Eur Spine J (2016) 25:3658–3665
DOI 10.1007/s00586-016-4621-2
ORIGINAL ARTICLE
Influence of high-heeled shoes on the sagittal balance of the spine
and the whole body
Tim Weitkunat1 • Florian M. Buck2 • Thorsten Jentzsch3 • Hans-Peter Simmen1
Clément M. L. Werner1 • Georg Osterhoff1
Received: 2 February 2016 / Revised: 27 March 2016 / Accepted: 16 May 2016 / Published online: 20 May 2016
Ó Springer-Verlag Berlin Heidelberg 2016
Abstract
Purpose Wearing high heels is associated with chronic
pain of the neck, lower back and knees. The mechanisms
behind this have not been fully understood. The purpose of
this study was to investigate the influence of high-heeled
shoes on the sagittal balance of the spine and the whole
body in non-habitual wearers of high heels.
Methods Lateral standing whole body low-dose radio-
graphs were obtained from 23 female participants (age
29 ± 6 years) with and without high heels and radiological
parameters describing the sagittal balance were quantified.
These were analyzed for differences between both condi-
tions in the total sample and in subgroups.
Results Standing in high heels was associated with an
increased femoral obliquity angle [difference (D)
3.0° ± 1.7°, p \ 0.0001], and increased knee (D
2.4° ± 2.9°, p = 0.0009) and ankle flexion (D
38.7° ± 3.4°, p \ 0.0001). The differences in C7 and
meatus vertical axis, cervical and lumbar lordosis, thoracic
kyphosis, spino-sacral angle, pelvic tilt, sacral slope, and
spinal tilt were not significant. Individuals adapting with
less-than-average knee flexion responded to high heels by
an additional increase in cervical lordosis (D 5.8° ± 10.7°
vs. 1.8° ± 5.3°).
& Georg Osterhoff
georg.osterhoff@usz.ch
1
Division of Trauma Surgery, University of Zurich,
Raemistrasse 100, 8091 Zurich, Switzerland
2
Department of Radiology, University of Zurich, Balgrist,
Forchstrasse 340, 8008 Zurich, Switzerland
3
Department of Orthopedics, University of Zurich, Balgrist,
Forchstrasse 340, 8008 Zurich, Switzerland
123
•
Conclusions In all participants, wearing high heels led to
increased flexion of the knees and to more ankle flexion.
While some participants responded to high heels primarily
through the lower extremities, others used increased cer-
vical lordosis to adapt to the shift of the body’s center of
gravity. This could explain the different patterns of pain in
the neck, lower back and knees seen in individuals wearing
high heels frequently.
Keywords Sagittal balance
High heels
Knee flexion

Cervical lordosis
Lumbar lordosis
Thoracic kyphosis
Introduction
Wearing high heels on a regular basis can lead to perma-
nent changes of posture in adolescents and young adults, as
well as malposition of the spine and the legs [1]. Nearly
two-third of habitual wearers of high heels report lumbar
back pain [2], compared to less than 20 % in the general
population [3].
In the past, this has been linked to changed biome-
chanics due to elevation of the hind foot. Some authors
suspected a compensatory increase in lumbar lordosis and
pelvic tilt as the main causes of lumbar pain [1, 2, 4, 5],
while others suspected muscle fatigue and increased lum-
bar vertebral compression through increased muscle
activity [6, 7], or increased forces transferred on the ground
when walking [8]. However, the somewhat suggestive
potential role of lumbar lordosis and pelvic tilt in the
development of lumbar back pain remains unclear and
under debate [1, 9–13].
When standing in high-heeled shoes, the body’s center
of gravity is being shifted cranially and anteriorly [2, 14].
Using photogrammetric measurements, Lee et al. could
Eur Spine J (2016) 25:3658–3665 3659

