G Model

GAIPOS-3449; No. of Pages 6

Gait & Posture xxx (2012) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Gait & Posture

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

Residual gait deviations in adolescents treated during infancy for unilateral

developmental dysplasia of the hip using Pemberton’s osteotomy
Chu-Fen Chang a,1, Ting-Ming Wang b,1, Jyh-Horng Wang b, Shier-Chieg Huang b, Tung-Wu Lu a,*
a
Institute of Biomedical Engineering, National Taiwan University,Taiwan, ROC
b
Department of Orthopaedic Surgery, National Taiwan University Hospital,Taiwan, ROC

A R T I C L E I N F O A B S T R A C T

Article history: Early reduction using Pemberton’s osteotomy has been suggested for treating DDH but no data on the
Received 21 June 2011 long-term residual gait changes in such patients are available in the literature. This study aimed to bridge
Received in revised form 5 November 2011 the gap by performing quantitative gait analysis on eleven females (age: 10.6  1.0 years) who were
Accepted 20 November 2011
treated for unilateral DDH using open reduction with Pemberton’s osteotomy at 1.6  0.5 years of age, and
eleven age-matched healthy controls. Walking at a normal speed, the Pemberton group displayed
Keywords: significantly more anterior tilt, hiking at the affected side and rotation towards the unaffected side of the
Developmental dysplasia of the hip (DDH)
pelvis, and more knee flexion and ankle dorsiflexion in the affected limb. With this asymmetrical gait, they
Pemberton’s osteotomy
appeared to reduce the demands on the hip flexors and abductors, and knee extensors in the affected limb,
Gait analysis
Joint kinematics which might have been involved during the osteotomy, but increased compensatory efforts from the hip
Joint kinetics extensors, ankle plantarflexors and knee flexors in the unaffected limb.
ß 2011 Elsevier B.V. All rights reserved.

1. Introduction
Developmental dysplasia of the hip (DDH) is characterized by a
poorly formed acetabulum and a displaced femoral head, affecting
predominantly females in a ratio of about 5–8:1 [1]. The ratio of
unilateral DDH to bilateral DDH is about 4–6:1 [2]. The insufficient
coverage and incongruence at such a pathological hip will result in
greater joint contact pressures, which will aggravate pain, hasten
degenerative changes and eventually lead to joint osteoarthritis
(OA) [3]. For patients with untreated DDH, approximately 20–50%
of dysplastic hips were found to have premature degenerative OA
[4] and the severity of degeneration was closely related to gait
abnormalities [5].
Abnormal gait in patients with unilateral DDH includes slower
than normal walking speed, shorter steps, less anterior pelvic tilt,
more hiking at the affected side and rotation towards the
unaffected side of the pelvis, reduced hip sagittal motion, but
greater knee and ankle flexion [5–7]. Reduced hip sagittal
moments, and hip and knee abductor moments, together with
decreased ground reaction forces (GRF) during the initial double
limb stance (DLS) were also found in these patients [5–8]. It has
been suggested that these compensations are a protective
mechanism for diminishing the loads and pain on the dysplastic
* Corresponding author. Tel.: +886 2 33653335; fax: +886 2 33653335.
E-mail address: twlu@ntu.edu.tw (T.-W. Lu).
1
These authors contributed equally to this work.
hip [5–8]. In general, the gait of patients with unilateral DDH is
asymmetrical and less efficient than that of healthy controls.
Early intervention for DDH is believed to be of vital importance
to mediate and sometimes prevent the development of sequelae
[6,9], such as avascular necrosis, acetabular dysplasia and OA [3,4],
and gait abnormalities. Pemberton’s osteotomy is one of the most
prevalent procedures for DDH. It re-orients the acetabulum
through the flexible triradiate cartilage [10] and has the capability
for adjusting the volume and direction of coverage of the dysplastic
hip [10,11]. Early treatment for DDH using Pemberton’s osteotomy
before the patient starts weight-bearing is considered to be
beneficial for decreasing the sequela of DDH [10] because the
dysplastic hip can be settled in a near normal loading condition
after this procedure. Early intervention before gait maturation is
also considered to be helpful for reducing residual gait deviations.
Previous quantitative gait studies have focused mainly on adult
or adolescent patients treated for hip dysplasia, reporting residual
gait deviations after periacetabular osteotomy [9,12,13] or after
total hip arthroplasty (THA) for hip OA primary or secondary to
DDH [8,14,15]. Only a limited number of studies have evaluated
gait patterns in children who had been surgically treated by open
reduction with soft tissue release at the ages between 2.4 and 20
months [16–18]. Moreover, the data reported were mainly limited
to the sagittal-plane kinematics and kinetics of the hip joint
[16,17]. No study has reported the changes in the three-
dimensional (3D) gait patterns in patients treated for DDH at an
early age using Pemberton’s osteotomy. On the other hand, a
recent study on adolescents who were treated for unilateral DDH

