Body composition determination by bioimpedance: an update

Michel Y. Jaffrin
Department of Biological Engineering, UMR CNRS
6600, University of Technology of Compiegne,
Compiegne, France
Correspondence to Michel Y. Jaffrin, Department of
Biological Engineering, UMR CNRS 6600, University
of Technology of Compiègne, BP 20529, 60205
Compiegne, France
Tel: +33 3 44 23 43 98; fax: +33 3 44 23 79 42;
e-mail: michel.jaffrin@utc.fr
Current Opinion in Clinical Nutrition and
Metabolic Care 2009, 12:482–486
Purpose of review
To review various methods for measuring body composition by bioimpedance and their
limitations, as well as available impedance meters, including body fat analyzers for home
use.
Recent findings
Bioimpedance spectroscopy, which requires multifrequency impedance meters, is
preferable for fluid volume measurements, especially extracellular fluid, whereas
bioimpedance analysis at 50 kHz is more widely used for measuring fat-free mass. A
method for using bioimpedance spectroscopy equations with 50 kHz impedance
meters has been recently proposed and successfully tested. Low cost foot-to-foot
impedance meters (body fat analyzers) with plantar electrodes on a body scale, that are
easy and fast to use, have been compared with medical impedance meters and with dual
X-ray absorptiometry measurements and found reasonably accurate, except for
individuals with very low or high BMI.
Summary
Body composition by bioimpedance is gaining acceptance in nutrition, hemodialysis,
gerontology and sports medicine. Body fat analyzers that have been validated by
comparison with dual x-ray absorptiometry could be useful to general practitioners,
nutritionists and cardiologists.

Keywords
bioimpedance analysis, bioimpedance spectroscopy, extracellular water, fat-free mass,
segmental bioimpedance, total body water

Curr Opin Clin Nutr Metab Care 12:482–486

ß 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins
1363-1950

Total body water (Vt) and extracellular (Ve) volumes by

Introduction bioimpedance analysis methods
The measurement of body composition parameters such They express Ve and Vt as linear functions of H2/R,
as fat tissue mass, lean body mass (LBM), body cell mass where H is the height of the individual, body weight
(BCM), total body water (TBW) and extracellular water W and sometimes H2/X, where R and X are respectively
(ECW) is important in clinical nutrition. Although dual the resistance and reactance at 50 kHz. These equations
X-ray absorptiometry (DXA) can give a detailed and fairly are based on comparisons with isotopic dilution
accurate distribution of fat tissue mass, fat-free mass measurements. We quote here a few correlations, giving
(FFM) and bone mineral content (BMC) in the body, TBW in liters if H is in centimeters, for men and
it is expensive and cannot be repeated frequently due to women:
radiation. Bioimpedance, which is noninvasive and rela-
tively inexpensive, has been used for measurements of Kushner and Schoeller [1]:
TBW, ECW, FFM, which is the sum of LBM and BMC,
but its methods are indirect and often empirical and may V t ¼ 0:5561 H 2 =R þ 0:0955 W þ 1:726 (1)
not be accurate in patients with abnormal morphology.
Hannan et al. [2]:
In this study, we review various bioimpedance methods
and instruments as well as recent developments for V t ¼ 0:446 H 2 =R þ 0:126 W þ 5:82 (2)
calculating body composition parameters.
Sun et al. [3]:
Principle of body composition measurement
V t ¼ 0:45 H 2 =R þ 0:18 W þ 1:20 in men (3)
by impedance
Impedance is mostly used for measuring body fluid
volumes, FFM and fat tissue mass. V t ¼ 0:45 H 2 =R þ 0:11 W þ 3:75 in women (4)

1363-1950 ß 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins DOI:10.1097/MCO.0b013e32832da22c

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

There are fewer bioimpedance analysis (BIA) corre-
lations for ECW, those of Sun et al. [3]:
V e ¼ 0:123 H 2 =R þ 0:0119 H 2 =X þ 6:15
Sergi et al. [4]:
V e ¼ 3:32 þ 0:2 H 2 =R þ 0:005 H 2 =X þ 0:08W
BIA methods are considered as appropriate for healthy
individuals and patients with stable electrolyte balance
and BMI between 17 and 34 kg/m2 [5].