show a compensatory posterior [2] tilt of the whole upper
body with high heels, while Drzal-Grabiec and Snela
observed a forward inclination of the trunk [13]. They did
not differentiate the underlying single skeletal dynamics
leading to this movement, though. On the contrary, the vast
majority of studies on high heels and their potential impact
on posture and whole body sagittal balance focus on lim-
ited aspects; most of them on lumbar lordosis and pelvic
tilt.
The purpose of this study was to investigate the influ-
ence of high-heeled shoes on the sagittal balance of not
only the spine but the whole body, based on a compre-
hensive set of radiologically precisely specified parameters
of sagittal balance, in upright standing healthy younger
female non-habitual wearers of high heels.
wearing high-heeled shoes. A clinical examination of the
spine and pelvis was performed and the negative result of a
pregnancy urine test was confirmed.
Two standing lateral radiographs of the whole body
were taken—once barefoot and once with high-heeled
shoes (Fig. 1). Radiographs were taken using a biplanar
low-dose radiographic imaging system (EOS imaging,
Paris, France), with both images together corresponding to
a total dose of 0.1–0.7 mSv per participant [17].
After a first radiographic examination without shoes,
participants selected standardized high-heeled shoes (heel
height 9 cm, heel diameter 1 cm) of their size and wore
them for 10 min, either standing or walking ad libitum. A
second set of standing lateral radiographs of the whole
body was then acquired with high heels.

Radiological quantification
Methods
The following parameters were assessed on the whole body
Patients radiographs: C7 and meatus sagittal vertical axis, cervical
and lumbar lordosis, thoracic kyphosis, spino-sacral angle,
The protocol of the present study was approved by the local
ethics committee (Kantonale Ethik-Kommission Zürich,
KEK-ZH-Nr. 2014-0014) and registered with clinicaltri-
als.gov (SNCTP000000049).
Female volunteers (age C21 years) were included given
that they were no frequent user of high-heeled shoes (i.e.,
no longer than 6 h per month with heels [3 cm during the
last 2 years). The study’s focus was the impact of high
heels on healthy women in contrast to habitual users of
which may already have developed different and varying
degrees of adaption or even irreversible degenerative
changes.
Exclusion criteria were osseous malignancies, missing
written informed consent, cognitive inability to consent on
their own, and pregnancy.
Prior to the study, preliminary EOS radiographs were
taken from the senior author with and without high heels,
showing a difference of 7° for the spino-sacral angle. The
literature reports a standard deviation of 8° for the spino-
sacral angle in the normal population [15]. Assuming a
type I error of 0.01 with a desired statistical power of 0.80,
a minimum sample size of 19 participants was calculated
using PS Power and Sample Size Calculations 3.0
(a = 0.05) [16]. Eventually 23 participants were included
between March and November 2014.

Data acquisition

After obtaining informed consent, the participants were
interviewed with special regard to their age, gender, height,
weight, shoe size, past medical history, occupational and
recreational activities, as well as duration and frequency of
Fig. 1 Data acquisition. By the use of a biplanar low-dose radio-
graphic imaging system, standing lateral radiographs of the whole
body were taken—once barefoot and once with high-heeled shoes.
The shown radiograph was acquired for preliminary studies and
depicts the senior author

