The Knee 21 (2014) 688–693

Contents lists available at ScienceDirect

The Knee

Effect of lower limb malalignment in the frontal plane on transverse

plane mechanics during gait in young individuals with varus
knee alignment
Felix Stief a,⁎, Harald Böhm b, Chakravarthy U. Dussa b, Christel Multerer b, Ansgar Schwirtz c,
Andreas B. Imhoff d, Leonhard Döderlein b
a
Orthopedic University Hospital Friedrichsheim gGmbH, Frankfurt/Main, Germany
b
Orthopedic Hospital for Children, Aschau i. Chiemgau, Germany
c
Department of Biomechanics in Sport, Technical University, Munich, Germany
d
Department of Orthopedic Sports Medicine, Technical University, Munich, Germany

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

Article history: Background: Varus knee alignment has been identified as a risk factor for the progression of medial knee osteo-
Received 20 June 2013 arthritis (OA). This study tested the hypothesis that not only frontal plane kinematics and kinetics but also trans-
Received in revised form 4 March 2014 verse plane lower extremity mechanics during gait are affected by varus malalignment of the knee.
Accepted 11 March 2014 Methods: Eighteen, otherwise healthy children and adolescents with varus malalignment of the knee were stud-
ied to examine the association between static varus malalignment and functional gait parameters. Kinematic data
Keywords:
were collected using a Vicon motion capture system (Vicon Motion Systems, Oxford, UK). Two AMTI force plates
Gait analysis
Varus malalignment
(Advanced Mechanical Technology, Inc., Watertown, MA, USA) were used to collect kinetic data.
Knee osteoarthritis Results: The results indicated that changes in transverse plane mechanics occur concomitantly with changes in
Knee adduction moment knee malalignment in the frontal plane. A mechanical consequence of varus knee malalignment is obviously
Internal tibia rotation an increased endorotation of the foot (internal foot placement) and an increased internal knee rotation (tibia
rotation) during stance phase. The linear correlation between the maximum external knee adduction moment
in terminal stance and the internal knee rotation in terminal stance (r = 0.823, p b 0.001) shows that this trans-
verse plane gait mechanics is directly in conjunction with intrinsic compressive load on the medial compartment
during gait.
Conclusions: Understanding factors that influence dynamic knee joint loading in healthy, varus malaligned knees
may help us to identify risk factors that lead to OA. Thus, three-dimensional gait analysis could be used for clinical
prognoses regarding the onset or progression of medial knee OA.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction adduction moment is used as the primary parameter for characterizing

Mechanical factors, such as varus malalignment of the knee in the
frontal plane, have been implicated in the progression of medial knee
OA [1–6]. A recent study by Hayashi et al. [6] has shown that varus
knee alignment is associated with increased risk of incidence and
enlarging bone marrow lesions. Quantitative gait analysis as an adjunct
to static radiographic measures of alignment provides an estimate of
tibiofemoral load through an inverse dynamic analysis and can be
used to gain a better understanding of the biomechanical factors that
contribute to the pathogenesis of knee OA [7,8]. The external knee
⁎ Corresponding author at: Orthopedic University Hospital Friedrichsheim gGmbH,
Marienburgstraße 2, 60528 Frankfurt/Main, Germany. Tel.: +49 69 6705 862.
E-mail addresses: felix.stief@gmx.de, f.stief@friedrichsheim.de (F. Stief).
intrinsic compressive load and articular cartilage degeneration in the
medial knee compartment during gait [4,5,9,10].
Previous studies have investigated biomechanical changes during
gait in the sagittal and frontal planes in patients with knee OA compared
to healthy controls [5,11,12]. For example, patients with medial knee OA
make initial contact with the ground with the knee in a more extended
position and have a smaller range of knee flexion during the stance
phase of walking [11]. However, these studies on patients with
established knee OA have made it difficult to determine if biomechani-
cal changes are involved in the development of disease, are in response
to degenerative changes in the joint, or are compensatory mechanisms
in response to these degenerative changes or to joint pain.
In children and adolescents with varus malalignment of the knee
without signs of knee OA abnormally increased knee internal rotation
and hip external rotation moments were detected [13]. While transverse