Bioimpedance spectroscopy methods
These methods require multifrequency impedance
meters such as the Xitron Hydra (Xitron Tech, San
Diego, California, USA). The bioimpedance spec-
troscopy (BIS) method [6] uses Re, the whole body
resistance extrapolated to zero frequency in the R–X
plane for calculating Ve from
V ¼ K b rae H 2 =R
where Kb is a shape coefficient considering the body as
the sum of five cylinders (four limbs and the trunk),
which was found equal to 4.3 from statistical measure-
ments made in adults [7]. This is logical, as low frequency
currents cannot penetrate intracellular water (ICW) as the
cell membrane acts as a capacitor. rae is the apparent
ECW resistivity due to the presence of nonconducting
tissues such as fat and bones, using the mixture conduc-
tivity theory of Hanai [8], is
rae ¼ re ½W =ðDb V e Þ3=2
where Db is the body density (1.05 kg/l). With the help of
(8), Eq. (7) gives
!2=3
H 2 W 1=2
V e ¼ ke (9)
Re
 
K b re
where ke ¼ 102 1=2 ¼ 0:306 for men and 0.316 for
Db
women, if re is in V cm.
A similar method was proposed by Fenech et al. [9] and
Jaffrin et al. [10] for expressing Vt as a function of Rinf, the
whole body resistance extrapolated to infinite frequency
in the R–X plane as
2=3
V t ¼ kt ðH 2 W 1=2 =Rinf Þ
with kt ¼ (Kbrt/Db1/2)2/3 ¼ 0.576 for men and 0.561 for
women. The Xitron Hydra 4200 impedance meter gives
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Body composition determination by bioimpedance Jaffrin 483
TBW with an equation similar to (10) with a resistivity
function of Re/Rinf to account for the ECW/TBW ratio
[11]. The BIS methods of Hanai [8] have a more scientific
(5) approach than BIA methods and seem more reliable, at
least for ECW. Intracellular volume Vi is calculated as the
difference between TBW and ECW. A quantitative
comparison of BIA and BIS methods listed above on a
(6) group of 57 individuals can be found in the study by
Jaffrin and Morel [12].
Fat free mass, fat mass and nutritional parameters
As FFM contains, in healthy individuals, an average of
72.9% of water, it is logical that BIA correlations for FFM
are similar to those for TBW. Sun et al. [3] have proposed
FFM ¼ 10:68 þ 0:65 H 2 =R þ 0:26 W
þ 0:02R in men (11a)
FFM ¼ 9:53 þ 0:69 H 2 =R þ 0:18 W
(7) þ 0:02R in women (11b)
While Kyle et al. [13] suggested
FFM ¼ 4:104 þ 0:518 H 2 =R þ 0:13 X þ 0:231 W
þ 4:23 sex (12)
with sex ¼ 1 in men and 0 in women. Fat tissue mass is
obtained as the difference between body weight and
FFM. BCM is calculated as
BCM ¼ 0:00835 TBK (13)
(8)
where the total body potassium concentration (TBK) is
expressed as a function of R and X at 50 kHz [14]. The
Xitron calculates FFM as
FFM ¼ 1:106V e þ 1:521 V i (14)
A bridge from bioimpedance analysis to bioimpedance
spectroscopy methods
As 50 kHz BIA impedance meters are more frequently
available than multifrequency ones, it is legitimate to
investigate the possibility to extrapolate Re and Rinf from
R50, the whole body resistance at 50 kHz, in order to
permit them to use BIS equations to calculate ECW and
TBW. This was done by measuring whole body resist-
ances R50, Re and Rinf with a Xitron 4200 in 58 healthy
individuals [15,16]. The ratio R50/Rinf was found to be
1.231  0.02 in men and 1.224  0.02 in women. Sim-
ilarly, the ratio R50/Re was equal to 0.806  0.02 in men
(10)
and to 0.833  0.02 in women. Thus, extrapolation of Re
and Rinf from R50 consisted in multiplying R50 by mean
values of these ratios. Values of Ve and Vt were then
calculated from equations (9) and (10) using these
 484 Assessment of nutritional status and analytical methods
extrapolated resistances. They were then compared with
values of Vex and Vtx obtained with Rex and Rinfx extrapo-
lated by the Xitron in a second group of 31 individuals
different from the first. Paired Student’s tests showed
that ECW and TBW measured at 50 kHz were not
significantly different from those using Rex and Rinfx.
