Full text of DOI:10.1177/107110070302400310

FOOT & ANKLE INTERNATIONAL
Copyright © 2003 by the American Orthopaedic Foot & Ankle Society, Inc.
The Effect of Leg Length Discrepancy on Foot Loading Patterns
and Contact Times
G.C. O’Toole, AFRCSI*; N.K. Makwana, FRCS (Orth.)**; J. Lunn, FRCSI*;
J. Harty, AFRCSI*; M.M. Stephens, FRCSI, MSc (Bioeng.)#
Dublin, Ireland
ABSTRACT
on the longer side.4 Using ground-reaction-force-vector
analysis, a discrepancy of more than 2 cm caused an
asymmetry in gait between longer and shorter limbs.4
Leg length discrepancy is associated with a
decreased stance time on the shorter side, decreased
walking velocity, increased cadence and decreased
step length on the shorter side.5 Studies on patients with
cerebral palsy have demonstrated compensatory
supination of the foot on the shorter side and pronation
of the foot on the longer side.6 In addition to these
changes, equinus of the foot and lowering of the pelvis
on the short side help to minimize the discrepancy.7
Whether leg length discrepancy be present in the
spastic setting of cerebral palsy or the dynamic setting
of a healthy volunteer little is known about the detri-
mental effect, if any, on the foot as a result of these
adaptive changes.
The effect of a simulated leg length discrepancy on foot
loading patterns and gait cycle times in normal individu-
als was investigated. Thirty feet of 15 normal volunteers
were evaluated. Leg length discrepancy was simulated
using flexible polyurethane soles. As leg length discrep-
ancy increased, the total loading increased from 35.31 to
37.99 kg/cm2/sec, forefoot loading increased from 15.58 to
19 kg/cm2/sec, hindfoot loading remained the same. All
subjects, except females with middle loading patterns,
demonstrated increased hallux loading. The contact
phase of gait decreased from 22% to 13%, the midstance
phase remained the same, the propulsion phase
increased from 44% to 50%. All findings were statistically
significant. Leg length discrepancy has significant
effects on the foot. Different adaptive loading patterns
amongst subsets of individuals are seen.
We evaluated the effect of simulated leg-length dis-
crepancy on foot loading patterns in normal individuals
to determine if any changes exist.
Key Words: Leg Length Discrepancy; Foot Adaptive
Changes.
MATERIALS AND METHODS
INTRODUCTION
Thirty feet of 15 normal volunteers were used in this
study. Each subject was screened to ensure that there
was no history of gait abnormality, a lower limb disorder
or a leg length discrepancy. There were seven males and
eight females studied, mean age 26.5 (range, 17 to 37)
years. The mean height was 1.67 m (range, 1.54 to 1.93)
and mean weight was 66.5 kg (range, 55 to 90).
It has been well documented that leg length discrep-
ancy can be associated with back pain,1 knee pain and
hip problems.2 Differences as little as 6 mm may be
important in individuals subjected to repetitive loading.3
A discrepancy of 1.3 to 2.5 cm has been associated
with the development of superolateral arthritis of the hip
Leg length discrepancy was simulated using flexible
polyurethane sole of 1-5 cm thickness that was secured
to the sole of a sandal and worn on the opposite foot.
A Musgrave footprint computerized pedobarograph
system (Musgrave Medical Ltd., Llangollen, U.K.) was
used to collect pressure patterns of the feet. The two
foot plates measure 650x297 mm and contain 2048
force sensing resisters. Each sensor size was 0.25 cm2.
The scanning time was 18 msec, and the operating fre-
quency was 56 Hz. This unit was calibrated in kg/cm2.
