b i o m e d i c a l j o u r n a l 4 3 ( 2 0 2 0 ) 4 6 9 e4 7 5

Available online at www.sciencedirect.com

ScienceDirect
Biomedical Journal
journal homepage: www.elsevier.com/locate/bj

Original Article

Association of age in motor function outcomes

after multilevel myofascial release in children with
cerebral palsy

Chia-Hsieh Chang a, Chia-Ling Chen b,c, Kuo-Kuang Yeh b,d, Ken N. Kuo e,f,*
a
Department of Pediatric Orthopedics, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan
b
Department of Physical Medicine and Rehabilitation, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan
c
Graduate Institute of Early Intervention, College of Medicine, Chang Gung University, Taoyuan, Taiwan
d
Department of Physical Therapy, Graduation Institute of Rehabilitation Science, College of Medicine, Chang Gung
University, Taoyuan, Taiwan
e
Cochrane Taiwan, Taipei Medical University, Taipei, Taiwan
f
Department of Orthopaedic Surgery, National Taiwan University Hospital, Taipei, Taiwan

article info abstract

Article history: Background: Many recommend deferring orthopedic surgery for cerebral palsy-related dis-
Received 18 October 2018 orders in young children. However, age is correlated with musculoskeletal deterioration,
Accepted 14 October 2019 and deferral may affect surgical outcomes. We aimed to clarify the relationships among
Available online 3 December 2020 age, degree of musculoskeletal disorder, and postoperative motor function change in
children with cerebral palsy.
Keywords: Methods: We prospectively evaluated children with cerebral palsy and a knee flexion gait
Age disorder who underwent multilevel myofascial release between June 2010 and July 2014.
Cerebral palsy The children were divided into younger (<10 years of age) and older (>10 years of age)
Contracture groups. Outcome measures included the Gross Motor Function Measure (GMFM), range of
Gross motor function motion, spasticity, and physical capacity. Preoperative factors and postoperative changes
Multilevel surgery were compared between the groups using the chi-squared, independent t-, and Mann
eWhitney tests. Significant factors were plotted by participant age to identify the re-
lationships between age and other variables.
Results: We analyzed 20 patients who underwent multilevel myofascial release (12 and 8 in
the younger and older groups, respectively). Whereas most preoperative factors were
comparable between the two groups, the older group had a higher range of motion limi-
tation score (44.4 vs. 36.1, p < 0.05). The older group also showed less improvement in the
GMFM (
0.3 vs. þ3.0, p < 0.05) and physical capacity (þ0 vs. þ1, p < 0.05) scores after 6
months of postoperative rehabilitation.
Conclusions: Age was positively correlated with the range of motion limitation and nega-
tively correlated with postoperative GMFM improvement. The less favored postoperative
rehabilitation course in older children needs to be considered for parents whose children
are amenable to surgeries.

* Corresponding author. Department of Orthopaedic Surgery, Taipei Medical University, 250, Wuxing St., Taipei 110, Taiwan.
E-mail address: [email protected] (K.N. Kuo).
Peer review under responsibility of Chang Gung University.
https://doi.org/10.1016/j.bj.2019.10.003
2319-4170/© 2019 Chang Gung University. Publishing services by Elsevier B.V. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
470 b i o m e d i c a l j o u r n a l 4 3 ( 2 0 2 0 ) 4 6 9 e4 7 5

disorder). We tested the null hypothesis that older children

At a glance of commentary with CP have the same postoperative motor outcomes as
younger children after multilevel myofascial release.
Scientific background on the subject