123
3660
pelvic tilt, sacral slope, spinal tilt, pelvic incidence,
femoral obliquity angle, knee flexion, and ankle flexion
(Fig. 2). Imaging software was used for radiographic
quantification (Surgimap, Version 2.1.8, Surgimap, New
York, NY, USA, and Impax, Version 6.6.6.1544, Agfa
HealthCare GmbH, Bonn, Germany).
The specifications adopted for C7 sagittal vertical axis,
thoracic kyphosis, and lumbar lordosis were those descri-
bed by Quian et al. [18]. Analogous to the C7 sagittal
vertical axis, the meatus sagittal vertical axis was deter-
mined by first establishing a straight line between the
centers of the acoustic meatuses and dropping a perpen-
dicular from the midpoint of that line. Next, a line repre-
senting the sagittal footprint was drawn connecting the
midpoint between both calcaneal tuberosities with the
midpoint between the centers of both first metatarsal heads.
The horizontal distance between the midpoint of this line
and the perpendicular quantified the meatus sagittal vertical
axis, which was recorded with a negative sign when the
plumb line was posterior of the midpoint and with a pos-
itive sign otherwise. For quantification of cervical lordosis,
the Cobb C2–C7 method was used [19]. In general, lordosis
was indicated with negative values and kyphosis with
Fig. 2 Radiographic parameter. Imaging software was used for
radiographic quantification of the following parameters on the whole
body radiographs: C7 sagittal vertical axis (C7-SVA), meatus sagittal
vertical axis (Meatus-SVA), spino-sacral angle (SSA), spinal tilt (ST),
123
Eur Spine J (2016) 25:3658–3665
positive ones. The definitions of pelvic incidence, pelvic
tilt and sacral slope were applied as described by Legaye
et al. [20]. Spino-sacral angle and spinal tilt were defined as
described by [15]. The definition of the femoral obliquity
angle deviated slightly from the one provided by Quian
et al. [18] as we were able to use the mechanical axis of the
entire femur, rather than only the proximal part (anatomical
femoral axis). Knee flexion was defined as the angle
between the mechanical femoral axis and the mechanical
tibial axis. Ankle flexion was defined as the angle between
the perpendicular on the tibial axis and a line paralleling
the footprint, positive values indicating plantarflexion of
the ankle joints.
Statistical analysis
For analysis, all data including the radiographic parameters
were recorded in Excel sheets (Microsoft, Seattle, WA,
USA) and then imported into JMP-SAS (version 11, SAS
Institute, Cary, NC, CA, USA). Differences in means
between the barefoot and the high-heeled radiographs were
assessed with paired t test. For subgroup comparisons of
these differences in means, 95 percent (%) confidence
femoral obliquity angle (FOA), knee flexion angle (KFA), and ankle
flexion angle (AFA). In addition, cervical and lumbar lordosis,
thoracic kyphosis, pelvic tilt, sacral slope, and pelvic incidence were
determined as described in the ‘‘Methods’’ section
Eur Spine J (2016) 25:3658–3665
intervals (CI) were calculated for describing the precision
of the effects found within and between subgroups, along
with p values for the subgroup differences based on
unpaired t tests. The level of significance was p \ 0.05 for
all tests.
Results
In total, 23 participants were included (Table 1). The data
of all participants were complete, with the exception of one
radiograph taken in the barefoot condition that inadver-
tently ended at mid femur, leading to missing values of the
meatus sagittal vertical axis, the femoral obliquity angle,
the knee flexion angle and the ankle flexion angle for that
participant.
High heels were worn 1.3 times per week on average,
for a weekly duration of (mean ± standard deviation)
1.3 ± 1.6 h. Average weekly exercise duration was
4.1 ± 3.5 h and about half of the sample (n = 11 partici-
pants) reported to predominantly work in a desk job.
Radiographic analysis
When compared to standing barefoot, standing in high
heels was associated with a less negative femoral obliquity
angle [difference (D) 3.0° ± 1.7°, p \ 0.0001], a less
negative knee flexion angle (D 2.4° ± 2.9°, p = 0.0009)
and a higher ankle flexion angle (D 38.7° ± 3.4°,
p \ 0.0001; Appendix 1). Thus, high heels led to increased
flexion of the hips and knees and to more ankle flexion.
The effects of all other differences between the barefoot
and high-heeled condition were relatively small with
respect to the inter-individual variability, thus not reaching
statistical significance (Fig. 3). The findings indicate that
an important mechanism of adaptation was increased knee
flexion in response to the anterior and cranial shift of the
center of gravity that is induced by high-heeled shoes.
With the exception of the meatus sagittal vertical axis
(r = 0.19) and the ankle flexion angle (r = 0.30), all
Table 1 Participants’ baseline characteristics (n = 23)
M SD Min. Max.
Age (years) 29.1 6.3 23.0 50.0
Height (m) 1.7 0.1 1.5 1.8
Weight (kg) 65.5 11.9 47.0 98.0
BMI (kg/m2) 23.9 4.0 19.8 37.3
Shoe size (European metric) 38.6 1.5 36.0 41.0
M mean, SD standard deviation, min minimum, max maximum,
m meters, kg kilograms, m2 square meters
3661
Fig. 3 Results. The graph is showing box-and-whisker plots of the
high heels minus barefoot differences (‘‘DELTA’’). Positive values
(to the right) represent an increase with wearing high heels. p values
are stated for parameters where the paired t test reached the level of
significance. Femoral obliquity angle, knee flexion angle and ankle
flexion angle show a significant increase. As pelvic incidence is a
static parameter independent of the experimental condition, it is not
contained in the figure. Increased lordosis is characterized by negative
and increased kyphosis by positive values
correlations between the barefoot and high-heeled condi-
tion were large ([0.50) or very large ([0.90) indicating a
good overall measurement precision (Appendix 1).
Large correlations and small differences between the
condition-specific means indicate a lack of sensitivity of
the corresponding parameter for high heels effects. Such a
pattern was seen most clearly in the pelvic incidence and
also to a lesser degree in the thoracic kyphosis, lumbar
lordosis, spino-sacral angle, pelvic tilt, sacral slope and
spinal tilt.
Large correlations and pronounced differences between
the conditions were found in the C7 sagittal vertical axis,
the cervical lordosis, the knee flexion angle and the femoral
