Electrical Velocimetry Research Articles

Impedance cardiography (electrical velocimetry) and transthoracic echocardiography for non-invasive cardiac output monitoring in pediatric intensive care patients: a prospective single-center observational study.

IntroductionElectrical velocimetry (EV) is a type of impedance cardiography, and is a non-invasive and continuously applicable method of cardiac output monitoring. Transthoracic echocardiography (TTE) is non-invasive but discontinuous.MethodsWe compared EV with TTE in pediatric intensive care patients in a prospective single-center observational study. Simultaneous, coupled, left ventricular stroke volume measurements were performed by EV using an Aesculon® monitor and TTE (either via trans-aortic valve flow velocity time integral [EVVTI], or via M-mode [EVMM]). H0: bias was less than 10% and the mean percentage error (MPE) was less than 30% in Bland–Altman analysis between EV and TTE. If appropriate, data were logarithmically transformed prior to Bland–Altman analysis.ResultsA total of 72 patients (age: 2 days to 17 years; weight: 0.8 to 86 kg) were analyzed. Patients were divided into subgroups: organ transplantation (OTX, n =28), sepsis or organ failure (SEPSIS, n =16), neurological patients (NEURO, n =9), and preterm infants (PREM, n =26); Bias/MPE for EVVTI was 7.81%/26.16%. In the EVVTI subgroup analysis for OTX, NEURO, and SEPSIS, bias and MPE were within the limits of H0, whereas the PREM subgroup had a bias/MPE of 39.00%/46.27%. Bias/MPE for EVMM was 8.07%/37.26% where the OTX and NEURO subgroups were within the range of H0, but the PREM and SEPSIS subgroups were outside the range. Mechanical ventilation, non-invasive continuous positive airway pressure ventilation, body weight, and secondary abdominal closure were factors that significantly affected comparison of the methods.ConclusionsThis study shows that EV is comparable with aortic flow-based TTE for pediatric patients.

Propofol effect on cerebral oxygenation in children with congenital heart disease.

Propofol is a short-acting, intravenously administered hypnotic agent which is used in procedural sedation in children. Propofol is known to decrease systemic vascular resistance, arterial blood pressure and can lead to desaturations and decreased systemic perfusion in children with cardiac shunting. This may result in a reduction in cerebral blood flow and oxygenation. Near-infrared spectroscopy (NIRS) can monitor cerebral tissue oxygenation in the frontal neocortex. The objective of our study was to measure the changes in cerebral oxygen and blood supply after Propofol infusion in children with congenital heart disease. Propofol infusion may reduce cerebral oxygenation in children with congenital heart disease. The study group consisted of 32 children (f:m=18:14), with median age of 49 (5-112) months and median weight of 15 (5-34) kg. We performed NIRS derived continuous measurement of cerebral oxygenation and cardiac output using Electrical velocimetry for 5min before and after sedation with Propofol (1-2mg/kg i.v.) for cardiac catheterization. Simultaneously, non-invasive arterial blood pressure and transcutaneous oxygen saturation were measured. Propofol sedation led to a significant decrease in mean arterial pressure (79±16 vs. 67±12mmHg) (p=0.01) and cardiac index (3.2±0.8 vs. 2.9±0.6ml/min/m(2)) (p=0.03). In contrast, cerebral tissue oxygenation index, increased significantly from 57±11 to 59±10% (p<0.05). Sedation with Propofol increased cerebral tissue oxygenation despite a decrease in cardiac index and arterial blood pressure. This may be caused by a decreased oxygen consumption of the sedated brain with intact cerebral auto-regulation.

Non-Invasive Cardiac Output Measurement in Low and Very Low Birth Weight: A Method Comparison

Background: Cardiac output (CO) measurement in low (LBW) and very low (VLBW) birth weight infants is difficult. Hitherto, sporadical transthoracic echocardiography (TTE) is the only non-invasive measurement method. Electrical velocimetry (EV) has been evaluated as an alternative in normal weight newborns.Objectives: The study was designed to evaluate if EV could be interchangeable with TTE even in LBW and VLBW infants.Methods: In 28 (17 LBW, 11 VLBW) pre-mature newborns, n = 228 simultaneous TTE (trans-aortic Doppler), and EV measurements (134 LBW, 94 VLBW) of stroke volume (SV) and heart rate (HR) were performed, thereof calculating body weight indexed SV (=SV*) and CO (=CO*) for all patients and the subgroups. Method comparison was performed by Bland–Altman plot, method precision expressed by calculation of the coefficient of variation (CV).Results: Mean CO* in all patients was 256.4 ± 44.8 (TTE) and 265.3 ± 48.8 (EV) ml/kg/min. Bias and precision were clinically acceptable, limits of agreement within the 30% criterion for method interchangeability (17). According to their different anatomic dimensions and pathophysiology, there were significant differences of SV(*), HR, and CO* for LBW and VLBW infants as well for inotropic treatment and ventilation mode.Conclusion: Extending recent publications on EV/TTE comparison in newborns, this study suggests that EV is also applicable in LWB/VLBW infants as a safe and easy to handle method for continuous CO monitoring in the NICU and PCICU.

