Potential errors in the application of mixture theory to multifrequency bioelectrical impedance analysis

Potential errors in the application of mixture theory to the analysis of multiple-frequency bioelectrical impedance data for the determination of body fluid volumes are assessed. Potential sources of error include: conductive length; tissue fluid resistivity; body density; weight and technical errors of measurement. Inclusion of inaccurate estimates of body density and weight introduce errors of typically < +/-3% but incorrect assumptions regarding conductive length or fluid resistivities may each incur errors of up to 20%.

Prediction of the chemical composition of lamb carcasses from multi-frequency impedance data

Multi-frequency bioimpedance analysis (MFBIA) was used to determine the impedance, reactance and resistance of 103 lamb carcasses (17.1-34.2 kg) immediately after slaughter and evisceration. Carcasses were halved, frozen and one half subsequently homogenized and analysed for water, crude protein and fat content. Three measures of carcass length were obtained. Diagonal length between the electrodes (right side biceps femoris to left side of neck) explained a greater proportion of the variance in water mass than did estimates of spinal length and was selected for use in the index L-2/Z to predict the mass of chemical components in the carcass. Use of impedance (Z) measured at the characteristic frequency (Z(c)) instead of 50 kHz (Z(50)) did not improve the power of the model to predict the mass of water, protein or fat in the carcass. While L-2/Z(50) explained a significant proportion of variation in the masses of body water (r(2) 0.64), protein (r(2) 0.34) and fat (r(2) 0.35), its inclusion in multi-variate indices offered small or no increases in predictive capacity when hot carcass weight (HCW) and a measure of rib fat-depth (GR) were present in the model. Optimized equations were able to account for 65-90 % of the variance observed in the weight of chemical components in the carcass. It is concluded that single frequency impedance data do not provide better prediction of carcass composition than can be obtained from measures of HCW and GR. Indices of intracellular water mass derived from impedance at zero frequency and the characteristic frequency explained a similar proportion of the variance in carcass protein mass as did the index L-2/Z(50).

A comparison of segmental and wrist-to-ankle methodologies of bioimpedance analysis

The common approach of bioelectrical impedance analysis to estimate body water uses a wrist-to-ankle methodology which, although not indicated by theory, has the advantage of ease of application particularly for clinical studies involving patients with debilitating diseases. A number of authors have suggested the use of a segmental protocol in which the impedances of the trunk and limbs are measured separately to provide a methodology more in keeping with basic theory. The segmental protocol hits not, however, been generally adopted, partly because of the increased complexity involved in its application, and partly because studies comparing the two methodologies have not clearly demonstrated a significant improvement from the segmental methodology. We have conducted a small pilot study involving ten subjects to investigate the efficacy of the two methodologies in a group of normal subjects. The study did not require the independent measure of body water, by for example isotope dilution, as the subjects were maintained in a state of constant hydration with only the distribution between limbs and trunk changing as a result of change in posture. The results demonstrate a significant difference between the two methodologies in predicting the expected constancy of body water in this study, with the segmental methodology indicating a mean percentage change in extracellular water of -2.2%; which was not significantly different from the expected null result, whereas the wrist-to-ankle methodology indicated a mean percentage change in extracellular water of -6.6%. This is significantly different from the null result, and from the value obtained from the segmental methodology (p = 0.006). Similar results were obtained using estimates of total body water from the two methodologies. (C) 1998 Elsevier Science Ltd. All rights reserved.

