Full text of DOI:10.1080/07315724.2002.10719230

This article was downloaded by: [University of Newcastle, Australia]
On: 30 December 2014, At: 16:24
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of the American College of Nutrition
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uacn20
Phantoms for Cross-Calibration of Dual Energy X-ray
Absorptiometry Measurements in Infants
Mouhanad Hammami MDa, Jean-Charles Picaud MD, PhDc, Christoph Fusch MDd, Elaine M.
Hockman PhDb, Jacques Rigo MDe & Winston W.K. Koo MBBS, FACNa
a Departments of Pediatrics, Obstetrics and Gynecology (M.H., W.W.K.K.), Wayne State
University, Detroit, Michigan
b Computing and Information Technology (E.M.H.), Wayne State University, Detroit,
Michigan
c Neonatal Unit and Human Nutrition Research Center, University of Lyon, Lyon, FRANCE
(J.-C.P.)
d Neonatal Unit, University Children’s Hospital, Greifswald, GERMANY (C.F.)
e Neonatal Unit, University of Liege, Liege, BELGIUM (J.R.)
Published online: 26 Jun 2013.
To cite this article: Mouhanad Hammami MD, Jean-Charles Picaud MD, PhD, Christoph Fusch MD, Elaine M. Hockman
PhD, Jacques Rigo MD & Winston W.K. Koo MBBS, FACN (2002) Phantoms for Cross-Calibration of Dual Energy X-
ray Absorptiometry Measurements in Infants, Journal of the American College of Nutrition, 21:4, 328-332, DOI:
10.1080/07315724.2002.10719230
To link to this article: http://dx.doi.org/10.1080/07315724.2002.10719230
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Original Research
Phantoms for Cross-Calibration of Dual Energy X-ray
Absorptiometry Measurements in Infants
Mouhanad Hammami, MD, Jean-Charles Picaud, MD, PhD, Christoph Fusch, MD, Elaine M. Hockman, PhD,
Jacques Rigo, MD, and Winston W.K. Koo, MBBS, FACN
Departments of Pediatrics, Obstetrics and Gynecology (M.H., W.W.K.K.), Computing and Information Technology (E.M.H.),
Wayne State University, Detroit, Michigan, Neonatal Unit and Human Nutrition Research Center, University of Lyon, Lyon,
FRANCE (J.-C.P.), Neonatal Unit, University Children’s Hospital, Greifswald, GERMANY (C.F.), Neonatal Unit, University of
Liege, Liege, BELGIUM (J.R.)
Key words: phantom, body composition, fat, lean, bone, infant
Objective: To test the suitability of phantoms to cross-calibrate body composition measurements in small
subjects among different dual energy X-ray absorptiometry (DXA) instruments.
Methods: A set of four phantoms with total weights 1520g, 3140g, 4650g and 7490g were made with low
cost and easily available materials. Each phantom was made from assembling polyethylene bottles (100 to 1000
mL) filled with either pure olive oil or electrolyte solution in different combinations, and borosilicate tubes (3
and 5 mL) and flexible polypropylene tubing filled with calcium carbonate. Triplicate measurements of each of
the four phantoms were performed with three pencil beam densitometers made by the same manufacturer
(Hologic Inc., Waltham, MA): two QDR 2000 (University of Liege, Liege, Belgium, and Wayne State
University, Detroit, Michigan) and a QDR1500 (University Children’s Hospital, Greifswald, Germany) using
infant whole body-scanning mode and analyzed with software V5.73P.
Results: DXA measured total weight, or bone, lean and fat masses, from one center were highly predictive
of DXA measurements from the other centers with an adjusted r2 of 0.94 to 1.00, p Ͻ 0.001. This was the case
whether the measurements from single scan or from average of triplicate scans were used in the analysis.
Conclusions: Systematic corrections, in the form of linear transformations, are possible to allow comparison
of clinical data generated from different centers. Different size phantoms can be made to accommodate the
varying range of weights and body composition of study subjects.
studies in infants, it is imperative to determine whether differ-
ent instruments used to measure body composition in infants
INTRODUCTION
can be cross-calibrated to allow meaningful comparison of data
generated from different institutions, although no such study
has been reported. We therefore aim to test the suitability of a
set of phantoms to cross-calibrate body composition measure-
ments in small subjects among different DXA instruments.