Comparison of the Hemodynamic Effects of the Induction Agents Ketamine, Etomidate and Sevoflurane Using the Model of Electrical Velocimetry Based Cardiac Output Monitoring in Pediatric Cardiac Surgical Patients

Objective: To compare the haemodynamic effects of the induction agents ketamine, etomidate and sevoflurane using the model of electrical velocimetry based cardiac output monitoring in paediatric cardiac surgical patients. Design: Prospective randomized study. Setting: Tertiary care hospital. Participants: 60 children < 2 years age undergoing cardiac surgery. Interventions: The patients were randomized into 3 equal groups to receive 1.5-2.5 mg/kg iv ketamine (group K), 0.2-0.3 mg/kg iv etomidate (group E) or upto 8% sevoflurane (group S) as the induction agent. Hemodynamic parameters were noted before and after induction of anaesthesia utilizing a noninvasive cardiac monitor based on the model of electrical velocimetry. Measurements and Main Results: The demographic characteristics of the patients were similar in the three groups. The HR decreased in all groups, least in group E (P ≤ 0.01) but the MAP decreased only in group S (P ≤ 0.001). In group S, the stroke volume improved from 9 ± 3.2 ml to 10 ± 3.2 ml (P ≤ 0.05) and the stroke volume variation decreased from 25% ± 6.4% to 13% ± 6.2% (P ≤ 0.001). The stroke index and systemic arterial saturation improved in all groups (P ≤ 0.01). The cardiac index and index of contractility were unchanged. The transthoracic fluid content reduced in groups E and S, but did not change in group K (P ≤ 0.05). Conclusions: Etomidate appeared to provide the most stable conditions for induction of anesthesia in children undergoing cardiac surgery, followed by ketamine and sevoflurane.

Atrioventricular delay programming in cardiac resynchronization therapy devices: fixed or adaptive?: A randomized monocenter trial

IntroductionCardiac resynchronization therapy devices are routinely programmed on fixed atrioventricular delays (AVD) under resting conditions based on echocardiographic techniques. Whether this AVD also ensures optimal exercise hemodynamics, is unclear. MethodsIn order to compare fixed-AVD with rate-adaptive AVD, 100 patients with cardiac resynchronization therapy systems and sinus rhythm were randomized to fixed-AVD or adaptive-AVD. The patients then underwent bicycle ergometry with noninvasive hemodynamic monitoring. At rest and at peak exercise, stroke volume, cardiac output, and cardiac index were determined using “electrical velocimetry.” ResultsThere were no significant differences in clinical characteristics and baseline hemodynamic parameters between fixed or adaptive AVD. In patients randomized to adaptive AVD, a trend towards higher stroke volume, cardiac output, and cardiac index at peak exercise was encountered. ConclusionsBased on the trend towards better exercise hemodynamics demonstrated by this pilot study, a randomized follow-up study with clinical end points appears to be justified to clarify this issue.

Electrical velocimetry as a tool for measuring cardiac output in small infants after heart surgery

Cardiac output (CO), the product of stroke volume (SV) and heart rate, is essential to guarantee organ perfusion, especially in the intensive care setting. As invasive measurement of CO bears the risk of complications there is a need for non-invasive alternatives. We investigated if electrical velocimetry (EV) and transthoracic Doppler (Doppler-TTE) are interchangeable for the non-invasive measurement of SV and able to reflect the post-surgical SV/CO trend. Comparison of SV measurements by EV and Doppler-TTE was performed in 24 newborns after switch operation (n = 240 measurements). Three subgroups of measurements (=periods) were created according to the patients' status in the course of post-surgical CO recovery. Bland-Altman analysis found acceptable bias and limits of agreement for the interchangeability of the two methods. Mean overall SV was 3.7 ml with a mean overall bias of 0.28 ml (=7.6 %). The mean percentage error of 29 % was acceptable according to the method of Critchley and Critchley. Overall precision expressed by the coefficient of variation (CV) was 6.6 % for SV(TTE) and 4.4 % for SV(EV). SV(TTE) and SV(EV) medians in the three periods were significantly different and documented the post-surgical CO trend. EV and Doppler-TTE are interchangeable for estimating SV. EV has the advantages of easy handling and allows continuous measurement.