0966-6362/$ – see front matter ß 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.gaitpost.2011.11.024

Please cite this article in press as: Chang C-F, et al. Residual gait deviations in adolescents treated during infancy for unilateral
developmental dysplasia of the hip using Pemberton’s osteotomy. Gait Posture (2012), doi:10.1016/j.gaitpost.2011.11.024
G Model
GAIPOS-3449; No. of Pages 6

2 C.-F. Chang et al. / Gait & Posture xxx (2012) xxx–xxx

Table 1
Means (standard deviations) of the temporal–spatial parameters and the loading rates of the GRF of the Pemberton and control groups.

Control group Pemberton group Pg Ps

Affected Unaffected Affected Unaffected

Temporal–spatial parameters
Gait speed (% LL/s) 151.4 (18.7) 149.1 (16.2) 0.922 –
Stride length (% LL) 149.7 (7.8) 154.1 (8.0) 0.224 –
Cadence (steps/min) 118.9 (11.3) 112.7 (11.5) 0.124 –
Step width (% LL) 12.0 (3.1) 13.9 (3.0) 0.131 –
Step length (% LL) 76.9 (3.8) 79.5 (3.4) 76.7 (3.8) 0.104 0.880 0.005*
Terminal DLS (% gait cycle) 13.7 (1.3) 10.5 (1.7) 12.8 (3.4) 0.002* 0.241 0.028*
Upward accelerations of the heel 12.9 (4.4) 9.7 (4.7) 6.6 (4.7) 0.036* <0.001* 0.182
at the end of the swing phase (m/s2)
Loading rate of the GRF (% BM/s) 1302.4 (287.4) 1887.6 (610.5) 1808.7 (480.2) 0.041* 0.016* 0.779

LL: leg length (cm); DLS: double limb stance; BM: body mass; Pg: comparisons between the Pemberton group and the control group using U-test; Ps: comparisons between
the two limbs in the Pemberton group using Wilcoxon signed ranks test.
*
Significant difference (p < 0.05).

with Pemberton’s osteotomy during infancy showed that these In a gait laboratory, each subject walked at a self-selected pace on an 8-m
walkway, wearing 28 retroreflective markers [20] to track the motions of the pelvis,
patients sustained greater than normal rates of repetitive loading
and each thigh, shank and foot. Three-dimensional trajectories of the markers were
around heel-strike in both hips, exposing themselves to a higher measured using a 7-camera motion analysis system (Vicon 512, Oxford Metrics,
risk of premature hip OA [19]. Since the loading between the foot U.K.) and the GRF were recorded from two forceplates (AMTI, U.S.A.).
and the ground is a result of the control of the whole body and the The locomotor system was modeled as a 7-link system with the pelvis, thighs,
way it makes contact with the floor, a complete 3D gait analysis shanks and feet as rigid links, each embedded with an orthogonal coordinate system
[20]. The center of rotation of the hip was estimated using a functional method [21],
will be helpful for a better picture of the mechanisms that may be and the centers of the knee and ankle joints were assumed to be on the mid-points
responsible for the altered loading condition [19]. between the femoral epicondyles and between the malleoli, respectively [7].
The purpose of this study was to evaluate the residual changes Segmental inertial properties were determined using Dempster’s coefficients [22].
in the joint kinematics and kinetics of the locomotor system during A cardanic rotation sequence (z–x–y) was used to describe the rotational joint
movements [23]. The toe-out angle at the end of single limb stance (SLS) was
level walking in adolescents who had received Pemberton’s
calculated as the angle between the line of progression and the line between the 3rd
osteotomy for unilateral DDH at an early age. It was hypothesized metatarsal head and the heel. Inverse dynamics analysis was performed to obtain
that, compared to healthy controls, these patients would display the internal joint moments. Loading rates of the GRF were calculated as the
different, asymmetric gait patterns, leading to increased loading maximum slopes of the vertical component curves during the initial DLS [19]. For
rates during the loading response. subsequent statistical analysis, peak values, as well as values of the joint angles and
moments, and pelvic angles at heel-strike, the beginning and end of SLS and toe-off,
were extracted. Temporal–spatial parameters were also calculated, including gait
speed, cadence, stride length, step length, and step width [19], which were
2. Materials and methods normalized to LL. The upward acceleration of the heel marker immediately before
initial contact was obtained to indicate how the swing foot was controlled for a
Eleven female adolescents who had received Pemberton’s osteotomy for smooth heel-strike.
unilateral DDH (age at operation: 1.6  0.5 years; age at experiment: 10.6  1.0 For between-subject comparisons, joint moments were normalized to BM and LL,
years; body mass (BM): 33.6  8.3 kg; height: 140.8  8.0 cm; leg length (LL): affected and the GRF were normalized to BM. For the control group, each of the gait variables
limb = 72.9  6.4 cm and unaffected limb = 72.7  6.7 cm) participated in this study was calculated for each limb and then averaged across limbs for each trial. The non-
with informed written consent of their parents or guardians. They had been neglected parametric Mann–Whitney U-test was used to detect the differences between the
cases and had undergone an index operation at walking age. Following soft tissue two groups and the Wilcoxon signed-ranks test was used to detect the differences
procedures and open reduction, a Pemberton’s osteotomy was performed if the between the affected and unaffected side in the Pemberton group. All significance
affected hip was stable in abduction and internal rotation, but was unstable in the levels were set at a = 0.05.
neutral position. At the time of experiment, they were community ambulatory, and
were free of pain, fixed flexion deformities that would limit their hip motion during
walking, or any other neuromuscular disease. Their frontal-plane acetabular coverages 3. Results
were improved to within normal range at six months post-surgery according to the X-
ray findings. Eleven healthy controls were matched with the Pemberton group
according to sex, age, height and body mass (age: 11.1  1.5 years; BM: 35.7  7.7 kg;
Most temporal–spatial parameters were not statistically
height: 143.5  9.8 cm; LL: 76.1  6.1 cm). Approval to conduct the study was granted different between groups but a significantly shorter terminal
by the Institutional Research Board. double limb stance (DLS) was found in the affected limb of the
Ipsilateral (-) / Contralateral (+) side
Downward Rotation (-) / Hiking (+)
Posterior (-) / Anterior (+) Tilt