Dual energy X-ray absorptiometry (DXA) is the recognized
standard for bone measurements and is increasingly being used
for the measurement of soft tissue body composition in adults
[1] and children [2] including infants [3]. However, it is well
known that DXA measurements varied with instruments from
different manufacturers [4–6] and even within the same man-
ufacturer [7] because of differences in scanner design, materials
used for calibration and analysis algorithms. Thus, in order to
determine the comparability of results generated from different
centers, it is essential to test all the properties of scanner
performance as a whole with the use of phantoms and/or human
subjects. With increasing availability of DXA technique for
METHODS
Phantoms
Pure olive oil (Salov North America Corp, Hackensack,
NJ), an electrolyte solution containing a mixture of sodium
Address reprint requests to: Dr. Winston Koo, Department of Pediatrics, Hutzel Hospital, 4707 St Antoine Blvd, Detroit, MI 48201. E-mail: wkoo@wayne.edu
Journal of the American College of Nutrition, Vol. 21, No. 4, 328–332 (2002)
Published by the American College of Nutrition
328

Phantoms for Dual Energy X-ray Absorptiometry
chloride and potassium mono-basic phosphate (Sigma Aldrich
Inc., St. Louis, MO) and calcium carbonate powder (Sigma
Aldrich Inc, St. Louis, MO) were used to mimic fat, lean and
bone. Polyethylene bottles (Nalge Nunc International, Roches-
ter, NY) of different shapes and capacities (100 to 1000 mL)
were filled with either pure olive oil or electrolyte solution.
Different size borosilicate tubes (3 and 5 mL, Becton Dickinson
Vacutainer Systems, Rutherford, NJ) and flexible polypro-
pylene tubing (Nalge Nunc International, Rochester, NY) were
filled with calcium carbonate. Bottles and tubes were taped
together in layers to form nine blocks. Each block contained
different quantities of oil, electrolyte solution and calcium
carbonate. The blocks were assembled contiguously with one
another in a predetermined fashion to form four phantoms with
total weights 1520g, 3140g, 4650g and 7490g as determined by
an electronic scale (Seca model 727, Toledo Scale Corp.,
Toledo, OH). The maximum dimensions of the four phantoms
varied between 28 to 58 cm in length, 12 to 32 cm in width and
11 to 15 cm thick. All blocks were kept in a box at room
temperature between DXA measurements.
and total weight. Repeated measures analysis of variance was
used to determine the equivalence of the triplicate DXA mea-
surements (within subject factor) among the four phantoms
(between subject factor) and whether there was interaction
between DXA measurements from different size phantoms
from different instruments.
Regression analyses were performed to determine the abil-
ity of DXA measurements from UL and UCH to predict the
DXA measurements of the same phantoms at WSU. Univariate
analysis of variance with Helmert contrasts was used to analyze
comparability of residuals from each prediction equation based
on UL and UCH data respectively. The same procedures were
repeated to determine the regression equation for prediction of
DXA measurements of the same phantoms at the other centers
from WSU DXA measurements.
The same procedures were repeated using the first of the
triplicate measurements to mimic the clinical situation of gen-
erating one satisfactory scan per subject. This was done to
determine whether the same relationships exist with the use of
data from one or three DXA scan. All statistical tests were
performed with SPSS 10.0 (SPSS Inc., Chicago, IL) for win-
dows at an adopted significance level of 0.05.