This system conforms well to the desired sensor char-
acteristics as outlined by Rosenbaum and Becker8 and
*
Registrar, Cappagh National Orthopaedic Hospital, Finglas, Dublin 11,
Ireland
** Foot and Ankle Fellow, Cappagh National Orthopaedic Hospital, Finglas,
Dublin 11, Ireland
#
Consultant Orthopaedic Surgeon, Cappagh National Orthopaedic
Hospital, Finglas, Dublin 11, Ireland
Corresponding Author:
Mr. Michael Stephens
Cappagh National Orthopaedic Hospital
Finglas, Dublin 11, Ireland
Phone: +353 1 834-1211
Fax: +353 1 834-1211
256
Downloaded from fai.sagepub.com at Flinders University on March 18, 2016

Foot & Ankle International/Vol. 24, No. 3/March 2003
LLD FOOT LOADING PATTERNS
257
helped ensure that our system conformed
well to the recommended technical
specifications of reproducibility and stabil-
ity, while proving that the system was
independent of subjective mood and
motivation.10
By analyzing the control foot loading
patterns of all subjects, it was possible to
further subcategorize the volunteers into
hallux or mid-forefoot loading patterns.
This was achieved by measuring the
maximum pressure under both the hallux
and the mid-forefoot and calculating the
resultant ratio of hallux/mid-forefoot pres-
sures. A subject achieving a ratio of
greater than 1.00 was considered to have
a hallux loading pattern, conversely a
ratio of less than 1.00 the subject was
categorized as having a mid-forefoot
loading pattern.
Fig. 1: The gait cycle time analysis of an individual showing the control pattern and the
pattern with a leg length discrepancy of 5 cm. Time in milliseconds is represented on the
x axis and load in kilograms on the y axis. Tracing of the total load (kg) is shown on the
top and the total pressure (kg/cm2) tracing is shown on the bottom. This tracing demon-
strates a decrease in the contact phase time (190 ms to 150 ms) and an increase in the
propulsion phase (170 ms to 210 ms) while the mid-stance phase (350 ms) and total time
(710 ms) remains. It also shows an increase in total forefoot load ie. area under the
curve during propulsion phase.
The maximum load from each sensor
was used for analysis of load as this rep-
resented the peak load in the temporal
profile of recorded loads. These records
were then analyzed to determine the
the operating frequency of 56 Hz is well below the rec-
ommended maximum sampling frequency of 100 Hz as
outlined by Schaff et al.9
magnitude and distribution of the loading on the foot.
RESULTS
Each subject was allowed to walk on the walkway
several times to accommodate to the surface and labo-
ratory area. The walkway was 20 m long and 1.5 m wide
with the footplates level with the floor. The subject’s
starting position was altered to ensure that, at their nor-
mal pace, the foot would make contact centrally on the
footplate with the subject looking straight ahead, without
having to look down to strike the plate.
Recordings of the foot pressures and load were first
recorded barefoot without leg length discrepancy (con-
trol) and then when simulating a leg length discrepan-
cy, pressure recordings were taken from the shorter
limb. The process was repeated until three recordings
of an entire foot pattern were achieved.
The test was repeated with each centimeter increment
and the trial repeated on the opposite limb. For each leg
and test condition the mean of the three trials was cal-
culated. The maximum load and pressure data for each
sensor were exported to a spreadsheet for analysis of
the load and load distribution across the foot. Statistical
analysis was by students t-test.
The test was repeated again several weeks later to
ensure that all the results were reproducible for an indi-
vidual and to ensure that the foot loading patterns did not
display any significant changes due to a subject’s famil-
iarity with the test process. Repeating the experiments
The patterns of recorded load were reproducible for
an individual whether repeated during the same test
session or when tested several weeks later.
The results show that as leg length discrepancy
increased the total loading on the foot of the short limb
increased from a mean of 35.31 (SD 7.19) to 37.99
(SD 7.92) kg/cm2/sec; forefoot loading increased from
a mean of 15.58 (SD 4.84) to 19.00 (SD 5.38)
kg/cm2/sec, whereas the hindfoot loading remained
the same.