http://dx.doi.org/10.1016/j.knee.2014.03.004
0968-0160/© 2014 Elsevier B.V. All rights reserved.
 F. Stief et al. / The Knee 21 (2014) 688–693
plane mechanics can initiate degenerative changes to the articular carti-
lage and have been implicated in the progression of knee OA [10], there is
a lack of research on the relationship between static varus malalignment
of the knee and transverse plane lower extremity mechanics during gait.
In particular, it is still not known which role the rotational gait profile
plays in the progression of knee OA. The purpose of the present study
was therefore to investigate the effect of pathological varus alignment
of the knee on transverse plane gait parameters of the lower extremity
in young individuals without signs of knee OA. The following question
should be answered and may have important implications for the pro-
gression of knee OA: Does a relationship exist between static lower ex-
tremity malalignment in the frontal plane and transverse plane lower
extremity mechanics during gait? We hypothesized that (1) the static
varus alignment of the knee correlates with internal knee rotation (tibial
rotation) and endorotation of the foot (internal foot placement) during
the stance phase of gait, and (2) the rotational gait profile has an influ-
ence on the external knee adduction moment and thus also leads to de-
generative changes in the knee joint in young patients with varus
malalignment of the knee.
2. Methods
2.1. Subjects
Eighteen, otherwise healthy children and adolescents with varus
malalignment of the knee 12–19 years of age were consecutively select-
ed during clinical visits between January 2008 and October 2012
(Table 1). Solely patients were included with a clinical indication for a
full-length standing anteroposterior radiograph and a pathological
varus alignment of at least one knee according to the mechanical axis
angle (MAA) of the lower limb [14]. Alignment was defined as patholog-
ical varus when the angle was more than 1.3° [14]. Patients were
excluded if they had signs of OA or rheumatoid arthritis, anterior cruci-
ate ligament (ACL) deficiency, neuromuscular dysfunction, achondro-
plasia, sagittal or transverse plane deformities of the leg, flexion
contractures in the knee or hip joint, leg length discrepancy of more
than 1 cm, avascular necrosis, history of major trauma or a sports injury
of the lower extremity, knee surgery within the last 12 months, chronic
joint infection, intraarticular corticosteroid injection, or morbid obesity
according to the body mass index [15].
Fifteen healthy subjects were recruited as control group (Table 1).
All subjects had undergone a clinical examination, including passive
hip, knee and ankle range of motion (Table 1). The patients and the
healthy subjects had normal strength and full range of motion in the
lower extremities.
All subjects, and their parents in the case of subjects under age 18,
were thoroughly familiarized with the gait analysis protocol. Subjects
18 and over gave written informed consent. Subjects younger than 18
Table 1
Study population characteristics (mean with standard deviation in parenthesis) and p-values.
Variable Patients Controls
n 18 15
Sex, no. female/no. male 11/7 9/6
Age, years 15.2 (1.8) 15.1 (4.3)
Height, m 1.64 (0.18) 1.64 (0.15)
Body mass, kg 55.7 (11.2) 52.8 (12.8)
Body mass index, kg/m2 20.6 (3.0) 19.2 (2.2)
Max. passive internal hip rotation 44.3 (13.7) 45.8 (3.6)
(prone, hip neutral, knee flexed 90°), °
Max. passive external hip rotation 37.0 (14.6) 41.3 (3.1)
(prone, hip neutral, knee flexed 90°), °
Max. passive tibial torsion −12.0 (10.5) −12.9 (4.5)
(prone, hip neutral, knee flexed 90°), °
Self-selected walking speed, m/s 1.25 (0.10) 1.28 (0.11)
Mechanical axis angle, ° 10.3 (7.0)
External foot rotation (tibial torsion) is negative.
689
gave verbal assent to study participation, and their parents provided
written informed consent to participate in this study, as approved by
the local ethics committee and in accordance with the Helsinki
Declaration.
2.2. Gait analysis methods
Three-dimensional gait analysis was carried out using a Vicon mo-
tion capture system (Vicon Motion Systems, Oxford, UK) operating
at a sampling rate of 200 Hz (Fig. 1). The level walkway was 15 m
long and viewed by eight infrared cameras. Two AMTI force plates
(Advanced Mechanical Technology, Inc., Watertown, MA, USA) were
situated at the mid-point of the walkway to collect kinetic data at
1000 Hz.
To improve the reliability and accuracy when analyzing frontal and
transverse plane gait data, a custom made lower body protocol
described in a previous investigation [16] was used. In addition to the
standardized Helen Hayes marker set [17], reflective markers on the
