Gait & Posture 64 (2018) 255–259

Contents lists available at ScienceDirect

Gait & Posture

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

Full length article

Dynamic spasticity determines hamstring length and knee flexion angle T

during gait in children with spastic cerebral palsy
⁎
Ja Young Choia, Eun Sook Parkb, Dongho Parkb, Dong-wook Rhab,
a
Department of Physical Medicine & Rehabilitation, Eulji University Hospital, Eulji University College of Medicine, Daejeon, Republic of Korea
b
Department of Rehabilitation Medicine, Severance Hospital, Research Institute of Rehabilitation Medicine, Yonsei University College of Medicine, Seoul, Republic of Korea

A R T I C LE I N FO A B S T R A C T

Keywords: Background: Previous researchers reported that popliteal angle did not correlate well with knee angle during gait
Cerebral palsy in individuals with spastic cerebral palsy (CP).
Spasticity Research question: To determine if hamstring spasticity, as measured by Modified Tardieu Scale (MTS) at rest, is
Tardieu scale associated with knee flexion angle at initial contact and midstance during gait.
Knee angle
Methods: Thirty ambulatory children (mean age 8.7 ± 2.4 years) diagnosed with spastic CP participated. The
Muscle length
hamstrings’ spasticity was assessed in the supine position with the MTS, measuring R1 (muscle reaction to
passive fast stretch), R2 (passive range of motion), and R2-R1 (dynamic component of spasticity). We conducted
3-dimensional computerized gait analysis and calculated semimembranosus muscle-tendon length and length-
ening velocity during gait using musculoskeletal modeling and inverse kinematic analysis by OpenSim. Pearson
correlation coefficients were calculated to estimate the association of MTS with biomechanical parameters
during gait.
Results: Knee flexion angle at initial contact and maximal knee extension angle during stance phase significantly
positively correlated with both R1 and ㅣR2 - R1ㅣ of MTS, but not with R2 angle. The length of semimem-
branosus at initial contact, end of swing, and minimal length during stance phase were strongly negatively
associated with R1, rather than R2 or ㅣR2 - R1ㅣ angles.
Significance: The R1 angle of MTS (muscle reaction to passive fast stretch) is more relevant correlate of knee
flexion angle during gait than the R2 (passive range of motion).

1. Introduction
Flexed knee gait is one of the most common gait abnormalities in
children with cerebral palsy (CP) and its frequency occurs as age in-
creases [1]. The etiology of flexed knee gait is multi-factorial, short and
spastic hamstrings are considered to be the main cause of this gait
abnormality [2]. Although many authors have reported that hamstring
lengthening is an effective treatment for improving flexed knee gait
[3,4], some previous studies have shown that hamstring lengths are not
actually shorter than usual in CP patients with flexed knee gait [5,6].
Measurement of the popliteal angle is a widely used clinical means
of assessing hamstring length in the supine position. However, the
popliteal angle did not correlate well with the knee angle during gait in
individuals with CP in several previous studies [5,7–9]. Assessment of
the popliteal angle is a static measurement in the supine position only;
therefore, it could be difficult to reflect the dynamic range of the knee
angle and hamstring length during gait. The medial hamstring muscle is
a biarticulate muscle that acts as both a hip extensor and a knee flexor,
and so it has a tremendous effect on the complex interaction of hip and
knee joints during gait.
Spasticity is commonly defined as a velocity dependent increase in
tonic stretch reflexes due to hyper‐excitability [10], while stiffness is a
mechanical resistance of the myotendinous tissue, as it is passively
lengthened. The Modified Tardieu Scale (MTS) is a clinical tool for
assessing spasticity that includes quantitative measurement [11]. Pas-
sive movements of MTS are tested at two speeds: as slow as possible
(R2), and as fast as possible (R1). As for knee flexor muscle, the R2
angle of MTS is equal to the popliteal angle. Although several studies
have investigated the correlation between popliteal angle and kine-
matic parameter of the knee joint during gait [5,7], there is a lack of
research on the relationship between dynamic assessment of spasticity
using MTS and gait analysis parameters.

