Gait & Posture 40 (2014) 38–42

Contents lists available at ScienceDirect

Gait & Posture

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

Gait changes in individuals with bilateral hallux valgus reduce first

metatarsophalangeal loading but increase knee abductor moments
Kao-Shang Shih a,b,c, Hui-Lien Chien a, Tung-Wu Lu a,d,*, Chu-Fen Chang a,e,
Chien-Chung Kuo a,f
a
Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan, ROC
b
Department of Orthopedic Surgery, Shin Kong Wu Ho-Su Memorial Hospital, Taipei, Taiwan, ROC
c
School of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan, ROC
d
Department of Orthopaedic Surgery, School of Medicine, National Taiwan University, Taipei, Taiwan, ROC
e
Department of Physical Therapy, Tzu-Chi University, Hua-Lien, Taiwan, ROC
f
Department of Orthopedics, China Medical University Hospital, Taichung, Taiwan, ROC

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

Article history: Hallux valgus (HV), one of the most common foot pathologies in modern society, affects not only the foot
Received 22 July 2013 itself, but also the other lower limb joints. The purpose of the study was to investigate the kinematic and
Received in revised form 26 January 2014 kinetic changes in the lower limb joints in patients with bilateral HV during level walking. Twelve female
Accepted 9 February 2014
patients with bilateral HV and 12 healthy female controls walked while three-dimensional kinematic
and kinetic data were measured. Patients with HV were found to shift their center of pressure (COP)
Keywords: laterally away from the 1st metatarsophalangeal joint (MPJ), which helped unload the joint during late
Gait analysis
stance. The lateral shift of the COP in these patients was associated with the reduced toe-out angles of the
Center of pressure
foot as a result of increased internal rotation of the hip. However, this strategy increased the abductor
Knee abductor moment, Toe-in gait, Knee
OA moments at the knee, an index closely correlated with the medial load at the knee and a predictor of the
onset and progression of medial OA. Early treatment of HV may be helpful not only for reducing foot pain
and deformity, but also for preventing the potentially harmful loading at the knee, especially in those at
risk of medial knee OA.
ß 2014 Elsevier B.V. All rights reserved.

1. Introduction
Hallux valgus (HV) is characterized by a lateral deviation of the
hallux relative to the first metatarsal. Being one of the most
common foot pathologies in modern society [1], HV affects
predominantly the female population with reported prevalence
estimates ranging from 30% to 58% as opposed to 13–25% in men
[2,3]. HV deformity usually occurs together with a decrease in
height of the transverse arch, a widening of the forefoot and altered
pressure distribution under the metatarsal heads [4]. Both foot and
knee pain have been reported with HV deformity [5]. The presence
of HV increases disability levels in women with knee osteoarthritis
(OA) [6] and HV was linked to the development of knee OA in a case
study [7]. These findings suggest that HV may affect not only the
* Corresponding author at: Institute of Biomedical Engineering, National Taiwan
University, No. 1, Sec. 1, Jen-Ai Road, Taipei 100, Taiwan, ROC.
Tel.: +886 2 33653335; fax: +886 2 33653335.
E-mail address: twlu@ntu.edu.tw (T.-W. Lu).
foot, but other joints, too, especially the knee. However, the
underlying mechanism has not been clearly described.
Previous studies on patients with HV have focused mainly on
the measurements of the plantar pressure distributions [8,9].
However, since the plantar pressure distributions are affected
directly by the motions of the proximal segments/joints, a more
complete picture of the compensatory changes in response to HV
deformity would be difficult to obtain without knowledge of the
kinematic and kinetic changes in the proximal segments.
Previous gait studies on patients with HV have focused mainly
on the description of the kinematics of the foot and ankle complex
[10,11]. Only a limited number of studies reported the influence of
HV deformity on the kinematics of the proximal segments/joints
[4,12,13]. One such study was on patients with a juvenile form of
hallux valgus, which is pathologically different from adult-onset
HV [12]. The patient group was found to show greater hip and knee
flexion during swing phase, lesser knee extension and greater
plantar flexion at the beginning of the stance phase. Another study
reported the motions of the lower limb joints in the sagittal and
frontal planes of six patients with HV [4]. The patient group