medial malleolus, medial femoral condyle and major trochanter were
applied to determine positions of joint centers of rotation for the hip,
knee and ankle. In contrast to the Helen Hayes marker set [17], the cen-
ters of rotation for the knee and ankle joints were defined statically as
the midpoint between the medial and lateral femoral condyle and
malleolus markers. This eliminates the reliance on the subjective palpa-
tion of the thigh and tibia wand markers, which is difficult to handle and
less reliable within or between therapists than manual palpation [18].
The center of the hip joint was calculated using a geometrical prediction
method [17]. The major trochanter marker was used to improve the
prediction of the hip joint center by immediately calculating the dis-
tance between the anterior superior iliac spine and the major trochanter
by anatomical landmarks [16]. Varus/valgus alignment of all knee joints
was determined by rotating the local shank coordinate system about the
y-axis of the thigh (vector from the lateral knee to the medial knee)
until the x-axis of the shank (vector from the knee center to the ankle
center) fell in line with the x-axis of the thigh (vector from the knee cen-
ter to the hip center). According to Hunt et al. [7] using the same marker
set for a similar purpose, this varus angulation provided an accurate,
marker-based measure of lower limb alignment in the frontal plane. It
has been shown that adding these few extra markers to the standard
Helen Hayes marker set increases the repeatability in three-
dimensional gait analysis, reduces the measurement errors when ana-
lyzing frontal and transverse plane gait data, and improves the accuracy
of the knee joint axis by reducing the knee axis cross-talk phenomenon
[16,19,20]. Furthermore, our lower body protocol enables the detection
of significant effects in the tibiofemoral angles and moments in the fron-
tal and transverse plane between three different patterns of movement
[21]. Further processing incorporating force plate data allowed external
moments (normalized to body weight) about the mathematically
derived joint centers to be calculated.
Discrete variables of interest (Table 2) were calculated for five bare-
foot trials at a self-selected speed and then averaged for further analysis
on the basis of complete marker trajectories and a clear foot-forceplate-
p -value contact. In the case of bilateral involvement, measurements were per-
formed only on the limb with greater malalignment. This ensures that
the statistical calculations were independent from each other. No signif-
0.941 icant differences were found in the discrete variables of interest
0.973
0.502
comparing the left and right sides in the control group. Therefore, only
0.144 one side – the left side was chosen here – was used for further analysis
0.690 of the controls.
After each acquisition session, missing frames were handled with
0.290
a fill-gap procedure using the Vicon-Nexus software version 1.7.1
0.763 (Vicon Motion Systems, Oxford, UK). The data were smoothed with a
Woltring filter and using a smoothing spline [22]. The external knee
0.431 adduction moment as well as the transverse plane gait parameters dur-
ing stance phase were automatically determined by a custom made
algorithm in Matlab 7.12.0 (The MathWorks, Inc., Natick, MA).
690 F. Stief et al. / The Knee 21 (2014) 688–693
a b
Fig. 1. Gait analysis methods: reflective markers (a), infrared camera (b), computerized body model (c).
According to Perry [23], loading response was defined as a subphase of
the gait cycle, i.e. 2–12% of the gait cycle. This phase follows the initial
contact of the foot with the floor and continues until the contralateral
limb is lifted for swing. Mid stance (12–31% of the gait cycle) begins
as the contralateral foot is lifted and continues until body weight is
aligned over the forefoot. Terminal stance (31–50% of the gait cycle) be-
gins with heel rise and continues until the contralateral foot strikes the
ground. To determine the foot progression angle, the angle of the long
axis of the foot segment in the global (lab) coordinate system relative
to the walking directions axis was computed.
2.3. Radiographic methods
Radiographic assessment of lower limb alignment was conducted on
the same day as the gait analysis. Patients stood barefoot in a forward
knee position with the patella centered over the femoral condyles and
feet straight ahead to control for any foot rotation effects [24] and to at-
tain a true full-length weight bearing anteroposterior radiograph [25].
On each radiograph, the joint centers of the hip, knee, and ankle were
identified. In accordance with Miyazaki et al. [4], Hunt et al. [7], and
