The Bergen Geriatric Research Group / Bergen's Oldest

No certain way of measuring muscle mass bedside

Ole Martin Steihaug and collaborators recently published an article in The jounal of Nutrition, Health and Aging adressing the methodological challenges regarding measurements of muscle mass in older individuals.

Ole Martin foto
Bergen Geriatric Research Group

Main content


Older hip fracture patients often have reduced muscle mass, which is associated with adverse outcomes. Dual energy X-ray absorptiometry (DXA) can determine muscle mass, but is not practical in the acute phase. In this study, the authors investigated bioelectrical impedance analysis (BIA) and anthropometry compared against DXA for detecting low muscle mass in hip fracture patients.


A total of 162 hip fracure patiens from two Norwegian Hospitals participated in this study where the mean age was ≥ 65 year. Appendicular lean mass (ALM) was determined by DXA, BIA and anthropometry 3 months after hip fracture. ALM by BIA was calculated by the Kyle, Janssen, Tengvall and Sergi equations, and ALM by anthropometry by the Heymsfield and Villani equations. The area under the receiver operating characteristic curve (AUC) was used to compare BIA and anthropometry for determining low ALM (≤5.67 kg/m2 for women and ≤7.25kg/m2 for men).


Mean ALM by DXA was 14.8 kg (SD 2.3) for women and 20.8 kg (SD 4.2) for men and 45% of women and 60% of men had low ALM. BIA (Kyle) in women (AUC 0.81, 95% confidence interval 0.72-0.89) and BIA (Sergi) in men (AUC 0.89, 95% CI 0.80-0.98) were best able to discriminate between low and normal ALM. Anthropometry (Heymsfield) was less accurate than BIA in women (AUC 0.64, 95% CI 0.54-0.75), and equal to BIA in men (AUC 0.72, 95% CI 0.72 0.56-0.87).


The article is currently  available as Online First at the website of The journal of Nutrition, Health and Aging.