Original article 175
Evaluating upper body movements during gait in healthy
children and children with diplegic cerebral palsy
Jacqueline Romkesa, Wietske Peetersb, Aidia M. Oosteromb, Sara Molenaarb,
Iris Bakelsb and Reinald Brunnera
Movements of the lower limbs during gait have been
analysed extensively whereas data on upper body
movements are scarce. The aim of this study was to
evaluate upper body movements during gait in nine
healthy children and 10 children with diplegic cerebral
palsy. Children were investigated using a full-body marker
set to calculate the upper body kinematics of trunk and
arms. When the healthy children were compared with the
children with cerebral palsy, the latter compensated more
for their gait deviations and were less stable. This was
expressed by their greater variability in arm movements
and increased movements at the thorax. The thorax
showed an increased forward tilt with greater range of
motion over the gait cycle. The shoulders were more
abducted with increased elbow flexion. Gait analysis with
Introduction
Cerebral palsy (CP) is a disorder of posture and motor
impairment that results from a malformation or damage to
the developing central nervous system. The damage can
occur either in utero, during delivery, or during the first 2
years of life [1]. The brain damage is nonprogressive
although the clinical manifestation may vary with time.
Spasticity is the most obvious manifestation of CP. At
first, children are often hypotonic, but during the
development of the nervous system, muscle tone will
change and spasticity will arise [2]. Other peripheral
motor manifestations of the neurological injury include
loss of selective motor control, muscle weakness, and
imbalances between agonist and antagonist muscle pairs
[3].
Patterns of spasticity, with resulting muscle imbalance
across the joints over time, can produce secondary
deformities such as joint subluxations or dislocations,
bony deformities, muscle-tendon shortenings, and joint
contractures. In many cases, dynamic deformities and
movement disorders are accentuated during ambulation
or activity such as increased hip flexion, hip adduction,
knee flexion, ankle equinus, hind foot valgus, and toe
flexion [1,4].
The most common form of CP is the subgroup diplegia.
Patients with diplegia show bilateral involvement with
the lower extremities usually more severely affected than
c 2007 Lippincott Williams & Wilkins
1060-152X !
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
the full-body marker set has offered prospects for a better
understanding of compensatory mechanisms for the
pathological gait pattern in children with diplegic cerebral
palsy. J Pediatr Orthop B 16:175–180 ! c 2007 Lippincott
Williams & Wilkins.
Journal of Pediatric Orthopaedics B 2007, 16:175–180
Keywords: cerebral palsy, diplegia, gait analysis, upper body kinematics
a
University Children’s Hospital Basel, Switzerland and bVU University Medical
Center Amsterdam, The Netherlands
Correspondence to Jacqueline Romkes, Laboratory for Gait Analysis,
University Children’s Hospital Basel, Burgfelderstrasse 101, CH-4055 Basel,
Switzerland
Tel: + 41 61 326 4540; e-mail: j.romkes@unibas.ch
the upper extremities. The trunk and neck muscles are
normotonic, and patients typically have no problems with
their trunk and head control. Yet, the upper extremities
are often slightly affected. Children with diplegic CP
start walking at an older age than their healthy peer
group. Usually, they will be able to walk with or without
assistive devices depending on the severity of the
disorder [2]. The rate of energy expenditure required
for walking, however, is increased and as a consequence
the children often complain of fatigue [5].
Movements of the lower limbs during gait have been
analysed extensively [6,7], whereas data on upper body
movements are scarce. As obvious manifestations of CP
are spasticity, loss of selective motor control, muscle
weakness, and imbalances between agonist and antago-
nist muscle pairs it is expected that the pathological gait
pattern of children with CP will not only affect the lower
extremities but will also have consequences on the upper
body. Therefore, the aim of this study was to investigate
upper body movements during gait in healthy children
and in a group of children with diplegic CP.
Materials and methods
Participants
A total of 10 children with diplegic CP (four girls, six
boys, 11.0 ± 3.2 years) and nine healthy children (six
girls, three boys, 13.6 ± 3.4 years) ranging in age from 8 to
18 years participated in the study. The average height and
weight of the diplegic CP group was 141 ± 17 cm and
 176 Journal of Pediatric Orthopaedics B 2007, Vol 16 No 3

