Clinical Nutrition 32 (2013) 579e584

Contents lists available at SciVerse ScienceDirect

Clinical Nutrition
journal homepage: http://www.elsevier.com/locate/clnu

Original article

The use of bioelectrical impedance analysis to estimate total body

water in young children with cerebral palsyq
Kristie L. Bell a, b, c, d, *, Roslyn N. Boyd a, b, Jacqueline L. Walker a, c, Richard D. Stevenson e,
Peter S.W. Davies c
a
Queensland Cerebral Palsy and Rehabilitation Research Centre, School of Medicine, The University of Queensland, Australia
b
Department of Paediatric Rehabilitation, Royal Children’s Hospital, Brisbane, Queensland, Australia
c
Children’s Nutrition Research Centre, School of Medicine, The University of Queensland, Australia
d
Queensland Children’s Medical Research Institute, The University of Queensland, Australia
e
Division of Developmental Pediatrics, University of Virginia (UVA) School of Medicine, and UVA Children’s Hospital, USA

a r t i c l e i n f o s u m m a r y

Article history: Background & aims: Body composition assessment is an essential component of nutritional evaluation in
Received 3 August 2012 children with cerebral palsy. This study aimed to validate bioelectrical impedance to estimate total body
Accepted 4 October 2012 water in young children with cerebral palsy and determine best electrode placement in unilateral
impairment.
Keywords: Methods: 55 young children with cerebral palsy across all functional ability levels were included. Height/
Cerebral palsy
length was measured or estimated from knee height. Total body water was estimated using a Bodystat
Children
1500MDD and three equations, and measured using the gold standard, deuterium dilution technique.
Body composition
Bioelectrical impedance analysis
Comparisons were made using Bland Altman analysis.
Results: For children with bilateral impairment, the Fjeld equation estimated total body water with the
least bias (limits of agreement): 0.0 L (
1.4 L to 1.5 L); the Pencharz equation produced the greatest: 2.7 L
(0.6 Le4.8 L). For children with unilateral impairment, differences between measured and estimated total
body water were lowest on the unimpaired side using the Fjeld equation 0.1 L (
1.5 L to 1.6 L)) and
greatest for the Pencharz equation.
Conclusions: The ability of bioelectrical impedance to estimate total body water depends on the equation
chosen. The Fjeld equation was the most accurate for the group, however, individual results varied by up
to 18%. A population specific equation was developed and may enhance the accuracy of estimates.
Australian New Zealand Clinical Trials Registry (ANZCTR) number: ACTRN12611000616976.
Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

1. Introduction complex.5 In the clinical setting, weight for height indices,

Poor nutrition and growth abnormalities are frequently re-
ported in children with cerebral palsy (CP)1,2 and have been linked
with increased health care utilisation and reduced societal partic-
ipation.3 Despite its importance, the assessment of nutritional
status in children with CP is not straightforward. Standard
anthropometric measures of height and weight may be difficult to
obtain in this population4 and interpretation of these measures is
q This work was presented (in part) at the Australasian Academy of Cerebral Palsy
and Developmental Medicine, Brisbane 2012, and the American Academy of Cere-
bral Palsy and Developmental Medicine, Toronto 2012.
* Corresponding author. Queensland Cerebral Palsy and Rehabilitation Research
Centre, 7th Floor, Block 6, Royal Brisbane and Women’s Hospital, Herston Rd,
Herston, Queensland 4029, Australia. Tel.: þ61 7 3636 5537; fax: þ61 7 3636 5480.
E-mail address: k.bell@uq.edu.au (K.L. Bell).
including the body mass index (BMI) are often used to estimate
body size and as an indicator of body fatness in different pop-
ulations.6 Due to known alterations in growth and body composi-
tion in children with CP, assumptions regarding the relationship
between weight for height indices and body fatness in typically
developing children are not appropriate for use in this population.5
A thorough clinical assessment of nutritional status in children with
CP requires an accurate and reliable measure of body composition.7
The measurement of body composition in children with CP in
the clinical setting is fraught with difficulty and there is little
consensus in the literature regarding the best method to utilise.5,8,9
Bioelectrical impedance analysis (BIA) is quick, non-invasive and
lends itself to routine clinical use.10 It is being used with increasing
frequency in studies of children with CP.9e15 BIA is based on elec-
trical theory where the volume of a cylindrical conductor is
proportional to its length2/impedance.16 In humans, this volume is