DXA Scans
Three densitometers from three centers were assessed in this
study. All densitometers were from the same manufacturer
(Hologic Inc., Waltham, MA): two QDR 2000 (one located at
University of Liege (UL), Liege, Belgium, and the other lo-
cated at Wayne State University (WSU), Detroit, MI, USA) and
a QDR1500 located at the Neonatal unit of University Chil-
dren’s Hospital (UCH), Greifswald, Germany. Quality control
scans for each densitometer were performed daily using a
manufacturer-supplied anthropomorphic spine phantom. The in
vitro coefficients of variation (CV) for Ͼ1 year for the deter-
mination of bone mineral content, bone area and bone mineral
density were 0.43%, 0.42% and 0.46%, respectively at UL,
were 0.38%, 0.30% and 0.34% at WSU, and were 0.35%,
0.35% and 0.31%, respectively at UCH.
All densitometers were operated in the pencil beam mode,
the only technique freely available for body composition stud-
ies in infants. The four phantoms were scanned in triplicate on
each densitometer using infant whole body-scanning mode and
analyzed with manufacturer-supplied software V5.73P. One
investigator (J.-C.P.) familiar with the agreed layout of the
phantoms was present at each site to insure the correct assem-
bly and placement of the phantoms for DXA measurements.
Each center used its own operator for scan acquisition and
analysis. Phantoms were transported personally or shipped
between centers via commercial courier.
RESULTS
A representative phantom and its corresponding scan are
shown in Fig. 1. DXA measurements of the four phantoms
from the study sites are shown in Table 1. DXA measurements
were highly correlated (adjusted r2 ϭ 0.96 to 1.00) with weight
of the components and total weight of each phantom. There was
no significant difference among triplicate DXA measurements
of the phantoms. Therefore averages across the three measure-
ments were used in further analyses. There was no interaction
among DXA measurements from different size phantoms using
the three instruments.
DXA measurements from UL and UCH were highly pre-
dictive (adjusted r2 ϭ 0.94 to 1.00, p Ͻ 0.001) of DXA
measurements of the same phantoms at WSU (Table 2).
Statistical Analysis
DXA measured total weight, lean mass, fat mass, bone
mineral content, bone area and bone mineral density were used
in data analysis. Percent of fat was presented as descriptive data
and not analyzed further since it was calculated from fat mass
Fig. 1. A 5 kg phantom assembled from various blocks (left) and the
resultant dual energy X-ray absorptiometry scan (right).
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
329

Phantoms for Dual Energy X-ray Absorptiometry
Table 1. Triplicate Measurements* of Phantoms Using Dual Energy X-ray Absorptiometry
UL
SD
UCH
SD
WSU
SD
Mean
CV%
Mean
CV%
Mean
CV%
Phantom 1.5 kg
Total g
Lean Mass g
Fat Mass g
BMC g
1507
982
490
34.4
170
0.202
1.2
6.9
6.4
0.15
2.2
0.08
0.70
1.30
0.42
1.32
0.76
1499
1034
432
32.9
165
2.3
0.15
2.07
5.15
4.23
4.23
0.58
1506
1037
435
34.4
163
1.6
0.11
1.46
3.71
5.20
5.05
1.25
21.4
22.3
1.39
7.0
15.1
16.1
1.79
8.2
Bone Area cm2
BMD g/cm2
Phantom 3 kg
Total g
0.002
0.199
0.001
0.211
0.003
3156
1965
1116
75.1
285
5.2
0.16
1.30
2.14
1.44
1.60
1.58
3147
2005
1074
68.9
271
4.9
0.15
3.35
6.58
1.80
1.00
1.64
3157
2049
1037
72.1
277
3.1
0.10
4.37
8.97
4.67
2.13
3.63
Lean Mass g
Fat Mass g
BMC g
25.5
23.9
1.08
4.6
67.1
70.7
1.24
2.7
89.4
93.0
3.37
5.9
Bone Area cm2
BMD g/cm2
Phantom 5 kg
Total g
0.264
0.004
0.254
0.004
0.260
0.009
4664
3178
1373
113.2
364
9.1
0.19
1.93
5.02
1.05
1.21
0.32
4677
3174
1395
107.4
341
3.6
0.08
3.17
7.25
0.87
1.49
0.80
4670
3349
1218
102.9
350
7.7
0.16
2.12
6.68
7.22
1.09
6.59
Lean Mass g
Fat Mass g
BMC g
61.2
68.9
1.19
4.4
100.5
101.1
0.94
5.1
71.0
81.4
7.43
3.8
Bone Area cm2
BMD g/cm2
Phantom 7 kg
Total g
0.311
0.001
0.316
0.003
0.294
0.019
7513
5364
1991
182.2
631
9.5
0.13
0.85
1.82
0.85†
0.49
0.73†
7521
5600
1755
165.7
588
9.9
0.13
1.78
5.20
0.96
1.61
1.44
7563
5695
1700
168.3
590
5.5
0.07
0.34
1.17
0.51
0.77
0.61
Lean Mass g
Fat Mass g
BMC g
45.7
36.3
1.54
3.1
99.8
91.2
1.58
9.5
19.3
19.9
0.87
4.6
Bone Area cm2
BMD g/cm2
0.290
0.002
0.282
0.004
0.285
0.002
Abbreviation: UL ϭ University of Liege, Belgium, UCH ϭ University Children Hospital, Germany, and WSU ϭ Wayne State University, USA, BMC ϭ bone mineral
content, BMD ϭ bone mineral density.