The success of this method is due to the fact that the
ratios R50/Rinf and R50/Re vary between narrow limits
among healthy individuals, because of the strong corre-
lation between ECW and TBW volumes, reported in the
studies by Hannan et al. [2] and Matthie et al. [17].
Various types of impedance meters
(1) Medical impedance meters are used in supine pos-
ition with four gel-type electrodes, two voltage and
two current ones, pasted on the right foot and wrist.
Examples of BIA devices are the RJL 101A, 101B
(RJL Systems, Detroit, Michigan, USA), which is
manufactured by Akern (Florence, Italy) in Europe
and the Bodystat 1500 and 1500 MDD (5/50 kHz)
(Bodystat, Douglas, Isle of Man, UK) and the Bio-
dynamics BIA 310 (Biodynamics Corporation, Seat-
tle, Washington, USA). A recent addition is the
BodyExplorer produced by Juwell Medical (Gauting,
Munich, Germany), easy to use, that calculates ECW
by the modified BIS method of Jaffrin [16]. Multi-
frequency impedance meters operate either at var-
ious frequencies between 5 and 1000 kHz, such as the
Xitron Hydra, and permit to use BIS methods or at
three to four frequencies of 5, 50, 100 and 200 kHz,
such as Akern B109 and the Bodystat QuadScan 4000.
The Xitron Hydra has been redesigned by Fresenius
Medical Care (Bad Homburg, Germany) and is now
commercialized under BCM name for monitoring
hemodialysis patients and for body composition.
(2) Body fat analyzers for the general public, also called
leg-to-leg or foot-to-foot impedance meters (FFIs),
first introduced by Tanita Corporation (Tokyo,
Japan), consist of a body scale with four contact
reusable electrodes under the feet. They are inex-
pensive and used in upright position. However, with
these devices, the current only circulates in the legs
and the lower part of the trunk. Thus, the FFM they
measure results from an extrapolation to the whole
body using a proprietary equation of resistance,
weight, height, age and sex obtained, in principle,
by comparison with DXA data. Their fat tissue mass
is equal to the difference between body weight and
FFM. Some devices, such as Tanita models, express
fat tissue mass as percentage of body mass, whereas
others, such as Tefal FFI (Tefal SAS, Rumilly,
France), give fat tissue mass in kilograms. Upper
body fat analyzers (Omron HBF 301; Omron, Vernon
Hill, Illinois, USA) have only digital electrodes for
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
finger contact, and measurements are taken with
arms extended horizontally.
Some FFIs, intended for professional use such as the
Tanita BC 418, have been equipped with electrodes
mounted on handles for the hands in addition to plantar
electrodes and can measure the resistances of the limbs
and trunk by connecting sequentially various combi-
nations of two current and two voltage electrodes. By
comparison with FFM and fat tissue mass in limbs and
trunk provided by DXA, it is possible to obtain equations
giving FFM and fat tissue mass in limbs and trunk from
their resistances [18]. Multifrequency systems, the
Inbody720 [19], Biospace (Seoul, Korea) and X-Scan,
Jawon (Kyungsan City, Korea), are also available. These
professional systems are, not surprisingly, much more
expensive than FFI for the general public.