http://dx.doi.org/10.1016/j.gaitpost.2014.02.011
0966-6362/ß 2014 Elsevier B.V. All rights reserved.
 K.-S. Shih et al. / Gait & Posture 40 (2014) 38–42
showed increased peak ankle plantarflexion at the beginning of the
stance phase and peak knee extension at the end of swing but
reduced peak hip abduction throughout the gait cycle. These
studies did not provide the relationships between HV and the
kinetic changes in the proximal joints. Komatsu et al. [13] reported
the kinetic changes in the joints of the lower extremities pre- and
post-operative treatment of the HV, but it was a case report on an
older patient with RA and HV. From a mechanical perspective,
altering the kinematics and kinetics at one joint can influence the
mechanics of other joints in a linkage [14]. Changes of the foot
progression angle, such as a result of foot diseases, also affect the
center of pressure (COP) and the mechanics of the proximal joints
[15]. However, no study has reported a complete account of the
influence of HV on the kinematics and kinetics of the proximal
joints in relation to those of the foot (e.g., progression angle and
COP position) during gait. A combined kinematic and kinetic
analysis would be helpful for a better understanding of the
movement strategies and compensations in response to HV
deformity.
The purpose of the study was to investigate the kinematic and
kinetic changes of the foot and the lower limb joints in patients
with bilateral HV as compared to matched healthy subjects during
level walking. It was hypothesized that patients with bilateral HV
would adopt compensatory gait patterns to reduce the loading to
the first hallux, resulting in increased loads at the knee.
2. Materials and methods
2.1. Subjects
Twelve female patients with bilateral HV (age (mean  standard
deviation or SD): 45.5  9.2 years; height: 157.0  4.3 cm; mass:
55.7  8.4 kg) and twelve age and sex-matched healthy controls (age:
46.8  9.8 years; height: 159.3  4.8 cm; mass: 63.6  14.4 kg)
participated in the study. An a priori power analysis based on pilot
results using GPOWER [16] determined that 11 subjects per group
would yield a power of 0.8 at a significance level of 0.05. Informed
written consent as approved by the Institutional Research Board was
obtained for all participants. Hallux valgus was diagnosed using
dorso-plantar weight-bearing X-rays, i.e., hallux valgus angle
(HVA) 3 158 and intermetatarsal angle (IMA) 3 98 [17]. Patients
with knee osteoarthritis (OA) were excluded. The mean (SD) HVA and
IMA of the patients were 30.5 (4.7)8 and 15.2 (3.4)8, respectively.
Healthy subjects were free from any neuromusculoskeletal pathology
that might have affected gait. For the patients with HV, foot pain was
assessed using 100-mm visual analog scales (VAS; 0 = no pain,
100 = worst imaginable pain).
2.2. Data collection
In a gait laboratory, each subject walked barefoot at a self-
selected pace on an 8-m walkway. The subjects wore T-shirt and
shorts, which were wrapped when necessary, to expose anatomi-
cal landmarks for the placement of the markers. Twenty-eight
retro-reflective skin markers were used to track the motion of the
pelvis (ASISs and PSISs), and thighs (greater trochanter, mid-thigh,
medial and lateral epicondyles), shanks (head of fibula, tibial
tuberosity, medial and lateral malleolus) and feet (navicular
tuberosity, fifth metatarsal base, heel and big toe) [18]. Additional
markers were also placed on the first and second metatarsal heads.
Three-dimensional trajectories of the markers were measured
using a 7-camera motion capture system (Vicon 512, Oxford
Metrics, UK) at a sampling rate of 120 Hz, and the ground reaction
force (GRF) was gathered from two forceplates (AMTI, USA) at a
frequency of 1080 Hz [18]. Six successful trials were obtained for
each subject.
39
2.3. Data analysis