* Average of triplicate measurements.
† Excluded one BMC and one BMD value at ϳ50% lower than the value generated from the other two scans.
Helmert contrasts in analysis of variance confirmed that the
residuals from UL and UCH taken together did not differ from
WSU (as zero), nor did UL and UCH residuals differ from each
other.
generated from different centers using different instruments. It
would be useful to have in vivo studies to cross-calibrate
different instruments as has been done with human adults
[4–7]. However, the use of human infants for this purpose is
impractical under most circumstances, especially since the
three instruments employed for this study are located in widely
separated geographic regions. Thus, the use of phantoms would
be the most practical means to cross-calibrate different instru-
ments in the study of body composition in small subjects
particularly infants. However, phantoms suitable for use in
older children [8] and adults [4–7] are inappropriate for the
assessment of body composition in small subjects because of
major differences in the acquisition and analysis of scans.
Our goal for this study was to develop a set of phantoms that
can be made easily and inexpensively and have sufficient
flexibility for any investigator to further modify or adjust the
components to create different size phantoms with varied body
composition. The primary purpose for the use of these phan-
toms is to determine the interrelationship of the DXA measure-
ments among different densitometers rather than the compari-
son of absolute accuracy of the DXA densitometers. This is
Similar results were found when using the first value instead
of the average of the triplicate DXA measurements for the
prediction of DXA measurements among the three institutions
with an adjusted r2 ϭ 0.94 to 1.00, p Ͻ 0.001. These findings
were applicable to the determination of predicted mean, stan-
dard error of estimate and the slope of the prediction equations
regardless of the DXA parameter measured. The differences in
the slope of the predicted equation were Ͻ1% in all cases
whether the first or average of triplicate measurements were
used in the prediction equations.
DISCUSSION
To advance the understanding of physiologic and pathologic
effects on body composition in infants and young children, it is
imperative that a system exists to determine the validity of data
330
VOL. 21, NO. 4

Phantoms for Dual Energy X-ray Absorptiometry
Table 2. Prediction of Dual Energy X-ray Absorptiometry (DXA) Measurements at Wayne State University (WSU), USA from
DXA Measurements Obtained from Densitometers at University of Liege (UL), Belgium, and University Children Hospital
(UCH), Germany, Using Average of Triplicate Scans of the Same Phantoms
UL
UCH
Total weight (Total) g
Adjusted r2
1.008 Total* Ϫ 21.5
1.000
1.005 Total* Ϫ 9.4
1.000
Predicted mean/SEE
Lean mass (LM) g
Adjusted r2
4224/11.4
1.065 LM* Ϫ 26.8
1.000
4224/18.7
1.021 LM* ϩ 17.4
0.999
Predicted mean/SEE
Fat mass (FM) g
3032/19.5
3032/75.3
0.839 FM† ϩ 54.9
0.993
0.924 FM† ϩ 22.6
0.977
Adjusted r2
Predicted mean/SEE
Bone mineral content (BMC) g
Adjusted r2
1097/44.1
1097/78.8
0.901 BMC† ϩ 3.25
0.999
0.995 BMC† ϩ 1.15
0.994
Predicted mean/SEE
Bone area (BA) cm2
Adjusted r2
94.4/2.0
0.923 BA* ϩ 10.5
0.999
94.4/4.3
1.006 BA* ϩ 1.8
0.999
Predicted mean/SEE
Bone mineral density (BMD) g/cm2
Adjusted r2
345/4.9
345/5.7
0.789 BMD† ϩ 0.052
0.990
0.744 BMD‡ ϩ 0.067
0.941
Predicted mean/SEE
0.263/0.004
0.263/0.009
* p Յ 0.001; † p Յ 0.01; ‡ p ϭ 0.02.