Pelvic Tilt Pelvic Obliquity Pelvic Rotation

15 15 15
10 10
10
Rotation towards

5 5
Degree

5
0 0
0
-5 -5
-5 -10 -10

-10 -15 0 -15

0 20 40 60 80 100 20 40 60 80 100 0 20 40 60 80 100
% Gait cycle % Gait cycle % Gait cycle

Fig. 1. Ensemble-averaged three-dimensional pelvic motion during the gait cycle of the affected limb (black, solid line) and unaffected limb (black, dotted line) of the
Pemberton group and of the normal group (gray, solid line). A black star indicates significant difference between the affected limb and normal group, a white star indicates
significant difference between the unaffected limb and normal group, and a gray star indicates significant difference between the affected and unaffected limb.

Please cite this article in press as: Chang C-F, et al. Residual gait deviations in adolescents treated during infancy for unilateral
developmental dysplasia of the hip using Pemberton’s osteotomy. Gait Posture (2012), doi:10.1016/j.gaitpost.2011.11.024
G Model
GAIPOS-3449; No. of Pages 6

C.-F. Chang et al. / Gait & Posture xxx (2012) xxx–xxx 3

Pemberton group (Table 1). The Pemberton group also walked with
significantly reduced upward acceleration of the heel marker at the
end of the swing phase and with greater GRF loading rates in both
limbs. Temporal–spatial variables and GRF loading rates were not
significantly different between the limbs in the Pemberton group,
except significantly greater step lengths and shorter terminal DLS
were found in the affected limb (Table 1).
The Pemberton group walked with significantly more anterior tilt
and upward hiking at the affected side of the pelvis, as well as more
pelvic rotation towards the unaffected side (Fig. 1 and Table 2).
While the two groups showed similar angular motion patterns of all
the joints in the sagittal plane, the Pemberton group walked with a
more flexed knee and dorsiflexed ankle in the affected limb during
most of the gait cycle, but with a more flexed knee only at the
beginning of SLS and an increased peak hip extension in the
unaffected limb (Table 3). Joint motion patterns in the frontal and
transverse planes were similar between groups, except for the
increased ankle abduction at the beginning of SLS in the affected
limb (affected limb: 3.3  3.98; control: 1.1  3.58, p = 0.005). The
Pemberton group also showed significantly greater toe-out angles in
the affected limb. Similar results for the affected limb were also found
when compared to the unaffected limb, except for knee flexion at the

Table 2
Means (standard deviations) of the pelvic motion (8) at four key events (T1: heel-strike; T2: beginning of SLS; T3: end of SLS; T4: toe-off) and the peak pelvic angles during the
stance and swing phase (P1 and P2, respectively).