The pelvis-leg apparatus was modeled as a 7-link system [18].
Coordinates of the markers gathered during a static calibration
trial were used to define the anatomical coordinate system of each
of the links (body segments), with the positive x-axis directed
anteriorly, the positive y-axis superiorly and the positive z-axis to
the right. A Cardanic rotation sequence (z–x–y) was used to
describe the rotational movements of each joint. In order to
minimize the errors owing to skin movement artifacts, a global
optimization method was used [19]. With the measured GRF and
kinematic data, inverse dynamics were used to calculate the
internal moments at the lower limb joints, which were normalized
to body weight (BW) and leg length (LL), the latter defined as the
distance between the ipsilateral ASIS and medial malleolus
measured using a tape measure. Inertial properties for each body
segment were obtained using an optimization method [20]. All the
modeling and inverse dynamics analysis procedures were imple-
mented in MATLAB (version 7.10, MathWorks, Inc., USA).
Gait speed, cadence, step width and stride length were
calculated. Stride length was defined as the distance between
the positions of the heel marker at two consecutive heel-strikes of
the same foot. Foot progression angle (FPA) was calculated as the
angle between the line of progression and the line connecting the
heel and the second metatarsal head [21]. A positive FPA indicated
toe-out. For subsequent statistical analysis, values of all the joint
angles and moments and FPA at heel-strike and toe-off of the
contralateral leg and toe-off of the ipsilateral leg were extracted for
each trial. These gait instances were selected because they are
related to the forefoot rocker, which is relevant to HV. Values were
also averaged over the stance phase for each of the joint angle and
moment component and FPA for each trial. Stance phase was the
period from ipsilateral heel-strike to toe-off defined using the
measured ground reaction forces. The positions of the point of
application of the GRF, also called the center of pressure (COP),
were calculated using the data measured by the forceplates. The
distance between the COP and center of the first metatarsopha-
langeal joint (MPJ) was calculated when the COP passed the line
joining the 1st and 2nd MPJ. This COP-MPJ distance was divided by
the distance between the 1st and 2nd MPJ to give the normalized
COP-MPJ distance. The extracted values and average values over
the stance phase were averaged across the trials and limbs for all
variables for each subject. All the variables calculated were tested
for normality using the Shapiro–Wilk test. Comparisons of the
variables between the HV and control groups were performed
using independent t-tests. Between-foot comparison of the VAS in
the HV group was performed using paired t-test. Pearson’s
correlation analysis was also performed to establish the correlation
between the foot progression angle and normalized COP-MPJ
distance. A correlation coefficient of 0.75 and higher indicates a
high to excellent correlation; 0.50–0.75 a moderate correlation,
0.25–0.5 a fair correlation, and 0.00–0.25 a poor or no correlation
[22]. All significance levels were set at a = 0.05 (SPSS, version 17.0,
SPSS Inc., USA).
3. Results
All the calculated variables were normally distributed. No
significant difference in the mean VAS for pain was found between
left (mean: 55.6 mm, SD: 5.8) and right foot (mean: 54.4 mm, SD:
7.7) in the HV group. No significant differences were found
between the two groups in age, body height and mass, and any
temporal-distance parameters (Table 1).
Moderate negative correlations were found between normal-
ized COP-MPJ distance and the foot toe-out angle during the gait
cycle (Table 2). The normalized COP-MPJ distances in the HV group
40 K.-S. Shih et al. / Gait & Posture 40 (2014) 38–42