Data analysis in multiple-frequency bioelectrical impedance analysis

The performance of three analytical methods for multiple-frequency bioelectrical impedance analysis (MFBIA) data was assessed. The methods were the established method of Cole and Cole, the newly proposed method of Siconolfi and co-workers and a modification of this procedure. Method performance was assessed from the adequacy of the curve fitting techniques, as judged by the correlation coefficient and standard error of the estimate, and the accuracy of the different methods in determining the theoretical values of impedance parameters describing a set of model electrical circuits. The experimental data were well fitted by all curve-fitting procedures (r = 0.9 with SEE 0.3 to 3.5% or better for most circuit-procedure combinations). Cole-Cole modelling provided the most accurate estimates of circuit impedance values, generally within 1-2% of the theoretical values, followed by the Siconolfi procedure using a sixth-order polynomial regression (1-6% variation). None of the methods, however, accurately estimated circuit parameters when the measured impedances were low (<20 Omega) reflecting the electronic limits of the impedance meter used. These data suggest that Cole-Cole modelling remains the preferred method for the analysis of MFBIA data.

Bioelectrical impedance analysis predicts outcome in patients with suspected bacteremia

Fluid shifts from intracellular to extracellular water (ICW to ECW) are a feature of sepsis, caused by increased vascular permeability and cell catabolism. Changes in ECW and total body water (TBW) were assessed in a prospective observational study of patients with bacteremia by a bedside technique, and its prognostic impact determined; In 78 hospital patients with fever, the resistance ratio (Rinf/RO) and estimated ECW/TBW ratio from multifrequency bioelectrical impedance analysis, and serum albumin concentration were measured. Rinf/RO and ECW/TBW ratios decreased from day 0 to 2 in patients with significant bacteremia (n = 31), but not in patients with doubtful or negative blood cultures (n = 22 and 25), Increased Rinf/RO at baseline, and further increase of ECW/TBW from day 0 to 2, were associated with lower rate of recovery after 1 week and with higher mortality. Baseline Rinf/RO above the median (0.75) had positive and negative predictive values of 0.31 and 0.95 for death. This prognostic effect was independent of underlying disease and blood culture result in a multivariate model. Hypoalbuminemia at baseline was predictive of outcome, but changes in albumin from day 0 to 2 were unrelated to blood culture results or outcome. In patients with bacteremia,fluid shifts from intracellular to extracellular,vater occur early are rapidly reversible by antibiotic treatment but are associated with adverse prognosis. Bioelectrical impedance deserves further study as a tool for bedside monitoring of patients with bacteremia.

Optimizing electrode sites for segmental bioimpedance measurements

Recent advances in the application of bioelectrical impedance analysis (BIA) have indicated that a more accurate approach to the estimation of total body water is to consider the impedance of the various body segments rather than simply that of the whole body. The segmental approach necessitates defining and locating the physical demarcation between both the trunk and leg and the trunk and arm. Despite the use of anatomical markers, these points of demarcation are difficult to locate with precision between subjects. There are also technical problems associated with the regional dispersion of the current distribution from one segment (cylinder) to another of different cross-sectional area. The concept of equipotentials in line with the proximal aspects of the upper land lower) limbs along the contralateral limbs was investigated and, in particular, the utility of this concept in the measurement of segmental bioimpedance. The variation of measured segmental impedance using electrode sites along these equipotentials was less than 2.0% for all of the commonly used impedance parameters. This variation is approximately equal to that expected from biological variation over the measurement time. It is recommended that the electrode sites, for the measurement of segmental bioelectrical impedance in humans, described herein are adopted in accordance with the proposals of the NM Technology Assessment Conference Statement.

Sensitivity of multiple frequency bioelectrical impedance analysis to changes in ion status

Bioelectrical impedance analysis has found extensive application as a simple noninvasive method for the assessment of body fluid volumes, The measured impedance is, however, not only related to the volume of fluid but also to its inherent resistivity. The primary determinant of the resistivities of body fluids is the concentration of ions. The aim of this study was to investigate the sensitivity of bioelectrical impedance analysis to bodily ion status. Whole body impedance over a range of frequencies (4-1012 kHz) of rats was measured during infusion of various concentrations of saline into rats concomitant with measurement of total body and intracellular water by tracer dilution techniques. Extracellular resistance (R-o), intracellular resistance (R-i) and impedance at the characteristic frequency (Z(c)) were calculated. R-o and Z(c) were used to predict extracellular and total body water respectively using previously published formulae. The results showed that whilst R-o and Z(c) decreased proportionately to the amount of NaCl infused, R-i increased only slightly. Impedances at the end of infusion predicted increases iu TBW and ECW of approximately 4-6% despite a volume increase of less than 0.5% in TBW due to the volume of fluid infused. These data are discussed in relation to the assumption of constant resistivity in the prediction of fluid volumes from impedance data.