Orthopedic surgery for gait disorder in cerebral palsy is

believed to be safer and better outcomes when per-
Methods
formed in older children. However soft tissue contrac-
Study design
ture and gait disorders often deteriorate with skeletal
growth. Optimal age of orthopedic surgery for gait dis-
We performed a prospective comparative study in our tertiary
orders is still controversial.
medical center between June 2010 and July 2014. This study
What this study adds to the field was approved by the Institutional Review Board of the au-
thors' affiliated institutions. We obtained written informed
Age was positively correlated with the range of motion consent for each child's participation from their parent or
limitation and negatively correlated with postoperative legal guardian.
improvement in gross motor function. Less and slower
postoperative motor function improvement was noted in Participants
children older than 10 years old. Results offer another
opinion for parents whose children are amenable to Children with spastic diplegic or quadriplegic CP who were
surgeries. scheduled to undergo bilateral multilevel soft tissue release
for a knee flexion gait disorder were included in the study.
Children who underwent osteotomy were excluded because
Cerebral palsy (CP) is a non-progressive neurological dis- the postoperative course of recovery after that procedure is
ease that leads to various musculoskeletal system disorders.
The hypertonic muscles cannot gain adequate length during
childhood skeletal growth, resulting in a progressively
decreased range of motion (ROM) and diminished motor
function [1]. In a longitudinal survey, Bell et al. reported that
walking function deteriorates after the ROM decreases in
ambulatory children with CP [2]. In adolescents with CP, sig-
nificant correlations have been reported between soft tissue
contracture and declining motor capacity [3]. Orthopedic
surgery is often indicated when conservative treatments can
no longer improve muscle and joint contractures [4e12].
In neurological diseases that cause progressive musculo-
skeletal system changes, age correlates with the deterioration
of musculoskeletal function [2,5] and could be a factor
affecting treatment outcomes [13]. Previous studies have re-
ported that joint deformity and muscle contracture recur-
rence were more common when surgery was performed in

children younger than 8 years old [14,15]. Svehlı́k et al. used
logistic regression to retrospectively analyze the gait deviation
index in a 10-year longitudinal study of children with CP. They
concluded that increased age at the time of surgery was the
most important predictor of a satisfactory long-term result
[16]. Additionally, they reported that the greatest improve-
ment in gait function was found in 10e12-year-old children
with a Gross Motor Function Classification System (GMFCS)
level III [17]. These findings suggest that among children with
CP, orthopedic surgery is safer and has better outcomes when
performed in older children. However, patients who undergo
surgery at an older age might have minor underlying neuro-
logical disorders and slower disease progression. This poten-
tial confounding factor should be controlled to address the
controversy about whether surgery should be performed
when indicated or deferred until the patient is older. There-
fore, we compared surgical outcomes in age-based groups of
children with CP with similar neurological involvement, gross
motor function, and surgical indications (knee flexion gait Fig. 1 Patient enrollment flowchart.
 b i o m e d i c a l j o u r n a l 4 3 ( 2 0 2 0 ) 4 6 9 e4 7 5 471

different from that following soft tissue release. Children with that for GMFCS level III was 36.1e65.4 [22]. Higher scores
a GMFCS level of I, IV, or V and hemiplegia were excluded to indicated better gross motor function.
prevent from ceiling and floor effects in the evaluation of We measured ROM limitation using the Spinal Alignment
motor function changes. Other exclusion criteria are listed in and Range of Motion Measure [23]. The bilateral hip (12 items),
Fig. 1. We defined knee flexion gait as a gait with a knee flexion knee (4 items), and ankle (4 items) scores were summed (0e80)
angle of 20 or higher throughout the stance phase or 30 or to represent the overall lower extremity ROM limitation;
higher at the terminal swing [18,19]. higher scores indicated more severe ROM limitations.
Children were divided in a younger (<10 years of age) and Spasticity was measured using the 5-point Modified Ash-
older (10 years of age) groups. The age 10-year division was worth Scale [24] which is scored as follows: 1, normal tone; 2,

