Thoracic electrical bioimpedance (TEB) relates to changes in electrical conductivity of aortic blood flow and can be obtained from the thoracic surface to determine stroke volume and cardiac output (CO) . According to the theory of TEB, erythrocytes change their random orientation in the descending aorta during diastole to an alignment at the beginning of the systole. A refined algorithm to calculate CO by TEB, referred to as electrical velocimetry, has been introduced recently [2,3]. The measurement is based on the maximum rate of change in TEB as the ohmic equivalent of mean aortic blood flow acceleration according to the Bernstein–Osypka equation, and is referred to as electrical velocimetry as implemented in the Aesculon monitor (Osypka Medical GmbH, Berlin, Germany).
The aim of this study was to compare CO values obtained by the Aesculon monitor with those obtained by subxiphoidal Doppler flow measurements.
Patients and methods
With approval of the hospital ethical committee as well as written parental or patient consent, children from birth to 16 years of age, scheduled for diagnostic or interventional cardiac catheterization, were enrolled into this study. Cardiac catheterization was performed under general anaesthesia, including tracheal intubation and artificial ventilation. Detailed indication for cardiac catheterization and underlying cardiac diagnosis is given in Table 1. CO measurements were performed at the end of the cardiac catheterization procedure under steady state haemodynamics. Prior to the CO measurements, residual intracardiac or extracardiac shunts and left outflow tract obstruction were excluded by transoesophageal echocardiography or angiography or both.
Standard ECG surface electrodes were attached side-to-side in a vertical direction to the patients' left middle and lower neck, and to the lower thorax at the left mid-axillary line, at the level of the heart and xiphoid process. The electrodes were connected to the Aesculon monitor (Fig. 1). A correct signal quality was verified by visualization of the ECG and the impedance waveform. CO was calculated by transformation to the ohmic equivalent of the mean aortic blood flow acceleration and heart rate correction, as described previously in detail [2–4]. Electrical velocimetry CO was continuously displayed on the monitor and recorded as an average value over 10 valid cardiac cycles.
CO by subxiphoidal Doppler flow measurements was determined from transthoracic echocardiography. The transverse diameter of the descending aorta (DAo) from inner edge to inner edge in early systole was assessed at the level just below the diaphragm. An estimate of the cross-sectional area (CSA) of the descending aorta (
) was calculated. Flow velocity (VA) evaluation in the descending aorta was performed in the subxiphoidal view with appropriate transducers (5.0 and 3.5 MHz) on Sonos 7500 (Philips, Eindhoven, Netherlands). Optimal signal was recognized as waveform with minimal spectral dispersion and loudest auditory sound. A single consultant cardiologist performed all subxiphoidal Doppler measurements. Doppler-derived CO was calculated by using the formula:
For each patient, three consecutive CO values were measured within 5 min by subxiphoidal Doppler flow and contemporary by the Aesculon monitor. Both operators were blinded for the CO values achieved by either technique.
The average of three simultaneous electrical velocimetry and subxiphoidal Doppler CO measurements within 5 min was calculated per patient for each technique and used for subsequent analysis. Whitney U-test, linear regression analysis and Bland–Altman analysis  were performed to compare median CO values obtained by the Aesculon with those obtained by subxiphoidal Doppler flow measurement. Bias was calculated from 36 paired averages as the mean difference, and precision as two SDs of differences between paired values of the two methods. Data are presented as median (range). A P value of less than 0.05 was considered significant.
A total of 36 children undergoing cardiac catheterization (19 girls, 17 boys) aged 0.5–16.0 (median: 5.7) years were investigated. Baseline characteristics are given in Table 1. A total of 108 paired CO measurements by Aesculon monitor and subxiphoidal Doppler flow were attained. The coefficient of variation of CO measurements was 4.8% for the Aesculon monitor and 4.3% for the subxiphoidal Doppler flow, respectively. CO values obtained by Aesculon monitor [0.55–5.58 (2.62) l min−1] and subxiphoidal Doppler flow measurements [0.62–6.27 (3.05) l min−1] differed significantly between both methods (P = 0.04). Simple regression analysis revealed moderate correlation between CO values obtained from both techniques (r2 = 0.5544, P < 0.001) (Fig. 2). Bias for CO values between the Aesculon monitor and subxiphoidal Doppler flow measurements was 0.31 l min−1 with a precision of ±1.92 l min−1 for all data pairs (Fig. 3).