Abbreviations: CP, cerebral palsy; MTS, modified Tardieu scale; ROM, range of motion; 3D, 3-dimensional
⁎
Corresponding author at: Department of Rehabilitation Medicine, Severance Hospital, Research Institute of Rehabilitation Medicine, Yonsei University College of Medicine, 50 Yonsei-
ro, Seodaemun-gu, Seoul, 03722, Republic of Korea.
E-mail address: MEDICUS@yuhs.ac (D.-w. Rha).

https://doi.org/10.1016/j.gaitpost.2018.06.163
Received 6 April 2018; Received in revised form 1 June 2018; Accepted 24 June 2018
0966-6362/ © 2018 Elsevier B.V. All rights reserved.
J.Y. Choi et al.
Therefore, the aim of this study was to investigate whether the use
of the MTS in children with CP to determine both dynamic spasticity
and mechanical resistance of hamstring muscles is associated with
hamstring length and knee flexion angle during gait.
2. Methods
This was a retrospective study conducted in a university-affiliated,
tertiary-care teaching hospital. Ethical approval was granted by the
institutional review board and ethics committee of our hospital (4-
2014-0516).
2.1. Participants
For this study, the medical records of children with CP, who were
referred to our motion analysis laboratory between December 2012 and
November 2014, were retrospectively reviewed. Inclusion criteria were
as follows: (1) able to walk independently without assistance (Gross
Motor Function Classification System [GMFCS] level I-II), (2) Modified
Ashworth Scale (MAS) at knee flexors ≥ 1+ with R2 − R1 angle of
MTS > 15 °, and (3) age 5–15 years. Exclusion criteria were as follows:
(1) chemodenervation therapy within the past 6 months, (2) previous
selective rhizotomy, intrathecal baclofen pump and orthopedic surgery,
(3) previous history of peripheral neuropathy or myopathy. In the case
of bilateral CP, we selected the more affected limb based on the MTS
score. Accordingly, we analyzed 30 limbs of 30 children with CP in this
study.
2.2. Assessments
2.2.1. Clinical measures of hamstring spasticity
The muscle tone of the hamstring was assessed with the MTS [11].
During the MTS, children were asked to relax in the supine position,
with the hip joint of the contralateral leg in a flexed position to elim-
inate residual anterior pelvic tilt [12,13]. The study leg was positioned
to 90-degree hip flexion with full-degree knee flexion. Then, the knee
was passively extended twice: the first time very slowly (> 5 s for the
entire range of motion), and the second time as fast as possible (< 1 s).
Two levels of the popliteal angle were measured by manual goniometry
referring to R2 and R1 angles, respectively. All clinical measures were
routinely recorded as part of the gait analysis by two experienced
physiatrists. The angle of muscle reaction (R1) referred to the point in
the range of motion (ROM) where a catch was first felt during a quick,
passive extension of the knee joint. By contrast, full range of motion
(R2) referred to the popliteal angle measured at the end of the move-
ment. The absolute difference between the two angles ( R2 − R1 ) re-
presented the dynamic components of spasticity [11]. Therefore, R2
represented only the mechanical resistance of hamstrings, whereas R1
was the summation of both mechanical resistance and the dynamic
spasticity of hamstrings.
2.2.2. Gait kinematics
Gait analysis was performed using a computerized 3D motion ana-
lysis (VICON MX-T10 Motion Analysis System, Oxford Metrics Inc.,
Oxford, UK) to measure kinematic data during the gait cycle. Subjects
were instrumented with 16 passive reflective markers according to
Helen Hayes marker set. Six digital videos were recorded simulta-
neously, from each video on the front and rear, and four videos on the