35.0 ± 10.8 kg and of the control group 158 ± 18 cm and overlying bony landmarks or on specific anatomical
47.1 ± 11.2 kg, respectively. All children were community positions. All children walked barefoot at a self selected
ambulators without assistive devices such as walkers or walking speed for a distance of 10 m. The measurements
crutches. Three children, however, generally walked with continued until kinematic information from at least eight
ankle foot ortheses and three children with Nancy Hylton trials was collected. Data collection and processing were
dynamic foot ortheses. Three children received surgery performed using a VICON 460 movement analysis system
more than 1 year before the investigation. Five children (Oxford Metrics, Oxford, UK) with six cameras at a
had botulinum toxin-A treatment in the past but the sampling rate of 120 Hz. The data were further analysed
injections were given at least 6 months before investiga- using the Workstation and Polygon software (Oxford
tion. Only data of barefoot walking were used for the Metrics). The data were expressed in degrees, and time
study. The parents of all children gave informed consent was normalized to percentage of gait cycle at 2% intervals.
for the scientific use of the data and the study was A gait cycle began and ended with two successive floor
approved by the local ethical committee. contacts of one foot. Average profiles over five trials were
generated at each of the 2% intervals of the gait cycle for
Kinematic data collection and analysis each participant. In cases for which more than five trials
A full-body marker set was used to evaluate kinematics of
pelvis, thorax, spine, shoulder, elbow, and wrist (Fig. 1).
The spine angles reflect the relative movement between Table 1 Summary of gait parameters in the sagittal plane
the pelvis and trunk and the angles of the thorax and Diplegic CP group Healthy group
pelvis segments reflect the absolute movement in space. Sagittal Mean ± 1 SD (1) Mean ± 1 SD (1) P values
The lower body was modelled according to the Helen
Pelvis ( + = anterior tilt)
Hayes Marker set [8] and the upper body was modelled Angle at IC 14.6 ± 8.7 12.3 ± 5.5 0.499
according to the Plug In Gait model (Vicon Motion Angle at TO 16.2 ± 7.5 12.0 ± 5.2 0.178
Systems) as described by Gutierrez et al. [9]. In total, 34 Peak angle in stance 19.6 ± 8.2 13.1 ± 5.6 0.064
% GC to peak in stance 20.6 ± 9.5 30.7 ± 13.2 0.071
reflective markers (+ 14 mm) were placed on the skin Peak angle in swing 18.8 ± 8.2 13.9 ± 4.5 0.135
% GC to peak in swing 77.0 ± 9.4 85.8 ± 8.8 0.051
ROM over GC 7.7 ± 3.1 2.7 ± 0.6 < 0.001
Thorax ( + = posterior tilt)
Fig. 1 Angle at IC – 12.1 ± 4.8 – 8.7 ± 3.5 0.104
Angle at TO – 11.0 ± 4.5 – 7.5 ± 3.8 0.088
Peak value – 9.1 ± 4.5 – 6.8 ± 4.2 0.267
Minimum value – 15.0 ± 4.2 – 10.0 ± 3.5 0.012
ROM over GC 5.9 ± 1.9 3.2 ± 0.9 < 0.001
SD over five trials 2.6 ± 2.2 1.6 ± 1.1 0.092
Spine ( + = anterior tilt)
Angle at IC – 3.7 ± 11.0 – 3.5 ± 7.6 0.966
Angle at TO – 6.3 ± 10.8 – 4.1 ± 7.2 0.620
Peak value – 0.5 ± 10.4 – 2.4 ± 7.4 0.659
Minimum value – 10.1 ± 11.2 – 5.7 ± 7.9 0.341
ROM over GC 9.6 ± 4.5 3.3 ± 1.3 < 0.001
SD over five trials 2.6 ± 2.2 1.8 ± 1.4 0.228
Shoulder ( + = flexion)
Angle at IC – 19.4 ± 13.4 – 16.3 ± 11.1 0.594
Angle at TO 1.7 ± 15.8 2.0 ± 7.1 0.954
Peak value 7.8 ± 13.6 6.5 ± 7.5 0.793
Time to peak (% GC) 44.4 ± 10.4 45.6 ± 4.0 0.758
Minimum value – 22.2 ± 12.4 – 17.4 ± 11.0 0.387
ROM over GC 30.0 ± 12.5 23.9 ± 15.6 0.352
SD over five trials 6.2 ± 3.7 3.2 ± 2.3 0.006
Elbow ( + = flexion)
Angle at IC 45.9 ± 12.6 31.5 ± 8.6 0.010
Angle at TO 59.6 ± 17.7 43.4 ± 9.4 0.025
Peak value 64.3 ± 14.1 45.9 ± 9.5 0.004
Time to peak (% GC) 53.6 ± 10.7 49.1 ± 6.3 0.288
Minimum value 41.9 ± 10.4 29.9 ± 8.3 0.013
ROM over GC 22.4 ± 9.1 16.1 ± 14.0 0.258
SD over five trials 7.9 ± 3.6 3.1 ± 1.5 < 0.001
Wrist ( + = extension)
Angle at IC 22.3 ± 15.0 13.0 ± 11.7 0.150
Angle at TO 32.4 ± 8.6 20.5 ± 6.3 0.003
Peak value 34.4 ± 7.8 21.2 ± 6.2 0.001
Time to peak (% GC) 55.4 ± 25.7 56.7 ± 12.7 0.895
Minimum value 16.4 ± 16.5 11.5 ± 11.9 0.478
ROM over GC 18.0 ± 12.7 9.7 ± 9.7 0.130
SD over five trials 10.8 ± 7.7 3.4 ± 2.8 < 0.001