obliquity angle. Even though this could not be statistically
confirmed for the first two of these parameters due to their
sizeable variability, this pattern of findings indicates a
pronounced high heels effect on all four parameters. As
indicated by the significant high heel-induced changes in
the femoral obliquity and knee flexion angles, one main
mechanism of adaptation was knee flexion. Based on the
findings mentioned above, a second important mechanism
would have been a change in cervical lordosis, even though
123
3662
the noticeable variability of this parameter (Fig. 2) made it
impossible to detect a significant difference.
It is conceivable, however, that not all mechanisms of
musculo-skeletal adaptation to the shift of the center of
gravity were used by all participants to the same degree. To
further explore the consistency of the data with the
hypothesis of two person-specific primary mechanisms of
adaptation, the sample was split at the mean of the knee
flexion angle of 2.4° (Appendix 1), rendering one subgroup
above-average knee flexion (n = 10) and one subgroup
with below-average knee flexion (n = 12).
The subgroup adapting to high heels predominantly by
higher knee flexion also showed a significantly more pro-
nounced change toward a higher femoral obliquity angle when
compared to the other subgroup (mean change 4.2° vs 2.0°, D
2.2, 95 % CI 1.0–3.4). In contrast, the increase in cervical
lordosis upon transition to high heels was clearly stronger in
the subgroup with below-average knee flexion (mean change
5.8° ± 10.7°) compared to the subgroup with above-average
knee flexion, who in fact responded with a slight kyphosis
(mean change 1.8° ± 5.3°, D 7.6, 95 % CI 0.3–15.0).
Discussion
This is the first study known to the authors using whole
body radiographs to investigate the influence of high-
heeled shoes on the sagittal balance of not only the spine
but the whole body in upright standing healthy female non-
habitual wearers of high heels. By deploying a modern
whole body low-dose radiographic imaging system, a
comprehensive set of radiological parameters could be
precisely quantified under conditions of upright barefoot
standing as well as standing in high heels.
We found that wearing high heels leads to increased
flexion of the knees and, of course, to more ankle plantar
flexion. Thus, most of the adaptive responses to the antero-
cranial shift of the center of gravity when standing on high
heels occurred in the lower extremity, especially the knees.
Some of the participants, though, used an increase of cer-
vical lordosis as an additional mechanism to shift the center
of gravity backwards.
These results support the need for investigating the
whole body rather than only a part of it, as most studies on
this topic have done so far [1, 2, 5, 7, 9, 13, 21].
No statistically significant changes were seen for thoracic
kyphosis, lumbar lordosis and the measures of pelvic sagittal
inclination. At first sight, this is surprising as many studies on
the influence of high heels on the sagittal balance focus on
these [1, 2, 4, 5, 11–13]. Standing on high-heeled shoes results
in an antero-cranial shift of the body’s center of gravity [2, 14].
It is known that, very similarly, degenerative changes of the
spine result in a forward shift of the body’s center of gravity
123
Eur Spine J (2016) 25:3658–3665
[21]. In the aged population, this entails a number of mecha-
nisms to re-balance the body’s center of gravity. These include
an increase of cervical and lumbar lordosis, a decrease of
thoracic kyphosis, a posterior tilt of the pelvis with hyperex-
tension of the hips, and increased flexion of the knees and the
ankle joints [20, 21]. The main difference is that the shifting
moment in the elderly is a degenerative loss of lumbar lor-
dosis, while in individuals on high heels it comes from below
the feet. In consequence, elderly patients will compensate
mainly in the axial skeleton while young women on high heels
mainly respond by knee and ankle flexion. Naturally, elderly
people do also have less mobility and muscular control around
their axial skeleton.
Although not the focus of this study, it is also possible
that changes in muscular tension at the autochthonous
musculature play an important role as a compensatory
mechanism of posture maintenance. Individuals, who may
not have the appropriate muscle strength in the spine or
legs to keep the body balanced, may have to use knee
flexion to keep balance. Of note, the participants of the
present study were no habitual wearers of high-heeled
shoes. This may have led to compensation mechanisms of
the knees, while more experienced high heels wearers may
have converted from using knee flexion, towards spinal
balancing mechanisms as described in previous studies.
The main limitations of this study include the fact that
even a whole body radiograph can only document a single
moment of a body’s position while balancing is always a
dynamic process. This might explain why the measurements
showed the higher variabilities the more cranially they were
measured (Fig. 2). This especially accounts for the site-
specific measures (cervical and lumbar lordosis, thoracic
kyphosis) and less for the global parameters (C7- and meatus
sagittal vertical axis). As the sample size calculation was
based on a global parameter, this could mean that the sample
of the present study was potentially smaller than needed. An
alternative explanation for the large variability of some
measures is that the participants might always have used a
combination of two or more re-balancing mechanisms at the
same time, some of them too subtle to detect with the study’s
sample size. With regard to the parameters not reaching
statistical significance in our study, it cannot be excluded that
a larger sample might have led to significant findings through
increased statistical power. This would also explain the
contradictory findings in the literature with regard to high
heels and lumbar lordosis [9]. A further potential explanation
for the changes in lumbar lordosis being not statistically
significant could be the high variability in pelvic incidence
found in our study sample (54° ± 10°, minimum 33°,
maximum 70°; see Appendix 1). As a high pelvic incidence
would be associated with a greater lordosis and a low pelvic
incidence with a small lumbar lordosis, this could explain the
high inter-individual variability.
Eur Spine J (2016) 25:3658–3665 3663