side, while the child walked barefoot on an 8 m pathway. Data from
three trials at a self-selected walking speed were collected for each
subject at a sampling rate of 100 Hz. We captured all data based on the
VICON Plug-in-Gait model; then we used NEXUS software version 1.8.5
to calculate joint kinematics, based on an average of three re-
presentative trials.
256
Gait & Posture 64 (2018) 255–259
2.2.3. Muscle-tendon length
We determined the muscle-tendon lengths and velocities of the
semimembranosus for each subject. The muscle-tendon length was es-
timated using a 3D computer model of the lower extremity (Lower Limb
Extremity Model 2010) [12]. We conducted inverse kinematic analysis
of motion capture data with this model using OpenSim, an open source
biomechanics simulation application [13]. In order to measure the
changes of muscle-tendon length during gait, we used the least-squares
formulation [14] to compute a set of desired joint angles for tracking
based on the marker trajectories from the gait analysis. We normalized
the muscle-tendon lengths based on the lengths when the hip and knee
were in the anatomic position, with all joint angles set at zero [6]. We
averaged data for three steps from a multi-step trial to analyze the
muscle-tendon length. We estimated muscle–tendon lengthening velo-
city by computing the numerical derivative of the muscle–tendon
length data with respect to time. We analyzed peak lengthening velo-
city during mid to end swing.
2.3. Statistical analysis
Statistical analysis was performed using the Statistical Package for
the Social Sciences for Windows (SPSS version 23.0, IBM SPSS
Incorporated, Chicago, IL, USA). Descriptive statistics, such as mean
and standard deviation (SD), were used to summarize patient demo-
graphics. Pearson correlation coefficients for parametric data and
Spearman’s rank correlation coefficients for nonparametric data were
computed to estimate the association of MTS with biomechanical
parameters. The level of significance was set to p-value < 0.05.
3. Results
In total, 30 ambulatory children with spastic cerebral palsy (18
unilateral and 12 bilateral; 16 boys and 14 girls), aged from 5 to 15
years (mean age of the children was 8.7 ± 2.4 years), whose GMFCS
level was I or II (GMFCS level I/II: 19/11) participated in this study.
The characteristics of the subjects are described in Table 1.
3.1. Correlation between MTS and the kinematic data during gait
Positive values of kinematic data indicate pelvic anterior tilt and
knee flexion, while negative values indicate pelvic posterior tilt and
knee extension. Table 2 shows the correlation coefficients between the
MTS and the sagittal knee angle on gait analysis. The absolute value of
R2 - R1 (ㅣR2 - R1ㅣ) of MTS moderately positively correlated with the
knee angle at initial contact (r = 0.482; p < 0.01), and knee angle at
end swing (r = 0.543; p < 0.01), and weakly correlated with maximal
knee extension during stance phase (rs = 0.387; p < 0.05). R1 of MTS
also showed a moderate positive correlation with the knee angle at
initial contact (r = 0.490; p < 0.01) and end swing (r = 0.582;
Table 1
Participant characteristics.
Characteristic No./Valuea
No. of participants 30
No. of legs 30
Most affected side, right/left 11 / 19
Gender, male/female 16 / 14
Age at gait analysis, years 8.7 ± 2.4 (5–15)
GMFCS level, I / II 19 / 11
Type of cerebral palsy, unilateral/ bilateral 18 / 12
Modified Tardieu scale of knee flexor
R1 angle, degrees 65.00 ± 20.55 (20–110)
R2 angle, degrees 36.83 ± 13.68 (5–60)
ㅣR2 - R1ㅣangle, degrees 28.17 ± 12.21 (15–70)
GMFCS: Gross motor functional classification system.
a
Values are mean ± standard deviation (range).
J.Y. Choi et al. Gait & Posture 64 (2018) 255–259