All intercepts had p Ͼ 0.05 although the intercept for the BMD prediction equation from UL had p ϭ 0.051.
SEE ϭ Standard error of the estimate.
consistent with the means to obtain standardized DXA mea-
surements of the spine using instruments from different man-
ufacturers that are known to provide different values for the
same subject [9]. The design of our phantoms also satisfied the
recommendations of the International DXA Standardization
Committee that cross-calibration among different instruments
should not be based on the use of a single phantom [9].
We have independently reported [10–12] the validity of the
pencil beam DXA technique for the measurement of body
composition using instruments from the same manufacturer
based on animal tissue studies, and it was not our intention to
reproduce the anatomically correct or exact duplication of body
composition of infants, since there are great differences among
infants and it would be prohibitively expensive and time con-
suming to achieve these goals. In any case, the physical dimen-
sions and body composition values of our phantoms can be
modified to accommodate the wide range of weights and body
composition in clinical subjects, thus allowing cross compari-
son of any clinical studies involving small subjects.
In this study, the strongly predictive relationships of DXA
measurements among the three instruments would support that
data generated from different densitometers made by the same
manufacturer employing the same DXA pencil beam technique
and the same software are comparable. Furthermore, systematic
corrections in the form of linear transformations are possible to
allow comparison of clinical data generated from different
studies. It is also possible that our system of phantoms can be
used to determine whether these relationships remain true for
data generated from the use of other DXA techniques or the use
of instruments from different manufacturers.
That the intercepts of the regression equations for the pre-
diction of DXA measurements among various centers were not
significantly different from zero would support the absence of
systematic difference among the densitometers tested, although
the intercept for BMD prediction equation derived from UL
approached significance. Even if there was a systematic differ-
ence in BMD measurements among different densitometers, it
is still possible to compare data among different centers since
the slope of the relationship in BMD measurements remain
significantly highly correlated with an adjusted r2 of Ն 0.94.
By way of clarification, the conversion of Celsius to Fahrenheit
or vice versa would show a different intercept, but the slope
would indicate that these two measurements are highly signif-
icantly related. In any case, BMD as an index of bone mass
measurement is inappropriate in growing individuals such as
infants [13,14]. Furthermore, since DXA bone mass measure-
ment was validated for adults based on the mass of hydroxy-
apatite or other material [15] and for infants was based on the
mass of carcass ash and calcium [10–12], i.e., not based on
density, the potential clinical significance of any discrepancy in
BMD measurements would be limited.
The same predictive ability among DXA measurements for
the various components and total weight of phantoms whether
using data from first or average of multiple DXA scans has
major clinical implications. Thus the generation of a single
good quality DXA scan, specifically without movement arti-
fact, is likely to be adequate for clinical studies in infants. This
results in reduction of radiation exposure, time and cost of
clinical studies. In contrast, it is theoretically possible that a
transient variability in the output X-ray source may have led to
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
331

Phantoms for Dual Energy X-ray Absorptiometry
the one outlier in the triplicate measurement of one phantom. If
this was the case, then the use of one DXA scan may be
inadequate to determine the existence of this problem during
any clinical study. The occurrence of the outlier measurement
was unlikely the result of operator error since no repositioning
was performed between the three scans. In any case, it is
critical to maintain the instrument in optimal operating condi-
tion, remain vigilant to the assurance of uninterrupted X-ray
energy output, consistent approach in data acquisition and
analysis, avoidance of motion artifact and strict adherence to all
aspects of quality assurance including duplicate scans on a
sample of subjects in any DXA study.