Control group Pemberton group Pg Ps

Affected Unaffected Affected Unaffected

Anterior tilt (+)/posterior tilt ( )

T1 4.3 (3.1) 8.1 (3.5) 8.4 (3.7) 0.03* 0.036* 0.160
T2 4.7 (3.3) 7.7 (2.6) 8.6 (3.3) 0.017* 0.025* 0.310
T3 3.9 (3.6) 8.5 (3.5) 8.8 (4.4) 0.011* 0.025* 0.866
T4 4.4 (3.4) 8.8 (3.2) 7.9 (3.5) 0.017* 0.052* 0.483
P1 5.5 (4.1) 9.0 (3.9) 9.1 (3.9) 0.016* 0.019* 0.202
P2 5.2 (3.5) 9.3 (4.2) 9.0 (3.7) 0.023* 0.042* 0.859
Hiking (+)/downward rotation ( )
T1 0.9 (0.7) 0.6 (2.5) 2.2 (2.3) 0.138 0.057 0.033*
T2 2.4 (1.0) 4.5 (2.2) 0.6 (2.5) 0.008* 0.008* 0.008*
T3 1.0 (0.7) 2.2 (2.2) 1.0 (2.3) 0.057 0.008* 0.013*
T4 2.7 (1.0) 1.0 (2.1) 4.8 (1.9) 0.004* 0.002* 0.006*
P1 4.1 (0.7) 5.8 (1.9) 2.1 (2.4) 0.002* 0.002* 0.008*
P2 0.1 (0.7) 2.7 (2.1) 1.0 (1.7) <0.001* 0.051 0.008*
Rotation towards contralateral side (+)/ipsilateral side ( )
T1 5.7 (2.4) 8.0 (3.6) 4.9 (3.0) 0.02* 0.269 0.041*
T2 4.3 (2.0) 6.3 (3.3) 4.0 (2.9) 0.032* 0.906 0.075
T3 5.4 (2.4) 4.6 (3.0) 7.8 (3.7) 0.335 0.018* 0.026*
T4 4.3 (1.9) 4.0 (3.0) 6.9 (2.4) 0.944 0.007* 0.062
P1 7.2 (1.8) 9.4 (3.3) 6.9 (2.8) 0.041* 0.944 0.110
P2 5.7 (2.4) 8.0 (3.6) 4.9 (3.0) 0.02* 0.269 0.041*

SLS: single limb stance; Pg: comparisons between the Pemberton group and the control group using U-test; Ps: comparisons between the two limbs in the Pemberton group
using Wilcoxon signed ranks test.
*
Significant difference (p < 0.05).

Table 3
Means (standard deviations) of the joint angles (degrees) at four key events (T1: heel-strike; T2: beginning of SLS; T3: end of SLS; T4: toe-off) and the peak angles during the
stance and swing phase (P1 and P2, respectively).

Control group Pemberton group Pg Ps

Affected Unaffected Affected Unaffected

Hip flexion (+)/extension ( )

T1 23.7 (4.0) 24.8 (3.8) 22.7 (4.5) 0.589 0.384 0.286
T2 19.4 (5.3) 21.0 (4.0) 20.6 (4.7) 0.384 0.760 0.790
T3 9.9 (4.1) 10.5 (5.7) 12.6 (3.6) 0.981 0.105 0.131
T4 3.9 (4.9) 3.8 (7.6) 7.2 (3.8) 0.906 0.051 0.091
P1 11.3 (4.5) 12.0 (5.4) 14.6 (3.1) 0.760 0.036* 0.110
P2 27.9 (3.6) 28.9 (3.4) 25.6 (4.3) 0.724 0.180 0.091
Knee flexion (+)/extension ( )
T1 5.7 (6.6) 9.7 (4.8) 6.24 (3.8) 0.057 0.869 0.091
T2 17.9 (7.2) 25.4 (5.3) 23.1 (6.4) 0.004* 0.036* 0.155
T3 15.1 (5.9) 21.6 (5.2) 16.4 (6.7) 0.014* 0.495 0.021*
T4 43.4 (6.0) 50.1 (9.2) 43.9 (5.3) 0.041* 0.796 0.091
P1 43.4 (6.0) 50.1 (9.2) 43.9 (5.3) 0.041* 0.796 0.091
P2 61.6 (7.5) 69.8 (7.7) 64.0 (7.0) 0.014* 0.410 0.046*
Ankle dorsiflexion (+)/plantarflexion ( )
T1 1.3 (3.6) 0.9 (5.7) 3.8 (7.0) 0.689 0.312 0.477
T2 3.8 (3.8) 1.6 (3.1) 3.4 (3.1) 0.070 0.832 0.155
T3 2.9 (5.4) 6.8 (4.8) 2.6 (4.2) 0.029* 0.689 0.033*
T4 17.4 (5.2) 12.7 (6.1) 17.6 (6.8) 0.018* 0.981 0.046*
P1 6.8 (3.7) 10.4 (4.3) 6.8 (4.1) 0.025* 0.981 0.033*
P2 3.8 (2.8) 5.0 (4.5) 3.0 (5.8) 0.290 0.290 0.328
Toe-out angles 6.3 (1.8) 10.2 (4.8) 9.6 (5.9) 0.02* 0.108 0.594

SLS: single limb stance; Pg: comparisons between the Pemberton group and the control group using U-test; Ps: comparisons between the two limbs in the Pemberton group
using Wilcoxon signed ranks test.
*
Significant difference (p < 0.05).