As leg length discrepancies increased the contact
phase of gait decreased from a mean of 22% (SD 4.34)
of the stance phase to 13% (SD 3.48), p<0.001. The
midstance phase remained the same whereas the
propulsion phase increased from 44% (SD 6.325) to
50% (SAD 8.58), p<0.003 (Fig. 1)
As leg length discrepancy increased, hallux loading
similarly increased in three of the four categories of sub-
jects, males with mid-forefoot loading, males with hallux
loading, and females with hallux loading (p<0.001). The
exceptions were those females who demonstrated a
pattern of mid-forefoot loading on their control foot pat-
tern. These females demonstrated further mid-forefoot
loading as leg length discrepancy increased (p=0.004)
(Figs. 2, 3).
Downloaded from fai.sagepub.com at Flinders University on March 18, 2016

258
O’TOOLE, MAKWANA, LUNN, HARTY AND STEPHENS
Foot & Ankle International/Vol. 24, No. 3/March 2003
DISCUSSION
show that contact phase decreased from 22% to 13%
(p<0.0001), the midstance phase remained the same
and the propulsion phase increased from 44% to 50%
(p<0.003). These changes are consistent with our find-
ings of increased total forefoot load because if the fore-
foot is in contact with the ground for an increased length
of time the total load being transferred is increased.
As leg length discrepancy increased there was not a
significant decrease in the overall time of the gait cycle
or any discernible shortening of the gait cycle time.
We confirmed that as the leg length discrepancy
increased there was a deviation of load towards the
medial side of the forefoot.6,7,13,14,15 This pronation or
forefoot adaptive change of the foot and transfer of the
load medially was seen in three of the four subgroups of
subjects in our study. We demonstrated a subset of
subjects with different adaptive patterns. The female
mid-forefoot loading did not conform to this.
Previous studies have demonstrated that footwear
increased loading on the medial aspect of the foot11,12 and
that increasing the heel height of the footwear decreased
the area of forefoot contact while causing a deviation of
load towards the medial side of the forefoot.13,14
The simulated leg length discrepancy as studied in
our paper can be considered analogous to raising the
heel height. As heel height increases the total forefoot
load remains but there is deviation towards the medial
side.12
A Musgrave Footprint Pedobarograph represents sig-
nificant advances in technology. This was used by
Corrigan et al.12 and is similar to the model used in our
study. Intervening advances in 3-D computer software
have allowed more detailed analysis of the entire foot
loading pattern.
As a result of this entire foot analysis, simultaneous
forefoot, midfoot and hindfoot adaptive patterns could
be detected.
The female mid-forefoot loading group increased
loading toward the midfoot in their adaptive patterns.
We recognize that our findings have limited applica-
tion. They apply to healthy volunteers who possess a
full capacity for adaptive changes in the shorter limb in
response to a leg length discrepancy. We have not stud-
ied persons with polio or ambulating patients with cere-
bral palsy in whom the capacity to adapt might be lost.
Further study is required on these patients before any
conclusions of this paper can be extrapolated to their
clinical setting.
We feel the categorization of subjects into four sub-
groups of male mid-forefoot loading, male hallux loading,
female mid-forefoot loading and female hallux loading is
simpler than the method of Hughes et al.15 They
described four main patterns of forefoot pressure distri-
bution based around first ray loading.15 Subjects with cen-
tral and lateral loading patterns as categorized by
Hughes et al. were labeled as mid-forefoot loading in our
study. Similarly, those subjects categorized by Hughes et
al. as having medial and medial/central forefoot loading
were categorized as hallux loading in this study.
This paper demonstrates that as the leg length dis-
crepancy increases there is an increase in total loading
on the whole foot 35.31 (SD 7.19) to 37.99 (SD 7.92)
kg/cm2/sec (p<0.0001). Furthermore, contrary to other
papers12 we also demonstrated an increase in the total
forefoot loading 15.58 (SD 4.84) to 19.00 (SD 5.38)
kg/cm2/sec patterns (p<0.0005).
However, we postulate that increasing forefoot load-
ing as leg length discrepancy increases might render a
subject more prone to forefoot pathology.
This study demonstrates that increasing leg length
discrepancy causes an increase in the total foot loading
and an increase in the total forefoot loading on the
shorter side. It also demonstrates that gait cycle times
are altered with a decrease in the contact phase and an
increase in the propulsion phase. This study demon-
strates a differing adaptive change in forefoot loading
patterns in different subject subgroups as the leg length
discrepancy increases.
