A comparison of body composition estimates using dual-energy X-ray absorptiometry and air-displacement plethysmography in South African neonates

• Published in: European journal of clinical nutrition Vol. 70; no. 11; pp. 1254 - 1258
• Main Authors: Wrottesley, S V, Pisa, P T, Micklesfield, L K, Pettifor, J M, Norris, S A
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.11.2016; Nature Publishing Group
• Subjects: 631/443/319/1642/393; 692/499; Absorptiometry, Photon; Body Composition; Body fat; Clinical Nutrition; Comparative analysis; Dual energy X-ray absorptiometry; Epidemiology; Estimates; Fat-free; Fat-free body mass; Female; Health aspects; Humans; Infant, Newborn; Infants; Internal Medicine; Longitudinal studies; Male; Medicine; Medicine & Public Health; Metabolic Diseases; Metabolism; Neonatal care; Neonates; Newborn babies; Nutritional Requirements; original-article; Plethysmography; Public Health; Reproducibility of Results; South Africa; X-rays; South Africa
• ISSN: 0954-3007; 1476-5640
• DOI: 10.1038/ejcn.2016.91

Background/Objectives: Neonatal body composition is an important predictor of future metabolic risk; however, the comparability of objective assessment techniques, particularly in African populations undergoing rapid health transition, is not known. This paper compares body composition estimates by air-displacement plethysmography (ADP) and dual-energy X-ray absorptiometry (DXA) in South African neonates. Subjects/Methods: Fat mass, fat-free mass and body fat percentage (%fat) estimates by ADP and DXA were compared in 88 urban, black South African neonates. The level of agreement between the techniques was assessed using Bland–Altman analyses. Results: Significant correlations were observed between ADP and DXA measurements of fat mass ( r =0.766), fat-free mass ( r =0.942) and %fat ( r =0.630); however, ADP estimates of fat mass (408±172 g vs 337±165 g; P< 0.001) and %fat (12.9±4.4% vs 9.9±4%; P< 0.001) were significantly higher and fat-free mass (2681±348 g vs 2969±375 g; P< 0.001) significantly lower than those by DXA. Fat-free mass estimates showed greater consistency in the level of agreement between the techniques compared with fat and %fat estimates where the differences between methods were less predictable. Conclusion: Although ADP and DXA body composition estimates are highly correlated in neonates, significant differences are observed between the techniques. This is particularly relevant for fat mass and %fat estimates, where differences are highly variable between methods. Further investigation is needed to minimise inter-method differences to ensure accurate and comparable assessment of body composition at birth and across longitudinal study follow-up.