0261-5614/$ e see front matter Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
http://dx.doi.org/10.1016/j.clnu.2012.10.005
580 K.L. Bell et al. / Clinical Nutrition 32 (2013) 579e584
equivalent to total body water (TBW). A number of equations to
estimate TBW from measures of height (or length) and impedance
have been developed in different populations.9,17e21 Criterion
validation studies in children with CP remain limited and have been
small in size with limited scope.9e11,13 BIA has been validated for
use in five to 12 year old children with mild motor impairment
(Gross Motor Function Classification System (GMFCS) I and II),10
and two to 21 year old children and adolescents with more
severe impairment (GMFCS IIIeV).9,11,13 Results indicate the ability
of BIA to estimate TBW in individuals with CP is dependent on the
equation used.10 Equations to estimate TBW in young children with
CP across the full spectrum of motor impairment have not been
investigated. Our previous work highlighted difficulties with the
use of the four electrode technique in children with unilateral
impairment where BIA consistently overestimated TBW by an
average of 1.1 kg (limits of agreement
0.3 to 2.5 kg) when
impedance was measured on the unimpaired side.10 Optimal
electrode placement in children with unilateral impairment
requires further investigation. The reliability of impedance
measures in children with CP has not been reported.
This study aimed to determine (a) the criterion validity of three
different equations to estimate TBW from impedance in preschool
children with CP across the full range of motor severity; (b) optimal
electrode placement in preschool children with unilateral CP and
(c) the reliability of impedance measurements in young children
with CP.
2. Methods
2.1. Participants
Children born between 1st September 2006 and 31st December
2009 with CP across all GMFCS Levels were recruited for the
study.22 Participants were excluded if they had a progressive or
neurodegenerative lesion or a chromosomal abnormality known to
affect growth and not historically considered CP.23
This study was conducted as part of a larger longitudinal inves-
tigation into the growth and health outcomes of preschool aged
children with CP (NHMRC569605) (Australian New Zealand Clinical
Trials Registry (ANZCTR) number: ACTRN12611000616976).22
Recruitment for this study commenced in April 2009. Suitable
candidates were recruited through the Queensland Children’s Health
Services District, the Queensland CP Health Service, the Queensland
CP Register, the Queensland CP League, and other regional hospitals
and health service districts throughout Queensland.
Written, informed consent was obtained from the parents or
legal guardians. Ethical approvals were received from The Univer-
sity of Queensland Medical Research Ethics Committee
(2008002260), the Children’s Health Services District Ethics
Committee (HREC08/QRCH/112/AM01), the CP League of Queens-
land (CPLQ 2008/2010 1029), Gold Coast Health Service District
Human Research Ethics Committee (HREC/09/QGC/88), the
Townsville Health Service District Human Research Ethics
Committee (HREC/09/QTHS/96), Central Queensland Human
Research Ethics Committee (SSA/10/QCQ/13), and the Mater Health
Services Human Research Ethics Committee (1520EC). Children and
parents/guardians attended their closest centre or outreach hub
location for appointments. The same study team visited the nine
geographic locations to collect data.
2.2. Methods
Functional severity was determined for each child using the
internationally accepted GMFCS.24 Type of CP (eg, spastic, dystonic,
or hypotonic) and motor distribution (unilateral, bilateral, number