GC, gait cycle; IC, initial contact; ROM, range of motion; SD, standard deviation;
Participant ready for data collection with a full-body marker set. TO, toe-off.
P < 0.05 = significant (written in bold).

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 Upper body movements during gait Romkes et al. 177

showed good data quality, the first five trials were taken Table 3 Summary of gait parameters in the transversal plane
for further analysis. The kinematic parameters analysed Diplegic CP group Healthy group
and the results are given in Tables 1–3. To describe the Transversal Mean ± 1 SD (1) Mean ± 1 SD (1) P values
variability of the gait pattern of one participant the Pelvis ( + = int. rotation)
standard deviation (SD) over five trials was calculated. Angle at IC 3.4 ± 10.9 5.5 ± 4.5 0.584
Angle at TO – 7.5 ± 7.8 – 3.4 ± 3.8 0.229
No significant differences were found when the right and Minimum angle – 9.4 ± 8.4 – 5.4 ± 4.1 0.221
left legs were compared within the groups; therefore, the % GC to min value 54.2 ± 17.3 60.1 ± 10.4 0.329
left leg was chosen for statistical analysis. The data sets of ROM over GC 16.7 ± 5.1 11.9 ± 7.8 0.124
Thorax ( + = ext. rotation)
the CP group were compared with the data sets of the Angle at IC – 2.0 ± 6.7 – 1.1 ± 3.3 0.712
control group with an independent Student’s t-test. Angle at TO 2.2 ± 8.5 2.4 ± 3.2 0.943
P-values of 0.05 or less were considered statistically Peak value 7.2 ± 7.1 3.5 ± 3.6 0.173
Minimum value – 5.3 ± 6.8 – 2.4 ± 3.4 0.265
significant. ROM over GC 12.5 ± 5.6 5.9 ± 2.0 0.005
SD over five trials 4.2 ± 2.2 2.4 ± 1.3 0.003
Spine ( + = int. rotation)
Angle at IC 5.6 ± 7.2 6.7 ± 5.2 0.729
Results Angle at TO – 9.1 ± 4.5 – 6.3 ± 5.0 0.212
The kinematic data (group average and SD) are described Peak value 8.7 ± 5.0 7.3 ± 5.1 0.553
Minimum value – 12.5 ± 5.2 – 7.5 ± 5.4 0.058
in Tables 1–3 for the sagittal, frontal, and transversal ROM over GC 21.1 ± 6.6 14.8 ± 8.2 0.078
plane. Figure 2 displays the group average and SD of the SD over five trials 3.1 ± 1.2 1.9 ± 1.2 0.005
Shoulder ( + = int. rotation)
kinematic movement curves during gait. Range of motion Angle at IC – 14.4 ± 14.5 2.1 ± 25.6 0.098
(ROM) of the pelvic tilt over the gait cycle was increased Angle at TO – 4.3 ± 10.5 7.9 ± 14.6 0.051
for patients, with the curves showing a ‘double-bump’ Peak value 1.5 ± 11.0 12.6 ± 16.3 0.098
Minimum value – 19.3 ± 14.4 – 1.8 ± 23.3 0.062
ROM over GC 20.8 ± 9.5 14.4 ± 11.2 0.193
SD over five trials 9.7 ± 7.2 4.4 ± 2.7 0.006
Table 2 Summary of gait parameters in the frontal plane
ext., external; GC, gait cycle; IC, initial contact; int., internal; ROM, range of
Diplegic CP group Healthy group motion; SD, standard deviation; TO, toe-off.
P < 0.05 = significant (written in bold).
Frontal Mean ± 1 SD (1) Mean ± 1 SD (1) P values