Stroke volume obtained by electrical interrogation of the brachial artery: transbrachial electrical bioimpedance velocimetry

The goal of this study is to measure left ventricular stroke volume (SV) from the brachial artery (BA) using electrical bioimpedance. Doppler-derived SV was used for comparison. Twenty-nine healthy adults were recruited for study. Doppler echocardiographic-derived SV was obtained from the product of distal left ventricular outflow tract cross-sectional area and systolic velocity integral. SV from the BA was obtained by transbrachial electrical bioimpedance velocimetry (TBEV). Application of a current field across the left brachium was effected by injection of a constant magnitude, high frequency, low amperage, alternating current. Therein, a static voltage (U0) and pulsatile voltage change (ΔU(t)) were measured and converted to their corresponding impedances, Z0 and ΔZ(t). TBEV-derived SV was obtained by multiplying a square root value of the normalized, acceleration-based, peak first time derivative of ΔZ(t) by a volume conductor and systolic flow time. Inter-method agreement was determined by the Bland–Altman method. To assess the contribution of blood resistivity variations to ΔZ(t), BA diameters were measured at end-diastole and peak systolic expansion. Results indicate that since the BA demonstrates parabolic, laminar flow, with minimal diameter changes, blood resistivity variations are likely responsible for the derived impedance changes. Bland–Altman analysis shows that SV is obtainable by TBEV from healthy humans at rest.

Left ventricular stroke volume measurement by impedance cardiography correlates with echocardiography in neonates

The aim of this study was to validate impedance cardiography (electrical velocimetry (EV)) as a continuous noninvasive cardiac output monitoring in neonates and infants. As the reference method, discontinuous transthoracic echocardiography (TTE) was used.

Continuous non-invasive cardiac output measurements in the neonate by electrical velocimetry: a comparison with echocardiography

ObjectiveElectrical velocimetry (EV) is a non-invasive method of continuous left cardiac output monitoring based on measurement of thoracic electrical bioimpedance. The objective was to validate EV by investigating the agreement...

A comparative evaluation of electrical velocimetry and inert gas rebreathing for the non-invasive assessment of cardiac output

When assessing the function of the cardiovascular system, cardiac output (CO) is a substantial parameter. For its determination, numerous non-invasive techniques have been proposed in the recent years including inert gas rebreathing (IGR) and impedance cardiography (ICG). The aim of our study was to evaluate whether a novel ICG algorithm (electrical velocimetry) and IGR can be used interchangeably in the clinical setting. A total of 120 consecutive stable patients were included resulting in two pairs of repeated non-invasive cardiac output measurements. The mean CO was 5.0±1.2l/min (range 2.6-8.6l/min) using IGR and 4.4±1.1l/min (1.7-7.4l/min) using ICG, respectively. Bland-Altman analysis revealed an acceptable agreement with a mean bias of 0.6±1.2l/min. We found a high reproducibility with a mean bias of 0.2±0.7l/min for IGR and 0.0±0.3l/min for ICG (p<0.001), respectively. There was a statistically significant difference for unphysiological circulatory conditions represented by values of 2.6-4.1l/min and 5.6-8.6l/min. Both non-invasive techniques are associated with low operating costs and require only a few expendable items for the rapid determination of cardiac function. We found an acceptable agreement between IGR and ICG as well as a high reproducibility, which was statistically significant higher for ICG. For cardiac output states exceeding the physiological range, we found a statistically significant difference. Consequently, values of cardiac function determined by either method should not be used interchangeably in the clinical setting.