chosen in accordance with the study by Svehlı́k et al. which mild spasticity with catching in limb movement or minimal
reported the greatest improvement in gait function in 10e12- resistance throughout the remainder (<50%) of the ROM; 3,
year-old children [17]. We intended to compare the short-term moderate spasticity with increased tone throughout most of
postoperative rehabilitation improvement in children the ROM; 4, severe spasticity with difficulty in passive motion;
younger and older than 10 years old. and 5, extreme spasticity with rigidity in flexion and exten-
sion. The scores for bilateral hamstring and gastrocnemius
Surgery and postoperative therapy muscles were summed to indicate the global spasticity in key
muscles responsible for knee flexion gait. The spasticity score
Single-event multilevel soft tissue releases of both lower ex- ranged from 020, and the scores increased with severity.
tremities were performed in all patients. The muscles that Physical capacity was evaluated on the basis of the number
required release were identified by preoperative clinical of times the participant could stand up from a squatting po-
evaluation of gait characteristics [18] and physical examina- sition in 30 s [25]. Light support was allowed for participants
tion. The surgeries varied among participants including pro- who might lose their balance while standing up. Physical ca-
cedures performed around the hips (tenotomy of the adductor pacity was graded as follows: 0, could not stand up even with
longus, gracilis, psoas, or combined), knees (myofascial support; 1, stood up once or twice with support; 2, stood up
release of the semimembranosus and semitendinosus mus- 3e8 times with support; 3, stood up > 8 times with support;
cles), and ankles (myofascial release of the gastrocnemius, and 4, stood up > 8 times without support; higher scores
tibialis posterior, or both). indicated better physical capacity.
Postoperatively, long leg splints were applied for 2 weeks We performed measurements of test-retest reliability at 2-
to maintain the knee flexion contracture correction and week intervals and assessed the interrater reliability for each
facilitate standing training. Non-articulated or ground re- of the four tests used to evaluate surgical outcomes.
action force ankle-foot orthoses were applied for gait
training. Physical therapy was conducted by two certified Statistical analysis
pediatric physical therapists. In the first 2 weeks, therapy
included standing, balance training, and strengthening of We compared preoperative baseline measurements and post-
the back and hip muscles. Thereafter, physical therapy operative changes between the two age groups. Baseline mea-
focused on strengthening the knee and hip muscles and gait surements included body mass index (BMI); GMFCS level; and
training using a walker. The children returned every 2 GMFM-66, ROM limitation, spasticity, and physical capacity
weeks for a checkup by research physical therapists for the scores. Surgical outcomes were measured by the changes in
first 6 weeks and then every 6 weeks until 6 months after GMFM-66, ROM limitation, spasticity, and physical capacity
surgery. scores. Chi-squared, independent t-, and ManneWhitney tests
Each patient underwent four assessments as follows: the were used for categorical (GMFCS level), continuous (BMI,
week before surgery and postoperatively at 6 weeks, 3 GMFM-66, ROM limitation), and ordinal (spasticity and physical
months, and 6 months to evaluate the changes following capacity scales) variables, respectively. Following the compar-
surgery and after rehabilitation. The outcome measures ison between age groups, significant factors were plotted by
included gross motor function, ROM limitation, spasticity, participant age to identify the relationships between age and
selective motor control, and physical capacity. These evalua- variables. Correlation between postoperative GMFM change
tions were performed by the same physical therapists who and preoperative baseline measurements was analyzed using
had provided postoperative physical therapy. Pearson’ correlation. The pre-operative factors that were
significantly correlated to GMFM change were entered multiple
Outcome measures regression test. We used SPSS Statistics for Windows, Version
20.0 (IBM Corp., Armonk, NY, US) for all analyses, and the level
The Gross Motor Function Measure (GMFM) is a standardized of significance was set at a p-value <0.05.
instrument used to quantify changes in gross motor ability in
children with CP [20,21]. We used the GMFM-66 that includes
66 items in 5 domains: lying & rolling, sitting, crawling & Results
kneeling, standing, and walking, running & jumping skills.
The GMFM-66 was modified from previous version of 88 items Participants
to improve the interpretability of changes following in-
terventions [21]. The maximal range of scores for children Fig. 1 shows the study flow diagram. Of 22 consenting par-
with GMFCS level II at age of 5e12 years was 44.8e92.2 and ticipants, two in the older group failed to attend all scheduled
472 b i o m e d i c a l j o u r n a l 4 3 ( 2 0 2 0 ) 4 6 9 e4 7 5

Table 1 Baseline data in 20 children with cerebral palsy.

No Age Sex BMI GMFCS GMFM ROM Spasticity Capacity Surgery
1 5.1 M 13.6 III 45.3 34 12 0 HKA
2 5.9 M 18.3 III 53.9 39 18 1 HKA
3 6.2 M 17.7 III 55.6 35 17 2 HK
4 6.9 M 18.2 III 50.6 42 18 0 HKA
5 7.7 M 21.2 II 74.2 27 15 4 HKA
6 8.1 F 13.6 III 54.6 44 16 3 KA
7 8.5 M 13.9 III 53.1 46 18 0 HKA
8 8.9 M 15.9 II 60.1 39 12 3 HKA
9 9.0 M 17.7 III 52.1 35 20 4 HKA
10 9.2 M 19.7 II 65 26 16 3 HK
11 9.3 M 14 II 76.8 32 19 3 KA
12 9.7 M 15.9 II 63.6 34 21 3 HKA
13 10.3 M 19.8 III 53.1 45 20 1 HKA
14 10.6 F 15.4 III 44.2 57 19 0 HKA
15 10.6 F 15.3 III 52.9 53 21 4 HKA
16 10.8 F 17 III 49.9 33 18 2 HK
17 11.6 M 16.9 II 65 32 17 3 HKA
18 11.8 M 14.9 II 75.3 36 23 3 HKA
19 11.8 M 14.8 II 68.9 51 12 4 HK
20 12.1 F 15.2 III 48.7 54 21 2 HKA