Pierson RN: A multicenter comparison of dual energy x-ray ab-
sorptiometers: in vivo and in vitro measurements of bone mineral
content and density. J Bone Min Res 11:275–285, 1996.
6. Diessel E, Fuerst T, Njeh CF, Tylavsky F, Cauley J, Dockrell M,
Genant HK: Evaluation of a new body composition phantom for
quality control and cross-calibration of DXA devices. J Appl
Physiol 89:599–605, 2000.
7. Abrahamsen B, Gram J, Hansen TB, Beck-Nielsen H: Cross cal-
ibration of QDR-2000 and QDR-1000 dual energy X-ray densito-
meters for bone mineral and soft tissue measurements. Bone 16:
385–390, 1995.
8. Shypailo RJ, Posada JK, Ellis KJ: Whole-body phantoms with
anthropomorphic-shaped skeletons for evaluation of dual energy
x-ray absorptiometry measurements. Appl Radiat Isot 49:503–505,
1998.
9. Steiger P for the Committee for Standards in DXA: Standardiza-
tion of spine BMD measurements. J Bone Miner Res 10:1602–
1603, 1995.
CONCLUSION
We have demonstrated that phantoms of various sizes can
be made from easily available and low cost materials and can
be used to cross-calibrate different DXA instruments to allow
meaningful comparison of data among different centers.
10. Koo WWK, Massom LR, Walters J: Validation of accuracy and
precision of dual energy x-ray absorptiometry for infants. J Bone
Miner Res 10:1111–1115, 1995.
11. Picaud JC, Rigo J, Nyamugabo K, Milet J, Senterre J: Evaluation
of dual energy X ray absorptiometry for body composition assess-
ment in piglets and term human neonates. Am J Clin Nutr 63:157–
163, 1996.
12. Fusch C, Slotboom J, Fuehrer U, Schumacher R, Keisker A,
Zimmermann W, Moessinger A, Boesch C, Blum J: Neonatal body
composition: dual energy X ray absorptiometry, magnetic reso-
nance imaging, and three dimensional chemical shift imaging
versus chemical analysis in piglets. Pediatr Res 46:465–473, 1999.
13. Nelson DA, Koo WWK: Interpretation of absorptiometric bone
mass measurements in the growing skeleton: issues and limita-
tions. Calcif Tiss Internat 65:1–3, 1999.
REFERENCES
1. Van Loan MD: Is dual energy X-ray absorptiometry ready for
prime time in the clinical evaluation of body composition? Am J
Clin Nutr 68:1155–1156, 1998.
2. Ellis KJ, Shypailo RJ, Abrams SA, Wong WW: The reference
child and adolescent models of body composition. A contemporary
comparison. Ann N Y Acad Sci 904:374–382, 2000.
14. Rauch F, Schoenau E: Changes in bone density during childhood
and adolescence: an approach based on bone’s biological organi-
zation. J Bone Miner Res 16:597–604, 2001.
3. Koo WWK: Body composition measurements during infancy. Ann
N Y Acad Sci 904:383–392, 2000.
4. Wahner HW, Looker A, Dunn WL, Walters LC, Hauser MF,
Novak C: Quality control of bone densitometry in a National
Health Survey (NHANES III) using three mobile examination
centers. J Bone Miner Res 9:951–960, 1994.
15. Kelly TL, Slovik DM, Neer RM: Calibration and standardization
of bone mineral densitometers. J Bone Miner Res 4:663–669,
1989.
5. Economos CD, Nelson ME, Fiatarone MA, Dallal GE, Heymsfield
SB, Wang J, Russel-Aulet M, Yaumura S, Ma R, Vaswani AN,
Received November 19, 2001; revision accepted May 24, 2002.
332
VOL. 21, NO. 4