Mündermann et al. [8] using the same method for a similar purpose,
the MAA of the lower extremity was used to quantify alignment in the
frontal plane and obtained with the analysis software DiagnostiX-32
(Gemed) by the same investigator. Alignment was measured as the
angle formed by the line from the femoral head center to the femoral
intercondylar notch center and the line from the ankle talus center to
the center of the tibial spine tips. In accordance with the study of
Specogna et al. [26], reliability of MAA measurements was high (ICC
= 0.97). The MAA measured from standing radiographs uses therefore
the same landmarks and definitions like the marker-based three-
dimensional gait analysis to measure lower limb alignment in the
Table 2
Correlation coefficients (r) to static mechanical axis angle and correlation p-values for
transverse plane kinematic and kinetic parameters during stance phase in the patient
group (n = 18).
Joint Parameter r p-value
Angles (°)
Foot Max. endorotation stance phase 0.584
Knee Max. internal rotation terminal stance 0.696
Hip Max. internal rotation stance phase −0.459
Moments (nm/kg)
Knee Max. internal rotation terminal stance 0.660
Hip Max. external rotation loading response/mid stance −0.661
Max. internal rotation terminal stance 0.040
Internal rotation is positive; external rotation is negative.
c
frontal plane. This enables comparison between these two measure-
ments and can be used as quality assurance of the marker-based
lower body protocol.
2.4. Statistical analysis
The normal distribution of the present sample was confirmed using
the Kolmogorov–Smirnov test. Means and standard deviations of the
variables of interest were calculated. Simple linear regression (Pearson
product–moment correlation coefficient; r) was used to examine the
association between static varus malalignment and gait parameters as
well as between the MAA measured from standing radiographs and
the static lower limb alignment measurements in the frontal plane
based on reflective markers. The significance level adopted in this
study was set at p b 0.05. We interpreted correlation values below
0.30 as low, between 0.30 and 0.65 as medium, and correlations above
0.65 as high [27]. Differences between groups were tested for signifi-
cance using a Student's t-test. The statistical calculations were carried
out using the SPSS version 18.0.0 (Chicago, IL, USA). Post hoc power
calculations were determined with the analysis software R (package
pwr, version 1.1.1) and according to Cohen [27].
3. Results
Subject group characteristics are summarized in Table 1. No significant differences
were observed across groups for self-selected walking speed, passive hip rotation, passive
tibial torsion and anthropometric parameters.
Post hoc power calculations revealed that a sample size of 18 patients provides 80%
power for detecting a correlation coefficient between the MAA and frontal/transverse
plane kinematic and kinetic parameters as small as r = 0.62 (p = 0.05, two sided).
Fig. 2 shows a strong linear relationship (r = 0.834, p b 0.001, power = 100%) between
the MAA measured from standing radiographs and the static lower limb alignment mea-
surements in the frontal plane based on reflective markers and the gait analysis system.
Linear correlations between the MAA measured from standing radiographs and gait pa-
rameters in the transverse plane are presented in Table 2. High correlations with increased
internal knee rotation (tibia rotation; r = 0.696, p = 0.004, power = 88%) and medium cor-
relations with increased endorotation of the foot (internal foot placement; r = 0.584, p =
0.022, power = 74%) during the stance phase of gait were detected. Regarding the kinetic
data, significant, high linear correlations with the static MAA are shown for the maximum in-
ternal knee rotation moment in terminal stance (r = 0.660, p = 0.007, power = 87%) and
the maximum external hip rotation moment in loading response/mid stance (r = −0.661,
p = 0.007, power = 87%).
0.022 Furthermore, the maximum external knee adduction moment in terminal stance
0.004 was linearly correlated with maximum internal knee rotation (r = 0.823, p b 0.001,
0.085 power = 94%) in terminal stance and maximum internal hip rotation during stance
phase (r = −0.742, p = 0.002, power = 90%).
Differences between the patient and control groups are shown in Fig. 3. Patients with
0.007
varus malalignment of the knee were walking with a significantly increased endorotation
0.007
of the foot compared to healthy controls. A post hoc power calculation revealed that the
0.886
difference in maximum foot endorotation between our study groups (18 patients, 15
controls) has a power of 95%.
 F. Stief et al. / The Knee 21 (2014) 688–693 691