Abbreviations: A: surgeries around the ankles; BMI: body mass index; F: female; GMFCS: gross motor function classification system level; GMFM:
gross motor function measure score; H: surgeries around the hips; K: surgeries around the knees; M: male; ROM: lower extremity range of
motion limitation score.

clinic appointments and were excluded. Therefore, we revealed a positive correlation between age and preoperative
analyzed 20 patients (mean age, 9.2 years; age range, 5.1e12.1 ROM limitation scores [Fig. 2], indicating that age >10 years
years). In these patients, the mean GMFM-66 score was 58.1 was associated with more severe soft tissue contractures.
(range, 44.2e76.8), and all had spastic gait with excessive knee Postoperatively, the young patients experienced a decrease
flexion. Observational gait pattern analysis showed that 16 of GMFM scores at 6 weeks, recovery to pre-operative condi-
and 4 children had equinus and crouch gait, respectively. tion at 3 months, and improvement at 6 months. A common
Children with crouch gait received multilevel soft tissue condition in the older patients was decreasing GMFM scores at
release, sparing the ankle (cases 3, 10, 16, and 19; Table 1). post-operative 6 weeks and 3 months. Some recovered to pre-
After grouping the children by age, there were 12 and 8 in operative condition after 6 months, and the others did not
the younger and older groups, respectively. Preoperative BMIs; [Table 3].
GMFCS levels; and GMFM, spasticity, and physical capacity Compared with the younger group, the older group showed
scores were comparable between the two groups. Children in less improvement in GMFM scores (change from preoperative
the older group had higher ROM limitation scores than did
those in the younger group (44.4 vs. 36.1, respectively; t-test,
p ¼ 0.028, Table 2). The scatter plot of data from all subjects

Table 2 Preoperative measurements in children with

cerebral palsy grouped by age.
<10 years old >10 years old p-value
(n ¼ 12) (n ¼ 8)
Age 7.9 (1.5) 11.2 (0.7)
c
GMFCS (level II:III) 5:7 3:5 0.852
BMIa 16.6 (2.6) 16.2 (1.7) 0.648
GMFM66 scoresa 58.7 (9.5) 57.2 (11.1) 0.750
ROM limitation 36.1 (6.2) 44.4 (9.3) 0.028*
scoresa
Spasticity scoresb 17.5 (15.25e18.75) 19.5 (17.25e21) 0.111
Capacity scoresb 3 (0.25e3) 2.5 (1.25e3.75) 0.811

Continuous variables are reported as means (standard deviation),

and ordinal variables, as medians (interquartile ranges).
*p < 0.05. Fig. 2 Scatter plot of age-related preoperative range of
a
Independent t-test. motion limitation (ROM) scores in children with cerebral
b
Mann-Whitney test.
c
palsy. A positive correlation suggests that the range of
Chi-squared test.
motion decreases with age in growing children.
 b i o m e d i c a l j o u r n a l 4 3 ( 2 0 2 0 ) 4 6 9 e4 7 5 473

Table 3 Change at post-operative 6 months in the 20

children.
No GMFM ROM Spasticity Capacity
1 2.6
3 0 2
2 9.7
6
2 2
3 0.5
6 0 1
4 1.3
12
6 3
5 7.7
7
4 0
6 1.6
13
4 0
7 0.8
11
8 3
8
0.8
2 2 0
9 2.8
14
4 0
10 2.4 7 5 0
11 3.2 13
2 1
12
0.9
7
9 1
13 0.8
18
4 1
14
3.8
10
3 0
15 2.3
19
5 0
16
1.2
7
6
1 Fig. 3 Scatter plot of age-related gross motor function
17 3.1
16
7 0 measure (GMFM) changes in children with cerebral palsy 6
18
2.7 2
9 0 months postoperatively. A negative correlation suggests that
19 1.1
7
1 0
the benefits of surgery decrease with age.
20
2.4 9
5 0