Reliability of multiple frequency bioelectrical impedance analysis: An intermachine comparison

The technical reliability (i.e., interinstrument and interoperator reliability) of three SEAC-swept frequency bioimpedance monitors was assessed for both errors of measurement and associated analyses. In addition, intraoperator and intrainstrument variability was evaluated for repeat measures over a 4-hour period. The measured impedance values from a range of resistance-capacitance circuits were accurate to within 3% of theoretical values over a range of 50-800 ohms. Similarly, phase was measured over the range 1 degrees-19 degrees with a maximum deviation of 1.3 degrees from the theoretical value. The extrapolated impedance at zero frequency was equally well determined (+/-3%). However, the accuracy of the extrapolated value at infinite frequency was decreased, particularly at impedances below 50 ohms (approaching the lower limit of the measurement range of the instrument). The interinstrument/operator variation for whole body measurements were recorded on human volunteers with biases of less than +/-1% for measured impedance values and less than 3% for phase. The variation in the extrapolated values of impedance at zero and infinite frequencies included variations due to operator choice of the analysis parameters but was still less than +/-0.5%. (C) 1997 Wiley-Liss, Inc.

Fluid shifts resulting from exercise in rats as detected by bioelectrical impedance

Purpose: This study was designed to investigate the immediate effect of exercise intensity and duration on body fluid volumes in rats throughout a 3-wk exercise program. Methods: Changes in the extracellular water (ECW) and total body water (TBW) volumes of rats were measured preexercise and postexercise using multiple frequency bioelectrical impedance analysis. Groups of rats were exercised at two intensities (6 m.min(-1) and 12 m.min(-1)) for two exercise times (60 min and 90 min) 5 d.wk(-1) during a 3-wk period. Changes in plasma electrolytes, glucose, and lactate resulting from the exercise were also measured on 3 d of each week. Results: Each group of animals showed significant losses in ECW and TBW as a direct result of daily exercise. The magnitude of fluid loss was directly related to the intensity of the exercise, bur not to exercise duration; although the magnitude of daily fluid loss at the higher intensity exercise (12 m.min(-1)) decreased as the study progressed, possibly indicating a training effect. Conclusion: At low-intensity exercise, there is a small bur significant loss in both TBW and ECW fluids, and the magnitude of these losses does not change throughout a 3-wk exercise program. At moderate levels of exercise intensity, there is a greater loss of both TBW and ECW fluids. However, the magnitudes of these losses decrease significantly during the 3-wk exercise program, thus demonstrating a training effect.

Noninvasive measurement of cerebral bioimpedance for detection of cerebral edema in the neonatal piglet

The association of sustained cerebral edema with poor neurological outcome following hypoxia-ischaemia in the neonate suggests that measurement of cerebral edema may allow early prediction of outcome in these infants. Direct measurements of cerebral impedance have been widely used in animal studies to monitor cerebral edema, but such invasive measurements are not possible in the human neonate. This study investigated the ability of noninvasive cerebral impedance measurements to detect cerebral edema following hypoxia-ischaemia. One-day-old piglets were anaesthetized, intubated and ventilated. Hypoxia was induced by reducing the inspired oxygen concentration to 4-6% O-2. Noninvasive cerebral bioimpedance was measured using gel electrodes attached to the scalp. Cerebral bioimpedance was also measured directly by insertion of two silver-silver chloride electrodes subdurally. Noninvasive and invasive measurements were made before, during and after hypoxia. Whole body impedance was measured to assess overall fluid movements. Intracranial pressure was measured continuously via a catheter inserted subdurally, as an index of cerebral edema. There was good agreement between noninvasive and invasive measurements of cerebral impedance although externally obtained responses were attenuated. Noninvasive measurements were also well correlated with intracranial pressure. Whole body impedance changes did not account for increases in noninvasively measured cerebral impedance. Results suggest that noninvasive cerebral impedance measurements do reflect intracranial events, and are able to detect cerebral edema following hypoxia-ischaemia in the neonate. (C) 2002 Elsevier Science B.V. All rights reserved.