Please cite this article in press as: Chang C-F, et al. Residual gait deviations in adolescents treated during infancy for unilateral
developmental dysplasia of the hip using Pemberton’s osteotomy. Gait Posture (2012), doi:10.1016/j.gaitpost.2011.11.024
G Model
GAIPOS-3449; No. of Pages 6

4 C.-F. Chang et al. / Gait & Posture xxx (2012) xxx–xxx

Table 4
Means (standard deviations) of the joint moments (% of BM  LL) at three key events (T1: heel-strike; T2: beginning of SLS; T3: end of SLS) and the peak moments during the
stance phase (P1).

Control group Pemberton group Pg Ps

Affected Unaffected Affected Unaffected

Sagittal plane
Hip extensor (+)/flexor ( )
T1 4.8 (2.1) 3.0 (2.2) 3.2 (2.7) 0.051 0.151 0.724
T2 4.7 (2.0) 5.3 (2.9) 7.1 (2.8) 0.724 0.018* 0.016*
T3 5.2 (1.6) 4.6 (2.6) 4.4 (2.0) 0.452 0.151 0.929
P1 (extensor) 11.3 (3.2) 12.7 (6.0) 13.5 (6.4) 0.796 0.359 0.091
P1 (flexor) 9.7 (2.0) 7.0 (2.4) 7.8 (3.6) 0.008* 0.009* 0.657
Knee extensor (+)/flexor ( )
T1 3.2 (1.1) 2.7 (1.2) 3.1 (1.3) 0.064 0.076 0.477
T2 7.3 (2.3) 5.6 (1.8) 3.6 (2.4) 0.041* <0.001* 0.021*
T3 1.4 (1.2) 2.0 (1.7) 3.3 (2.5) <0.001* <0.001* 0.016*
P1 8.9 (2.7) 6.7 (1.8) 5.8 (2.1) 0.046* 0.004* 0.213
Ankle plantarflexor (+)/dorsiflexor ( )
T1 0.1 (0.1) 0.1 (0.1) 0.01 (0.3) 0.289 0.204 0.594
T2 0.5 (2.3) 0.4 (1.3) 1.5 (1.8) 0.259 0.023* 0.477
T3 15.4 (1.7) 16.1 (2.8) 16.0 (3.2) 0.906 0.796 0.534
P1 16.3 (1.6) 17.2 (2.7) 17.1 (3.4) 0.269 0.589 0.929
Frontal plane
Hip abductor (+)/adductor ( )
T1 1.0 (1.0) 0.5 (0.7) 0.4 (1.6) 0.180 0.151 0.790
T2 7.2 (1.3) 7.8 (1.9) 6.5 (3.0) 0.541 0.588 0.168
T3 8.8 (1.3) 7.3 (2.1) 7.4 (2.0) 0.041* 0.115 1.000
Knee abductor (+)/adductor ( )
T1 0.7 (0.4) 0.4 (0.5) 0.5 (0.4) 0.138 0.323 0.328
T2 1.4 (1.1) 1.5 (1.4) 1.1 (1.7) 0.869 0.981 0.213
T3 2.4 (1.0) 2.3 (0.9) 3.1 (1.4) 0.424 0.165 0.062
Ankle abductor (+)/adductor ( )
T1 0.05 (0.1) 0.02 (0.1) 0.03 (0.1) 0.396 0.796 0.919
T2 0.3 (0.5) 0.93 (0.9) 0.7 (0.9) 0.027* 0.078 0.286
T3 0.4 (1.2) 0.28 (1.4) 0.4 (2.0) 0.724 0.371 0.110

SLS: single limb stance; Pg: comparisons between the Pemberton group and the control group using U-test; Ps: comparisons between the two limbs in the Pemberton group
using Wilcoxon signed ranks test.
*
Significant difference (p < 0.05).