Abbreviations: GMFM: Gross motor function measure score; ROM:

lower extremity range of motion limitation score.
respectively). The Spinal Alignment and Range of Motion
Measure test-retest and interrater reliabilities were good (ICC,
0.95e0.97 and 0.89, respectively). The Modified Ashworth
measurement,
0.3 vs. þ2.6 in the older vs. the younger group;
Scale test-retest and interrater reliabilities (ICC, 0.73e0.79 and
t-test, p ¼ 0.045). After 6 months of rehabilitation, the older
0.86, respectively) were moderate to good. The physical ca-
group showed lesser improvement in physical capacity than
pacity evaluation test-retest (ICC, 1.0) and interrater re-
the younger group (change from preoperative measurement,
liabilities (ICC, 1.0) were good.
0 vs. þ1 in the older vs. younger group; ManneWhitney test,
p ¼ 0.024; Table 4). The scatter plot of data from all subjects
revealed an inverse relationship between age and post-
Discussion
operative improvement in GMFM scores [Fig. 3].
The Pearson's correlation showed the postoperative GMFM
We compared the postoperative results after multilevel lower
change was significantly correlated to age (r ¼
0.535,
extremity soft tissue releases in age-based groups of children
p < 0.001), BMI (r ¼ 0.430, p < 0.05), and pre-Op ROM limitation
with CP and found that both age and muscle contracture affect
scores (r ¼
0.386, p < 0.05). The multiple regression showed
surgical outcomes in this population. The short-term outcome
age was the only significant factor for the post-operative
was not comparable to the long-term outcomes in the study
GMFM change (GMFM change ¼
0.45  age þ 1.86, p ¼ 0.042). 
by Svehlı́k [16,17]. However, the less favored rehabilitation
course after surgery in older children needs to be considered
Test-retest and interrater reliabilities when discussing surgical interventions with parents.
The relationships among age, motor function, and ROM
The GMFM test-retest and interrater reliabilities were excel- limitation are important in children with CP. Previous studies
lent (intraclass correlation coefficients [ICCs], 0.997 and 0.998, have revealed that GMFM scores increase with age through
the first few years of life [22,26]. In adolescents, gross motor
function may decline, and ROM limitation is the most signif-
Table 4 Changes 6 months after bilateral multilevel lower icant factor leading to this decline [3]. In our analysis, we
extremity soft tissue releases. found an inverse relationship between age and ROM. Older
<10 years old >10 years old p-value age and substantial ROM limitation were associated with
(n ¼ 12) (n ¼ 8) worse surgical outcomes. The relationships among increasing
GMFM66 scoresa 2.6 (3.2)
0.3 (2.5) 0.045* age, diminishing ROM, declining motor function, and less
ROM limitation scoresa
5.1 (8.1)
7.5 (9.9) 0.557 optimal surgical outcomes should be considered when a child
Spasticity scaleb
3 (
5.5e0.25)
5 (
6.75e3.25) 0.162 is amenable to surgical intervention.
Capacity scaleb 1 (0e2) 0 (0e0) 0.024* This study had different results from those of the studies
Continuous variables are reported as means (standard deviation), 
by Svehlı́k 
et al. [16,17]. Svehlı́k et al. retrospectively analyzed
and ordinal variables, as medians (interquartile ranges). 10-year longitudinal data of the gait deviation index and
*p < 0.05. concluded that the greatest improvement in gait function was
a
Independent t-test.
b found in 10e12-year-old children. This study showed that
Mann-Whitney test.
children older than 10 years old had less postoperative
474 b i o m e d i c a l j o u r n a l 4 3 ( 2 0 2 0 ) 4 6 9 e4 7 5

improvement in GMFM and physical capacity at 6 months universal and basic measurement of gross motor function and
postoperatively than did children who had surgery when has been used to record the post-operative change in children
younger than 10 years old. The different timing of assessment, with different GMFCS levels [9]. In this study we focused on
6 months postoperatively vs. 10 years postoperatively, and the the effects from age on short-term change in GMFM.
different outcome measures, motor function/capacity vs. gait In conclusion, our findings disprove the hypothesis that
deviation index, might be responsible for the different con- multilevel surgery at an older age has the same result as that