Table 1 and greater ankle pronator and external rotator moments at toe-off
Means (standard deviations) of temporal-distance parameters in the hallux valgus
(Table 4).
and control groups.

Parameters Hallux valgus Control p

4. Discussion
Gait speed (m/s) 1.06 (0.10) 1.12 (0.86) 0.137
Stride length (m) 1.12 (0.62) 1.16 (0.57) 0.125 The purpose of the study was to investigate the kinematic and
Step width (cm) 8.13 (1.87) 8.34 (4.20) 0.875
kinetic changes of the lower limb joints during level walking in
Stride time (s) 1.07 (0.52) 1.05 (0.48) 0.329
Cadence (steps/min) 108.7 (7.5) 108.2 (17.0) 0.918 patients with bilateral HV as compared to age-matched healthy
Stance time (%) 64.8 (1.7) 63.9 (1.0) 0.119 controls. The HV group was found to adopt compensatory changes,
Double leg suport time (%) 13.5 (2.3) 12.8 (1.0) 0.344 primarily increased hip internal rotation and reduced toe-out
Single leg support time (%) 38.4 (2.6) 37.2 (2.6) 0.290
angle, shifting the GRF away from the 1st MPJ at the expense of
significantly increased knee abductor moments. The knee abductor
were also significantly greater than those in the controls (Table 2). moment is an often-used predictor of knee joint loading and has
Patients with HV showed significantly reduced toe-out angles been associated with the onset and progression of OA [23]. The
throughout the stance phase with significantly increased mean results suggest that early treatment of HV may be helpful not only
internal rotations at the hip (Tables 2 and 3). Patients with HV also to reduce foot pain and deformity [24], but also to prevent
showed significantly reduced flexion or increased extension at the potentially harmful loading at the knee, especially in those at risk
hip at the key instances and significantly reduced knee abductions of medial knee OA.
at toe-off and contralateral initial contact (Table 3). Patients with HV were found to shift their GRF laterally away
For joint moments, the HV group sustained significantly greater from the 1st MPJ as indicated by the increased distance between
mean knee abductor moments during stance phase, significantly the COP and 1st MPJ, which helped reduce the loading and
greater knee internal rotator moments at contralateral heel-strike, associated pain at the 1st MPJ during late stance. The lateral shift of

Table 2
Means (standard deviations) of the normalized COP-MPJ distances and their correlations with the foot progression angle at toe-off (TO), contralateral toe-off (CTO), and
contralateral heel-strike (CHS), as well as their average values during stance phase in the hallux valgus (HV) and control groups.

Hallux valgus Control p Correlations

r p

Normalized COP-MPJ distances (%) 80.90 (6.91) 71.12 (11.37) 0.018*

Foot progression angle: toe-out (+)/toe-in ( ) angle (8)

CTO 1.1 (4.0) 7.8 (4.2) 0.001* 0.543 0.006*
CHS 2.8 (3.8) 11.4 (6.4) 0.001* 0.586 0.003*
TO 2.8 (8.9) 15.5 (11.7) 0.007* 0.414 0.044*
Average 2.0 (3.9) 9.5 (5.3) 0.001* 0.577 0.003*
*
p < 0.05.

Table 3
Means (standard deviations) of the joint angles at toe-off (TO), contralateral toe-off (CTO), and contralateral heel-strike (CHS), as well as their average values during stance
phase in the hallux valgus (HV) and control groups. Units:8.

Hip Knee Ankle

HV Control p HV Control p HV Control p

Flexion (+)/extension ( ) Flexion (+)/extension ( ) Dorsiflexion (+)/plantarflexion

CTO 19.3 (4.2) 23.3 (5.7) 0.061 15.1 (5.9) 15.8 (6.8) 0.767 4.3 (2.8) 2.4 (2.4) 0.101
CHS 11.6 (4.5) 7.2 (5.2) 0.035* 13.1 (4.4) 14.6 (4.2) 0.385 3.9 (3.1) 5.1 (2.5) 0.314
TO 5.2 (4.1) 1.7 (4.0) 0.045* 43.2 (5.8) 42.2 (4.5) 0.630 16.1 (4.6) 14.4 (4.2) 0.373