Accuracy and limitations of foot-to-foot impedance
meters and other regional impedance devices
Lukaski and Siders [20] have tested a Tanita TBF 604
FFI and an Omron HBF 301 impedance meter with hand
electrodes against a 50 kHz RJL medical type impedance
meter. DXA measurements were also taken with a Holo-
gic QDR 2000-W (Hologic, Bedford, Massachusetts,
USA). They observed that FFI data were closer to those
of DXA than those of Omron. The TBF 604 under-
estimated fat tissue mass percentage by 2.6 points in
women versus DXA and overestimated it by 1.2 points in
men, whereas the HBF 301 underestimated it by 6.3
points in women and by 2.3 points in men. They attrib-
uted the higher accuracy of the FFI to the fact that
plantar electrodes offer a consistent and full contact with
feet under the action of weight, whereas hand electrodes
require voluntary squeezing of the grips, which may be
variable and dissymmetric.
Several studies [21–24] have compared data of FFI built
by Tanita (Japan) and Tefal (France) with DXA and
medical impedance meters and showed that they were
reliable for a healthy population. Jebb et al. [22] evaluated
the accuracy of the Tanita 305 FFI in 104 men and 101
women against a four-compartment (4-C) model combin-
ing measurements by DXA, underwater weighing and by
deuterium dilution for TBW. They found that mean fat
tissue mass measured by the Tanita was larger by
0.8  3.95 kg than that from the 4-C model. In a later
study [23], the same authors reported the same compari-
son on overweight and obese women during a weight-loss
period of 12 weeks followed by a weight gain during
40 weeks between the Tanita FFI, a 1500 Bodystat BIA,
DXA and the 4-C model. Mean values for percentage of
fat tissue mass calculated by 4-C model and measured by
DXA, BIA and FFI together with percentage of fat tissue
mass changes during the weight loss and gain periods are
given in Table 1. This table shows that FFI estimations

Table 1 Mean percentage of fat tissue mass measured by different methods in overweight and obese women at start of study after a
12-week weight-loss therapy and after 52 weeks of follow-up
Mean %FM
Method At start After 12 weeks of therapy
4-C model 37.8 30.2
DXA 38.0 29.9
Bodystat 35.5 27.4
Tanita 37.0 29.7
Mean %FM changes during the first 12 weeks and the last 40 weeks. 4-C, four compartment; %FM, percentage of fat tissue mass; DXA, dual X-ray
absorptiometry. Adapted with permission [23].
are generally closer to values of 4-C model and DXA than
those of BIA. Thus, a well calibrated FFI can be more
accurate than a medical BIA impedance meter. Two FFI,
a Tanita BF 625 and a Tefal Bodymaster Vision (Tefal
SAS) and a whole body BIA (RJL 101) were compared
with a Hologic DXA by Lazzer et al. [24] in 53 obese
adolescents. They found, by comparison with DXA, that
both Tanita and Tefal devices underestimated fat tissue
mass less than whole body impedance. They concluded
that foot-to-foot impedance could be acceptable to assess
body composition in groups of overweight adolescents.
The Tefal FFI has also been tested against three reference
methods, DXA, underwater weighing and deuterium
dilution, for measuring TBW in 841 individuals in a multi-
centric and multiethnic study [25]. These authors found
that fat tissue mass measured by the Tefal device was well
correlated with DXA measurements with a mean overall
difference of 0.5  3.67 kg. Jaffrin et al. [26] showed that
the Tefal FFI, because of a patented square signal, could
provide a low and a high frequency resistance and calcu-
lated ECW volumes from this low resistance by a modified
BIS method. In 2008, Jaffrin and Moreno [27] extended
this method to obtain TBW from the high frequency
resistance also used for FFM. Both values of ECW and
TBW were shown to be not significantly different from
corresponding values obtained using a multifrequency
Xitron 4200 and equations (9) and (10).
