H. G. Goovaerts

Measurement of transcellular fluid shift during haemodialysis

A method is presented to measure transcellular fluid shifts during haemodialysis based on a simplified model of the electrical admittance of biological tissues. It allows for the measurement of intracellular and extracellular conductivities and their ratios. The method is noninvasive, clean and harmless, and can be easily computerised in order to be performed continuously. A typical example is given of a recording during haemodialysis.

Estimation of non-cardiogenic pulmonary oedema using dual-frequency electrical impedance

The study investigates the effects of non-cardiogenic oedema, especially the accumulation of protein in extracellular fluid, on thoracic impedance and proposes a new method of oedema measurement based on an impedance ratio from a dual-frequency measurement. In vitro measurements in a cell containing an albumin-in-saline solution yield a resistance increase when the albumin concentration increases. Subsequently, 13 patients having acute respiratory failure are measured. The single-frequency Z0 measurements and the proposed impedance ratio are compared with extravascular lung water (EVLW) determined by the double indicator dilution method. The single-frequency measurement correlates poorly with EVLW (r=−0.24, p=0.56). In some patients, a total thoracic impedance increase is found with increasing EVLW. The correlation between the impedance ratio and EVLW is r=−0.79 (p<0.0005). The ratio decreases as EVLW increases. Thus, when oedema is measured using bio-impedance, cardiogenic and noncardiogenic oedema yield different results. It is well recognised that cardiogenic oedema decreases total thoracic impedance. In non-cardiogenic oedema, however, protein accumulation causes an impedance increase. The decrease in the impedance ratio as EVLW increases can be explained by the accumulation of albumin in the extracellular compartment.

Thoracic geometry and its relation to electrical current distribution: consequences for electrode placement in electrical impedance cardiography

In thoracic impedance cardiography (TIC) measurements the neck electrodes are often positioned at the basis of the neck, close to the neck-thorax transition. Theoretically, this neck-thorax transition will cause inhomogeneities in the current density and potential distribution. This was simulated using a 3D finite element method, solely representing the geometrical neck-thorax transition. The specific conductivity was 7 10−3 (Ωcm)−1 and the injected current was 1 mA. As expected, the model generated inhomogeneities in the current distribution at the neck-thorax transition, which reached as far as 5 cm into the neck and 20 cm into the thorax. These results are supported by in vivo measurements performed in 10 young male subjects, in which the position of the neck electrodes was varied. A two-way ANOVA revealed that the stroke volume of the lowest neck position was significantly different from the other positions. Small shifts in the position of the neck electrode resulted in large changes in impedance and stroke volume (127 to 82 ml for the Kubicek equation). To standardise the electrode position, the authors strongly recommend placement of the neck electrodes at least 6 cm above the clavicula.

The inaccuracy of Kubicek's one-cylinder model in thoracic impedance cardiography

The validity of a one- and a two-cylinder model, underlying thoracic impedance cardiography (TIC), was investigated by studying the length dependence of the impedance parameters Z/sub 0/, (dZ/dt)/sub min/, and stroke volume (SV). It can be shown that, within a one-cylinder model, all parameters are directly proportional to the length, whereas, if the volume conduction of the thorax and the neck are modeled separately, Z/sub 0/ and (dZ/dt)/sub min/ are expected to be linear dependent and SV will he nonlinear dependent upon the length. The expectations were compared to results from in vivo measurements. Two electrode arrays were studied, in which the caudal recording electrode position was varied; SV was calculated using Kubiceks equation. Except for small distances, the results showed a nearly linear relation between the parameters and the length. Regression analysis of the linear part revealed statistically significant intercepts (p<0.05). Neither the intercept nor the nonlinear part can be explained by a one-cylinder model, whereas a model consisting of two cylinders serially connected describes the experimental results accurately. Thus SV estimation based on a one-cylinder model is biased due to the invalid one-cylinder model. Corrections for the Kubicek-equation need to be developed in future research using this two-cylinder model.

The influence of extravascular lung water on cardiac output measurements using thoracic impedance cardiography

The purpose of this study was to investigate the influence of pulmonary oedema as measured with the double indicator dilution technique on the accuracy of cardiac output (CO) measurement using thoracic impedance cardiography (TIC) compared with thermodilution in thirteen sepsis patients. Differences in the Kubicek and Sramek-Bernstein equation with respect to pulmonary oedema were explored theoretically and experimentally. From a parallel two cylinder model a hypothesis can be derived that CO determined with the Kubicek equation is oedema independent, whereas CO determined using the Sramek-Bernstein equation is oedema dependent. Experimentally, CO determined using Kubiceks equation correlated better with thermodilution CO (r = 0.75) than CO determined with the Sramek-Bernstein equation (r = 0.25). The effect of oedema on the accuracy of TIC was investigated by comparing the differences in the CO of impedance and thermodilution to the extravascular lung water index. For the Kubicek equation the difference was not influenced by oedema (r = 0.04, p = 0.84), whereas for the Sramek-Bernstein equation the difference was affected by oedema (r = 0.39, p = 0.05). Thus, the effects of pulmonary oedema on the accuracy of TIC measurements can better be understood with the parallel cylinder model. Moreover, the Kubicek equation still holds when pulmonary oedema is present, in contrast to the Sramek-Bernstein equation.