The present study compared CO measurements derived from the Aesculon monitor and subxiphoidal Doppler flow measurements in paediatric patients. The main finding was that the Aesculon monitor does not reliably reflect CO values when compared with subxiphoidal Doppler flow measurements.
A noninvasive monitoring of the CO in children is of important interest in the paediatric intensive care unit and in critically ill children during paediatric anaesthesia. Small vessel size, or complex cardiovascular abnormalities often preclude an assessment of haemodynamics by pulmonary artery catheter (PAC) or pulse contour analysis . Moreover, these techniques are invasive, central lines bear potential risks [7,8] and repeated fluid injection during PAC thermodilution technique may lead to fluid overload particularly in infants.
The Aesculon monitor is an easy to attach, noninvasive device combined with other monitoring features (blood pressure, heart rate, saturation, temperature). The device is small, compact, as well as easy to understand and to install. To date, CO measurements derived from the Aesculon monitor were compared with transoesophageal Doppler echocardiography , PAC thermodilution technique [10,11] and the Fick-oxygen principle , indicating clinically acceptable agreement in adults but not in children (Table 2). However, in a longitudinal study in piglets  comparing PAC thermodilution technique with the Aesculon monitor at different settings demonstrated a good agreement for continuous trend monitoring.
Traditionally, new CO monitors were evaluated using thermodilution technique by a PAC or using the Fick principle. Both techniques measure blood flow through the lung and have their limitations . The strength of the presented study is that the electrical velocimetry, a correlate of aortic blood flow, is compared with aortic Doppler flow measurements. Assessment of CO by Doppler flow has intensively been done from the oesophagus. A review from 25 published reports  has indicated good correlation between CO derived from oesophageal Doppler probes and the PAC thermodilution technique in adults. Studies in paediatric patients investigating the accuracy of oesophageal Doppler flow measurements are limited, but found clinically accurate estimates of CO [16,17].
Unfortunately, we were not able to demonstrate an acceptable agreement between CO values obtained by the Aesculon monitor and subxiphoidal Doppler flow measurement. Potential explanations for this disagreement are technical, patient-dependent or user-dependent factors. Assessment of the electrical velocimetry using the Aesculon monitor may be altered by left outflow tract obstructions. However, left outflow tract obstructions were excluded by transoesophageal echocardiography or angiography or both prior to the CO measurements in our study cohort. Further sources of patient-dependent errors are related to the principle of measuring CO by the electrical velocimetry. Inaccuracy may result from lung expansion caused by intermittent positive pressure ventilation, thoracic volume overload due to pleural effusion or pulmonary oedema, arrhythmias or other ECG abnormalities, as well as hyperdynamic sepsis. New approaches for optimized electrode position on the patient's surface less dependent on thoracic cavity and lung volume changes are under investigation.
The CO assessment by Doppler flow measurement itself may bear technical and patient-dependent sources of error, especially in the assessment of the correct aortic diameter. Small differences in diameter produce a large difference in CSA. Aortic diameters and changes in the aortic shape in patients with congenital heart disease may alter flow patterns and influence subxiphoidal Doppler flow measurements, which may be considered as a limitation of this study. The challenge to accurately measure the aortic diameter in adults due to low sonographic window quality did not account for our paediatric patient population. We encountered optimal sonographic windows in paralysed patients; Doppler signals were generally easy to obtain and to recognize. However, user-dependent factors may be taken into account. Whereas the Aesculon monitor is easy to apply and user-independent, the Doppler method is highly dependent on the user and his/her skill. To limit the impact of technical and user-dependent errors for the Doppler method, a single, well experienced cardiologist performed subxiphoidal measurements.