of limbs involved) was assessed according to the criteria of Sanger25
and the internationally accepted classification system on the
European CP Register.26 All motor assessments were conducted by
two trained, independent physiotherapists. When initial consensus
on classification was not reached, agreement was reached through
case discussion.
Weight was measured to the nearest 100 g using chair scales
(Seca Ltd, Germany) or portable electronic scales (Homemaker Ltd,
Australia). Height or length (for children under two years of age or
children unable to stand) was measured to the last completed
millimetre with a portable stadiometer/length measuring board
(Shorr Productions, LLC, Maryland, USA). Knee height was
measured with an anthropometer and the left hand side of the body
or the least impaired side for children with unilateral impairment
(Holtain Ltd, Dyfed, UK). For children where a measure of height or
length was not possible, height was estimated from knee height
using population specific published validated equations.4
Impedance (Ohm) was measured using a Body Stat 1500MDD
(Isle of Mann, UK) at 800 mA and a fixed frequency of 50 kHz.
Children were required to lie in a supine position with arms and legs
slightly abducted from the trunk. All jewellery and metallic objects
were removed prior to measurement. The electrical current was
applied through two non-polarising surface electrodes placed at the
dorsal surfaces of the hand and foot over the distal aspect of the
second and third metacarpals and metatarsals. The voltage drop
was measured by two further electrodes placed at the right pisiform
prominence of the wrist and between the lateral and medial mal-
leoli of the ankle. The proximal and distal electrodes were placed
a minimum of 5 cm apart. All measurements were conducted on
both sides of the body in duplicate, with a third measurement taken
if the difference between the first two measurements was greater
than 5 U. The mean of the two closest values was used for analysis.
TBW was measured non-invasively, using the deuterium-
dilution technique.27 Children were given a dose of deuterium in
the form of water either orally or via feeding tube. Any spillage was
collected in an absorbent cloth which was weighed before and after
dosing to accurately determine how much fluid was lost.28 A single
baseline urine sample was collected prior to administration of the
dose to determine natural baseline enrichments of the isotopes and
a second urine sample was collected at approximately five hours
after dosing to enable calculation of the body water pool using
standard equations.27 Collection of urine samples from children
with poor or no bladder control involved the use of urine bags or
the inclusion of an absorbent liner in their diaper from which urine
was extracted for analysis.28
2.3. Data analysis
Age corrected for gestational age was used for those children
under two years of chronological age who were born prior to 37
weeks gestation. Chronological age was used for all other children.
TBW was estimated from impedance using three previously
published equations.17,18,20 These equations were selected as they
were all developed in groups containing children of preschool age
and are all based on electrical theory.
Kushner equation18: TBW ¼ 0.04 þ 0.593 (height (m)2/
Impedance) þ 0.065 weight (kg)
Pencharz equation17: TBW ¼ 2.99 þ 0.649 (height (m)2/
Impedance)
Fjeld equation20: TBW ¼ 0.76 þ 0.180 (height (m)2/
Impedance) þ 0.390 weight (kg)
For children with unilateral impairment, differences between
TBW estimated from impedance on the impaired and unimpaired
 K.L. Bell et al. / Clinical Nutrition 32 (2013) 579e584 581