By measuring segmental FFM in 40 healthy individuals
using an eight-electrode Tanita BC-418 and a DXA,
Pietrobelli et al. [18] reported values of appendicular
FFM and fat tissue mass percentages not significantly
different from those by DXA. Volgyi et al. [19] compared
percentage of fat tissue mass measured with a Tanita BC-
418, an InBody720 and DXA. They reported a systematic
underestimation of percentage of fat tissue mass by the
BIA devices as compared with DXA, of four points
(Tanita) and 5.8 points (InBody) in normal individuals.
In obese patients, the differences fell to two and 1.6
points, respectively. They attributed the better accuracy
of Tanita to the inclusion of age and degree of physical
accuracy in its software. Gibson et al. [28] tested an
Inbody720 on a cohort of Hispanic, black and white
individuals against a 4-C model and confirmed the
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Body composition determination by bioimpedance Jaffrin 485
Mean %FM change
After 52 weeks After 12 weeks Weight gain in last 40 weeks
34.9 7.6 4.6
34.5 8.2 4.6
32.5 7.6 5.1
34.5 7.3 4.6
Inbody720 significant percentage of fat tissue mass
underestimation in comparison with the 4-C model.
Neovius et al. [29] evaluated the ability of the Tanita
BC-418 to assess changes in body fat in obese women and
reported that although the BC-418 underestimated per-
centage of fat tissue mass by an average of five points as
compared with DXA, percentage of fat tissue mass
changes after a 6-week exercise period were in good
agreement with DXA data, except in the case of large
fat losses. A prototype eight-electrode FFI using the
electronics of the Tefal FFI was tested on 257 healthy
individuals in comparison with DXA measurements [30].
The first group of 170 individuals was used to determine
FFM and fat tissue mass equations, and the remaining
individuals were used for cross-validation of these
equations. Limbs and trunk FFM and fat tissue mass
were found to be not significantly different from DXA
values in both the first and validation groups.
Conclusion
The measurement of body composition by bioimpedance
is gaining acceptance in nutrition [31–33], sports medi-
cine [34] and hemodialysis for the prediction of ideal
patient dry weight and estimation of hydration state
[35,36]. The accuracy for multiracial applications will
increase with the development of specific ethnical
equations [37,38]. FFIs with medical and scientific refer-
ences are increasingly used in weight management pro-
grams [23,39]. Some models seem to underestimate fat
tissue mass, perhaps for commercial reasons. Segmental
eight-electrode models, which are much easier and faster
to use than traditional impedance meters, may open the
way to new research and clinical opportunities such as
measurement of appendicular skeletal muscles and
regional fat distribution.
References and recommended reading
Papers of particular interest, published within the annual period of review, have
been highlighted as:
 of special interest
 of outstanding interest
Additional references related to this topic can also be found in the Current
World Literature section in this issue (pp. 545–546).
1 Kushner RF, Schoeller DA. Estimation of total body water by bioelectrical
impedance analysis. Am J Clin Nutr 1986; 44:417–424.
 486 Assessment of nutritional status and analytical methods
2 Hannan WJ, Cowen SJ, Fearon KC, et al. Evaluation of multifrequency
bio-impedance analysis for the assessment of extracellular and total body
water in surgical patients. Clin Sci 1994; 86:479–485.
3 Sun SS, Chumlea WC, Heymsfield SB, et al. Development of bioelectrical
impedance prediction equations from body composition with the use of a
multicomponent model for use in epidemiologic surveys. Am J Clin Nutr 2003;
77:331–340.
4 Sergi G, Bussolotto M, Perini P, et al. Accuracy of bioelectrical bioimpedance
analysis for the assessment of extracellular space in healthy subjects and in
fluid retention states. Ann Nutr Metab 1994; 38:158–165.
5 Kyle UG, Bosaeus I, De Lorenzo AD, et al. Bioelectrical impedance analysis-
part II: utilization in clinical practice. Clin Nutr 2004; 23:1430–1453.
6 De Lorenzo A, Andreoli A, Matthie JR, Withers P. Predicting body cell mass
with bio-impedance by using theoretical methods: a technological review.