Average 3.6 (4.2) 8.6 (4.5) 0.010* 14.2 (5.2) 15.2 (5.1) 0.636 0.3 (2.5) 1.1 (2.1) 0.170

Hip Knee Ankle

HV Control p HV Control p HV Control p

Adduction (+)/abduction ( ) Adduction (+)/abduction ( ) Supination (+)/pronation ( )

CTO 6.4 (2.7) 5.5 (2.4) 0.407 3.2 (2.0) 4.2 (3.3) 0.382 0.5 (1.7) 0.7 (1.9) 0.792
CHS 5.9 (2.9) 5.3 (2.2) 0.562 2.8 (1.9) 5.2 (2.8) 0.023* 1.6 (2.6) 0.5 (1.9) 0.227
TO 0.7 (2.6) 2.5 (2.5) 0.097 10.1 (3.5) 13.9 (5.1) 0.046* 8.8 (3.4) 8.2 (3.7) 0.690

Average 5.7 (2.0) 4.9 (1.5) 0.257 2.9 (1.7) 4.7 (2.8) 0.080 0.6 (1.7) 0.1 (1.5) 0.411

Hip Knee Ankle

HV Control p HV Control p HV Control p

Internal (+)/external ( ) rotation Internal (+)/external ( ) rotation Internal (+)/external ( ) rotation

CTO 0.2 (2.2) 1.8 (3.1) 0.078 2.0 (2.1) 3.4 (3.0) 0.188 1.1 (1.4) 1.2 (1.9) 0.909
CHS 1.8 (3.1) 0.8 (1.9) 0.345 3.2 (2.7) 3.2 (3.0) 0.955 2.1 (1.6) 2.7 (1.7) 0.376
TO 2.8 (2.9) 5.0 (4.4) 0.147 5.6 (2.9) 6.9 (3.4) 0.338 6.1 (1.7) 6.7 (1.9) 0.454

Average 0.6 (2.3) 1.3 (1.9) 0.036* 0.2 (1.6) 0.8 (2.1) 0.463 0.8 (1.2) 1.1 (1.6) 0.586
*
p < 0.05.
 K.-S. Shih et al. / Gait & Posture 40 (2014) 38–42 41

Table 4
Means (standard deviations) of the joint moments at toe-off (TO), contralateral toe-off (CTO), and contralateral heel-strike (CHS), as well as their average values during stance
phase in the hallux valgus (HV) and control groups. Units: %BW
LL.

Hip Knee Ankle

HV Control. p HV Control p HV Control p

Extensor (+)/flexor ( ) Extensor (+)/flexor ( ) Plantarflexor (+)/dorsiflexor ( )

CTO 3.92 (2.10) 4.96 (1.75) 0.203 6.47 (2.14) 6.17 (1.94) 0.718 0.30 (0.80) 0.55 (0.70) 0.430
CHS 4.36 (1.82) 4.41 (3.79) 0.969 1.49 (1.36) 2.23 (2.33) 0.349 14.31 (1.01) 13.97 (0.85) 0.384
TO 3.10 (1.12) 3.39 (1.01) 0.508 0.29 (0.25) 0.24 (0.42) 0.754 0.07 (0.11) 0.13 (0.13) 0.287

Average 0.52 (1.23) 0.14 (2.15) 0.366 2.60 (1.13) 2.76 (1.28) 0.744 5.85 (0.91) 5.91 (0.62) 0.834

Hip Knee Ankle

HV Control. p HV Control p HV Control p

Abductor (+)/adductor ( ) Abductor (+)/adductor ( ) Pronator (+)/supinator ( )

CTO 7.59 (1.13) 7.63 (2.01) 0.950 2.01 (1.34) 0.84 (0.88) 0.020* 0.40 (0.32) 0.30 (0.29) 0.454
CHS 9.86 (0.82) 10.43 (1.33) 0.225 3.64 (1.54) 2.25 (1.14) 0.019* 0.37 (1.16) 0.14 (0.92) 0.596
TO 0.96 (0.39) 1.32 (0.62) 0.107 0.36 (0.14) 0.43 (0.18) 0.295 0.01 (0.03) 0.03 (0.04) 0.007*