sides of the body was determined using a paired t-test. Agree- Table 2
ment between TBW estimated from impedance using each of Bland Altman analysis of three equations to estimate total body water (TBW) from
impendence using isotope dilution as the reference method in preschool aged
these equations and TBW measured using the deuterium dilution children with bilateral impairment (measurements conducted on the left hand side
technique was determined using the method described by Bland of the body only) (n ¼ 38).
and Altman.29 Agreement between repeated impedance
Method Estimated Mean of estimated Bias (L) Limits of
measurements for each electrode configuration was determined TBW  SD (L) and measured agreement (L)
using Bland and Altman29 analysis, technical errors and coeffi- TBW  SD (L)
cient of variation. The relationship between height2/impedance Deuterium 7.6  1.4
and TBW measured using deuterium dilution was examined using Kushner18 7.6  1.4 7.6  1.5 0.08
1.9 to 2.0
regression analysis. Pencharz17 10.4  1.5 9.0  1.4 2.7 0.6 to 4.8
Fjeld20 7.6  1.1 7.6  1.2 0.0
1.4 to 1.5

3. Results Correlations between the mean of the measurements and the difference were not
statistically significant for any of the equations.

Data for both TBW using deuterium dilution and impedance

measures were available for 55 children (38 boys) aged 2.40  0.59
years from 65 potentials. Descriptive data is included in Table 1.
Gross motor function distribution was as follows: I ¼ 28, II ¼ 4,
III ¼ 9, IV ¼ 5, and V ¼ 9. Predominant motor impairments were:
bilateral spasticity (n ¼ 30), unilateral spasticity (n ¼ 17), dystonia
(n ¼ 1), hypotonia (n ¼ 5) and athetosis (n ¼ 2). Stature was
measured in 22 children, length in 25 and height was estimated
from knee height for 8 children. Stature or length was measured
directly for all children with unilateral impairment. Children with
unilateral impairment were heavier (t ¼
3.01, p ¼ 0.004), taller
(t ¼
2.60, p ¼ 0.01) and had greater TBW (t ¼
2.92, p ¼ 0.005)
than those with bilateral impairment.
3.1. Validity of TBW estimates from impedance in bilateral
impairment
The mean difference and limits of agreement for TBW estimated
using each equation in comparison to TBW measured using the
isotope dilution technique for children with bilateral impairment
are shown in Table 2. Mean estimated TBW, differences between
estimated and measured TBW and the 95% limits of agreement
were greatest for the Pencharz17 equations and lowest for the
Fjeld20 equation. The Bland and Altman plot of the difference
between measured and predicted TBW and the mean of the
measurements for the Fjeld20 equation is shown in Fig. 1. The
relationship between TBW and height2/Impedance for children
with bilateral impairment is shown in Fig. 2.
3.2. Validity of TBW estimates from impedance and electrode
placement in unilateral impairment
For children with unilateral impairment, results for TBW esti-
mated from impedance measurements on the impaired versus the
unimpaired sides of the body are shown in Table 3. Estimated TBW
was lowest on the impaired compared to the unimpaired sides of
the body for all three equations. Mean differences between the
impaired and unimpaired sides were: for the Kushner18 equation:
0.3 L (3%) (t ¼ 2.55, p ¼ 0.02); for the Pencharz17 equation 0.5 L (4%)
(t ¼ 4.53, p ¼ 0.00); and for the Fjeld20 equation: 0.2 L (2%) (t ¼ 4.52,
p ¼ 0.00). There was no significant correlation between the bias and
TBW for any of the equations as the difference was consistent across
the range of measurements.
3.3. Reliability of impedance measurements
Duplicate measurements were available for 50 children (91%) on
the left hand side of the body and 49 children (89%) on the right
hand side of the body. Duplicate measures were not available for all
children for all configurations due to difficulty for the children to lie
still for sufficient time. Mean differences (SD) between first and
second measurements are shown in Table 4. A third measurement
was required on 22 (44%) occasions for the left hand side of the
body and 17 (35%) occasions for the right had side of the body due
to differences between the first and second measurements being
greater than five Ohms. Five Ohms was selected as the difference
that was acceptable based on previous reports of variations of
repeated measurements.30 Technical errors and coefficient of
variations for repeated measures were small and are shown in
Table 4.
4. Discussion
This study aimed to examine the validity and reliability of
bioelectrical impedance analysis (BIA) for the estimation of total
body water (TBW) in young children with CP. The results demon-
strate that the ability of BIA to assess TBW accurately depends on
the equation chosen. The Fjeld equation proved highly accurate at
the population level for both children with bilateral and unilateral
impairment. It estimated TBW, on average, with no mean bias for
the children with bilateral impairment; however, estimates for
individuals varied by up to 20% from the actual measured value. For
the children with unilateral impairment, estimates using the Fjeld
equation were most accurate when conducted on the unimpaired
side of the body with an average bias of 0.1 L (1%) compared to the
average bias on the impaired side of 0.2 L (2%). Individual results
varied by up to 17% of the actual measured value. In both groups,

Table 1
Descriptive data for preschool aged children with cerebral palsy by Gross Motor Function Classification System (unilateral versus bilateral impairment).