Non-invasive cardiac output and oxygen delivery measurement in an infant with critical anemia

To assess the combination of a non-invasive blood oxygen content (CaO(2)) monitor and a non-invasive cardiac output (CO) monitor to continuously measure oxygen delivery (DO(2); DO(2) = CaO(2) × CO). DO(2) was assessed during blood transfusions in an infant with acute hemolytic anemia following admission (~48 h). CaO(2) was measured by Pulse Co-Oximetry, which also provides estimates of hemoglobin (Hgb) concentration and percent oxygen saturation. CO was measured by Electrical Velocimetry, which also provides an estimate of stroke volume (SV). Lactate levels, an indirect measure of adequate DO(2), were assessed during the initial 8 h following admission. Incremental blood transfusions during the first 36 h increased Hgb from 2.7 to 9.5 g/dL during which time heart rate (HR) normalized from 156 to 115 beats/min. Lactate levels decreased from 20 to 0.8 mmol/L in the first 7 h. Non-invasive Hgb and CaO(2) measurements were well correlated with invasive Hgb and CaO(2) measures (r (2) = 0.88; P = 0.019; r (2) = 0.86; P = 0.0074, respectively). CO decreased from 2.47 ± 0.06 to 1.28 ± 0.02 L/min and SV decreased from 15.9 ± 0.4 to 11.1 ± 0.2 mL/beat. Mean arterial blood pressure was stable throughout the admission with systemic vascular resistance increasing from 407.6 ± 15.2 to 887.7 ± 30.1 dynes-s/cm(5). DO(2) was estimated to increase from 120.2 ± 18.9 to 182.4 ± 5.6 mL O(2)/min. Non-invasive continuous CO and CaO(2) monitors are shown in this single case to provide continuous DO(2) measurement. The ability to assess DO(2) may improve hemodynamic monitoring during goal directed therapies.

Cardiac output measured by electrical velocimetry in the CT suite correlates with coronary artery enhancement: A feasibility study

Cardiac output (CO) is inversely related to vascular contrast medium (CM) enhancement during computed tomography (CT). Impedance cardiography with a new technique, electrical velocimetry (EV), may create opportunities to measure CO pre-examination for adaptation of CM injection parameters. To relate CO(EV) measured by radiology staff to aortic attenuation as a measure of coronary artery attenuation during CT coronary angiography (CTCA), and to formulate a tentative statistical model to adapt CM injection parameters to CO. CO(EV) was measured immediately before 100 kVp CTCA (64-multirow detector) in 27 patients with presumed coronary artery disease. For CTCA, 260 mg I/kg (maximum dosage weight: 80/90 kg for women/men) was injected intravenously during 12 s. Simple linear regression analysis was performed to explore the correlation between aortic attenuation (Hounsfield units, HU) and body weight, the influence of CO(EV) on aortic attenuation adjusted to injected CM dose rate (HU per mg I/kg/s), and to establish a tentative formula on how to adapt CM injection parameters to CO(EV) and desired aortic attenuation. The correlation between aortic attenuation and body weight was weak and non-significant (r=-0.14 after outlier exclusion). A significant negative correlation (r=-0.63) was found between aortic attenuation adjusted to injected CM dose rate (HU per mg I/kg/s) and CO(EV). The resulting formula, CM dose rate=CO(EV)×(aortic attenuation-240)/55, made it possible to calculate CM volumes and injection rates at various COs and, for example, the present mean aortic attenuation (438 HU), injection time (12 s), CM concentration (320 mg I/ml), and a certain body weight. EV makes it possible to measure CO in the CT suite before vascular examinations. Hence, CM doses may be decreased in low CO states to reduce the risk of CM-induced nephropathy without jeopardizing diagnostic quality and may be increased in high CO states to avoid poor enhancement.

Physical stress testing of bovine jugular veins using magnetic resonance imaging, echocardiography and electrical velocimetry☆☆☆

Bovine jugular veins (BJVs) (Contegra) are valve-bearing pulmonary artery substitutes. Their valves have higher profiles than human pulmonary valves; this might result in less optimal performance. Therefore, we investigated the impact of stress and undersizing on conduit performance with ergometry, echocardiography and magnetic resonance imaging (MRI). Between April 2007 and June 2008, 20 BJV recipients (age 7.9-19.6 years) underwent spiroergometry and subsequent echocardiography; after due rest, ergometry was repeated and followed by MRI during recovery. A year later, exams were repeated. Data was evaluated as follows: comparison of stress related maximal individual valve performance changes (magnetic resonance: exercise induced average stroke volume changes by 61+/-49%; mean insufficiency increased by 2% in patients with <1% rest insufficiency and by 8% after rest insufficiency of >10%; the average rest gradient of 24+/-11 mmHg rose to 40+/-20 mmHg), and stratification of pooled observations by regurgitation fraction, insufficiency grades and z-values (insufficiency rose with increasing heart rate and decreasing stroke volume; undersizing increased gradients during recovery by 7+/-0.7 mmHg/z-value). Contegras high-profile valves tolerate stress without performance drop. Stress induced changes of insufficiency and gradient were clinically not significant, but sufficient to distort examination results; therefore, constant examination conditions are indispensable for a correct follow-up.