J Appl Physiol 1997; 82:1542–1558.
7 Van Loan MD, Withers PO, Matthie JR, et al. Use of bio-impedance spectro-
scopy to determine extracellular fluid, intracellular fluid, total body water and
fat-free mass. In: Ellis KJ, Eastman JD, editors. Human body composition: in
vivo methods, models and assessment. New York: Plenum; 1993. pp. 67–70.
8 Hanai T. Electrical properties of emulsions. In: Sherman D, editor. Emulsions
science. London: Academic Press; 1968. pp. 354–477.
9 Fenech M, Jaffrin MY, Dabaja Z. Validation of total body water measured by
multifrequency impedance. Arch Physiol Biochem 2003; 111:21.
10 Jaffrin MY, Fenech M, Moreno MV, Kieffer R. Total body water measurement
by a modification of the bioimpedance spectroscopy method. Med Biol Eng
Comput 2006; 44:873–882.
11 Matthie JR. Second generation mixture theory equation for estimating intra-
cellular water using bioimpedance spectroscopy. J Appl Physiol 2005;
99:780–781.
12 Jaffrin MY, Morel H. Body fluid measurements by bioimpedance: a review of
 bioimpedance spectroscopy (BIS) and bioimpedance analysis (BIA) methods.
Med Eng Phys 2008; 30:1257–1269.
This study provides a comparison of ECW and TBW volumes calculated by
different BIA and BIS methods on the same individuals.
13 Kyle UG, Genton L, Karsegard L, et al. Single prediction equation for
bioelectrical impedance analysis in adults aged 20–94 years. Nutrition
2001; 17:248–283.
14 Kotler DP, Burastero S, Wang J, Pierson RN. Prediction of body cell mass, fat-
free mass and total body water with bioelectrical impedance analysis: effects
of race, sex and disease. Am J Clin Nutr 1996; 64:489S–497S.
15 Morel H, Jaffrin MY. A bridge from bioimpedance spectroscopy to 50 kHz
bioimpedance analysis: application to total body water measurements. Phy-
siol Meas 2008; 29:S465–S478.
16 Jaffrin MY. Morel, extracellular volume measurements using bioimpedance
 spectroscopy (BIS)-Hanai method and wrist-ankle resistance at 50 kHz. Med
Biol Eng Comput 2009; 47:77–84.
A method for calculating ECW from 50 kHz resistance using the BIS equation with
different coefficient. It is tested against BIS method of Xitron 4200.
17 Matthie JR, Zaroutz B, De Lorenzo A, et al. Analytic assessment of various
bioimpedance methods used to estimate body water. J Appl Physiol 1998;
84:1801–1816.
18 Pietrobelli D, Rubiano F, St Onge MP, Heymsfield SB. New bioimpedance
analysis system: improved phenotyping with whole-body analysis. Eur J Clin
Nut 2004; 58:1479–1484.
19 Volgyi E, Tylavsky FA, Lyytikainen A, et al. Assessing body composition with
 DXA and bioimpedance: effects of obesity, physical activity and age. Obesity
2008; 16:700–705.
This study reports an interesting comparison between two eight-electrode BIA
devices and DXA for whole body FFM and fat tissue mass.
20 Lukaski HC, Siders WA. Validity and accuracy of regional bioelectrical impe-
dance devices to determine whole-body fatness. Nutrition 2003; 19:851–857.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
21 Nunez C, Gallagher D, Visser M, et al. Bioimpedance analysis: evaluation of
leg-to-leg system based on pressure contact foot-pad electrodes. Med Sci
Sports Exerc 1997; 29:524–531.
22 Jebb SA, Cole TJ, Doman D, et al. Evaluation of the novel Tanita body-fat
analyser to measure body composition by comparison with a four-compart-
ment model. Brit J Nutr 2000; 83:115–122.