Unilateral Bilateral
GMFCS I I II III IV V Total
(n ¼ 17) (n ¼ 11) (n ¼ 4) (n ¼ 9) (n ¼ 5) (n ¼ 9) (n ¼ 37)
Age (yrs) 2.65  0.60 2.38  0.63 2.32  0.59 2.06  0.57 2.38  0.60 2.50  0.52 2.34  0.60
Height (cm) 91.9  5.2 88.7  7.3 87.7  4.1 83.9  5.3 86.2  9.5 87.9  6.3 87.0  6.7a
Weight (kg) 14.2  2.0 12.8  2.0 12.9  1.7 11.6  1.2 12.5  2.8 12.0  2.7 12.3  2.1a
TBW (L) 8.7  1.0 8.5  1.4 8.3  1.0 7.3  0.9 7.8  2.3 6.9  1.3 7.6  1.4a

GMFCS: Gross Motor Function Classification System.

TBW: total body water.
a
Significantly different to the children with unilateral impairment.
582 K.L. Bell et al. / Clinical Nutrition 32 (2013) 579e584

2.5 Table 3
Difference between TBW estimated and TBW

Bland Altman analysis of three equations to estimate total body water (TBW) from
2 impedance using isotope dilution as the reference method in preschool aged chil-
dren with unilateral impairment (n ¼ 17).
1.5 + 2 SD
Method Estimated Mean of estimated Bias Limits of
1 TBW  SD and measured (L) agreement (L)
measured (L)

(L) TBW  SD (L)

0.5
Deuterium 8.7  1.0
0 bias Kushner18 (unimpaired) 9.1  1.0 9.0  0.8 0.2
2.7 to 2.2
Kushner18 (impaired) 8.8  0.9a 8.8  0.8 0.1
2.2 to 2.1
-0.5 17
Pencharz (unimpaired) 11.9  0.9 10.4  0.8 3.1 1.3 to 4.9
-1 Pencharz17 (impaired) 11.4  0.9a 10.1  0.9 2.6 1.1 to 4.1
Fjeld20 (unimpaired) 8.8  1.0 8.8  0.9 0.1
1.5 to 1.6
-1.5 Fjeld20 (impaired) 8.6  0.9a 8.7  0.9 0.2
1.1 to 1.5
- 2 SD
-2 Correlations between the mean of the measurements and the difference were not
statistically significant for any of the equations.
-2.5 a
Significantly different to unimpaired side (p < 0.5) by paired t test.
0 2 4 6 8 10 12
Mean of TBW estimated and measured (L)
evaluated in the current study as tibial length was not measured in
Fig. 1. Bland Altman plot for the difference between total body water (TBW) (L) esti- our subjects.
mated using the Fjeld equation and TBW (L) measured using deuterium dilution in The Pencharz equation estimated TBW with the greatest bias in
children with bilateral (,) and unilateral (-) impairment.
our study population and this is consistent with our previous
work.10 This equation overestimated TBW by an average of 35% with
the equations had excellent predictive value at the population level, a range in individuals of up to 63%. Use of this equation in the clinical
however, must be used with caution in individuals. setting will result in an enormous and unacceptable overestimation
The estimation equations included in this study were selected of TBW and hence fat free mass and underestimation of body fat. The
for two reasons: firstly, all three were derived from groups of Kushner equation performed well in our current study, estimating
children including children of preschool age and secondly, all are TBW to within 1% of the measured value on average, with a range in
based on the fundamental assumption of electrical theory. Kushner individuals of up to 26%. Interestingly, the Kushner equation esti-
and colleagues developed their equation from a group of 116 mated TBW with the least bias in our previous work investigating
subjects including 29 preschool aged children and cross validated it the accuracy of BIA for use in children aged five to 12 years with mild
in a smaller group of 59 subjects of which 15 were preschool aged.18 motor impairment.10 In this study, it estimated TBW to within 2% on
The Fjeld equation was developed from a group of 30 children aged average, with a range in individuals of up to 20%.10 These two studies
three to 30 months including both well nourished and malnour- suggest that the Kushner equation has the potential to provide
ished subjects.20 It was cross validated in a second group of 14 accurate estimates of TBW in different populations of children with
similarly aged children. The Pencharz equation was developed from CP, across broad age ranges and GMFCS levels. The convenience of
data obtained from a group of 101 malnourished children and a single equation with such broad application is obvious. The use of
adolescents before and after refeeding.17 This equation has since this equation in older children with severe gross motor impairment
been validated for use in severely impaired children with spastic and adolescents across the full spectrum of motor severity requires
quadriplegia.11 More recently, Reiken et al9 developed an estima- further investigation.
tion equation including tibial length as an alternative for direct Consistent overestimation in TBW from measures of impedance
measurement of height/length from a group of 61 children with conducted on the unimpaired sides of the body in children with
severe CP aged two to 18 years. This equation was not able to be unilateral impairment identified in our previous work10 were not
confirmed in the current study. Differences between estimates
12 from the impaired and unimpaired sides of the body, whilst
statistically significant, were very small (2% for the Fjeld equation)
and would have little impact on the clinical assessment of nutri-
10 tional status. Therefore, results obtained from measurements
conducted on either side of the body can be used to determine TBW
in young children with unilateral impairment. Potentially, differ-
Total Body Water (L)