In addition, this study’s results do not allow for con- Conclusions

clusions on the biomechanics of walking on high heels.
However, even though the forces acting on our musculo-
skeletal system are considerably stronger during walking,
we spent more than double as much time standing (5 h per
day) than stepping (2 h per day) [22].
However, by the use of sequential whole body radio-
graphs, the present study was able to obtain a compre-
hensive picture of intra-individual changes in sagittal
balance when standing in high heels and found that knee
flexion, ankle plantar flexion and—in some patients—cer-
vical lordosis were the most pronounced effects.
In line with our findings, previous studies suspected a
link between wearing high heels and patella-femoral pain
[23] and could show that wearing high heels constitutes a
predisposing factor for the development of knee
osteoarthritis [24].
It is possible that women responding to high heels pri-
marily with knee flexion are predisposed to developing
knee conditions. Likewise, women primarily responding by
increased cervical lordosis might be at higher risk of
developing cervical clinical conditions, including muscular
hardening and neck pain as it has been shown that wearing
high heels is associated with an increased activity of the
cervical paraspinal muscles [25, 26]. Future studies may
investigate these possible associations between person-
specific primary mechanisms of adaptation and specific
clinical conditions by case–control studies conducted in
women regularly wearing high heels.
In all participants, wearing high heels led to increased
flexion of the knees and to more ankle flexion. While some
participants responded to high heels primarily through the
lower extremities, others used increased cervical lordosis to
adapt to the shift of the body’s center of gravity. This could
explain the different patterns of pain in the neck, lower
back and knees seen in individuals wearing high heels
frequently.
Compliance with ethical standards
Conflict of interest The authors declare that they have no competing
interests.
Funding This study was not founded by any external source.
Appendix 1: measured and derived (high heels
(HH) minus barefoot (SL) differences) radiological
parameters
The table below summarizes the quantified radiological
parameters under barefoot and high heels conditions, as
well as the parameters derived from the intra-individual
high heels minus barefoot differences. The rightmost col-
umn contains the between-conditions correlation coeffi-
cients [r(BF, HH)] as well as the paired t test p values of
the differences between the conditions.