Pelvic obliquity ( + = up)

Angle at IC 0.0 ± 4.6 1.4 ± 2.7 0.461
Angle at TO – 4.1 ± 4.0 – 5.0 ± 1.7 0.554 pattern with two peaks in mid-stance and mid-swing
Peak angle stance 4.6 ± 4.3 4.1 ± 2.3 0.752
% GC to peak in 16.6 ± 3.5 12.9 ± 4.8 0.070
phase. For the thorax, the CP group demonstrated a
stance significantly greater ROM in all three planes compared
Minimum angle – 5.6 ± 3.6 – 5.5 ± 1.6 0.945 with the healthy control group and the thorax was more
% GC to min value 61.6 ± 13.2 63.8 ± 4.1 0.641
ROM over GC 10.2 ± 3.2 9.6 ± 2.5 0.655 tilted towards anterior in the sagittal plane. The ROM of
Thorax lateroflexion ( + = contralateral) the spine angles (i.e. relative motion between pelvis and
Angle at IC – 1.5 ± 4.6 – 0.8 ± 2.1 0.675
Angle at TO 3.6 ± 4.1 0.8 ± 2.3 0.090
thorax) was significantly elevated in the CP group in the
Peak value 4.5 ± 4.4 1.3 ± 1.9 0.060 sagittal and frontal planes. The CP group showed
Minimum value – 4.5 ± 6.2 – 1.7 ± 2.0 0.219 significantly more abduction of the arms as described by
ROM over GC 8.9 ± 6.9 3.0 ± 1.0 0.020
SD over five trials 1.7 ± 0.8 0.9 ± 0.7 0.002
the increased peak shoulder angle and ROM in the
Spine lateroflexion ( + = ipsilateral) frontal plane. The elbow showed more flexion as
Angle at IC 1.6 ± 5.9 2.3 ± 2.5 0.7617 expressed by the significantly increased flexion angles
Angle at TO – 7.9 ± 6.1 – 5.8 ± 2.6 0.361
Peak value 8.9 ± 7.5 5.4 ± 2.6 0.222 at initial contact and toe-off. In addition, the minimum
Minimum value – 9.6 ± 5.4 – 6.3 ± 2.6 0.115 value reached in the gait cycle showed more flexion.
ROM over GC 18.3 ± 7.2 11.6 ± 2.4 0.017
SD over five trials 1.7 ± 0.6 1.1 ± 0.7 0.008
ROM, however, was equal for both groups. Variability
Shoulder ( + = abduction) between trials (i.e. SD over five trials) within a patient
Angle at IC 9.6 ± 9.9 6.2 ± 7.9 0.419 with CP was on average significantly increased for all
Angle at TO 17.9 ± 10.0 11.4 ± 7.9 0.139
Peak value 22.5 ± 10.0 13.6 ± 8.0 0.049 upper body parameters that were analysed except for the
Time to peak (% GC) 47.6 ± 5.0 45.6 ± 3.0 0.298 thorax and spine in the sagittal plane. Note that the
Minimum value 7.4 ± 9.6 5.9 ± 7.8 0.719 group SDs for some angles are rather large, indicating
ROM over GC 15.1 ± 7.7 7.7 ± 4.1 0.020
SD over five trials 5.2 ± 2.5 1.9 ± 1.4 < 0.001 larger differences between participants and patients for
Wrist ( + = ulnar) these angles.
Angle at IC 14.8 ± 18.5 16.4 ± 13.9 0.839
Angle at TO 11.6 ± 14.2 17.9 ± 11.1 0.297
Peak value 18.3 ± 18.1 20.5 ± 12.7 0.766
Minimum value
ROM over GC
8.3 ± 14.3
10.0 ± 5.4
12.2 ± 10.7
8.3 ± 6.4
0.520
0.554
Discussion
SD over five trials 10.8 ± 5.6 4.0 ± 5.0 < 0.001 Lower body kinematics of pathological cerebral palsied
gait has been studied frequently. It is, however, expected
GC, gait cycle; IC, initial contact; ROM, range of motion; SD, standard deviation;
TO, toe-off. that the altered gait pattern will influence upper body
P < 0.05 = significant (written in bold). movements as well by means of compensation and for

