Home page - Search Previous issues
23 Percentage body fat in Irish adults measured by bioelectrical impedance analysis compared with waist circumference measurements and body fat estimated from prediction equations
C.M. Murphy, M.A. Galvin, A. Flynn, S.N. McCarthy and M. Kiely
24 The use of bioelectrical impedance analysis to estimate total body water in children with cerebral palsy
and non-disabled children
K.L. Bell and P.S.W. Davies
25 The relationship between mammographic density and body composition
Z. Chen, L.K. Staten, G. Maskarinec, L. Arendell, R. Bruhn, J.K. Nicholas and J. Marshall
26 Reliability of body circumference and skinfold measurements by observers trained in groups
Jack Wang, Glenn Bartsch, Sai Subhasree Rahgavan, Teresa Yurik, Grace Peng, Li Chen, Dale LeSueur,
Judith C Shlay, for the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA)
27 Proof-of-principle to measure potassium in the human brain: a feasibility study
L. Wielopolski, L.M. Ramirez, P.K. Coyle, Z.M. Wang and S.B. Heymsfield
28 Does the combination of hand-to-hand and foot-to-foot impedance measurements improve the
prediction of body composition in African women?
A. Gartner1, A. Dioum2, F. Delpeuch1, B. Maire1 and Y. Schutz3
International Journal of Body Composition Research 2004, Vol. 2 No. 1: 3-13
C.M. Murphy1, M.A. Galvin1, A. Flynn1, S.N. McCarthy2 and M. Kiely1
1Department of Food and Nutritional Sciences, University College Cork, Ireland; 2Department of Clinical Medicine,
Trinity Centre for Health Sciences, St James’s Hospital, Dublin 8, Ireland.
The aim of this study was to compare percentage body fat (%BF) measured by bioelectrical impedance analysis (BIA) with %BF derived from published prediction equations and to assess waist circumference as an alternative measurement of body fatness in a representative sample of 18 to 64-y-old Irish adults. Percent body fat was measured using the Bodystat 1500 impedance analyser and standard procedures were used to measure weight, height and waist circumference in 1098 (495 men; 603 women) 18 to 64-yr-old adults who participated in the North South Ireland Food Consumption Survey (NSIFCS). Anthropometric data for each individual were applied to five age- and sex-specific prediction equations to estimate %BF. Mean %BF from BIA was significantly lower (P<0.005) than mean %BF estimated by the equations, with two exceptions: van Loan & Mayclin (1987) in women aged 18–64 y and Womersley & Durnin (1977) in women aged 18–50 y. The mean difference (Bland & Altman technique) between %BF from BIA and %BF from the equations was lower in women (–1.5%) than in men (–4.7%). %BF was highly correlated (P<0.001) with %BF from the equations (r=0.70–0.97). The %BF from BIA was most similar to %BF from the van Loan & Mayclin (1987) equation (mean difference ± SD: men–3.6±1.9%; women –0.2±2.6%) and was least similar to %BF from the Deurenberg et al. (1991) equation using impedance (mean difference: men –7.1±1.7%; women –5.0±2.2%). Agreement between quartiles of %BF and quartiles of waist circumference was not strong (men 43–70%; women 40–68%). The study highlights the difference in %BF between methods to determine body composition and the need to validate the BIA method in the population group of interest. The waist circumference measurement is not a suitable alternative to assess the body fatness of this population group.
International Journal of Body Composition Research 2004, Vol. 2 No. 1: 15-22
K.L. Bell1,2 and P.S.W. Davies1
1Children’s Nutrition Research Centre, Department of Paediatrics and Child Health, University of Queensland,
Royal Children’s Hospital, Brisbane, Queensland, Australia. 2Queensland Paediatric Rehabilitation Service,
Royal Children’s Hospital, Brisbane, Queensland, Australia
Measurement of body composition in children with cerebral palsy (CP) is essential since under-nutrition as well as overweight and obesity are frequently reported. Bioelectrical impedance analysis (BIA) is a quick, noninvasive, and precise technique for the assessment of total body water (TBW) that lends itself to use in such children. Validation studies have been conducted in various populations including healthy, non-disabled children and non-ambulatory children with spastic quadriplegic CP. The aim of the current study was to evaluate the use of BIA to predict TBW in ambulatory children with CP and non-disabled children. Impedance was measured with a Bodystat 1500 in 15 ambulatory children with CP and 16 non-disabled controls, aged 5–12 years. TBW was predicted from impedance using previously published equations and the Bodystat algorithm. TBW was measured using oxygen-18 and standard procedures as the reference method. The Bodystat algorithm overestimated TBW with the largest bias (5.8 ±1.1 kg for the children with CP; 4.6 ± 1.5 kg for the control group). The Kushner equation predicted TBW with the least bias (0.3 ± 1.3 kg for the children with CP; –0.6 ± 1.3 kg for the control group); however, the limits of agreement remained large (–2.3–2.9 kg for the CP group,–3.2–2.0 kg for the control group). There was no statistically significant relationship between the bias and TBW except for the Bodystat algorithm, indicating that in the other equations the bias was consistent across the range of measurements. The ability of BIA to predict TBW in children with CP and non-disabled children depends on the equation chosen.
International Journal of Body Composition Research 2004, Vol. 2 No. 1: 23-29
Z. Chen,1 L.K. Staten1, G. Maskarinec2, L. Arendell1, R. Bruhn1, J.K. Nicholas1 and J. Marshall1,3
1Mel and Enid Zuckerman Arizona College of Public Health, University of Arizona; 2Cancer Research Center of
Hawaii, University of Hawaii; 3Roswell Park Cancer Institute, Buffalo, New York, USA
The purpose of this study was to investigate associations between mammographic density measurements and
various body composition indices. The participants (n =79) were healthy Hispanic and non-Hispanic white
women age 49+ years. Mammographic density was assessed for both breasts using a computer-assisted method.