8 ences in body composition between the two sides of the body

become more apparent as children with CP reach school age.
Optimal electrode placement requires further investigation in older
6
children.
Reliability of impedance measurements at the population level
4 was good with minimal difference in duplicate measurements
(0.5% variation on the left side of the body and 0.1% variation on the
right side). However, there was significant variation within
2 a number of individuals and a third measurement was required in
44% of children for the left side of the body due to the duplicate
measures being greater than five ohms apart. For this reason,
0 duplicate measures should be performed in routine practice to
0 5 10 15 20
ensure accuracy.
Height (cm)2/ Impedance (Ohms)
Separate regression equations relating measured TBW to height2/
Fig. 2. Height (cm)2/Impedance (Ohms) versus Total Body Water (TBW) (L) for children Impedance, derived from our data set, for children with bilateral
with bilateral (,) and unilateral (-) impairment. impairment and children with unilateral impairment are shown in
 K.L. Bell et al. / Clinical Nutrition 32 (2013) 579e584
Table 4
Agreement between duplicate measurements of impedance for preschool aged children with cerebral palsy (Ohms).
Side of body Mean impedance Mean difference
(Ohms)  SD (Ohms)
Left (n ¼ 50) 667  98
3
Right (n ¼ 49) 668  97
1
Unimpaired (n ¼ 16) 623  86 8
15 to 16
Impaired (n ¼ 16) 662  91 5
9 to 12
Fig. 2. The two equations were remarkably similar and no significant
differences were found between either the slopes or intercepts using
regression analysis.31 A single equation using data from both groups
of children was subsequently generated and, as expected, was not
significantly different to the separate equations. The correlation was
strongest for the combined data and standard errors of the estimate
for all three equations were similar (0.8e0.9 L).
5. Conclusion
This study is the first to investigate the accuracy of BIA for the
estimation of TBW in young children with CP across the full spec-
trum of motor impairment and to investigate optimal electrode
placement in children with unilateral impairment. The results have
shown that BIA is a useful and accurate tool for the assessment of
body composition in preschool aged children with CP when an
appropriate, validated equation is utilised. The importance of using
an equation that has been validated in the population under
investigation has been highlighted. BIA provides an estimate of
body composition that is not limited to assessment of regional body
fat stores (as skinfold thickness measurements are). This has
significant implications for clinical nutrition assessment of children
with CP. The equation derived in the current study population
requires validation in another population of young children
with CP.
Sources of funding
This study was funded by the Australian National Health and
Medical Research Council Project Grant (569605). RNB is funded by
a National Health and Medical Research Council Career Develop-
ment Grant (473480), the Royal Children’s Hospital Foundation and
a Smart State Fellowship through the Queensland Department of
Innovation.
Statement of authorship
KLB, PSWD, RNB and RDS contributed to the study design,
protocol, and grant writing. JLW analysed samples, and contributed
to publication preparation. KLB wrote the manuscript and con-
ducted statistical analysis with input and advice from all authors.
All read and approved the final manuscript.
Conflict of interest
No conflicts of interest are declared by the authors.
Acknowledgements
We wish to thank our research assistants and students for their
contributions to this project: Joanne McMah, Stina Oftedal,
Katherine Benfer, Christine Finn, Rachel Jordan, Paula Luck, Lauren
Forbes, Fiona Caristo, Kelly Weir.
583
Limits of agreement Mean difference Technical Coefficient of
(Ohms) as a percentage error (Ohms) variation (%)
of mean (%)