Extra-cellular volume estimation by electrical impedance - phase measurement or curve fitting: a comparative study

In order to determine body fluid shifts between the intra- and extra-cellular spaces, multifrequency impedance measurement is performed. According to the Cole-Cole extrapolation, lumped values of intra- and extra-cellular conduction can be estimated which are commonly expressed in resistances Ri and Re respectively. For this purpose the magnitude and phase of the impedance under study are determined at a number of frequencies in the range between 5 kHz and 1 MHz. An approach to determine intra- and extra-cellular conduction on the basis of Bode analysis is presented in this article. On this basis, estimation of the ratio between intra- and extra-cellular conduction could be performed by phase measurement only, midrange in the bandwidth of interest. An important feature is that the relation between intra- and extra-cellular conduction can be continuously monitored by phase measurement and no curve fitting whatsoever is required. Based on a two frequency measurement determining Re at 4 kHz and phi(max) at 64 kHz it proved possible to estimate extra-cellular volume (ECV) more accurately compared with the estimation based on extrapolation according to the Cole-Cole model in 26 patients. Reference values of ECV were determined by sodium bromide. The results show a correlation of 0.90 with the reference method. The average error of ECV estimation was -3.6% (SD 8.4), whereas the Cole-Cole extrapolation showed an error of 13.2% (SD 9.5). An important feature of the proposed approach is that the relation between intra- and extra-cellular conduction can be continuously monitored by phase measurement and no curve fitting whatsoever is required.

A wideband high common mode rejection ratio amplifier and phase-locked loop demodulator for multifrequency impedance measurement

Design considerations and implementation of a multifrequency measuring channel for application in the field of bio-impedance measurement are discussed in this paper. The input amplifier has a differential configuration which is electrically isolated from the remaining circuits. Transformer coupling provides improved common mode rejection when compared to non-isolated input stages. The frequency characteristic of the section between input and demodulator is flat within ±0.1 dB between 4kHz and 1024 kHz. The synchronous demodulator is based on a wideband switched video amplifier. In contrast to commonly used lock-in techniques, the carrier for demodulation is recovered from the input signal by means of a phase-locked loop. This method ensures zero phase shift with respect to the input signal and improves the accuracy of measurement. The system has been developed primarily for thoracic impedance cardiography (TIC) but has also succesfully been applied in the field of total body bio-impedance analysis (BIA). At present an electrical impedance tomograph is under development based on the instrumentation described. Results regarding the measurement range and accuracy are given and some recordings of patient data are shown.

The influence of pulsatile flow on blood resistivity in impedance cardiography

The purpose of the study was to investigate the resistivity change over the cardiac cycle. This is important for the correct application of thoracic impedance cardiography (TIC). The ratio of spatial mean velocity over the vessel radius of the ascending and descending aorta of two female and eight male subjects (age ranging from 23 to 69 years) were measured in supine position using MRI. Based on Vissers (1989) equation the relative resistivity change was calculated. In all subjects the authors found a change of less than 15%, which is smaller than rigid tube experiments predicted. However, the peak resistivity change occurs at the same time as the peak in the impedance signal. Thus, the effects of resistivity changes on stroke volume calculation in TIC needs further investigation.

An electrically isolated balanced wideband current source: basic considerations and design.

At relatively high frequencies, the application of an alternating current through the body or a body segment results in electromagnetic stray fields which reduce the amount of current actually injected into the tissue under study. This radiation effect can be reduced by use of a symmetrical configuration current source. The symmetry of such an arrangement, however, depends on the stray capacitances of the source with respect to surrouding equipment. To minimise these effects, it is required that the source is electrically isolated from the surrounding equipment and the subject under study. In this manner stray capacitances with respect to elements of the current source are reduced. In such a configuration common mode voltages to the input amplifier of the measuring system are also reduced. The paper describes design considerations and the implementation of a wideband current source capable of injecting alternating current in the order of 300μARMS into biological tissue having impedances up to 1kΩ. Current stabilisation is obtained by means of a control circuit which measures the actual current passing through the tissue under study. Leakage currents arising from shielding and stray capacitances are compensated for. The usable frequency range is between 4 kHz and 1024 kHz and current stability is better than 0.2%. Through the use of a symmetrical, floating circuit a configuration is obtained which substantially reduces stray effects. The current source is connected to other circuits by means of two isolation ports: (1) a transformer coupling for the carrier frequency; and (2) an opto-coupler to transfer a phase reference signal obtained from current measurement. The current amplitude can be modulated by controlling the reference input to the control loop by means of a third auxiliary isolation port for transfer of the modulating signal.