clusions in this study and the study by Svehlı́k et al. at a younger age in children with CP from a viewpoint of
Younger children have greater motor development poten- postoperative recovery and rehabilitation. Age was positively
tial than older children, and this could be an underlying factor associated with decreased ROM and negatively correlated
for the better postoperative outcomes we found in the with postoperative motor function improvement. This study
younger group. Performance of motor function could be sup- offers another opinion for parents whose children are
pressed by muscle contracture and spasticity in the pre- amenable to surgeries.
operative assessment. Motor development potential had the
chance to express properly after removing the suppression.
Data from this study suggested that young children can reach Funding
their expected motor potential after surgical treatment, while
in older children, they only plateaued at their existed preop- This study was funded by a research grant from Chang Gung
erative motor function. Evidence from a randomized Medical Foundation (CMRPG391531).
controlled trial indicated that compared with physical therapy
alone, surgery provided greater improvement in motor func-
tion in children with CP-related disorders who were amenable Conflicts of interest
to surgery [12].
Another factor that could have affected postoperative The authors declare no conflicts of interest.
improvement in our patients was compliance with physical
therapy. We found that younger patients participated in
postoperative physical therapy more regularly than older references
children did. The children in the older age group spent more
time participating in school activities, and two older group
patients did not complete the study. The improvement in the [1] Crawford GN. An experimental study of muscle growth in
the rabbit. J Bone Joint Surg Br 1954;36:294e303.
GMFM (activity/participation) shown in this study was related
[2] Bell KJ, Ounpuu S, DeLuca PA, Romness MJ. Natural
to spastic diplegic CP (health condition), ROM limitation (body progression of gait in children with cerebral palsy. J Pediatr
structure), age (personal factor), and school life (environ- Orthop 2002;22:677e82.
mental factor), in accordance with the model of the Interna- [3] Bartlett DJ, Hanna SE, Avery L, Stevenson RD, Galuppi B.
tional Classification of Functioning, Disability, and Health. Correlates of decline in gross motor capacity in adolescents
This study had several limitations. First, our postoperative with cerebral palsy in Gross Motor Function Classification
System levels III to V: an exploratory study. Dev Med Child
follow-up was for only 6 months. With a longer follow-up,
Neurol 2010;52:e155e60.
physical capacity and GMFM in the older group may have
[4] Kerr GH, Selber P. Musculoskeletal aspects of cerebral palsy. J
improved. Nonetheless, the most intensive rehabilitation was Bone Joint Surg Br 2003;85:157e66.
administered in the first 6 months, and most of the post- [5] Gough M, Eve LC, Robinson RO, Shortland AP. Short-term
operative improvement occurred during this period. Second, outcome of multilevel surgical intervention in spastic
the number of participants in each group was small after diplegic cerebral palsy compared with the natural history.
applying our exclusion factors. The responses to myofascial Dev Med Child Neurol 2004;46:91e7.
[6] Browne AO, McManus F. One-session surgery for bilateral
release were variable among children with cerebral palsy.
correction of lower limb deformities in spastic diplegia. J
Variable responses in a small number of study subjects can
Pediatr Orthop 1987;7:259e61.
lead to an unstable statistical result. The association between [7] Abel MF, Damiano DL, Pannunzio M, Bush J. Muscle-tendon
GMFM change and age became insignificant when the two surgery in diplegic cerebral palsy: functional and mechanical
cases (Nos. 2 and 5) with a greater improvement in the changes. J Pediatr Orthop 1999;19:366e75.
younger age group were regarded as outliers. Therefore, [8] Damiano DL, Abel MF, Pannunzio M, Romano JP.
greater sample size is needed. Third, the Ashworth Scale for Interrelationships of strength and gait before and after
hamstrings lengthening. J Pediatr Orthop 1999;19:352e8.
spasticity classifies ordinal variables. This study summed the
[9] Kondo I, Hosokawa K, Iwata M, Oda A, Nomura T, Ikeda K,
bilateral knee and ankle scores to characterize the condition et al. Effectiveness of selective muscle-release surgery for
of the entire lower extremity and effects of multilevel soft children with cerebral palsy: longitudinal and stratified
tissue releases. Fourth, physical capacity was assessed by analysis. Dev Med Child Neurol 2004;46:540e7.
squat-to-stand in this study. Walking speed, endurance, and [10] Nene AV, Evans GA, Patrick JH. Simultaneous multiple
other spatial gait parameters were not performed. Fifth, the operations for spastic diplegia. Outcome and functional
assessment of walking in 18 patients. J Bone Joint Surg Br
surgical indication was a knee flexion gait disorder. However,
1993;75:488e94.
we did not include a gait analysis to document the post-
[11] Saraph V, Zwick EB, Zwick G, Steinwender C, Steinwender G,
operative improvement in knee flexion angle. Further studies Linhart W. Multilevel surgery in spastic diplegia: evaluation
using the Gillette gait index or gait deviation index to measure by physical examination and gait analysis in 25 children. J
the change in gait pathology are desirable [27]. GMFM is a Pediatr Orthop 2002;22:150e7.
 b i o m e d i c a l j o u r n a l 4 3 ( 2 0 2 0 ) 4 6 9 e4 7 5 475