beginning of SLS where no significant between-limb differences were
found (Table 3).
Compared to the controls, the Pemberton group showed
significantly reduced peak flexor moments of both hips (Table
4). At the beginning of SLS, while both the affected and unaffected
limbs reduced extensor moments at the knee, the unaffected limb
had increased additionally the hip extensor and ankle plantarflexor
moments. Ankle adductor moment was also increased at the
affected limb. At the end of SLS, the Pemberton group increased
flexor moments at the knee in both limbs, with reduced hip
abductor moments in the affected limb. Compared to the affected
limb, the unaffected limb showed significantly greater hip
extensor and ankle plantarflexor moments but reduced knee
extensor moments at the beginning of SLS, and increased knee
flexor moments at the end of SLS (Table 4).
4. Discussion
At normal walking speeds, the Pemberton group displayed
significantly more anterior tilt, hiking at the affected side and
rotation towards the unaffected side of the pelvis, and a more
flexed posture at the knee and ankle in the affected limb. With this
asymmetrical gait pattern, they appeared to reduce the demands
on the muscles involved in the osteotomy in the affected limb, with
increased possibly compensatory effort from some of the muscle
groups of the unaffected limb. However, while reducing the
muscular contribution to the joint loads in the affected limb, the
observed residual gait deviations during late swing and early initial
DLS suggest a mechanism responsible for the greater loading rate
at the hip in both limbs [19].
Early open reduction with Pemberton’s osteotomy for unilateral
DDH appeared to be helpful for the overall gait efficiency as
indicated by the patients’ normal gait speeds and stride lengths
(Table 1). This is in contrast to patients with DDH who underwent
an osteotomy during their adolescence (mean age: 16.1 years) [13]
and during adulthood (mean age: 30 years) [12], who walked with
significantly reduced speed compared to normal. Apart from the
condition of the congruency of the articular surfaces at the treated
hip, the muscles involved in the surgery, such as hip abductors and
flexors, were also suggested to be relevant to the residual gait
deviations [13]. While gait speeds and stride lengths were
improved after early reduction, the Pemberton group was still
left with residual gait deviations that might not be beneficial to
their joints.
Deviations in the pelvic motions seemed to be a major
contributor to the residual gait deviations in the Pemberton
group. Increased anterior pelvic tilt helped to displace the body’s
center of mass (COM) anteriorly, and thus the lines of action of the
GRF, leading to a more flexed knee and ankle, and a significant
reduction in the peaks of the hip flexor and knee extensor
moments in the affected limb (Table 4). Increased pelvic hiking and
toe-out angle on the affected side helped displace the lines of
action of the GRF laterally, reducing the ipsilateral hip abductor
moments at the end of SLS, in agreement with the literature
[24,25]. On the other hand, greater pelvic rotation towards the
unaffected side also contributed to the increased ankle dorsiflexion
but helped maintain a normal step length of the affected limb.
These changes in the affected limb seem to be related to the
reduction in the demands on the muscles involved in the surgery.
After Pemberton’s osteotomy, weakness of the hip flexors and
abductors, e.g., iliopsoas and gluteus medius, as a result of surgical