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 178 Journal of Pediatric Orthopaedics B 2007, Vol 16 No 3

Fig. 2

Sagittal plane Fontal plane Transversal plane

40 20 30
Pelvic tilt Pelvic obliquity Pelvic rotation
35 15 20
30 10

Down degrees up
Post degrees ant

ext degrees int

25 10
5
20
0 0
15 0 20 40 60 80 100 0 20 40 60 80 100
−5
10 −10
5 −10
−20
0 −15
0 20 40 60 80 100
−5 −20 −30
10 15 15
Thorax tilt Thorax lateroflexion Thorax rotation
5 10 10

ipsi degrees contra

ant degrees post

int degrees ext

0 20 40 60 80 100 5 5
−5
0 0
−10 0 20 40 60 80 100 0 20 40 60 80 100
−5 −5
−15

−20 −10 − 10

−25 −15 − 10
20 20 20
spine tilt Spine lateroflexion Spine rotation
15 15 15
10
Contra degrees ipsi

10 10
Post degrees ant

ext degrees int

5
5 5
0
0 20 40 60 80 100 0 0
−5 0 20 40 60 80 100 0 20 40 60 80 100
−5 −5
−10
−15 −10 − 10
−20 −15 − 15
−25 −20 − 20
20 35 40
Shoulder flexion Shoulder abduction Shoulder rotation
30 30
10
25 20
Add degrees abd
ext degrees flex

ext degrees int

0 20
0 20 40 60 80 100 10
15
−10 0
10 0 20 40 60 80 100
−10
−20 5
0 −20
−30 0 20 40 60 80 100
−5 − 30
−40 −10 − 40
90 (%) Gait cycle
Elbow flexion
80
70
Degrees (flex)

60
50
40
30
20
10
0
45 35
Wrist deviation
Wrist extension 30
40
Diplegic CP group (mean ± 1SD)
35 25
Radial degrees ulnar

20
Degrees (ext)

30
Healthy control group (mean ± 1SD)
25 15
20 10
15 5
10 0
0 20 40 60 80 100
5 −5
0 − 10
0 20 40 60 80 100 (%) Gait cycle
(%) Gait cycle