Physical stress testing of bovine jugular veins in pulmonary position using MRI, echocardiography and electrical velocimetry

Physical stress testing of bovine jugular veins in pulmonary position using MRI, echocardiography and electrical velocimetry

Comparison of electrical velocimetry and transthoracic thermodilution technique for cardiac output assessment in critically ill patients

The results of studies validating the assessment of cardiac output by pulmonary artery thermodilution and a modified algorithm using thoracic bioimpedance/electrical velocimetry in patients undergoing elective cardiac surgery are conflicting. The present observational study was designed to detect whether electrical velocimetry measurements are comparable to transthoracic thermodilution in septic patients after major general surgery. Cardiac output was assessed simultaneously by thoracic bioimpedance measurement/electrical velocimetry and transthoracic thermodilution technique (PiCCO) in 30 patients with severe systemic inflammatory response syndrome or sepsis with haemodynamic instability being treated in the surgical intensive care unit of an university hospital. Thirty simultaneous measurements were taken with both methods. The Bland-Altman analysis of agreement revealed a bias of -0.3 l min(-1) with a precision of +/-1.9 l min(-1) and wide limits of agreement (-4.1-3.5 l min(-1)). The percentage error was 54%. There was poor agreement between the values of cardiac output estimation by transthoracic thermodilution and those by electrical velocimetry. Electrical velocimetry could not replace invasive monitoring in this trial.

Evaluation of the Aesculon cardiac output monitor by subxiphoidal Doppler flow measurement in children with congenital heart defects

To evaluate the noninvasive electrical velocimetry (Aesculon) monitor for cardiac output (CO) by subxiphoidal Doppler flow measurement in children. CO was determined at the end of diagnostic or interventional cardiac catheterization for congenital heart defects. Standard ECG surface electrodes were attached in a vertical direction to the patients' left middle and lower neck, and lower thorax at the level of the heart and xiphoid process. Aesculon CO data were compared with a simultaneously measured CO by the subxiphoidal Doppler flow measurement technique. For each patient, measurements were repeated three times within 5 min. Whitney U-test, simple regression and Bland-Altman analysis were performed to compare CO values obtained by the two techniques. Data are given as range (median). A total of 36 children aged 5.7 (0.5-16.0) years were investigated. CO values obtained by Aesculon monitor [0.55-5.58 (2.62) l min] and subxiphoidal Doppler flow measurements [0.62-6.27 (3.05) l min] differed significantly between both methods (P = 0.04). Simple regression analysis revealed moderate correlation between CO values obtained from the two techniques (r = 0.5544, P < 0.001). Bias between the two methods was 0.31 l min with a precision of 1.92 l min. We conclude that electrical velocimetry using the Aesculon monitor does not reliably reflect absolute CO values as compared with subxiphoidal Doppler flow measurement.

The basic pacing rate in CRT patients: the higher the better?

To maximize the hemodynamic benefit of cardiac resynchronization therapy (CRT), echocardiographic AV interval optimization is routinely performed, complemented by VV interval optimization especially in non-responders. Programming of the basic pacing rate, however, is largely empirical in these patients. Therefore, the present study aimed to systematically evaluate the impact of basic pacing rate on hemodynamic parameters in CRT patients with sinus bradycardia. We included 70 consecutive patients with moderate to severe heart failure, LV ejection fraction <or=35%, left bundle branch block or a QRS duration >120 ms combined with echocardiographic evidence of ventricular dyssynchrony. All patients were on optimal heart failure medication, with CRT-ICD devices implanted at least 6 months before inclusion into the study. All patients were in sinus rhythm with a spontaneous heart rate <40 bpm. In all patients, cardiac output (CO) and stroke volume (SV) were determined using electrical velocimetry (EV) (Aesculon, Osypka Medical, Berlin, Germany). EV provides a new algorithm to calculate CO based on variations in thoracic electrical bioimpedance, which has been recently validated. Hemodynamic measurements were performed at four different pacing rates ranging from 40 to 70 bpm. A stepwise increase in CO was encountered with increasing heart rates, reaching statistical significance when comparing 70 with 40 bpm. SV remained unchanged throughout all pacing rates. In the range between 40 and 70 bpm, an increase in basic pacing rate enhances CO without reducing SV. According to this pilot study, a basic pacing rate between 60 and 70 bpm would appear reasonable.