Please cite this article in press as: Chang C-F, et al. Residual gait deviations in adolescents treated during infancy for unilateral
developmental dysplasia of the hip using Pemberton’s osteotomy. Gait Posture (2012), doi:10.1016/j.gaitpost.2011.11.024
G Model
GAIPOS-3449; No. of Pages 6
C.-F. Chang et al. / Gait & Posture xxx (2012) xxx–xxx
procedures [11,13,16–18,26] and/or altered mechanical properties
following bony realignment [13,26,27] may lead to a reduction of
the corresponding moments at the hip. Smaller than normal knee
extensor moments might also be necessary for weak knee
extensors, such as following the detachment and reflection of
the rectus femoris using an anterior approach [11,13,26].
Secondary kinematic and kinetic changes were found in the
Pemberton group in some of the non-surgically involved muscles
and joints, not only in the affected limb, but also in the unaffected
limb. Similar to the affected limb, increased anterior tilt of the
pelvis also led to increased knee flexion at the beginning of SLS and
reduced peaks of the hip flexor and knee extensor moments in the
unaffected limb (Tables 3 and 4). However, in contrast to the
affected limb, the additional increase in rotation towards the
unaffected side helped maintain a normal knee and ankle position
during late stance. Such pelvic deviations also helped draw the
body’s COM, and thus the line of action of the GRF, more anteriorly
during the end of SLS, increasing the knee flexor moments of the
affected limb while the unaffected limb also required an additional
increase of the hip extensor and ankle plantarflexor moments.
These pelvic deviations also contributed to other compensatory
changes, including increased ankle abduction with increased
adductor moments in the affected limb at the beginning of
SLS, and increased peak hip extension in the unaffected limb
(Tables 3 and 4).
Abnormal motion of the pelvis existed not only in patients
without treatment for unilateral DDH [7], but also in individuals
after osteotomy and THA for adult hip dysplasia [8,12], and in
children who had been treated using soft tissue release for
unilateral DDH [18]. While early intervention using Pemberton’s
osteotomy is considered beneficial for decreasing the sequela of
DDH and gait deviations, residual gait deviations were observed in
the current Pemberton group. The asymmetrical pelvic motion
appeared to be the primary change in response to the likely altered
muscular and joint conditions around the affected hip, with
obvious asymmetrical compensatory gait deviations in the limbs.
The residual deviations of the motion of the affected limb in the
Pemberton group may be responsible for the increased loading rate
at the hip in both limbs reported in the literature [19]. During mid-
swing, the affected limb significantly increased its peak knee
flexion (Table 3), which required a rapid knee extension to reach a
near-normal pose around heel-strike with significantly reduced
upward acceleration of the heel (Table 1). This reduced braking
ability of the affected limb subsequently led to increased braking
impulsive loading from the GRF. Around the heel-strike of the
affected limb, the unaffected limb maintained a normal motion
pattern, suggesting that the increased impact loadings in the
affected limb at this instant in time (Table 3) were not a result of
movement deviations of the unaffected limb. In contrast, the
affected limb showed significant deviations around the heel-strike
of the unaffected limb. These residual deviations appeared to
contribute to significantly shorter terminal DLS of the affected limb
(Table 1), indicating that these patients unloaded their affected
limb faster than did healthy controls [19], leading to an increased
impulsive braking force from the GRF of the unaffected limb. For
both limbs, the increased impulsive braking force from the GRF
may lead to increased loading rates of the axial forces at both hips
as increased loading rates of the GRF were correlated to increased
loading rates at the hips [19,28]. Since some hip muscles are often
involved in Pemberton’s osteotomy, and muscular response is too
slow to attenuate impulsive loads, increased loading rates at both
hips may expose the hips to a higher risk of premature OA [19].
While the current Pemberton group successfully reduced the hip
moments, and thus the muscular loads, the accompanying gait
deviations appeared to increase the harmful loading rates at both
hip joints.
Please cite this article in press as: Chang C-F, et al. Residual gait deviations in adolescents treated during infancy for unilateral
developmental dysplasia of the hip using Pemberton’s osteotomy. Gait Posture (2012), doi:10.1016/j.gaitpost.2011.11.024
5
In conclusion, although the Pemberton group showed a more
efficient gait with a nearly normal walking speed, they still
displayed asymmetrical residual gait deviations. These deviations
appeared to be a result of the altered pelvic motions in order to
reduce loads of those muscles in the affected hip that are often
involved during the surgery, despite the possibility of increasing
harmful loading rates at both hip joints [19]. Future studies may be
dedicated to evaluating the efficacy of pelvic control exercises and
strengthening of the hip flexors, knee extensors and hip abductors
on the improvement of gait patterns and the reduction of increased
loading rates at the hips in patients treated for unilateral DDH
using Pemberton’s osteotomy. The role of trunk motion in the
observed deviations also needs further investigation.
Acknowledgements
The authors gratefully acknowledge the support from the
National Science Council of Taiwan (NSC 90–2314-B-002–223).
Conflict of interest
We, Chu-Fen Chang, Ting-Ming Wang, Jyh-Horng Wang, Shier-
Chieg Huang and Tung-Wu Lu, declare that we have no proprietary,
financial, professional or other personal interest of any nature or
kind in any product, service and/or company that could be
construed as influencing the position presented in, or the view of,
the manuscript.
References
[1] Mellerowicz HH, Matussek J, Baum C. Long-term results of Salter and Chiari hip
osteotomies in developmental hip dysplasia. A survey of over 10 years follow-up
with a new hip evaluation score. Arch Orthop Trauma Surg 1998;117:222–7.
[2] Bowen JR, Kotzias-Neto A. Developmental dysplasia of the hip Maryland: data.
Trace Publishing Company; 2006.
[3] Michaeli DA, Murphy SB, Hipp JA. Comparison of predicted and measured
contact pressures in normal and dysplastic hips. Med Eng Phys 1997;19:180–6.
[4] Bohm P, Brzuske A. Salter innominate osteotomy for the treatment of devel-
opmental dysplasia of the hip in children results of seventy-three consecutive
osteotomies after twenty-six to thirty-five years of follow-up. J Bone Joint Surg
Am 2002;84-A:178–86.
[5] Lai, Lin KA, Su CJFC. Gait analysis of adult patients with complete congenital
dislocation of the hip. J Formos Med Assoc 1997;96:740–4.
[6] Pedersen ENG, Simonsen EB, Alkjaer T, Soballe K. Walking pattern in adults
with congenital hip dysplasia: 14 women examined by inverse dynamics. Acta
Orthop Scand 2004;75:2–9.
[7] Romano CL, Frigo C, Randelli G, Pedotti A. Analysis of the gait of adults who had
residua of congenital dysplasia of the hip. J Bone Joint Surg Am 1996;78:
1468–79.
[8] Lai KA, Lin CJ, Jou IM, Su FC. Gait analysis after total hip arthroplasty with leg-
length equalization in women with unilateral congenital complete dislocation of
the hip – comparison with untreated patients. J Orthop Res 2001;19:1147–52.
[9] Pedersen ENG, Alkjaer T, Soballe K, Simonsen EB. Walking pattern in 9 women
with hip dysplasia 18 months after periacetabular osteotomy. Acta Orthop
2006;77:203–8.
[10] Pemberton PA. Pericapsular osteotomy of the ilium for treatment of congenital
subluxation and dislocation of the hip. J Bone Joint Surg Am 1965;47:65–86.
[11] Morrissy RT, Weinstein SL. Lovell and Winter’s pediatric orthopaedics. Phila-
delphia: Lippincott Williams & Wilkins; 2006.
[12] Endo H, Mitani S, Senda M, Kawai A, McCown C, Umeda M, et al. Three-
dimensional gait analysis of adults with hip dysplasia after rotational acetab-
ular osteotomy. J Orthop Sci 2003;8:762–71.
[13] Sucato DJ, Tulchin K, Shrader MW, DeLaRocha A, Gist T, Sheu G. Gait, hip
strength and functional outcomes after a Ganz periacetabular osteotomy for
adolescent hip dysplasia. J Pediatr Orthop 2010;30:344–50.
[14] Benedetti MG, Montanari E, Catani F, Vicenzi G, Leardini A. Pre-operative
planning and gait analysis of total hip replacement following hip fusion.
Comput Meth Programs Biomed 2003;70:215–21.
[15] Catani F, Benedetti MG, Binazzi R, De Zerbi M, Leardini A, Giannini S. Func-
tional evaluation in total hip replacement patients. Chir Organi Mov
1998;83:349–57.
[16] Bassett GS, Engsberg JR, McAlister WH, Gordon JE, Schoenecker PL. Fate of the
psoas muscle after open reduction for developmental dislocation of the hip
(DDH). J Pediatr Orthop 1999;19:425–32.
[17] Konigsberg DE, Karol LA, Colby S, O’Brien S. Results of medial open reduction of
the hip in infants with developmental dislocation of the hip. J Pediatr Orthop
2003;23:1–9.
G Model
GAIPOS-3449; No. of Pages 6