20.0

17.5

15.0
MAA, radiograph [°]

12.5

10.0

7.5

5.0

2.5

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0

Lower limb alignment in the frontal plane, reflective markers [°]

Fig. 2. Scatter diagram of the relationship between the radiographic mechanical axis angle (MAA) and the static lower limb alignment measurements in the frontal plane based on reflec-
tive markers and the gait analysis system (r = 0.834, p b 0.001, power = 100%) for patients with varus malalignment of the knee (n = 18).

4. Discussion
The primary objective of the present study was to investigate the
correlation between static varus malalignment of the knee and dynamic
measures of the lower extremity during gait in the transverse plane. The
results suggest that a linear association exists between lower extremity
varus malalignment and an increased endorotation of the foot as well as
an increased maximum internal knee rotation (tibia rotation). Patients
with varus malalignment of the knee showed a significantly increased
internal foot placement (9.0°) during stance phase compared to the
control subjects without varus malalignment (1.3°), which could be
caused by the tendential increased tibia rotation in the patient group
(7.9°) compared to the control group (3.7°). Due to the fact that the
passive hip rotation and passive tibial torsion were not significantly
different between groups (Table 1), the observed differences in
transverse plane rotations during gait may either be a potential gait
mechanism to compensate for excessive varus malalignment of the
knee or essential in response to the altered MAA. As a result, this mod-
ified gait pattern in patients with varus malalignment may explain the
increased internal knee rotation moments compared to normally
aligned controls in a previous study [13]. Astephen et al. [28] also re-
ported a greater knee internal rotation moment coupled with a net
hip external rotation moment in patients with knee OA compared to a
healthy control group.
Regarding the accuracy of three-dimensional gait parameters, the
strong linear relationship between the MAA (radiologically) and the
MAA (based on reflective markers) suggests the suitability of the used
marker-based lower body protocol to analyze patients with varus
malalignment of the knee. No significant differences were observed
between patients and controls for self-selected walking speed. Thus,

Foot Endorotation Knee Internal Rotation Hip Internal Rotation

Stance Phase Terminal Stance Stance Phase
30
p = 0.854
p = 0.306
25

20 p = 0.003
13.6 14.1
15
Angle [°]

9.0 7.9
10
3.7
5
1.3
0

-5

-10
Patients (n = 18) Controls (n = 15)