Kinematic data in the sagittal, frontal, and transversal plane with the group averages (——) and 1 standard deviation (- - - -) for the diplegic cerebral
palsy (CP) group and the healthy control group.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

keeping balance and stability. Upper body kinematics has
been studied in normal individuals [10–13] and in
patients with pathology, in particular, myelomeningocele
[14,15] and patients with idiopathic scoliosis [16] or low
back pain [17]. For patients suffering from CP, however,
data are lacking; therefore, the aim of this study was to
evaluate upper body movements during gait in healthy
children and children with diplegic CP.
The results showed that the group of children with CP
had greater movements, that is increased ROM, of the
thorax (absolute angle in space) and spine (relative angle
between pelvis and thorax) especially in the sagittal and
frontal planes. The overall movement pattern of the
thorax and spine over the gait cycle of the healthy control
group is comparable with the normal data reported by
Nguyen and Baker [15]. Peak angles, that is maximum
and minimum values, occurred in both studies approxi-
mately at the same percentages of the gait cycle. Some
differences, however, in the magnitude of the angles were
noticed and can possibly be explained by using different
computer models.
Compared with the healthy control group, the CP group
showed an increase in forward thorax tilt. As the position
of the spine was not different between the two groups,
the pelvis in the CP group was more tilted towards
anterior as well.
At the moment, there is no consensus in terms of
segment definition or technique for recording and
analysing upper body kinematics. The Plug In Gait
model (Vicon Motion Systems) used in this study has
been described and validated by Gutierrez et al. [9] to
study the centre of mass during gait in children with
myelomeningocele. The model was chosen while it is
applicable in a clinical setting and involves both
kinematics of the thorax and the arms. In combination
with a lower body model, movements of the whole body
can be studied. The model, however, is not adequate to
analyse the relationship between spine mobility and gait.
Spine deformities, especially in the lumbar spine, in
ambulatory patients with spastic diplegic CP have been
reported in the literature [18–20]. Hyperlordosis has
been associated with increased anterior pelvic tilt [18],
whereas McCarthy and Betz [21] related hypolordosis to
tight hamstring muscles. For studying spine movement
more accurately in the future, it is suggested to place
subsequent markers on the spine as has been described
by, for example, Frigo et al. [11]. The increased complex-
ity and time needed for data collection, however, will
make a full-body analysis including a detailed spinal
analysis probably less clinically applicable.
The greater movements observed in the CP group, do not
seem to be accompanied by any time shifts. It is solely
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Upper body movements during gait Romkes et al. 179
the amplitude of the peak values that are increased and
no phase shifts between segments compared with the
healthy control group were observed (see Fig 2). This
implies that the coordination between segments does not
seem to be affected. In other pathologies, for example
chronic low back pain, changes in coordination by phase
shifts and timing were observed [17].
The children with CP showed more abduction in the
shoulders with increased elbow flexion. The spreading of
the arms outward can be compared to a tightrope artist
balancing on a rope. The arms are kept outward to gain
stability and balance. Furthermore, the data of the CP
group showed increased intra-subject variability, with the
exception of thorax and spine in the sagittal plane, as
expressed by an increase in SD calculated over five trials
within a participant. The hands have an open-end in
space, that is are not attached or influenced by external
forces except gravity. Therefore by positioning the arms
outward and further away from the body, the possibilities