Body composition, including total and regional body fat and lean soft tissue mass as well as bone mineral density
(BMD), was measured using dual-energy X-ray absorptiometry. Odds ratios (OR) and 95% confidence intervals
(95% CI) of having high % mammographic density (50th percentile as the cutoff) by different body
composition categories were calculated using logistic regression analysis. Percent mammographic density was
inversely related to % body fat (P = 0.013, trend test). In comparison to women in the lowest tertile of % trunk
fat, the ORs of having high % mammographic density were 0.40 (95%CI = 0.13, 1.25) and 0.11 (95% CI = 0.03,
0.39) respectively for women in the second and third tertiles of % trunk fat. Women with higher % trunk lean
soft tissue were four to seven times more likely to have high % mammographic density (P<0.05). The relationships
between BMD and % mammographic density were not statistically significant. It is concluded that body
soft tissue compositions are significantly related to % mammographic density. Future studies should take this
association into consideration when using mammographic density as a marker of breast-cancer risk.
International Journal of Body Composition Research 2004, Vol. 2 No. 1: 31-36
Jack Wang1, Glenn Bartsch2, Sai Subhasree Rahgavan3, Teresa Yurik2, Grace Peng2, Li Chen2, Dale LeSueur2,
Judith C Shlay4, for the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA)
1Body Composition Unit of St. Lukes-Roosevelt Hospital, Columbia University College of Physicians and
Surgeons, New York, NY, USA; 2CPCRA Statistical Center, University of Minnesota, Minneapolis, MN, USA;
3Harlem Hospital, Columbia University College of Physicians and Surgeons, New York, NY, USA; 4Denver
Community Programs for Clinical Research on AIDS, University of Colorado Health Sciences Center, Denver,
CO, USA
Anthropometry is suitable for studies involving large samples and multiple sites, allowing estimations of body fat content and distribution from models that utilize body circumferences and skinfolds. However, the technique requires well-trained observers and the use of standardized measurement instruments and protocols. As part of the multicenter metabolic study conducted by the Community Program for Clinical Research on AIDS (CPCRA 061), 26 nurses from clinical centers across the United States were trained in groups using a standardized training program consisting of reviewing illustrations on measurement protocols, observing demonstrations by a trainer, and practicing measurements for four body circumferences (BC) and five skinfolds (SF) on at least ten volunteers using the trainer’s readings as the standards. Criteria for passing the training was that the average of the absolute percent difference (%AD) between the readings of the trainer and the trainee on another five volunteers were ≤2% for BC and ≤20% for SF. Nineteen trainees passed the training with the overall mean of %AD being 1.9% for BC and 17.9% for SF. This study established a model for training observers in groups with 73% of trainees achieving skills similar to that of individually-trained observers for performing anthropometric measurements.
International Journal of Body Composition Research 2004, Vol. 2 No. 1: 37-43
L. Wielopolski1, L.M. Ramirez1, P.K. Coyle2, Z.M. Wang3 and S.B. Heymsfield3
1Brookhaven National Laboratory, Upton, NY; 2 University Hospital at Stony Brook, Neurology Department,
Stony Brook, NY; 3St Luke’s/Roosevelt Hospital, New York, NY, USA
We describe the results of a proof-of-principle to measure the potassium content in the human brain using the natural radioisotope 40K that is in equilibrium with the stable isotopes of potassium, 39K and 41K. A fixed relationship exists between radioactive potassium and the total potassium in the brain, which in turn reflects the brain’s cell mass and intracellular water compartment. Accordingly, we explored whether measurements of brain potassium could serve as possible indicators of intracellular cerebral edema. We designed, built, and then calibrated our system using a spherical phantom containing KCl salt dissolved in water at levels comparable to those in the human brain. Emitted radiation was detected using sodium iodide (NaI) and high-purity germanium (HP-Ge) detectors. Our results with phantoms and with five volunteers demonstrate the feasibility of measuring potassium at the levels normally present in human brain tissue. We plan to extend the system to detect the onset of brain edema in patients with multiple sclerosis.
International Journal of Body Composition Research 2004, Vol. 2 No. 1: 45-48
A. Gartner1, A. Dioum2, F. Delpeuch1, B. Maire1 and Y. Schutz3
1Nutrition Unit, UR 106 (WHO Collaborating Centre for Nutrition), IRD (Institut de Recherche pour le
Développement), Montpellier, France; 2Equipe de Nutrition, Laboratoire de Physiologie, Département de
Biologie Animale, Faculté des Sciences et Techniques, Université Cheikh Anta Diop de Dakar, Sénégal, West
Africa; 3Institute of Physiology, Faculty of Medicine, University of Lausanne, Switzerland
Hand-to-hand (HHI) or foot-to-foot (FFI) impedance methods are both easy to use and inexpensive, and they offer a rare opportunity to evaluate body composition in Africa. However, the standard algorithm supplied with each device is not valid in African subjects. We explored whether a combination of both methods may be valuable for improving the prediction of body composition in 170 African women as measured by air-displacement plethysmography (ADP). Simply taking the arithmetic mean of estimations from FFI and HHI still gave a marked and significant bias. Estimated values were regressed against the reference value from ADP. The mean of the two predicted values, or another predicted value combining HHI and FFI results in the same regression, suggested that there is no clear advantage to using the two methods together. The use of these field methods could gain in value if they were to provide raw impedance values, thus allowing to develop new predictive equations combining arm and leg impedance in order to track both limbs.