Average 6.78 (0.41) 7.16 (0.64) 0.099 2.26 (0.98) 1.30 (0.69) 0.011* 0.35 (0.55) 0.28 (0.44) 0.742

Hip Knee Ankle

HV Control. p HV Control p HV Control p

Internal (+)/external ( ) rotator Internal (+)/external ( ) rotator Internal (+)/external ( ) rotator

CTO 3.07 (0.87) 2.81 (0.75) 0.439 0.21 (0.19) 0.13 (0.26) 0.388 0.03 (0.25) 0.08 (0.24) 0.298
CHS 1.44 (0.59) 0.99 (0.82) 0.135 1.21 (0.57) 0.79 (0.36) 0.040* 3.09 (1.20) 3.60 (1.31) 0.340
TO 0.62 (0.21) 0.73 (0.22) 0.222 0.05 (0.04) 0.04 (0.04) 0.658 0.02 (0.05) 0.02 (0.04) 0.048*

Average 1.60 (0.43) 1.39 (0.43) 0.253 0.40 (0.25) 0.26 (0.17) 0.126 1.17 (0.50) 1.41 (0.57) 0.281
*
p < 0.05.

the COP away from the 1st MPJ in patients with HV was moderately
associated with the reduced toe-out angles of the foot as a result of
increased internal rotation of the hip (Fig. 1A). Greater out-toe
angles were shown to increase the medial pressure of the foot
during gait in various patient groups in the literature [15,25]. For
example, increased toe-out angle tends to induce a higher load on
the medial foot, and reduced toe-out angle was found to shift the
loading to the lateral side of the foot in children with cerebral palsy
[(Fig._1)TD$IG][25]. Foot progression angle was also found to be significantly
correlated with the time spent on the medial side of the foot in
individuals with diabetes mellitus (DM), peripheral neuropathy
(PN), or a forefoot ulcer [15]. In the current study, patients with HV
appeared to adopt primarily a hip internal rotation and toe-in
strategy to shift the GRF away from the 1st MPJ to reduce the
loading at the joint.
While a reduced toe-out angle helped reduce the load at the 1st
MPJ in the current patients with HV, they were found to sustain
increased abductor moments at the knee, similar in magnitude to