to move in space increase and therefore intra-subject
variability is likely to be higher.
In conclusion, gait analysis with the full-body marker set
has offered prospects for a better understanding of
compensatory mechanisms for the pathological gait
pattern in children with diplegic CP. Movements of the
upper extremities are used as compensation for gait
deviation and/or for fine-tuning of balance control. To
make this mechanism more efficient, the arms are moved
away from the body. Future research can investigate the
influence of treatment interventions (e.g. orthotics,
surgery, and botulinum toxin-A) on upper body move-
ments.
Acknowledgements
The authors acknowledge the Foundation for Movement
Disorders (Stiftung für Bewegungsstörungen) and the
Stichting Anna Fonds for their financial support.
References
1 Koman LA, Smith BP, Shilt JS. Cerebral palsy. Lancet 2004; 363:
1619–1631.
2 Hefti F. Children’s orthopaedics in practice [German]. Berlin, Heidelberg:
Springer-Verlag; 1998.
3 Johnson DC, Damiano DL, Abel MF. The evolution of gait in childhood and
adolescent cerebral palsy. J Pediatr Orthop 1997; 17:392–396.
4 Steinwender G, Saraph V, Zwick EB, Steinwender C, Linhart W. Hip
locomotion mechanisms in cerebral palsy crouch gait. Gait Posture 2001;
13:78–85.
5 Waters RL, Mulroy S. The energy expenditure of normal and pathologic gait.
Gait Posture 1999; 9:207–231.
6 Perry J. Gait analysis: normal and pathological function. Thorofare, New
Jersey: Slack; 1992.
7 Gage JR, Deluca PA, Renshaw TS. Gait analysis: principles and
applications. J Bone Joint Surg 1995; 77-A:1607–1623.
8 Kadaba MP, Ramakrishnan HK, Wooten ME. Measurement of lower
extremity kinematics during level walking. J Orthop Res 1990; 8:
383–392.
9 Gutierrez EM, Bartonek Å, Haglund-Åkerlind Y, Saraste H. Centre of mass
motion during gait in persons with myelomeningocele. Gait Posture 2003;
18:37–46.
 180 Journal of Pediatric Orthopaedics B 2007, Vol 16 No 3
10 Van Emmerik REA, McDermott WJ, Haddad JM, Van Wegen EEH.
Age-related changes in upper body adaptation to walking speed in
human locomotion. Gait Posture 2005; 22:233–239.
11 Frigo C, Carabalona R, Dalla Mura M, Negrini S. The upper body segmental
movements during walking by young females. Clin Biomech 2003; 18:
419–425.
12 Thorstensson A, Carlson H, Zomlefer MR, Nilsson J. Lumbar back muscle
activity in relation to trunk movements during locomotion in man. Acta
Physiol Scand 1982; 116:13–20.
13 Cromwell RL, Aadland-Monahan TK, Nelson AT, Stern-Sylvestre SM,
Seder B. Sagittal plane analysis of head, neck, and trunk kinematics and
electromyographic activity during locomotion. J Orthop Sports Phys Ther
2001; 31:255–262.
14 Bartonek Å, Saraste H, Eriksson M, Knutson L, Cresswell AG. Upper
body movements during walking in children with lumbo-sacral
myelomeningocele. Gait Posture 2002; 15:120–129.
15 Nguyen TC, Baker R. Two methods of calculating thorax kinematics
in children with myelomeningocele. Clin Biomech 2004; 19:
1060–1065.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
16 Kramers-de Quervain IA, Müller R, Stacoff A, Grob D, Stüssi E. Gait
analysis in patients with idiopathic scoliosis. Eur Spine J 2004; 13:
449–456.
17 Lamoth CJC, Meijer OG, Daffertshofer A, Wuisman PIJM, Beek PJ.
Effects of chronic low back pain on trunk coordination and back
muscle activity during walking: changes in motor control. Eur Spine J 2006;
15:23–40.
18 Johnson MB, Goldstein L, Thomas SS, Piatt J, Aiona M, Sussman M.
Spinal deformity after selective dorsal rhizotomy in ambulatory patients with
cerebral palsy. J Pediatr Orthop 2004; 24:529–536.
19 Spiegel DA, Loder RT, Alley KA, Rowley S, Gutknecht S, Smith-Wright DL,
Dunn ME. Spinal deformity following selective dorsal rhizotomy. J Pediatr
Orthop 2004; 24:30–36.
20 Harada T, Ebara S, Anwar MM, Kajiura I, Oshita S, Hiroshima K, Ono K.
The lumbar spine in spastic diplegia. J Bone Joint Surg Br 1993; 75:
534–537.
21 McCarthy JJ, Betz RR. The relationship between tight hamstrings and
lumbar hypolordosis in children with cerebral palsy. Spine 2000; 25:
211–213.