[12] Thomason P, Baker R, Dodd K, Taylor N, Selber P, Wolfe R, [20] Palisano RJ, Hanna SE, Rosenbaum PL, Russell DJ, Walter SD,
et al. Single-event multilevel surgery in children with spastic Wood EP, et al. Validation of a model of gross motor
diplegia: a pilot randomized controlled trial. J Bone Joint Surg function for children with cerebral palsy. Phys Ther
Am 2011;93:451e60. 2000;80:974e85.
[13] Chen HC, Kang LJ, Chen CL, Lin KC, Chen FC, Wu KP. Younger [21] Russell DJ, Avery LM, Rosenbaum PL, Raina PS, Walter SD,
children with cerebral palsy respond better than older ones Palisano RJ. Improved scaling of the gross motor function
to therapist-based constraint-induced therapy at home on measure for children with cerebral palsy: evidence of
functional outcomes and motor control. Phys Occup Ther reliability and validity. Phys Ther 2000;80:873e85.
Pediatr 2016;36:171e85. [22] Hanna SE, Bartlett DJ, Rivard LM, Russel DJ. Reference curves
[14] Borton DC, Walker K, Pirpiris M, Nattrass GR, Graham HK. for the Gross Motor Function Measure: percentiles for clinical
Isolated calf lengthening in cerebral palsy. Outcome analysis description and tracking over time among children with
of risk factors. J Bone Joint Surg Br 2001;83:364e70. cerebral palsy. Phys Ther 2008;88:1e12.
[15] Joo SY, Knowtharapu DN, Rogers KJ, Holmes Jr L, Miller F. [23] Bartlett D, Purdie B. Testing of the spinal alignment and
Recurrence after surgery for equinus foot deformity in range of motion measure: a discriminative measure of
children with cerebral palsy: assessment of predisposing posture and flexibility for children with cerebral palsy. Dev
factors for recurrence in a long-term follow-up study. J Child Med Child Neurol 2005;47:739e43.
Orthop 2011;5:289e96. [24] Bohannon RW, Smith MB. Interrater reliability of a modified
[16] Svehlı́k M, Steinwender G, Kraus T, Saraph V, Lehmann T, Ashworth scale of muscle spasticity. Phys Ther
Linhart WE, et al. The influence of age at single-event 1987;67:206e7.
multilevel surgery on outcome in children with cerebral [25] Voorman JM, Dallmeijer AJ, Knol DL, Lankhorst GJ, Becher JG.
palsy who walk with flexed knee gait. Dev Med Child Neurol Prospective longitudinal study of gross motor function in
2011;53:730e5. children with cerebral palsy. Arch Phys Med Rehabil

[17] Svehlı́k M, Steinwender G, Lehmann T, Kraus T. Predictors of 2007;88:871e6.
outcome after single-event multilevel surgery in children [26] Rosenbaum PL, Walter SD, Hanna SE, Palisano RJ, Russell DJ,
with cerebral palsy: a retrospective ten-year follow-up study. Raina P, et al. Prognosis for gross motor function in cerebral
Bone Joint Lett J 2016;98:278e81. palsy: creation of motor development curves. J Am Med
[18] Rodda JM, Graham HK, Carson L, Galea MP, Wolfe R. Sagittal gait Assoc 2002;288:1357e63.
patterns in spastic diplegia. J Bone Joint Surg Br 2004;86:251e8. [27] McMulkin ML, Mac Williams BA. Application of the Gillette
[19] Sutherland D, Davids JR. Common gait abnormalities of the gait index, gait deviation index and gait profile score to
knee in cerebral palsy. Clin Orthop Relat Res multiple clinical pediatric populations. Gait Posture
1993;288:139e47. 2015;41:608e12.