Parameter N M SD Min. Max. [r(BF, HH)] p

SL C7 sagittal vertical axis (cm) 23 -0.09 1.90 -3.4 3.2

HH C7 sagittal vertical axis (cm) 23 -0.62 2.28 -4.1 3.7 [0.59]
D C7 sagittal vertical axis (cm) 23 -0.53 1.93 -3.4 3.6 0.2008
SL Meatus sag. vert. axis (cm) 22 -1.61 2.35 -6.7 1.7
HH Meatus sag. vert. axis (cm) 23 -1.09 2.66 -5.1 5.7 [0.19]
D Meatus sag. vert. axis (cm) 22 0.64 3.20 -4.7 8.1 0.3582
SL Cervical lordosis (°) 23 0.52 9.87 -18.0 17.0
HH Cervical lordosis (°) 23 -1.70 14.83 -37.0 22.0 [0.80]
D Cervical lordosis (°) 23 -2.22 9.17 -25.0 12.0 0.2589
SL Thoracic kyphosis (°) 23 32.91 7.90 13.0 49.0
HH Thoracic kyphosis (°) 23 32.26 8.52 12.0 48.0 [0.84]
D Thoracic kyphosis (°) 23 -0.65 4.65 -8.0 8.0 0.5080
SL Lumbar lordosis (°) 23 -58.87 9.47 -77.0 -46.0
HH Lumbar lordosis (°) 23 -58.83 10.18 -77.0 -42.0 [0.91]
D Lumbar lordosis (°) 23 0.04 4.26 -8.0 10.0 0.9614
SL Spino-sacral angle (°) 23 133.83 7.51 120.0 156.0
HH Spino-sacral angle (°) 23 134.22 6.97 120.0 150.0 [0.90]
D Spino-sacral angle (°) 23 0.39 3.31 -7.0 5.0 0.5768
SL Pelvic incidence (°) 23 54.26 10.04 33.0 70.0
HH Pelvic incidence (°) 23 54.26 10.04 33.0 70.0 [1.00]

123
3664 Eur Spine J (2016) 25:3658–3665

Parameter N M SD Min. Max. [r(BF, HH)] p

D Pelvic incidence (°) 23 0.00 0.00 0.0 0.0 n.a.

SL Pelvic tilt (°) 23 11.78 6.88 -1.0 21.0
HH Pelvic tilt (°) 23 12.57 6.01 1.0 22.0 [0.90]
D Pelvic tilt (°) 23 0.78 3.00 -5.0 7.0 0.2239
SL Sacral slope (°) 23 42.48 7.19 32.0 62.0
HH Sacral slope (°) 23 41.65 7.42 28.0 55.0 [0.92]
D Sacral slope (°) 23 -0.83 2.99 -7.0 6.0 0.1994
SL Spinal tilt (°) 23 91.78 2.56 87.0 96.0
HH Spinal tilt (°) 23 92.74 3.22 86.0 98.0 [0.58]
D Spinal tilt (°) 23 0.96 2.72 -4.0 7.0 0.1060
SL Femoral obliquity angle (°) 22 -6.77 2.16 -11.0 -1.0
HH Femoral obliquity angle (°) 23 -3.78 2.19 -8.0 0.0 [0.72]
D Fem. ob. angle (°) 22 3.00 1.66 0.0 7.0 \0.0001
SL Knee flexion angle (°) 22 -5.50 3.62 -12.0 2.0
HH Knee flexion angle (°) 23 -3.26 3.91 -10.0 3.0 [0.72]
D Knee flexion angle (°) 22 2.36 2.87 -2.0 12.0 0.0009
SL Ankle flexion angle (°) 22 -1.50 1.92 -4.0 2.0
HH Ankle flexion angle (°) 23 37.13 3.39 31.0 43.0 [0.30]
D Ankle flexion angle (°) 22 38.73 3.40 31.0 44.0 \0.0001
N number of individuals, M mean, SD standard deviation, Min. minimum, Max. maximum

Appendix 2: subgroup analysis
Description of the high heels minus barefoot difference
values of the radiological parameters in the two subgroups
obtained by splitting the sample at the mean of the knee
flexion angle (2.36°). The column ‘‘Adaptation Knee’’
signifies below-average (n = 12) by ‘‘No’’, and above-
average knee flexion (n = 10) by ‘‘Yes’’. The difference of
the means of the two subgroups is contained in the dif-
ference column, along with its 95 % confidence interval
(CI confidence interval, LL lower limit, UL upper limit).
Statistical significance at the local 5 %-level can be con-
cluded from the CIs when they do not cover the null dif-
ference. For convenience, unpaired t test p values are
provided in brackets for the differences in means.