As leg length discrepancy increases we have demon-
strated that gait cycle times also change. Our findings
Fig. 2: The pattern seen in an individual categorized as having hal-
lux loading pattern. As the leg length discrepancy increases loading
of the hallux similarly increases.
Fig. 3: A female middle loading adaptive pattern. As the leg length
discrepancy increased from 0 cm to 5 cm females increasingly
loaded toward the middle.
Downloaded from fai.sagepub.com at Flinders University on March 18, 2016

Foot & Ankle International/Vol. 24, No. 3/March 2003
LLD FOOT LOADING PATTERNS
259
in the uninvolved lower limb in children with spastic hemiplegia;
The effect of actual and functional leg length discrepancy. J Paed
Orth, 20:88-92, 2000.
Editor’s Note
Although the current paper presents a study using
only two subjects, it is a technique which opens the door
to future studies. It is therefore presented as a tech-
nique which may have great promise to expand our
understanding of the workings of the heel/plantar fascia
complex. As many studies are, it is a work in progress
and should be viewed as such.
7. O’Connell, PA; D’Souza, L; Dudeney, S; Stephens, M: Foot defor-
mities in children with Cerebral Palsy. J Paed Orth, 18:743-747, 1998.
8. Rosenbaum, D; Becker, H-P: Plantar pressure distribution meas-
urements. Technical background and clinical applications. Foot
Ankle Surg, 3:1-14, 1997.
9. Schaff, P; Reinisch, P; Mitternacht, J; Senner, V: Praktikabilität
contra M¿ßtechnik. Orthopädie-Technik, 787-795, 1991.
10.Brand, RA: Assessment of musculoskeletal disorders by locomo-
tion analysis: a critical historical and epistemological review.
Biolocomotion: A Century of Research Using Moving Pictures.
Roma, Italy. Promograph, 227-242, 1992.
11. Soames, RW: Foot pressure patterns during gait. J Biomed Eng,
7:120-126, 1985.
12.Corrigan, JP; Moore, DP; Stephens, MM: Effect of heel height
on forefoot loading. Foot Ankle, 14:148-152, 1993.
13.McBride, ID; Wyss, UP; Cooke, TDV; Murphy, L; Philips, J;
Olney, SJ: First metatarsophalangeal joint reaction forces during
high heal gait. Foot Ankle, 11:282-288, 1991.
14.Schwartz, RP; Heath, AI; Morgan, DW; Towns, RC: A quantita-
tive analysis of recorded variables in the walking patterns of nor-
mal adults. J Bone Joint Surg, 46A:324-334, 1964.
REFERENCES
1. Giles, LGF; Taylor, JR: Low back pain associated with leg length
inequality. Spine, 6:510-521, 1981.
2. Gofton, JP; Trueman, GE: Studies in osteoarthritis of the hip:
Part II osteoarthritis of the hip and leg length discrepancy. Can
Med Asso J, 104:791-799, 1971.
3. Subotnick, SI: Limb length discrepancies of the lower extremity
(short leg syndrome). J Orthop Sports Phys Ther, 3:11-16, 1981.
4. Kafman, KR; Miller, LS; Sutherland, DH: Gait asymmetry in
patients with limb-length inequality. J Paed Orth, 16:144-150, 1996.
5. Bhave, A; Dror, P; Herzenberg, J: Improvement in gait parame-
ters after lengthening for the treatment of limb-length discrepancy.
J Bone Joint Surg, 81A:529-534, 1999.
15.Hughes, J; Clark, P; Jagoe, JR; Gerber, C; Klenerman, L: The
pattern of pressure distribution under the weightbearing forefoot.
Foot, 1:117-124, 1991.
6. Allen, P; Jenkinson, A; Stephens, M; O’Brien, T: Abnormalities
Downloaded from fai.sagepub.com at Flinders University on March 18, 2016