users with improved comfort and less and impulse compared to the generic crutch model and increased
fatigue, which are factors that would be useful to evaluate in the peak propulsive force and impulse compared to the two other
future. crutch models during the crutch stance phase of swing-through
Finally, future research is needed to evaluate whether a damper crutch gait. Although we did not observe any clinically significant
system requires more energy during gait compared to other crutch differences as a result of these kinetic changes (e.g. change in self-
models. Although kinetic energy of the system would be lost selected gait velocity), the forearm crutch with a damper system
through the use of the damper system, the recoiling of the may help to conserve forward momentum when walking on level
 M.K. MacGillivray et al. / Medical Engineering and Physics 38 (2016) 275–279
surfaces. Future research is needed to explore the effect of this
crutch design on gait kinematics and energy expenditure.
Conflicts of interest
The authors declare that the crutches studied were provided by
the “Sidestix” company. They had no involvement in the study de-
sign, analyses, or writing of this manuscript.
Funding
The study was funded by a federal granting program (MITACS)
and the University of British Columbia’s Summer Student Research
Program.
Ethical approval
University of British Columbia’s Clinical Research Ethics Review
Board #H013-0243.
Acknowledgements
We acknowledge the financial support of the Mitacs Accelerate
program, the National Research Council of Canada’s Industrial Re-
search Assistance Program, and the University of British Columbia’s
Summer Student Research Program.
References
[1] Epstein S. The classic: art, history, and the crutch. Clin Orthop Relat Res
1972;89:4–9.
[2] Shoup TE, Fletcher LS, Merrill BR. Biomechanics of crutch locomotion. J
Biomech 1974;7:11–19.
[3] Poddar SB, Gitelis S, Heydemann PT, Piasecki P. Bilateral predominant radial
nerve crutch palsy. A case report. Clin Orthop Relat Res 1993:245–6.
[4] Feldman DR, Vujic I, McKay D, Callcott F, Uflacker R. Crutch-induced axillary
artery injury. Cardiovascular Interventional Radiol 1995;18:296–9.
[5] Raikin S, Froimson MI. Bilateral brachial plexus compressive neuropathy
(crutch palsy). J Orthop Trauma 1997;11:136–8.
[6] Opila KA, Nicol AC, Paul JP. Upper limb loadings of gait with crutches. J
Biomech Eng 1987;109:285–90.
[7] Amin A, Singh V, Saifuddin A, Briggs TW. Ulnar stress reaction from crutch use
following amputation for tibial osteosarcoma. Skeletal Radiol 2004;33:541–4.
279
[8] McGoldrick F, O’Brien TM. Bilateral stress fractures of the ulna. Injury
1988;19:360–1.
[9] Suarez GG, Garcia GJ, Perez CL. Stress fracture of the ulna associated with
crutch use. J Orthop Trauma 2001;15:524–5.
[10] Malkan DH. Bilateral ulnar neuropraxia: a complication of elbow crutches. In-
jury 1992;23:426.
[11] Ginanneschi F, Filippou G, Milani P, Biasella A, Rossi A. Ulnar nerve compres-
sion neuropathy at Guyon’s canal caused by crutch walking: case report with
ultrasonographic nerve imaging. Arch Phys Med Rehabil 2009;90:522–4.
[12] Fisher SV, Patterson RP. Energy cost of ambulation with crutches. Arch Phys
Med Rehabil 1981;62:250–6.
[13] McBeath AA, Bahrke M, Balke B. Efficiency of assisted ambulation determined
by oxygen consumption measurement. J Bone Joint Surg Am 1974;56:994–
1000.
[14] Thys H, Willems PA, Saels P. Energy cost, mechanical work and muscular effi-
ciency in swing-through gait with elbow crutches. J Biomech 1996;29:1473–82.
[15] Slavens BA, Frantz J, Sturm PF, Harris GF. Upper extremity dynamics dur-
ing Lofstrand crutch-assisted gait in children with myelomeningocele. J Spinal
Cord Med 2007;30(Suppl 1):S165–71.
[16] LeBlanc M, Carlson LE, Nauenberg T. A quantitative comparison of four exper-
imental axillary crutches. J Prosthet Orthot 1993;5:20–8.
[17] Parziale JR, Daniels JD. The mechanical performance of ambulation using
spring-loaded axillary crutches. A preliminary report. Am J Phys Med Rehabil
1989;68:192–5.
[18] Seeley MK, Hunter I, Bateman T, Roggia A, Larson BJ, Draper DO. A kine-
matic comparison of spring-loaded and traditional crutches. J Sport Rehabil
2011;20:198–206.
[19] Shortell D, Kucer J, Neeley WL, LeBlanc M. The design of a compliant compos-
ite crutch. J Rehabil Res Dev 2001;38:23–32.
[20] Zhang Y, Liu G, Xie S, Liger A. Biomechanical evaluation of an innova-
tive spring-loaded axillary crutch design. Assistive Technol Off J RESNA
2011;23:225–31.
[21] Shoup TE. Design and testing of a child’s crutch with conservative energy stor-
age. Trans ASME 1980;102:672–6.
[22] Segura A, Piazza SJ. Mechanics of ambulation with standard and spring-loaded
crutches. Arch Phys Med Rehabil 2007;88:1159–63.
[23] Doherty S, Perreur-Lloyd K, Johnson W, Moonie G, Foreman C. Assistive mobil-
ity device. Patent # 20110240077, 2011.
[24] Joyce BM, Kirby RL. Canes, crutches and walkers. Am Family Physician
1991;43:535–42.
[25] Van Hook FW, Demonbreun D, Weiss BD. Ambulatory devices for chronic gait
disorders in the elderly. Am Family Physician 2003;67:1717–24.
[26] Haubert LL, Gutierrez DD, Newsam CJ, Gronley JK, Mulroy SJ, Perry J. A com-
parison of shoulder joint forces during ambulation with crutches versus a
walker in persons with incomplete spinal cord injury. Arch Phys Med Reha-
bil 2006;87:63–70.
[27] Orendurff MS, Schoen JA, Bernatz GC, Segal AD, Klute GK. How humans
walk: bout duration, steps per bout, and rest duration. J Rehabil Res Dev
2008;45:1077–89.