The relationship of anthropometry and bio-electrical impedance to dual-energy X-ray absorptiometry in elderly men and women

Objective: evaluation of the accuracy of fat-free mass (FFM) measured by anthropometry or bio-electrical impedance and by dual-energy X-ray absorptiometry (DXA). \udMethods: 34 subjects (17 males) aged over 75 years, selected from the patients of a group general practice and from a luncheon club and a day centre and all living indepe,ndently at home. FFM was measured by anthropometry (height, demi-span, knee height, weight and triceps skinfold thickness), bio-electrical impedance and DXA. \udResults: An equation for FFM provided by the manufacturer of the bio-electrical impedance analyser gave r2 as 0.900, standard error (SE) as 2.28 kg and systematic overestimate as 6.26 kg. FFM calculated from anthropometry gave an r2 of 0.944, SE of 22.5 kg and a systematic underestimate of 6.26 kg. FFM estimated from height, weight and gender gave an r2 of 0.900, SE of 3.0 and systematic overestimate of 3.0 kg. New equations derived from the data gave a narrower SE for equations derived from bio-electrical impedance (1.2 kg), anthropometry (1.95 kg) and bio-electrical impedance with height, weight and sex (1.45 kg). Substitution of demi-span or knee height for height in equations involving bio-electrical impedance, height, weight and sex gave an r2 of 0.965 and SE of 1.87 for demi-span and an r2 of 0.975 and SE of 1.51 for knee height. \udConclusions: by comparing calculations of FFM from bio-electrical impedance and anthropometry and by DXA we have been able to produce equations from anthropometry and bio-electrical impedance which provide a more accurate estimate of FFM in individuals over the age of 75 years