Fig. 3. Average values with standard deviation for maximum foot, knee and hip angles during stance phase in the transverse plane.
692 F. Stief et al. / The Knee 21 (2014) 688–693
differences in gait patterns could not be attributed to differences in
walking speed.
The external knee adduction moment was linearly correlated with in-
ternal knee (positive correlation) and hip rotation (negative correlation)
during the stance phase. This suggests that transverse plane gait me-
chanics in patients with varus malalignment of the knee but no signs of
knee OA are directly in conjunction with the external knee adduction
moment which is a contributing factor to articular cartilage degeneration
and disease progression in the medial knee compartment [4,5,9,10] (sec-
ond part of the hypothesis). This assumption is confirmed with the stud-
ies of Davids et al. [29] and MacWilliams et al. [30] that demonstrated
that an increased knee rotation in children with symptomatic intoeing
gait leads to an increased knee adduction moment in the coronal plane
during the stance phase of gait. These findings lead to the conclusion
that transverse plane bone morphology is associated with coronal knee
loading during gait, which may have important implications for the pro-
gression of OA in this patient group. It has been suggested that changes in
transverse plane mechanics at the knee can initiate degenerative chang-
es by placing altered loads on regions of the articular cartilage that were
previously conditioned for different load levels [10]. Internal tibia rota-
tion is thought to increase compressive and shear loading in the medial
compartment of the knee [31] and strong associations have been
shown between the foot progression angle and medial knee loading [9,
29,30,32]. In particular, internal foot placement shifts the loads more
onto the knee's medial compartment [9,29,30], which may create an un-
comfortable environment in the knee joint. Studies of patients with ACL
injury support the observation that rotational changes are a major factor
in the progression of knee OA [33–35]. In ACL deficient knees, the tibia
remained internally rotated throughout the stance phase compared to
the control limb. This offset led to loading of cartilage regions not typical-
ly loaded prior to the ACL injury (for an overview see e.g., Vincent et al.,
[35]). In addition to the varus malalignment of the knee, we believe
therefore that the transverse plane gait mechanics in the analyzed pa-
tient group later lead to increased ligament forces and degenerative
changes in the knee joint.
However, our results should be treated with some caution. Due to
the small sample size, some results of the present study may be under-
powered and the results could be significant with greater sample sizes.
Further studies with an enlarged sample size should also investigate the
effect of surgical interventions to correct excessive varus alignment, es-
pecially on transverse plane mechanics during gait. If a preoperatively
increased internal knee rotation were not reduced after surgery, the
correction of the rotational gait profile should be taken into account
for surgical planning of varus knees in young patients. Davids et al.
[29] and MacWilliams et al. [30] have shown that the correction of the
pathologically increased internal knee rotation by rotation osteotomy
normalized the knee adduction moment. A non-invasive gait modifica-
tion to reduce the knee adduction moment is a toe out foot position [9,
36]. This mechanism of gait compensation reduces the moment arm
length of the net ground reaction force vector with respect to the knee
joint center in the frontal plane and shifts the load onto the non-
diseased compartment. This may also help to reduce the knee adduction
moments at this early stage and could be a potential strategy to slow
disease progression and delay the evolution of later medial knee OA [4].
Understanding factors that influence dynamic knee joint loading in
healthy, varus malaligned knees may help us to identify risk factors
that lead to OA. Recent literature has been shown that the amount of
varus thrust measured quantitatively using skin markers correlates