23 Jebb SA, Servio M, Murgatroyd PR, et al. Validity of the leg-to-leg bioimpe-
dance to estimate changes in body fat during weight loss and regain in
overweight women: a comparison with multicompartment models. Intern J
Obes 2007; 31:756–762.
24 Lazzer S, Boirie Y, Meyer M, Vermorel M. Evaluation of two foot-to-foot
bioelectric impedance analysers to assess body composition in overweight
and obese adolescents. Brit J Nutr 2003; 90:987–992.
25 Boulier A, Chumlea WC, De Lorenzo A, et al. Body composition estimation
using leg-to-leg bioelectrical impedance: a six-site international cross valida-
tion study. Intern J Body Compos Res 2005; 3:31–39.
26 Jaffrin MY, Kieffer R, Moreno MV. Evaluation of a foot-to-foot impedance meter
measuring extracellular fluid volume in addition to fat-free mass and fat tissue
mass. Nutrition 2005; 21:815–824.
27 Jaffrin MY, Moreno MV. Measurements of total body water with a foot-to-foot
impedancemeter. Med Eng Phys 2008; 30:483–489.
28 Gibson AL, Holmes JC, Desautels RL, et al. Ability of new octapolar bioim-
pedance spectroscopy analysers to predict 4-component-model percentage
body fat in Hispanic, black and white adults. Am J Clin Nutr 2008; 87:332–
338.
29 Neovius M, Udden J, Hemmingsson E. Assessment of change in body fat
percentage with DXA and eight-electrode BIA in centrally obese women. Med
Sci Sports Exerc 2007; 39:2199–2203.
30 Jaffrin MY, Morel H, Lavielle Y. Body composition measurements in limbs
using eight-electrode bioimpedance. Scharfetter H, Mirva R, editors. In: 13th
ICEBI, IFMBE Proceedings 17; 29 August–2 September 2007; Graz,
Austria. Berlin, Heidelberg: Springer-Verlag; 2007. pp. 771–774.
31 Misra M, Soyka LA, Miller KM, et al. Regional body composition in adolescents
with anorexia nervosa and changes with weight recovery. Am J Clin Nutr
2003; 77:1361–1367.
32 Mika C, Herpertz-Dalhmann B, Heer M, Holtkamp K. Improvement of nutri-
tional status as assessed by multifrequency BIA during 15 weeks of refeeding
in adolescents with anorexia nervosa. J Nutr 2004; 134:3026–3030.
33 Moreno MV, Djeddi D, Jaffrin MY. Assessment of body composition in
adolescent patients with anorexia nervosa by bioimpedance. Med Eng Phys
2008; 30:783–791.
34 Utter AC, Nieman DC, Mulford GJ, et al. Evaluation of leg-to-leg BIA in
assessing body composition of high-school wrestlers. Med Sci Sports Exerc
2005; 37:1395–1400.
35 Zhu F, Leonard EF, Levin NW. Extracellular fluid redistribution during hemo-
 dialysis: bioimpedance measurement and model. Physiol Meas 2008;
29:S491–S502.
Continuous segmental ECW measurements in limbs and leg showed that a lower
initial leg ECW volume predisposes to hypotension.
36 Zhu F, Kuhlmann MK, Kotanko P, et al. A method for the estimation of
hydration state during hemodialysis using a calf bioimpedance technique.
Physiol Meas 2008; 29:S503–S516.
37 Prins M, Hawkesworth S, Wright A, et al. Use of bioelectrical impedance
analysis to assess body composition in rural Gambian children. Eur J Clin Nutr
2008; 62:1065–1074.
38 Wickramasinghe VP, Lamabadusuriya SP, Cleghorn GJ, et al. Assessment of
body composition in Sri Lankan children: validation of a bioelectrical impe-
dance prediction equation. Eur J Clin Nutr 2008; 62:1170–1177.
39 Chouinard LE, Schoeller DA, Watras AC, et al. Bioimpedance vs four-
compartment model to assess body fat change in overweight adults. Obesity
2007; 15:85–92.