6 C.-F. Chang et al. / Gait & Posture xxx (2012) xxx–xxx

[18] Omeroglu H, Yavuzer G, Bicimoglu A, Agus H, Tumer Y. No detectable major
changes in gait analysis after soft tissue release in DDH. Clin Orthop Rel Res
2008;466:856–61.
[19] Chang C-F, Wang T-M, Wang J-H, Huang S-C, Lu T-W. Adolescents after
Pemberton’s osteotomy for developmental dysplasia of the hip displayed
greater joint loading than healthy controls in affected and unaffected limbs
during gait. J Orthop Res 2011;29:1034–41.
[20] Lu T-W, Yen H-C, Chen H-L. Comparisons of the inter-joint coordination
between leading and trailing limbs when crossing obstacles of different
heights. Gait Posture 2008;27:309–15.
[21] Leardini A, Cappozzo A, Catani F, Toksvig-Larsen S, Petitto A, Sforza V, et al.
Validation of a functional method for the estimation of hip joint centre
location. J Biomech 1999;32:99–103.
[22] Dempster WT, Gabel WC, Felts WJ. The anthropometry of the manual work
space for the seated subject. Am J Phys Anthropol 1959;17:289–317.
[23] Grood ES, Suntay WJ. A joint coordinate system for the clinical description of
three-dimensional motions: application to the knee. J Biomech Eng
1983;105:136–44.
[24] Perry J. Pelvis and trunk pathological gait. In: Perry J, editor. Gait analysis:
normal and pathological function. Thorofare N.J.: Slack Incorporated; 1992.
pp. 265–279.
[25] Bowsher KA, Vaughan CL. Effect of foot-progression angle on hip joint
moments during gait. J Biomech 1995;28:759–62.
[26] Herring JA. Developmental dysplasia of the hip. In: Herring JA, editor.
Tachdjian’s pediatric orthopaedics: from the Texas Scottish Rite Hospital
for children. Philadelphia: Saunders/Elsevier; 2008. pp. 637–770.
[27] Rab GT. Biomechanical aspects of Salter osteotomy. Clin Orthop Relat Res
1978;132:82–7.
[28] Rohrle H, Scholten R, Sigolotto C, Sollbach W, Kellner H. Joint forces in the
human pelvis–leg skeleton during walking. J Biomech 1984;17:409–24.

Please cite this article in press as: Chang C-F, et al. Residual gait deviations in adolescents treated during infancy for unilateral
developmental dysplasia of the hip using Pemberton’s osteotomy. Gait Posture (2012), doi:10.1016/j.gaitpost.2011.11.024