Table 2
correlation coefficients between knee flexor Modified Tardieu Scale and kinematic data of sagittal plane.
Knee angle Pelvis angle

Parameters Initial contact Max Ext stance Max Fl swing End Anterior tilting at initial contact Maximal Anterior tilting
swing anterior tilting at End swing

R1 of MTS
Correlation .490b .387a .069 .582b −.282 −.175 −.311
p-value .006 .034 .719 .001 .131 .354 .094
R2 of MTS
Correlation .306 .160 .073 .390a −.054 −.007 −.104
p-value .100 .398 .703 .033 .778 .972 .584
ㅣR2 - R1ㅣ of MTS
Correlation .482b .387c .034 .543b −.415a −.288 −.408a
p-value .007 .035 .858 .002 .023 .123 .025

MTS, modified Tardieu scale; Max, maximal; Ext, extension; Fl, flexion.
Positive values of kinematic data indicate pelvic anterior tilt and knee flexion, while negative values indicate pelvic posterior tilt and knee extension.
a
Correlation is significant at the 0.05 level by Pearson’s correlation coefficient.
b
Correlation is significant at the 0.01 level by Pearson’s correlation coefficient.
c
Correlation is significant at the 0.05 level by Spearman’s rank correlation coefficient.

p < 0.01), and a weak positive correlation with maximal knee exten-
sion during stance phase (r = 0.387; p < 0.05). R2 only weakly posi-
tively correlated with the knee flexion angle at end swing (r = 0.390;
p < 0.05). Maximal knee flexion angle during the swing phase did not
correlate with MTS.
For pelvic kinematics, ㅣR2 - R1ㅣ negatively correlated with the
pelvic anterior tilting at initial contact (r = -0.415; p < 0.05) and end
swing (r = −0.408; p < 0.05). This result indicates that a higher
dynamic spasticity of hamstring was related to more posterior pelvic tilt
(Table 2).
A scatterplot showing the significant relationship between the R1 of
MTS and the knee angle at initial contact with a regression line is shown
in Fig. 1.
3.2. Correlation between MTS and the muscle-tendon length during gait
As for semimembranosus, R1 had the highest correlation with
muscle length, compared to R2 or R2 - R1 (Table 3). R1 of MTS showed
a moderate negative correlation with the muscle length of semimem-
branosus at initial contact (r = -0.405; p < 0.05) (Fig. 1), and minimal
length during stance phase (r = −0.494; p < 0.05) and end swing (r
= -0.443; p < 0.05), while R2 showed only a weak negative correla-
tion at end swing (r = −0.386; p < 0.05). By contrast, ㅣR2 - R1ㅣ of
MTS demonstrated a moderate negative correlation with the semi-
membranosus muscle length at only minimal stance phase (r = -0.430;
p < 0.05). In addition, the lengthening velocity of semimembranosus
during the swing phase did not significantly correlate with the MTS of
the hamstring.
4. Discussion
This study demonstrated that the R1 or R2 - R1 of the hamstring
MTS, which represent the dynamic component of spasticity, sig-
nificantly correlated with knee angle and hamstring length during gait,
unlike the R2, which represents the passive range of motion at the knee
joint.
We tested the MTS of the hamstring with the hip in the 90-degree
flexed position, placing the hamstrings at their maximal stretch across
the hip and knee [15]. The spasticity is activated when the muscle is
stretched to the R1 position; and then the viscoelasticity of the soft
tissues and joints and the component of contracture may have come
into play until the muscle is stretched to the R2 position [16]. Thus, the

Fig. 1. Correlation between the R1 angle of the Modified Tardieu Scale and the knee flexion angle (A) and semimembranosus length (B) at initial contact. Knee
flexion angle and semimembranosus length at initial contact were significantly correlated with the R1 angle.

257
J.Y. Choi et al. Gait & Posture 64 (2018) 255–259

Table 3 gait than static contracture.