Fig. 1. (A) Hip internal rotation (uhip) reduces toe-out angle (utoe-out). (B) Reduced toe-out angle shifts the ground reaction force (GRF) vector away from the knee joint center
and thus increases the GRF moment arm to the knee joint center, and thus the knee abductor moment. Gray and black lines indicate conditions before and after reducing the
toe-out angle.
42 K.-S. Shih et al. / Gait & Posture 40 (2014) 38–42
those in patients with mild knee OA [18]. From a mechanical
viewpoint considering the alignment of the lower extremity,
increased toe-out angle will shift the GRF vector laterally toward
the knee joint center and thus decreases the moment arm about the
knee, thereby reducing the knee abductor moment (Fig. 1B).
Previous studies have shown that increased toe-out angle reduces
the knee abductor moment during the late stance phase of gait in
individuals with healthy knees and in those with knee OA of a wide
range of severity [21,26]. In contrast to patients with knee OA, the
current patients with HV reduced the toe-out angles in order to shift
the GRF away from the first MPJ at the price of increased abductor
moments at the knee. Since knee abductor moments are closely
correlated with the medial load at the knee and are a predictor of
medial OA progression [23], prolonged increased knee abductor
moments in patients with HV may be harmful to the health of the
cartilage and may contribute to the development of medial OA.
Therefore, early treament for HV is needed to prevent overloading at
knee joints, especially in those at risk of medial knee OA.
Reduced toe-out angle via increased internal rotation of the hip
may also have negative effects on the patellofemoral joint.
Excessive hip internal rotation is thought to contribute to
lower-extremity torsional and patellofemoral joint malalignment
[27], as well as increased patellofemoral joint stress [28], which are
related to the development of anterior and peripatellar knee pain.
Patients with unilateral patellofemoral pain syndrome were found
to have significantly different HV angles between affected and
unaffected sides [29]. It appears that increased hip internal
rotation may increase the risk of patellofemoral pain syndrome in
patients with HV. However, quantification of the effects of HV on
the dynamic alignment of the patellofemoral joint using advanced
measurement techniques such as 3D fluoroscopy methods [30] is
needed to provide more direct evidence.
The current study described the compensatory changes patients
adopted in response to bilateral HV, and the accompanied potential
negative effects at the knee. However, since walking involves both
limbs, whether the same strategy would be adopted by patients with
unilateral HV requires further investigation. Further study is also
needed to quantify the effects of the severity of HV on the primary
and secondary changes in the lower limb joints.
5. Conclusions
Participants with HV were found to increase the internal
rotation at the hip, thus reducing the toe-out angle of the foot. This
shifted the GRF laterally, away from the 1st MPJ, at the expense of
increased knee abductor moments, a known stressor associated
with the onset and progression of OA. It appears that early
treatment of HV may be helpful not only to reduce the foot pain
and deformity, but also to prevent the potentially harmful loading
at the knee, especially in those at risk of medial knee OA.
Acknowledgement
The authors are grateful for the financial support from the
National Science Council of Taiwan (NSC 100-2221-E-341-001)
and China Medical University Hospital (DMR-97-073).
Conflict of interest statement
The authors declare that they 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 the manuscript.
References
[1] Eustace S, O’Byrne J, Stack J, Stephens MM. Radiographic features that enable
assessment of first metatarsal rotation: the role of pronation in hallux valgus.
Skeletal Radiol 1993;22:153–6.
[2] Nix S, Smith M, Vicenzino B. Prevalence of hallux valgus in the general
population: a systematic review and meta-analysis. J Foot Ankle Res
2010;3:21.
[3] Nguyen US, Hillstrom HJ, Li W, Dufour AB, Kiel DP, Procter-Gray E, et al.
Factors associated with hallux valgus in a population-based study of older
women and men: the MOBILIZE Boston Study. Osteoarthritis Cartilage
2010;18:41–6.
[4] Kozáková J, Janura M, Svoboda Z, Elfmark M, Klugar M. The influence of hallux
