RESUMEN
Simultaneous with a 12 channel ECG, body composition was analysed by segmental multi-frequency impedance analysis in 101 healthy subjects and in 118 patients with chronic heart failure (CHF, n= 40), chronic renal failure with haemodialysis (HD, n= 20), and miscellaneous internal diseases (n= 58). Whole body DXA and sodium bromide dilution were used as reference methods for total body water (TBW), extracellular fluid (ECF), appendicular muscle mass (AppMM) and fat mass (FM). Empirical prediction equations were developed in a randomized evaluation sample and then evaluated in unknowns. TBW, ECF, AppMM and FM could be predicted with regression coefficients of 0.96, 0.90, 0.95 and 0.93, respectively, all with p< 0.001. Only segmental impedances and height, but not age, sex, weight and BMI contributed to the prediction of water compartments. About half the patients with CHF and half of those on HD showed increased ECF/ICF ratio in relation to % FM at the legs but not at the thorax. The predicted AppMM was additionally corrected for increased ECF to determine "dry AppMM", which is markedly lower than the misleading reference DXA. This methodology shows promise as a combination of routine ECG with measurement of body composition, assessment of sarcopenia and detection of overhydration.