Therapeutic Optimization of Atrioventricular Delay in Cardiosurgical ICU Patients by Noninvasive Cardiac Output Measurements versus Pulse Contour Analysis

Optimizing atrioventricular (AV) delay improves cardiac output and postoperative outcome. Impedance cardiography (ICG) is a non-invasive method for CO measurement. This study evaluates the ability of two ICG methods to determine the optimal AV delay (OAVD) and to compare ICG with invasive PICCO measurements. In 14 cardiosurgical ICU patients (age 70.4 +/- 12.0 yrs) with temporary pacing wires, OAVD was determined by pulse contour analysis (PICCO) and ICG (conventional ICG [CI] and electrical velocimetry [EV] ICG monitors). Cardiac output (CO) and stroke volume (SV) were measured during DDD pacing with AVD varying from 70 to 270 ms in 20-ms increments. Measured OAV showed a linear correlation between PICCO and ICG: CI (r = 0.82, P < 0.0002) and EV (r = 0.84, P < 0.0002). The mean OAVD deviation between PICCO and ICG was 15.7 +/- 21.0 ms (CI) and 17.1 +/- 20.5 ms (EV). Hemodynamic parameters (SV increase OAVD against worst case) improved significantly (+ 11.7 +/- 7.2 %, P < 0.0001). Inappropriate selection of AVD can compromise the hemodynamic situation of cardiosurgical patients. As it is totally noninvasive, ICG is a reliable and effective tool for tailoring AVD. Both systems (CI and EV) offer valid OAV determination.

Cardiac output measurement in children: comparison of Aesculon® cardiac output monitor and thermodilution

BackgroundWe compared cardiac output (CO) measurements by the non-invasive electrical velocimetry (Aesculon®) monitor with the pulmonary artery catheter (PAC) thermodilution method in children. MethodsCO values using the Aesculon® monitor and PAC thermodilution were simultaneously recorded during cardiac catheterization in children. Measurements were performed under general anaesthesia. To compare, three consecutive measurements for each patient within 3 min were obtained. The means of the three values were compared using simple regression and Bland–Altman analysis. Data were presented as mean (sd). A mean percentage of <30% was defined to indicate clinical useful reliability of the Aesculon® monitor. ResultsA total of 50 patients with a median (range) age of 7.5 (0.5–16.5) yr were enrolled in the study. Mean CO values were 3.7 (1.5) litre min−1 (PAC thermodilution) and 3.1 (1.7) litre min−1 (Aesculon® monitor). Analysis for CO measurement showed a good correlation between the two methods (r=0.894; P<0.0001). The bias between the two methods was 0.66 litre min−1 with a precision of 1.49 litre min−1. The mean percentage error for CO measurements was 48.9% for the Aesculon® monitor when compared with PAC thermodilution. ConclusionsElectrical velocimetry using the Aesculon® monitor did not provide reliable CO values when compared with PAC thermodilution. Whether the Aesculon® monitor can be used as a CO trend monitor has to be assessed by further investigations in patients with changing haemodynamics.

Electrical velocimetry for measuring cardiac output in children with congenital heart disease

The purpose of this study was to evaluate the agreement of cardiac output measurements obtained by electrical velocimetry (CO(EV)) and those that derived from the direct Fick-oxygen principle (CO(F)) in infants and children with congenital heart defects. Simultaneous measurements of CO(EV) and CO(F) were compared in 32 paediatric patients, aged 11 days to 17.8 yr, undergoing diagnostic right and left heart catheterization. For non-invasive measurements of cardiac output by electrical velocimetry, which is a variation of impedance cardiography, standard surface electrodes were applied to the left side of the neck and the left side of the thorax at the level of the xiphoid process. Cardiac output determined using direct Fick-oxygen principle was calculated by direct measurement of oxygen consumption (VO2) and invasive determination of the arterio-venous oxygen content difference. An excellent correlation (r=0.97) was found between CO(EV) and CO(F) (P<0.001). The slope of the regression equation [0.96 (SD 0.04)] was not significantly different from the line of identity. The bias between the two methods (CO(EV)-CO(F)) was 0.01 litre min(-1) and the limits of agreement, defined as the bias (2 SD), were -0.47 and +0.45 litre min(-1). CO(EV) demonstrates acceptable agreement with data derived from CO(F) in infants and children with congenital heart disease. The new technique is simple, completely non-invasive, and provides beat-to-beat estimation of CO.