correlation coefficients between knee flexor Modified Tardieu Scale and the The hamstring is a biarticular muscle that affect both the knee and
length of semimembranosus. hip joints. In addition, the hip angle itself can have important mod-
Semimembranosus length Lengthening ulating effects on hamstring stretch reflexes [22]. Therefore, we as-
velocity sessed the pelvic angle during gait and only the absolute value of R2 -
Parameters Initial Min stance End swing during end swing R1 negatively correlated with the pelvic angle in the sagittal pelvic
contact
kinematic analysis, meaning the dynamic spasticity of hamstrings
R1 of MTS promoted posterior pelvic tilting, especially at initial contact and end
Correlation −.405a −.494b −.443a −.060 swing.
p-value .026 .006 .014 .753
R2 of MTS
4.2. Correlation between MTS and the muscle-tendon length during gait
Correlation −.340 −.358 −.386a .041
p-value .066 .052 .035 .830
ㅣR2 - R1ㅣ of In the muscle-tendon length study, the R1 angle of MTS showed a
MTS moderate negative correlation with semimembranosus length at both
Correlation −.178 −.430a −.178 −.190 the stance and swing phase. This means that muscle reaction to ex-
p-value .346 .018 .348 .315
aggerated stretch reflexes is associated with a short hamstring length
MTS, modified Tardieu scale. during gait. In contrast, the popliteal angle, R2 of MTS, only weakly
Min stance, minimal length during stance. correlated with semimembranosus length, and only at the end swing.
a
Correlation is significant at the 0.05 level by Pearson’s correlation coeffi- There is a paucity of studies on the correlation between popliteal angle
cient. and hamstring length during gait. Delp et al. [5] concluded that the
b
Correlation is significant at the 0.01 level by Pearson’s correlation coeffi- popliteal angle was not a valid indicator of maximum hamstring length
cient. during the gait cycle, whereas Thompson et al. [7] showed that the
popliteal angle significantly correlated with maximal hamstring length
R2 - R1 represents the dynamic spasticity and R1 represents the sum- during gait, but only when performed by the most reliable examiner
mation of both mechanical resistance and the dynamic spasticity of and when using a modified method. In their study, the modified po-
hamstrings. pliteal angle of > 40° used by the most reliable examiner may indicate
that the medial hamstrings are short. Our study is different because: (1)
4.1. Correlation between MTS and the kinematic data during gait the sample size was more than two times greater; (2) we used the
modified method assessing the popliteal angle by a reliable examiner.
This study found no significant correlation between the popliteal In the modified method, the contralateral hip is flexed to eliminate
angle and the knee flexion angle during stance phase, and only a weak anterior pelvic tilt caused by hip flexor tightness [7]. Although the
correlation was noted at the end swing. Although most previous studies modified method of measuring the popliteal angle is the more reliable
showed no significant correlation between the popliteal angle and the technique to reflect hamstring length, most previous studies used the
knee angle during the gait cycle [8,9], some studies still suggest that conventional method to measure the popliteal angle, with the contra-
such a relationship does exist. Desloovere et al. [17] showed a negative lateral hip in a neutral position [9] or not mentioned at all [8,17].
correlation between the popliteal angle and maximal knee extension in The hamstring muscle is stretched to its maximal length at initial
the stance phase, but the correlation was weak (r = -0.28, P < 0.01). contact and the terminal swing phase. This may be the reason that the
Faber et al. [18], by contrast, showed that the popliteal angle (R2), not popliteal angle showed a weak correlation with hamstring length at
the stretch restricted angle (R1), was strongly associated with the only the end swing in our study. During the swing phase of normal gait,
maximum knee extension at the end swing, suggesting that mechanical the hamstring muscles are rapidly stretched, through the combined
resistance of the hamstrings is more important for knee angle during action of knee extension and hip flexion. Hamstring muscles were re-
gait. This result partly agreed with our finding of a weak correlation ported to lengthen slowly in the second half of the swing phase in the
between the popliteal angle (R2) and the knee flexion angle only at end children with CP, and it may be caused by their shortening or spasticity
swing phase. However, R1 and R2 - R1 showed a higher correlation [23]. However, in our study there was no significant correlation be-
with knee flexion angles during the gait than R2 did in our study. tween MTS and the peak lengthening velocity of semimembranosus
Previous studies measured clinical parameters similar to R1, named during the swing phase. This means that the lengthening velocity is
“initial” or “first” popliteal angle [9,19]. These terms were defined as related not only to dynamic spasticity and muscle contracture, but also
the muscle reaction during stretch; however, the stretch velocity was to other factors, including co-contraction or selective motor control.
not as fast as the R1 we measured in this study. McMulkin et al. [9] Selective motor control is required, especially during the terminal
investigated the correlation between initial and final popliteal angles swing, when the combination of hip flexion and knee extension motion
and knee flexion angle in the stance phase, revealing a Pearson’s cor- is needed [24]. Impaired selective motor control was reported to in-
relation coefficient of −0.49. Whereas Cooney et al. [19] declared the fluence knee position at initial contact, which is associated with de-
initial popliteal angle was significantly correlated with knee extension creased knee extension during the swing phase [25].
at the terminal swing. The Cooney et al. study partially supports our Additionally, the popliteal angle did not show a significant corre-
findings, although the first resistance point in their study was not equal lation with hamstring length during the stance phase in our study.
to the R1 of MTS in our study. Spastic hamstrings have been found to Hoffinger et al. [6] found that the hamstring lengths during the stance
have lower activation thresholds during passive stretch [20]. Perry and phase in most patients with crouch gait were longer than the resting
Newsam [21] showed that patients with a flexed knee gait exhibit length and that hamstrings functioned as hip extensors during a sig-
premature firing of the hamstrings during the swing phase. This pre- nificant portion of the stance phase. This issue of the impact of ham-
mature activation of the hamstring is related to R1 of the MTS and is a string lengthening on knee flexion during stance is very much debated
dynamic component of spasticity. Reduced knee extension at the [3,4]. Previous studies show that many children with flexed knee gait
terminal swing might be explained by a premature onset of the ham- do not have shortened hamstrings, and our study findings also support
string activity due to exaggerated stretch reflexes. This study is the first this statement [5].
to report the significant correlation of R2 - R1, a dynamic component of This study indicates that a neural component of spasticity, other
MTS, with the knee flexion angle. Our study revealed that dynamic than static shortening or contracture, plays critical role in knee angle
spasticity had a stronger correlation with knee kinematics during the and hamstring length during the gait. In the same context, a child with

