Determining body fat mass using DXA and BIA

Nutrition Research Newsletter, Feb, 2005

Information on body composition, including percentage body fat (¿), is used frequently to evaluate growth and nutrition in children, and nutritional status in various disease states. Body composition assessment also has many clinical uses, such as assessing disease progression or treatment efficacy. There are multiple technologies available to measure body composition. Currently, the underwater weighing method is the gold standard. Air-displacement plethysmography and dual-energy X-ray absorptiometry (DXA) are the two relatively new reference methods. The use of these methods is limited due to the inaccessibility and the high cost of equipment. Therefore, simple methods such as bioelectrical impedance analysis (BIA) and skinfold-thickness measurements are still the norm in field studies and for public use. The BIA method has been extensively used in clinics, in sports medicine, and in weight reduction programs. Multiple studies have compared predictions of ¿ by BIA with measurements made by reference methods, but the results are contradictory.

In an attempt to answer questions regarding the accuracy and validity of the BIA, researchers performed parallel measurements of ¿ by suing both multifrequency BIA and DXA in a large sample of the healthy population in Newfoundland and Labrador. Subjects included 591 men and women between the ages of 19 and 60 years, without serious medical problems.

All measurements were performed after the subjects had fasted for 12 hr. The subjects were weighed in light clothes. DXA was used for the measurement of whole-body composition, including fat mass, lean body mass, and bone mineral densities. ¿ was calculated from entire body mass. BIA measurements were done immediately after the DXA analysis.

Correlations between BIA and DXA were 0.88 for the whole population, 0.78 for men, and 0.85 for women. The mean ¿ determined by BIA (32.89 [ or -] 8.00%) was significantly lower than that measured by DXA (34.72 [ or -] 8.66%). The cutoffs were sex specific. BIA overestimated ¿ by 3.03% and 4.40% when ¿ was <15% in men and <25% in women, respectively, and underestimated ¿ by 4.32% and 2.71% when ¿ was > 25% in men and > 33% in women, respectively.

BIA appears to be a good alternative for estimating ¿ when subjects are within a normal body weight range. BIA tends to overestimate ¿ in lean subjects and underestimate in obese subjects.

G. Sun, C. French, G. Martin, et al. Comparison of multifrequency bioelectrical impedance analysis with dual-energy X-ray absorptiometry for assessment of percentage body fat in a large, healthy population. Am J Clin Nutr;81:7478 (January, 2005). [Correspondence: G Sun, Discipline of Genetics, Faculty of Medicine, Memorial University of Newfoundland, Health Science Center, 300 Prince Phillip Drive, St John's, NL A1B 3V6 Canada. E-mail: gsun@mun.ca].