Poster session V: Development of body composition methodology and its use in clinical practice

International Journal of Obesity (2011) 35, S56–S62 © 2011 Macmillan Publishers Limited All rights reserved 0307-0565/11 www.nature.com/ijo Poster session V: Development of body composition methodology and its use in clinical practice International Journal of Obesity (2011) 35, S56–S62; doi:10.1038/ijo.2011.76 P108 Validation of Air Displacement Plethysmography for Body Composition in Living Piglets Anne Frondas-Chauty1,2, Isabelle Huton3, Isabelle Le Huerou-Luron3, Jean-Christophe Roze1,2 and Dominique Darmaun2 1 Neonatal Medicine, Nantes University Hospital, Nantes, France; 2 INRA, UMR 1280 PHAN, Nantes University, Nantes, France; 3 Institut National de la Recherche Agronomique, Saint-Gilles, France Objective: This study evaluated the precision of measurements of body fat mass (FM, % of body weight) in comparison with biochemical analyses, with living subjects. Methods: Three groups of 12 crossbred piglets [Piétrain x (Large White x Landrace)] from the experimental herd of INRA (St Gilles, France), were separated from sows at 2, 7 and 21 days of life. And underwent 4 consecutive body composition assessments by ADP using a PEAPOD instrument under general anaesthesia and spontaneous ventilation, after they had been shaved, washed and dried. After the last measure, piglets were euthanized, and whole body chemical analysis was performed using solvent extraction for lipid content. Results: The average body weight of piglets was 1490, 2210, and 5720 g at d2, d7, and d21, respectively. Mean % FM values determined by biochemical analysis and PEA-POD were 8.63±4.079% and 8.01±4.03% respectively. The mean difference (-0.66±1.72%) was not significantly different from zero. Linear regression and Bland Altman analyses of % FM obtained by both methods, indicated good agreement. The regression analysis indicated a low SEE (1.71) and a high R2 (0.827). As the PEA POD used pediatrics formulas, data were re-analyzed using equations from appropriate to piglets from literature. The linear regression was very similar when using modifications of the equations to take into account body surface area and fat mass density. The mean coefficient of variation (CV; 4 determinations for each piglet) was 13.4±10.5%. The mean CV for each group of piglet were: 24.6%±11.1; 10.5%±5.1; 8.6%±8.2 at d2, d7, and d21, respectively. CV was linked with % FM with biochemical analysis, body mass, and age of the piglet. Conclusion: This study confirms that ADP using the PEAPOD instrument provides highly precise and accurate measurements of fat mass in living animals that are in the same range of body weight and fat mass as human infants included low birth weight infants. P109 Validation of Bioimpedance Against Isotope Techniques for Determination of Body Composition in HIV-Infected Children A Sewnath1,*, MK Chhagan1, S Arpadi2 and M Ahdikari1 1 Department of Paediatrics and Child Health, Nelson R Mandela School of Medicine, University of KwaZulu- Natal, Durban, South Africa; 2St Luke’s Roosevelt Hospital, New York, NY, USA *Corresponding Author Objective: 1) To establish whether BIA equations currently used for deriving total body water (TBW) and Fat Free Mass (FFM) in Nigerian and HIV-infected American children are precise to be applied to HIV-infected children in South Africa. 2) To determine feasibility of blood and saliva specimens in isotope studies. 3) To compare performance of blood and saliva using the isotope dilution method in children aged 3 to 6 years. Methods: Anthropometry and Bioelectric impedance measurements were performed on 42 HIV-infected children using a 50-kHz tetrapolar BIA device. TBW was measured by deuterium oxide (D2O) dilution. TBW enrichment in blood and saliva was assessed using a Fourier–Transfer Infrared Spectrometer. ECW was measured by sodium bromide dilution and assessed using HPLC. Results: Good correlations were seen between measured and predicted values for TBW and FFM, and between saliva and blood assays of TBW. Bland-Altman plots showed poor agreement between measured and predicted values of TBW. Bland-Altman plots also showed saliva over estimated TBW. Conclusion: Co-morbidities included oral problems coupled with opportunistic diseases may have caused measurement variation in salivary samples. Altered mucosal permeability may have affected results by altering absorption and/or equilibration time of the isotope. Addition of variables such as reactance improved TBW. P110 Bioimpedance Vector Analysis (BIVA) to Monitor Changes in Body Composition in Patients with Anorexia Nervosa Verena Haas1, Andrea Riedl2, Tobias Hofmann2, Annette Nischan2, Michael Boschmann1 and Burghard Klapp2 1 Franz-Volhard-Centrum für Klinische Forschung, ECRC Charité Campus Buch, Universitätsmedizin Berlin, Germany; 2 Med. Klinik mit Schwerpunkt Psychosomatik, Charité Campus Mitte, Universitätsmedizin Berlin, Germany Objective: to describe changes in body composition before and during refeeding of female adult patients with Anorexia Nervosa (AN) using bioelectrical impedance analysis (BIA). Methods: Retrospective chart review and analysis of bioimpedance data was carried out on 57 female AN patients (mean age 26±8 years, mean BMI 13.6±2.0 kg/m2) on admission to inpatient therapy, and of 27 and 13 of these patients during short term (4 weeks) and medium term (3 months) nutritional recovery. Body composition was characterized with two different prediction equations (classical quantitative BIA) as well as with qualitative Bioimpedance Vector Analysis (BIVA). Results: Classical BIA produced implausible results (i.e. a negative fat mass) in up to 47% of the study population. In contrast BIVA demonstrated a significant loss of cell mass, as well as a highly variable hydration status ranging from dehydration to fluid overload, on admission to inpatient treatment. Interpretation of vector movement during refeeding suggested that short term weight gain predominantly consisted of fluid accumulation, while medium term weight gain resulted in increased hydrated soft tissue. Abstracts S57 Conclusion: the use of BIVA offers a more suitable alternative over classical BIA to monitor body composition changes in the treatment of patients with AN. P111 An Improved CT-based Method of Gradient Co-occurrence Matrix Analysis in Cirrhotic Liver Identification Huiyan Jiang1 and Tingting Sun2 1 Software College, Northeastern University, Shenyang, China; 2 Sino-Dutch Biomedical and Information Engineering School, Northeastern University, Shenyang, China Objective: Liver cirrhosis is a diffuse liver disease, which can be distinguished from normal liver by gradient-based textural information. To improve the efficiency and performance of traditional CT gradient co-occurrence matrix analysis for textural feature extraction in cirrhotic liver identification, an improved method is proposed. Methods: The new method is derived from the traditional gradient co-occurrence matrix, which simplifies the measuring of gradient between pixel pa (...truncated)