258
J.Y. Choi et al.
a dynamic but not yet static hamstring contracture may be better
treated with chemodenervation, including botulinum toxin or in-
trathecal baclofen, rather than orthopedic surgery, to improve knee
extension at the end swing and the concatenate initial contact [26]. We
need further studies with more subjects to confirm the effects of ham-
string muscle botulinum toxin injection on gait functions of children
with spastic CP.
4.3. Limitations
There are several limitations of this study. First, some children with
CP have bone deformities, but the musculoskeletal model used to esti-
mate the hamstring lengths in this study didn’t consider the effects of
variable bone deformities in each child on alteration of moment arm
and length of hamstring muscle. Future studies should develop the
musculoskeletal model to reflect these skeletal deformities.
Another limitation of our study is the standardization of the MTS.
Although reliable hamstring MTS in children with CP have been re-
ported [27], it can be difficult to determine the accurate angle of catch
in R1 measurements. Recently, instrumented clinical tests have been
developed to analyze biomechanical and electrophysiological muscle
responses during passive stretch [28]. However, our study is mean-
ingful to the clinician in an office setting who wishes to estimate the
biomechanical changes during gait based on the more feasible grading
of MTS.
5. Conclusion
The popliteal angle (R2 of MTS) did not significantly correlate with
the knee flexion angle or with hamstring length during stance phase,
and only weakly correlated at the end swing. By contrast, the muscle
reaction angle to exaggerated stretch reflexes (R1 of MTS) is a more
relevant parameter with the knee flexion angle during gait than passive
ROM of R2. In addition, the dynamic component of spasticity (R2 - R1)
is associated with the knee flexion and pelvic tilting angles during gait.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
This study was supported by a faculty research grant of Yonsei
University College of Medicine for 2014 (6-2014-0065).
References
[1] G.E. Rose, K.A. Lightbody, R.G. Ferguson, J.C. Walsh, J.E. Robb, Natural history of
flexed knee gait in diplegic cerebral palsy evaluated by gait analysis in children who
have not had surgery, Gait Posture 31 (3) (2010) 351–354.
[2] S.H. Dhawlikar, L. Root, R.L. Mann, Distal lengthening of the hamstrings in patients
who have cerebral palsy. Long-term retrospective analysis, J. Bone Jt. Surg. Am. 74
(9) (1992) 1385–1391.
[3] W.N. Chang, A.I. Tsirikos, F. Miller, N. Lennon, J. Schuyler, L. Kerstetter, et al.,
Distal hamstring lengthening in ambulatory children with cerebral palsy: primary
versus revision procedures, Gait Posture 19 (3) (2004) 298–304.
259
Gait & Posture 64 (2018) 255–259
[4] P.A. DeLuca, S. Ounpuu, R.B. Davis, J.H. Walsh, Effect of hamstring and psoas
lengthening on pelvic tilt in patients with spastic diplegic cerebral palsy, J. Pediatr.
Orthop. 18 (6) (1998) 712–718.
[5] S.L. Delp, A.S. Arnold, R.A. Speers, C.A. Moore, Hamstrings and psoas lengths
during normal and crouch gait: implications for muscle-tendon surgery, J. Orthop.
Res. 14 (1) (1996) 144–151.
[6] S.A. Hoffinger, G.T. Rab, H. Abou-Ghaida, Hamstrings in cerebral palsy crouch gait,
J. Pediatr. Orthop. 13 (6) (1993) 722–726.
[7] N.S. Thompson, R.J. Baker, A.P. Cosgrove, J.L. Saunders, T.C. Taylor, Relevance of
the popliteal angle to hamstring length in cerebral palsy crouch gait, J. Pediatr.
Orthop. 21 (3) (2001) 383–387.
[8] M.S. Orendurff, J.S. Chung, R.A. Pierce, Limits to passive range of joint motion and
the effect on crouch gait in children with cerebral palsy, Gait Posture 7 (2) (1998)
165.
[9] M.L. McMulkin, J.J. Gulliford, R.V. Williamson, R.L. Ferguson, Correlation of static
to dynamic measures of lower extremity range of motion in cerebral palsy and