with X-ray joint degeneration and thus appears to offer a simple dy-
namic assessment of knee OA severity in clinics [37]. Clinical gait analy-
sis is therefore a powerful tool for clinical prognoses regarding the onset
or progression of medial knee OA. Nevertheless, to understand the com-
plexity of knee OA and to develop earlier and more appropriate treat-
ment strategies for the disease will require continued study on the
interrelationships between risk factors for the disease and appropriate
models of disease progression.
In conclusion, the results of the present study indicated that biome-
chanical changes in the transverse plane occur concomitantly with
changes in knee malalignment in the frontal plane independent from
the presence of knee OA. A mechanical consequence of varus knee
malalignment is obviously an internal foot placement and an internal
knee rotation in conjunction with an increase of the internal knee rota-
tion moment and external knee adduction moment during the stance
phase of gait. Therefore, varus malalignment of the knee should not be
viewed as an isolated problem in the frontal plane.
Conflict of interest statement
The authors disclose any financial and personal relationships with
other people or organizations that could inappropriately influence
their work.
References
[1] Brouwer GM, van Tol AW, Bergink AP, Belo JN, Bernsen RM, Reijman M, et al. Asso-
ciation between valgus and varus alignment and the development and progression
of radiographic osteoarthritis of the knee. Arthritis Rheum 2007;56(4):1204–11.
[2] Andriacchi TP, Mündermann A, Smith RL, Alexander EJ, Dyrby CO, Koo S. A frame-
work for the in vivo pathomechanics of osteoarthritis at the knee. Ann Biomed
Eng 2004;32(3):447–57.
[3] McNicholas MJ, Gibbs S, Linskell JR, Barker S, McGurty D, Rowley DI. The influence of
external knee moments on the outcome of total meniscectomy. A comparison of ra-
diological and 3-D gait analysis measurements. Gait Posture 2000;11(3):233–8.
[4] Miyazaki T, Wada M, Kawahara H, Sato M, Baba H, Shimada S. Dynamic load at base-
line can predict radiographic disease progression in medial compartment knee oste-
oarthritis. Ann Rheum Dis 2002;61(7):617–22.
[5] Sharma L, Hurwitz DE, Thonar EJA, Sum JA, Lenz ME, Dunlop DD. Knee adduction
moment, serum hyaluronan level, and disease severity in medial tibiofemoral oste-
oarthritis. Arthritis Rheum 1998;41(7):1233–40.
[6] Hayashi D, Englund M, Roemer FW, Niu J, Sharma L, Felson DT, et al. Knee
malalignment is associated with an increased risk for incident and enlarging bone
marrow lesions in the more loaded compartments: the MOST study. Osteoarthritis
Cartilage 2012;20(11):1227–33.
[7] Hunt MA, Birmingham TB, Jenkyn TR, Giffin JR, Jones IC. Measures of frontal plane
lower limb alignment obtained from static radiographs and dynamic gait analysis.
Gait Posture 2008;27(4):635–40.
[8] Mündermann A, Dyrby CO, Andriacchi TP. A comparison of measuring mechanical
axis alignment using three-dimensional position capture with skin markers and
radiographic measurements in patients with bilateral medial compartment knee
osteoarthritis. Knee 2008;15(6):480–5.
[9] Hurwitz DE, Ryals AR, Case JP, Block JA, Andriacchi TP. The knee adduction moment
during gait in subjects with knee osteoarthritis is more closely correlated with static
alignment than radiographic disease severity, toe out angle and pain. J Orthop Res
2002;20(1):101–7.
[10] Andriacchi TP, Mündermann A. The role of ambulatory mechanics in the initiation
and progression of knee osteoarthritis. Curr Opin Rheumatol 2006;18(5):514–8.
[11] Mündermann A, Dyrby CO, Andriacchi TP. Secondary gait changes in patients with
medial compartment knee osteoarthritis: increased load at the ankle, knee, and
hip during walking. Arthritis Rheum 2005;52(9):2835–44.
[12] Turcot K, Armand S, Lübbeke A, Fritschy D, Hoffmeyer P, Suvà D. Does knee align-
ment influence gait in patients with severe knee osteoarthritis? Clin Biomech
(Bristol, Avon) 2013;28(1):34–9.
[13] Stief F, Böhm H, Schwirtz A, Dussa CU, Döderlein L. Dynamic loading of the knee and