The Aesculon monitor allows continuous online measurement and trend analysis, whereas the Doppler method allows an intermittent measurement to assess CO. On the basis of our results and limitations mentioned above, the Aesculon monitor, as a beat-to-beat monitor, provides an estimate on the haemodynamic situation with a graphic display over time. However, it does not seem a valuable tool to accurately determine CO values in paediatric patients.
CO measurements derived by the Aesculon monitor do not reliably reflect subxiphoidal Doppler-derived CO measurements in children with congenital heart defects. However, this noninvasive technique applied by minimal user expertise even in resuscitation situations may be used for an objective assessment of haemodynamic trends, such as detection of deterioration and monitoring of the efficacy of interventions. These potential benefits may compensate for precise determination of CO values with invasive and more time-intense techniques.
The present study was supported by EMDO Foundation (Zurich), Theodor-Ida-Herzog-Egli Foundation (Zurich) and by the UBS Donation (University Children's Hospital Zurich).
There are no conflicts of interest.
1 Bernstein DP. A new stroke volume equation for thoracic bioimpedance: theory and rationale. Crit Care Med 1986; 14:904–909.
2 Bernstein DP, Osypka MJ. Apparatus and method for determining an approximation of the stroke volume and the cardiac output
of the heart. US Pat. 6,511,438 B2; 28 Jan 2003.
3 Bernstein DP, Lemmens HJM. Stroke volume equation for impedance cardiography. Med Biol Eng Comput 2005; 43:443–450.
4 Bernstein DP. Bernstein–Osypka stroke volume equation for impedance cardiography: citation correction. Intensive Care Med 2007; 33:923.
5 Bland J, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310.
6 Fakler U, Pauli Ch, Balling G, et al
. Cardiac index monitoring by pulse contour analysis and thermodilution after pediatric cardiac surgery. J Thorac Cardiovasc Surg 2007; 133:224–228.
7 Bussières JS. Iatrogenic pulmonary artery rupture. Curr Opin Anaesthesiol 2007; 20:48–53.
8 Smith-Wright DL, Green TP, Lock JE, et al
. Complications of vascular catheterization in critically ill children. Crit Care Med 1984; 12:1015–1017.
9 Schmidt C, Theilmeier G, Van Aken H, et al
. Comparison of electrical velocimetry and transoesophageal Doppler echocardiography for measuring stroke volume and cardiac output
. Br J Anaesth 2005; 95:603–610.
10 Suttner S, Schöllhorn T, Boldt J, et al
. Noninvasive assessment of cardiac output
using thoracic electrical bioimpedance in hemodynamically stable and unstable patients after cardiac surgery: a comparison with pulmonary artery thermodilution. Intensive Care Med 2006; 32:2053–2058.
11 Tomaske M, Knirsch W, Kretschmar O, et al
, Working Group on Noninvasive Haemodynamic Monitoring in Paediatrics. Cardiac output
measurement in children: comparison of Aesculon cardiac output
monitor and thermodilution. Br J Anaesth 2008; 100:517–520.
12 Norozi K, Beck C, Osthaus WA, et al
. Electrical velocimetry for measuring cardiac output
in children with congenital heart disease. Br J Anaesth 2008; 100:88–94.
13 Osthaus WA, Huber D, Beck C, et al
. Comparison of electrical velocimetry and transpulmonary thermodilution for measuring cardiac output
in piglets. Paediatr Anaesth 2007; 17:749–755.
14 Critchley LAH, Critchley JAJH. A meta-analysis of studies using bias and precision to compare cardiac output
measurement techniques. J Clin Monit 1999; 15:85–91.
15 Laupland KB, Bands CJ. Utility of oesophageal Doppler as a minimally invasive haemodynamic monitor: a review. Can J Anaesth 2002; 49:393–401.
16 Murdoch IA, Marsh MJ, Tibby SM, McLuckie A. Continuous haemodynamic monitoring in children: use of transoesophageal Doppler. Acta Paediatr 1995; 84:761–764.
17 Tibby SM, Hatherill M, Murdoch IA. Use of transoesophageal Doppler ultrasonography in ventilated paediatric patients: derivation of cardiac output
. Crit Care Med 2000; 28:2045–2050.