control populations, J. Pediatr. Orthop. 20 (3) (2000) 366–369.
[10] J.W. Lance, The control of muscle tone, reflexes, and movement: Robert Wartenberg
Lecture, Neurology 30 (12) (1980) 1303–1313.
[11] R.N. Boyd, H.K. Graham, Objective measurement of clinical findings in the use of
botulinum toxin type a for the management of children with cerebral palsy, Eur. J.
Neurol. 6 (S4) (1999) s23–s35.
[12] E.M. Arnold, S.R. Ward, R.L. Lieber, S.L. Delp, A model of the lower limb for
analysis of human movement, Ann. Biomed. Eng. 38 (2) (2010) 269–279.
[13] S.L. Delp, F.C. Anderson, A.S. Arnold, P. Loan, A. Habib, C.T. John, et al., OpenSim:
open-source software to create and analyze dynamic simulations of movement,
IEEE Trans. Biomed. Eng. 54 (11) (2007) 1940–1950.
[14] T.W. Lu, J.J. O’Connor, Bone position estimation from skin marker co-ordinates
using global optimisation with joint constraints, J. Biomech. 32 (2) (1999)
129–134.
[15] J.L. Young, J. Rodda, P. Selber, E. Rutz, H.K. Graham, Management of the knee in
spastic diplegia: what is the dose? Orthop. Clin. North. Am. 41 (4) (2010) 561–577.
[16] W.K. Yam, M.S. Leung, Interrater reliability of modified ashworth scale and mod-
ified tardieu scale in children with spastic cerebral palsy, J. Child Neurol. 21 (12)
(2006) 1031–1035.
[17] K. Desloovere, G. Molenaers, H. Feys, C. Huenaerts, B. Callewaert, P. Van de Walle,
Do dynamic and static clinical measurements correlate with gait analysis para-
meters in children with cerebral palsy? Gait Posture 24 (3) (2006) 302–313.
[18] I.R. Faber, B. Nienhuis, N.P. Rijs, A.C. Geurts, J. Duysens, Is the modified tardieu
scale in semi-standing position better associated with knee extension and hamstring
activity in terminal swing than the supine Tardieu? Dev. Med. Child Neurol. 50 (5)
(2008) 382–387.
[19] K.M. Cooney, J.O. Sanders, M.C. Concha, F.L. Buczek, Novel biomechanics de-
monstrate gait dysfunction due to hamstring tightness, Clin. Biomech. (Bristol,
Avon) 21 (1) (2006) 59–66.
[20] L. Bar-On, E. Aertbelien, G. Molenaers, K. Desloovere, Muscle activation patterns
when passively stretching spastic lower limb muscles of children with cerebral
palsy, PLoS One 9 (3) (2014) e91759.
[21] J. Perry, C. Newsam, Function of the hamstrings in cerebral palsy, in: Sussman
(Ed.), The Diplegic Child: Evaluation and Management, American Academy of
Orthopaedic Surgeons, Rosemont, IL, 1992, pp. 299–307.
[22] J.J. Visser, J.E. Hoogkamer, M.F. Bobbert, P.A. Huijing, Length and moment arm of
human leg muscles as a function of knee and hip-joint angles, Eur. J. Appl. Physiol.
Occup. Physiol. 61 (5-6) (1990) 453–4560.
[23] P. Crenna, Spasticity and ‘spastic’ gait in children with cerebral palsy, Neurosci.
Biobehav. Rev. 22 (4) (1998) 571–578.
[24] E.G. Fowler, E.J. Goldberg, The effect of lower extremity selective voluntary motor
control on interjoint coordination during gait in children with spastic diplegic
cerebral palsy, Gait Posture 29 (1) (2009) 102–107.
[25] D.W. Rha, K. Cahill-Rowley, J. Young, L. Torburn, K. Stephenson, J. Rose,
Biomechanical and clinical correlates of stance-phase knee flexion in persons with
spastic cerebral palsy, PMR (2015), http://dx.doi.org/10.1016/j.pmrj.2015.06.003.
[26] D.L. Damiano, Rehabilitative therapies in cerebral palsy: the good, the not as good,
and the possible, J. Child Neurol. 24 (9) (2009) 1200–1204.
[27] J.M. Gracies, K. Burke, N.J. Clegg, R. Browne, C. Rushing, D. Fehlings, et al.,
Reliability of the Tardieu Scale for assessing spasticity in children with cerebral
palsy, Arch. Phys. Med. Rehabil. 91 (3) (2010) 421–428.
[28] L. Bar-On, E. Aertbelien, H. Wambacq, D. Severijns, K. Lambrecht, B. Dan, et al., A
clinical measurement to quantify spasticity in children with cerebral palsy by in-
tegration of multidimensional signals, Gait Posture 38 (1) (2013) 141–147.