25 to 19 0.4 7.9 1.2

21 to 19 0.2 7.4 1.1
1.2 5.2 0.8
0.8 3.6 0.6
References
1. Stevenson RD, Hayes RP, Cater LV, Blackman JA. Clinical correlates of linear
growth in children with cerebral palsy. Developmental Medicine and Child
Neurology 1994;36:135e42.
2. Brooks J, Day S, Shavelle R, Strauss D. Low weight, morbidity, and mortality in
children with cerebral palsy: new clinical growth charts. Pediatrics 2011;128:
e299e307.
3. Stevenson RD, Conaway M, Chumlea WC, Rosenbaum P, Fung E, Henderson CJ,
Worley G, Liptak GS, O’Donnell M, Samson Fang L, Stallings VA. Growth and
health in children with moderate to severe cerebral palsy. Pediatrics 2006;118:
1010e8.
4. Stevenson RD. Use of segmental measures to estimate stature in children with
cerebral palsy. Archives of Pediatrics and Adolescent Medicine 1995;149:658e62.
5. Kuperminc MN, Gurka MJ, Bennis JA, Busby MG, Grossberg RI, Henderson RC,
Stevenson RD. Anthropometric measures: poor predictors of body fat in chil-
dren with moderate to severe cerebral palsy. Developmental Medicine & Child
Neurology 2010;52:824e30.
6. Mei Z, Grummer-Strawn LM, Pietrobelli A, Goulding A, Goran MI, Dietz WH.
Validity of body mass index compared with other body-composition screening
indexes for the assessment of body fatness in children and adolescents.
American Journal of Clinical Nutrition 2002;75:978e85.
7. Kuperminc M, Stevenson RD. Growth and nutrition disorders in children with
cerebral palsy. Developmental Disabilities Research Reviews 2008;14:137e46.
8. Rieken R, Calis EAC, Tibboel D, Evenhuis HM, Penning C. Validation of skinfold
measurements and bioelectrical impedance analysis in children with severe
cerebral palsy: a review. Clinical Nutrition 2010;29:217e21.
9. Rieken R, van Goudoever JB, Schierbeek H, Willemsen SP, Calis EAC, Tibboel D,
Evenhuis HM, Penning C. Measuring body composition and energy expenditure
in children with severe neurologic impairment and intellectual disability.
American Journal of Clinical Nutrition 2011;94:759e66.
10. Bell KL, Davies PSW. The use of bioelectrical impedance analysis in children
with cerebral palsy and non-disabled children. International Journal of Body
Composition Research 2004;2:15e22.
11. Azcue MP, Zello GA, Levy LD, Pencharz PB. Energy expenditure and body
composition in children with spastic quadriplegic cerebral palsy. Journal of
Pediatrics 1996;129:870e6.
12. Hildreth HG, Johnson RK, Goran MI, Contompasis SH. Body composition in adults
with cerebral palsy by dual-energy X-ray absorptiometry, bioelectrical imped-
ance analysis, and skinfold anthropometry compared with the 18o isotope-dilu-
tion technique. American Journal of Clinical Nutrition 1997;66:1436e42.
13. Liu L, Roberts R, Moyer-Mileur L, Samson-Fang L. Determination of body
composition in children with cerebral palsy: bioelectrical impedance analysis
and anthropometry vs dual energy X-ray absorptiometry. Journal of the
American Dietetic Association 2005;105:794e7.
14. Tomoum HY, Badawy NB, Hassan NE, Alian KM. Anthropometry and body
composition analysis in children with cerebral palsy. Clinical Nutrition 2010;29:
477e81.
15. Veugelers R, Penning C, van Gulik ME, Tibboel D, Evenhuis HM. Feasibility of
bioelectrical impedance analysis in children with a severe generalized cerebral
palsy. Nutrition 2006;22:16e22.
16. Goran MI, Karskoun MC, Carpenter WH, Poehlman ET, Ravussin E, Fontvieille A.
Estimating body composition of young children using bioelectrical resistance.
Journal of Applied Physiology 1993;75:1776e80.
17. Pencharz PB, Azcue M. Use of bioelectrical impedance analysis measurements
in the clinical management of malnutrition. American Journal of Clinical
Nutrition 1996;64:485Se8S.
18. Kushner RF, Schoeller DA, Fjeld CR. Is the Impedance Index (Ht2/R) significant
in predicting total body water? American Journal of Clinical Nutrition 1992;56:
835e9.
19. Kushner RF, Schoeller DA. Estimation of total body water by bioelectrical
impedance analysis. The American Journal of Clinical Nutrition 1986;44:417e24.
20. Fjeld C, Freundt-Thurne J, Schoeller D. Total body water measurement by 18O
dilution and bioelectrical impedance in well and malnourished children.
Pediatr Res 1990;27:98e102.
21. Davies PSW, Jagger SE, Reilly JJ. A relationship between bioelectrical impedance
and total body water in young adults. Annals of Human Biology 1990;17:445e8.
22. Bell KL, Boyd RN, Tweedy SM, Weir KA, Stevenson RD, Davies PSW. A
prospective, longitudinal study of growth, nutrition and sedentary behaviou in
young children with cerebral palsy. BMC Public Health 2010;10:179.
584 K.L. Bell et al. / Clinical Nutrition 32 (2013) 579e584
23. Badawi N, Watson L, Petterson B, Blair E, Slee J, Haan E, Stanley F. What
constitutes cerebral palsy? Developmental Medicine and Child Neurology
1998;40:520e7.
24. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development
and reliability of a system to classify gross motor function in children with
cerebral palsy. Developmental Medicine and Child Neurology 1997;39:214e23.
25. Sanger T, Delgado M, Gaebler Spira D, Hallett M, Mink J. Classification and
definition of disorders causing hypertonia in childhood. Pediatrics 2003;111:
e89e97.
26. Surveillance of Cerebral Palsy in Europe. Surveillance of cerebral palsy in
Europe: a collaboration of cerebral palsy surveys and registers. Developmental
Medicine and Child Neurology 2000;42:816e24.
27. Halliday D, Miller AG. Precise measurement of total body water using trace
quantities of deuterium oxide. Biomedical Mass Spectrometry 1977;4:82e7.
28. Atkin LM, Davies PSW. Diet composition and body composition in preschool
children. American Journal of Clinical Nutrition 2000;72:15e21.
29. Bland JM, Altman DG. Statistical methods for assessing agreement between
two methods of clinical measurement. Lancet 1986;1:307e10.
30. Lukaski HC, Johnson PE, Bolonchuk WW, Lykken GI. Assessment of fat-free
mass using bioelectrical impedance measurements of the human body.
American Journal of Clinical Nutrition 1985;41:810e7.
31. Davies PSW, Cole TJ. The adjustment of measures of energy expenditure for
body weight and body composition. International Journal of Body Composition
Research 2003;1:45e50.