hip joint and compensatory strategies in children and adolescents with varus
malalignment. Gait Posture 2011;33(3):490–5.
[14] Moreland JR, Bassett LW, Hanker GJ. Radiographic analysis of the axial alignment of
the lower extremity. J Bone Joint Surg Am 1987;69(5):745–9.
[15] Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition
for child overweight and obesity worldwide: international survey. BMJ
2000;320(7244):1240–3.
[16] Stief F, Böhm H, Michel K, Schwirtz A, Döderlein L. Reliability and accuracy in three-
dimensional gait analysis: a comparison of two lower body protocols. J Appl
Biomech 2013;29(1):105–11.
[17] Davis RB, Ounpuu S, Tyburski D, Gage J. A gait data collection and reduction tech-
nique. Hum Mov Sci 1991;10(5):575–87.
[18] Serfling DM, Hooke AW, Bernhardt KA, Kaufman KR. Comparison of techniques for
finding the knee joint center. Proceedings of the Gait and Clinical Movement Analy-
sis Society. Denver; 2009. p. 43.
[19] Biagi F, Desloovere K, Benedetti MG, Ingrosso S, Wong P, Leardini A. Measuring
three-dimensional knee rotations with skin markers. Gait Posture 2008;28(Suppl.
2):75–6.
[20] Böhm H, Senner V. The effect of ski boot settings on tibio-femoral abduction and
rotation during standing and simulated skiing. J Biomech 2008;41(3):498–505.
[21] Stief F, Kleindienst FI, Wiemeyer J, Wedel F, Campe S, Krabbe B. Inverse dynamic
analysis of the lower extremities during Nordic walking, walking, and running. J
Appl Biomech 2008;24(4):351–9.
 F. Stief et al. / The Knee 21 (2014) 688–693
[22] Woltring HJ. Representation and calculation of 3-D joint movements. Hum Mov Sci
1991;10(5):603–16.
[23] Perry J. Gait analysis. normal and pathological function. Second Edition. Thorofare:
SLACK Incorporated; 2010.
[24] Hunt MA, Fowler PJ, Birmingham TB, Jenkyn TR, Giffin JR. Foot rotational effects on
radiographic measures of lower limb alignment. Can J Surg 2006;49(6):401–6.
[25] Paley D. Principles of deformity correction. New York: Springer; 2002.
[26] Specogna AV, Birmingham TB, DaSilva JJ, Milner JS, Kerr J, Hunt MA, et al. Reliability
of lower limb frontal plane alignment measurements using plain radiographs and
digitized images. J Knee Surg 2004;17(4):203–10.
[27] Cohen J. Statistical power analysis for the behavioural sciences. Philadelphia: Law-
rence Erlbaum Associates; 1988.
[28] Astephen JL, Deluzio KJ, Caldwell GE, Dunbar MJ, Hubley-Kozey CL. Gait and neuro-
muscular pattern changes are associated with differences in knee osteoarthritis se-
verity levels. J Biomech 2008;41(4):868–76.
[29] Davids JR, Davis RB, Jameson LC, Westberry DE, Hardin JW. Surgical management of
persistent intoeing gait due to increased internal tibial torsion in children. J Pediatr
Orthop 2014. http://dx.doi.org/10.1097/BPO.0000000000000173 [in press].
[30] MacWilliams BA, McMulkin ML, Baird GO, Stevens PM. Distal tibial rotation osteotomies
normalize frontal plane knee moments. J Bone Joint Surg Am 2010;92(17):2835–42.
693
[31] Turner MS. The association between tibial torsion and knee joint pathology. Clin
Orthop Relat Res 1994;302:47–51.
[32] Krackow KA, Mandeville DS, Rachala SR, Bayers-Thering M. Torsion deformity and
joint loading for medial knee osteoarthritis. Gait Posture 2011;33(4):625–9.
[33] Andriacchi TP, Dyrby CO. Interactions between kinematics and loading during walk-
ing for the normal and ACL deficient knee. J Biomech 2005;38(2):293–8.
[34] Georgoulis AD, Papadonikolakis A, Papageorgiou CD, Mitsou A, Stergiou N. Three-
dimensional tibiofemoral kinematics of the anterior cruciate ligament-deficient
and reconstructed knee during walking. Am J Sports Med 2003;31(1):75–9.
[35] Vincent KR, Conrad BP, Fregly BJ, Vincent HK. The pathophysiology of osteoarthritis:
a mechanical perspective on the knee joint. PM R 2012;4(5):S3–9.
[36] Chang A, Hurwitz D, Dunlop D, Song J, Cahue S, Hayes K, et al. The relationship be-
tween toe-out angle during gait and progression of medial tibiofemoral osteoarthri-
tis. Ann Rheum Dis 2007;66(10):1271–5.
[37] Kuroyanagi Y, Nagura T, Kiriyama Y, Matsumoto H, Otani T, Toyama Y, et al. A quan-
titative assessment of varus thrust in patients with medial knee osteoarthritis. Knee
2012;19(2):130–4.