Parameter Adaptation knee N M SD 95 % CI Difference (p) 95 % CI

LL UL LL UL

D C7 sagittal vertical axis (cm) No 12 0.01 1.96 -1.24 1.25 0.96 -0.72 2.63
Yes 10 -0.95 1.80 -2.24 0.34 (0.2464)
D Meatus sagittal vertical axis (cm) No 12 1.13 2.84 -0.67 2.94 1.08 -1.87 4.04
Yes 10 0.05 3.65 -2.56 2.66 (0.4567)
D Cervical lordosis (°) No 12 -5.83 10.74 -12.66 0.99 -7.63 -14.99 -0.28
Yes 10 1.80 5.31 -2.00 5.60 (0.0459)
D Thoracic kyphosis (°) No 12 -1.42 4.74 -4.43 1.59 -1.92 -6.15 2.32
Yes 10 0.50 4.74 -2.89 3.89 (0.3560)
D Lumbar lordosis (°) No 12 1.33 3.92 -1.15 3.82 2.63 -1.19 6.46
Yes 10 -1.30 4.57 -4.57 1.97 (0.1700)
D Spino-sacral angle (°) No 12 -0.33 3.55 -2.59 1.92 -1.13 -3.98 1.72
Yes 10 0.80 2.86 -1.25 2.85 (0.4186)
D Pelvic tilt (°) No 12 0.42 3.68 -1.92 2.75 -0.58 -3.22 2.05
Yes 10 1.00 2.16 -0.55 2.55 (0.6516)

123
Eur Spine J (2016) 25:3658–3665 3665

Parameter Adaptation knee N M SD 95 % CI Difference (p) 95 % CI

LL UL LL UL

D Sacral slope (°) No 12 -0.50 3.66 -2.82 1.82 0.40 -2.19 2.99
Yes 10 -0.90 2.08 -2.39 0.59 (0.7518)
D Spinal tilt (°) No 12 0.08 2.35 -1.41 1.58 -1.62 -4.02 0.79
Yes 10 1.70 2.95 -0.41 3.81 (0.1782)
D Femoral obliquity angle (°) No 12 2.00 1.04 1.34 2.66 -2.20 -3.36 -1.04
Yes 10 4.20 1.48 3.14 5.26 (0.0013)
D Ankle flexion angle (°) No 12 39.42 3.03 37.49 41.34 1.52 -1.58 4.62
Yes 10 37.90 3.79 35.20 40.61 (0.3200)