valgus on pelvis and lower extremity movement during gait. Acta Univ Palacki
Olomuc Gymn 2011;41:49–54.
[5] Roddy E, Zhang W, Doherty M. Prevalence and associations of hallux valgus in a
primary care population. Arthritis Rheum 2008;59:857–62.
[6] Guler H, Karazincir S, Turhanoglu AD, Sahin G, Balci A, Ozer C. Effect of
coexisting foot deformity on disability in women with knee osteoarthritis. J
Am Podiatr Med Assoc 2009;99:23–7.
[7] Ozguclu E, Kilic E, Kaymak B. A knee osteoarthritis connected with hallux
valgus-related pes planus. J Biomech 2008;41:3523–4.
[8] Yamamoto H, Muneta T, Asahina S, Furuya K. Forefoot pressures during
walking in feet afficted with hallux valgus. Clin Orthop Rel Res
1996;323:247–53.
[9] Plank MJ. The pattern of forefoot pressure distribution in hallux valgus. Foot
1995;5:8–14.
[10] Canseco K, Long J, Smedberg T, Tarima S, Marks RM, Harris GF. Multisegmental
foot and ankle motion analysis after hallux valgus surgery. Foot Ankle Int
2012;33:141–7.
[11] Deschamps K, Birch I, Desloovere K, Matricali GA. The impact of hallux valgus
on foot kinematics: a cross-sectional, comparative study. Gait Posture
2010;32:102–6.
[12] Miroslav J, Lee C, Zdenek S, Jitka K, Anna G. Kinematic analysis of gait in
patients with juvenile hallux valgus deformity. J Biomech Sci Eng 2008;3:
390–8.
[13] Komatsu K, Nakao S, Masuda Y, Okahisa T, Kondou S, Nishikawa K, et al. Effect
of hallux valgus during gait. Rigakuryoho Kagaku 2011;26:717–22.
[14] Guo M, Axe MJ, Manal K. The influence of foot progression angle on the knee
adduction moment during walking and stair climbing in pain free individuals
with knee osteoarthritis. Gait Posture 2007;26:436–41.
[15] Hastings MK, Gelber JR, Isaac EJ, Bohnert KL, Strube MJ, Sinacore DR. Foot
progression angle and medial loading in individuals with diabetes mellitus,
peripheral neuropathy, and a foot ulcer. Gait Posture 2010;32:237–41.
[16] Erdfelder E, Faul F, Buchner A. GPOWER: a general power analysis program.
Behav Res Methods Instrum C 1996;28:1–11.
[17] Coughlin MJ. Hallux valgus. J Bone Joint Surg Am 1996;78:932–66.
[18] Huang SC, Wei IP, Chien HL, Wang TM, Liu YH, Chen HL, et al. Effects of severity
of degeneration on gait patterns in patients with medial knee osteoarthritis.
Med Eng Phys 2008;30:997–1003.
[19] Lu TW, O’Connor JJ. Bone position estimation from skin marker co-ordinates
using global optimisation with joint constraints. J Biomech 1999;32:129–34.
[20] Chen SC, Hsieh HJ, Lu TW, Tseng CH. A method for estimating subject-specific
body segment inertial parameters in human movement analysis. Gait Posture
2011;33:695–700.
[21] Rutherford DJ, Hubley-Kozey CL, Deluzio KJ, Stanish WD, Dunbar M. Foot
progression angle and the knee adduction moment: a cross-sectional investi-
gation in knee osteoarthritis. Osteoarthritis Cartilage 2008;16:883–9.
[22] Portney LG, Watkins MP. Foundations of clinical research: applications to
practice. Upper Saddle River, N.J.: Pearson/Prentice Hall; 2009.
[23] Andriacchi TP. Dynamics of knee malalignment. Orthop Clin N Am
1994;25:395–403.
[24] Robinson AH, Limbers JP. Modern concepts in the treatment of hallux valgus. J
Bone Joint Surg Br 2005;87:1038–45.
[25] Chang WN, Tsirikos AI, Miller F, Schuyler J, Glutting J. Impact of changing foot
progression angle on foot pressure measurement in children with neuromus-
cular diseases. Gait Posture 2004;20:14–9.
[26] Chang A, Hurwitz D, Dunlop D, Song J, Cahue S, Hayes K, et al. The relationship
between toe-out angle during gait and progression of medial tibiofemoral
osteoarthritis. Ann Rheum Dis 2007;66:1271–5.
[27] Souza RB, Powers CM. Predictors of hip internal rotation during running: an
evaluation of hip strength and femoral structure in women with and without
patellofemoral pain. Am J Sports Med 2009;37:579–87.
[28] Lee K-B, Hur C-I, Chung J-Y, Jung S-T. Outcome of unilateral versus simulta-
neous correction for hallux valgus. Foot Ankle Int 2009;30:120–3.
[29] Kaya D, Atay OA, Callaghan MJ, Cil A, Caglar O, Citaker S, et al. Hallux valgus in
patients with patellofemoral pain syndrome. Knee Surg Sports Traumatol
Arthrosc 2009;17:1364–7.
[30] Tsai TY, Lu TW, Chen CM, Kuo MY, Hsu HC. A volumetric model-based 2D to 3D
registration method for measuring kinematics of natural knees with single-
plane fluoroscopy. Med Phys 2010;37:1273–84.