GH treatment in adults with chronic liver disease: A randomized, double-blind, placebo-controlled, cross-over study

Patients with chronic liver disease (CLD) are catabolic and GH-resistant. The effects of supraphysiological recombinant human GH (rhGH; 0.2 IU.kg(-1).d(-1)) treatment in adults with CLD were assessed in a randomized, double-blind, placebo-controlled cross-over trial (4-wk dietary run-in, 4-wk treatment, and 2-wk wash-out phases). Nine adults with mild- to moderate-severity CLD participated (median age, 49 yr; three males and six females; Child's classification A in six and B in three). Biopsy-proven etiologies were: alcohol (four patients), primary biliary cirrhosis (three patients), non-A, non-B, non-C hepatitis (one patient), and cryptogenic (one patient). Treatment with rhGH increased serum IGF-I (median increase over placebo, +93 mug.liter(-1); P = 0.004), IGF-binding protein-3 (+0.9 mg.liter(-1): P = 0.004), and acid labile subunit (+10.7 nM; P = 0.004). Total body potassium (+8.0 g; P = 0.023), body weight (+1.6 kg; P = 0.008), and total body water (by bioelectrical impedance; +4.9 kg; P = 0.004) increased. Resting metabolic rate (+313 ml.kg(-1).min(-1); P = 0.004) and lipid oxidation (+1072.0 kcal.d(-1); P = 0.032) increased. Metabolic changes included increased fasting plasma glucose (+1.2 mm; P = 0.008), insulin (+33.8 mU.liter(-1); P = 0.004), C-peptide (+0.7 nM; P = 0.004), and free-fatty acids (+0.1 mEq.liter(-1); P = 0.04). Clinical side effects included worsening edema and ascites. Hepatocellular function did not change. Therefore, rbGH treatment in CLD: 1) overcame hepatic GH resistance; 2) may have improved whole-body protein catabolism; 3) increased lipolysis and lipid oxidation; 4) increased insulin resistance; and 5) had potent antinatriuretic effects. Long-term safety and efficacy require further assessment.

Cerebral impedance and neurological outcome following a mild or severe hypoxic/ischemic episode in neonatal piglets

Multi-frequency bio-impedance has the potential to identify infants at risk of poor neurodevelopmental outcome following hypoxia by detecting cerebral edema. This study investigated the relationship between the severity of an hypoxic/ischemic episode, neurological outcome following the hypoxia and non-invasively measured cerebral bioelectrical impedance in piglets. One-day-old piglets were anaesthetised and ventilated. Hypoxia was induced by reducing the inspired oxygen concentration to 3-5%. Severe hypoxia was defined as hypoxia resulting in at least 30 min of low amplitude EEG (

Assessment of volume depletion in children with malaria

Background The degree of volume depletion in severe malaria is currently unknown, although knowledge of fluid compartment volumes can guide therapy. To assist management of severely ill children, and to test the hypothesis that volume changes in fluid compartments reflect disease severity, we measured body compartment volumes in Gabonese children with malaria. Methods and Findings Total body water volume (TBW) and extracellular water volume (ECW) were estimated in children with severe or moderate malaria and in convalescence by tracer dilution with heavy water and bromide, respectively. Intracellular water volume (ICW) was derived from these parameters. Bioelectrical impedance analysis estimates of TBW and ECW were calibrated and bioelectrical impedance analysis measurements were taken daily against dilution methods, until discharge. Sixteen children had severe and 19 moderate malaria. Severe childhood malaria was associated with depletion of TBW (mean [SD] of 37 [33] ml/kg, or 6.7% [6.0%]) relative to measurement at discharge. This is defined as mild dehydration in other conditions. ECW measurements were normal on admission in children with severe malaria and did not rise in the first few days of admission. Volumes in different compartments (TBW, ECW, and ICW) were not related to hyperlactataemia or other clinical and laboratory markers of disease severity. Moderate malaria was not associated with a depletion of TBW. Conclusions Significant hypovolaemia does not exacerbate complications of severe or moderate malaria. As rapid rehydration of children with malaria may have risks, we suggest that fluid replacement regimens should aim to correct fluid losses over 12-24 h.