References
1. Silva AM, de Siqueira GR, da Silva GA (2013) Implications of
high-heeled shoes on body posture of adolescents. Rev Paul
Pediatr 31:265–271
2. Lee C-M, Jeonga E-H, Freivalds A (2001) Biomechanical effects
of wearing high-heeled shoes. Int J Ind Ergon 28:321–326
3. Hoy D, Brooks P, Blyth F, Buchbinder R (2010) The epidemi-
ology of low back pain. Best Pract Res Clin Rheumatol
24:769–781
4. Pezzan, PA, João SM, Ribeiro AP, Manfio EF (2011) Postural
assessment of lumbar lordosis and pelvic alignment angles in
adolescent users and nonusers of high-heeled shoes. J Manip
Physiol Ther 34(9):614–621. doi:10.1016/j.jmpt.2011.09.006
5. Dai M, Li X, Zhou X, Hu Y, Luo Q, Zhou S (2015) High-heeled-
related alterations in the static sagittal profile of the spino-pelvic
structure in young women. Eur Spine J 24:1274–1281
6. Cronin NJ (2014) The effects of high heeled shoes on female gait:
a review. J Electromyogr Kinesiol 24:258–263
7. Mika A, Oleksy L, Mika P, Marchewka A, Clark BC (2012) The
effect of walking in high- and low-heeled shoes on erector spinae
activity and pelvis kinematics during gait. Am J Phys Med
Rehabil 91:425–434
8. Barton CJ, Coyle JA, Tinley P (2009) The effect of heel lifts on
trunk muscle activation during gait: a study of young healthy
females. J Electromyogr Kinesiol 19:598–606
9. Russell BS (2010) The effect of high-heeled shoes on lumbar
lordosis: a narrative review and discussion of the disconnect
between Internet content and peer-reviewed literature. J Chiropr
Med 9:166–173
10. Russell BS (2015) Letter to the Editor concerning ‘‘High-heeled-
related alterations in the static sagittal profile of the spino-pelvic
structure in young women’’ by Min Dai et al [Eur Spine J]. Eur
Spine J 24:1826–1827. doi:10.1007/s00586-015-3857-6
11. Opila KA, Wagner SS, Schiowitz S, Chen J (1988) Postural
alignment in barefoot and high-heeled stance. Spine (Phila Pa
1976) 13:542–547
12. Russell BS, Muhlenkamp KA, Hoiriis KT, Desimone CM (2012)
Measurement of lumbar lordosis in static standing posture with
and without high-heeled shoes. J Chiropr Med 11:145–153
13. Drzal-Grabiec J, Snela S (2013) Effect of high-heeled shoes on
the parameters of body posture. Spine (Phila Pa 1976)
38:1785–1789
14. Snow RE, Williams KR (1994) High heeled shoes: their effect on
center of mass position, posture, three-dimensional kinematics,
rearfoot motion, and ground reaction forces. Arch Phys Med
Rehabil 75:568–576
15. Roussouly P, Gollogly S, Noseda O, Berthonnaud E, Dimnet J
(2006) The vertical projection of the sum of the ground reactive
forces of a standing patient is not the same as the C7 plumb line:
a radiographic study of the sagittal alignment of 153 asymp-
tomatic volunteers. Spine (Phila Pa 1976) 31:E320–E325
16. Dupont WD, Plummer WD (1998) Power and sample size cal-
culations for studies involving linear regression. Control Clin
Trials 19(6):589–601
17. McKenna C, Wade R, Faria R, Yang H, Stirk L, Gummerson N,
Sculpher M, Woolacott N (2012) EOS 2D/3D X-ray imaging
system: a systematic review and economic evaluation. Health
Technol Assess 16:1–188
18. Quian BP, Jiang J, Qui Y, Wang B, Yu Y, Zhu Z (2014) The
presence of a negative sacral slope in patients with ankylosing
spondylitis with severe thoracolumbar kyphosis. J Bone Jt Surg
Am 96:1–6
19. Scheer JK, Tang JA, Smith JS, Acosta FL Jr, Protopsaltis TS,
Blondel B, Bess S, Shaffrey CI, Deviren V, Lafage V, Schwab F,
Ames CP, International Spine Study G (2013) Cervical spine
alignment, sagittal deformity, and clinical implications: a review.
J Neurosurg Spine 19:141–159
20. Legaye J, Duval-Beaupere G, Hecquet J, Marty C (1998) Pelvic
incidence: a fundamental pelvic parameter for three-dimensional
regulation of spinal sagittal curves. Eur Spine J 7:99–103
21. Barrey C, Roussouly P, Le Huec JC, D’Acunzi G, Perrin G
(2013) Compensatory mechanisms contributing to keep the
sagittal balance of the spine. Eur Spine J 22(Suppl 6):834–841
22. Healy GN, Winkler EA, Owen N, Anuradha S, Dunstan DW
(2015) Replacing sitting time with standing or stepping: associ-
ations with cardio-metabolic risk biomarkers. Eur Heart J
36:2643–2649
23. Ho KY, Blanchette MG, Powers CM (2012) The influence of heel
height on patellofemoral joint kinetics during walking. Gait
Posture 36:271–275
24. Kerrigan DC, Todd MK, Riley PO (1998) Knee osteoarthritis and
high-heeled shoes. Lancet 351:1399–1401
25. Mika A, Oleksy L, Mikolajczyk E, Marchewka A, Mika P (2011)
Changes of bioelectrical activity in cervical paraspinal muscle
during gait in low and high heel shoes. Acta Bioeng Biomech
13:27–33
26. Han D (2015) Muscle activation of paraspinal muscles in dif-
ferent types of high heels during standing. J Phys Ther Sci
27:67–69

123