Application of a dual energy x-ray absorptiometry derived 4-compartment body composition model: Non-discriminatory against leanness and sex. (December 2020)

• Record Type: Journal Article
• Authors: Smith-Ryan, Abbie E.; Blue, Malia N.M.; Hirsch, Katie R.; Brewer, Gabrielle J.
• Is Part Of: Clinical nutrition ESPEN. Volume 40(2020)
• Page Start: 401
• Page End: 405
• Publication Date: 2020-12
• Subjects: Fat mass -- Muscle mass -- Percent body fat -- Fat free mass -- Validity; Nutritionally induced diseases -- Periodicals; Metabolism -- Disorders -- Periodicals; 616.39005
• DOI: 10.1016/j.clnesp.2020.09.033
• Languages: English
• ISSNs: 2405-4577

Abstract

Summary: Background & aims: Body composition assessment has large variability. A dual-energy x-ray absorptiometry (DXA) derived four compartment (4C) method has been developed as an accurate and reliable method for assessing body composition in overweight/obese adults. This investigation was aimed at understanding the validity of the DXA-derived 4C equation for use in normal weight individuals, stratified by sex, and with varied levels of lean mass. Values were also compared against DXA alone. Methods: 78 men and women (68% female; Mean ± SD; Age: 19.2 ± 1.2 yrs; Ht: 168.8 ± 9.1 cm; Wt: 62.8 ± kg) completed a traditional 4C body composition reference assessment. Body composition was also assessed using a DXA-4C model. Validity was evaluated from total error (TE), constant error, and standard error of the estimate (SEE). Proportional bias was identified with Bland Altman plots. Results: Although significantly different (p < 0.05) the DXA-4C model produced ideal TE and SEE compared to the 4C criterion for all body composition outcomes of fat mass (TE: 2.1 kg; SEE: 1.9 kg), lean mass (TE: 2.1 kg; SEE: 1.8 kg), and percent body fat (TE: 3.6%; SEE: 3.4%). Validity results did not differ for men vs. women. DXA-4C estimates were slightly better in individuals with higher lean mass. DXA alone resulted in significantly greater error than DXA-4C (p < 0.05). Conclusion: Body composition assessed from DXA-4C is an accurate approach, particularly in those with high levels of lean mass. This model appears to be more accurate than DXA alone.