Validity of BMI as a measure of obesity in Australian white Caucasian and Australian Sri lankan children

Background: Body mass index ( BMI) is used to diagnose obesity. However, its ability to predict the percentage fat mass (% FM) reliably is doubtful. Therefore validity of BMI as a diagnostic tool of obesity is questioned. Aim: This study is focused on determining the ability of BMI- based cut- off values in diagnosing obesity among Australian children of white Caucasian and Sri Lankan origin. Subjects and methods: Height and weight was measured and BMI ( W/H-2) calculated. Total body water was determined by deuterium dilution technique and fat free mass and hence fat mass derived using age- and gender- specific constants. A % FM of 30% for girls and 20% for boys was considered as the criterion cut- off level for obesity. BMI- based obesity cut- offs described by the International Obesity Task Force ( IOTF), CDC/ NCHS centile charts and BMI- Z were validated against the criterion method. Results: There were 96 white Caucasian and 42 Sri Lankan children. Of the white Caucasians, 19 ( 36%) girls and 29 ( 66%) boys, and of the Sri Lankans 7 ( 46%) girls and 16 ( 63%) boys, were obese based on % FM. The FM and BMI were closely associated in both Caucasians ( r = 0.81, P < 0.001) and Sri Lankans ( r = 0.92, P< 0.001). Percentage FM and BMI also had a lower but significant association. Obesity cut- off values recommended by IOTF failed to detect a single case of obesity in either group. However, NCHS and BMI- Z cut- offs detected cases of obesity with low sensitivity. Conclusions: BMI is a poor indicator of percentage fat and the commonly used cut- off values were not sensitive enough to detect cases of childhood obesity in this study. In order to improve the diagnosis of obesity, either BMI cut- off values should be revised to increase the sensitivity or the possibility of using other indirect methods of estimating the % FM should be explored.

Ability of Bioelectrical Impedance to Predict Percentage Fat Mass in Children of Two Different Ethnic Origins

Detailed analysis of body composition in children has helped to understand changes that occur in growth and disease. Bioelectrical impedance analysis (BIA) has gained popularity as a simple, non-invasive and inexpensive tool of body composition assessment. Being an indirect technique, prediction equations have to be used in the assessment of body composition. There are many prediction equations available in the literature for the assessment of body composition from BIA. This study aims to cross-validate some of those prediction equations to determine the suitability of their use on Australian children of white Caucasian and Sri Lankan origins. Height, weight and BIA were measured. Total body water was measured using the isotope dilution method (D2O). Fat-mass (FM) and %FM were estimated from BIA using ten prediction equations described in the literature. Five to 14.99-year-old healthy, 96 white Caucasians and 42 Sri Lankan children were studied. The equation of Schaefer et al was the most suitable prediction equation for this group with the lowest mean bias for %FM assessment in both Caucasian (–1.0±9.6%) and Sri Lankan (1.6±5.2%) children and the fat content of the individuals did not influence the predictions by this equation. Impedance index (height2/impedance) explained for 80% of TBW in white Caucasians and 93% in Sri Lankans and figures were similar for the prediction of FFM. We conclude that BIA can be used effectively in the assessment of body composition in children. However, for the assessment of body composition using BIA, either prediction equations should be derived to suit the local populations or existing equations should be cross-validated to determine their suitability before their application.