Department of Anaesthesia and Intensive Care, The Chinese University of Hong Kong, Shatin, Hong Kong, China
* Corresponding author. Department of Anaesthesia and Intensive Care, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong, China. E-mail: hcritchley{at}cuhk.edu.hk
Accepted for publication May 31, 2005.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. In eight anaesthetized and ventilated dogs (weight 1522 kg) a high-precision flow probe was placed on the ascending aorta via a left thoracotomy incision and the direct cardiac output (COFP) was measured. Simultaneous COIC measurements were made using a RheoCardioMonitor (ACMA, Singapore). Lung oedema was induced by intravenous oleic acid 0.1 mg kg1. Lung fluid was assessed by the decrease in basal thoracic impedance (Zb). Percentage errors between the two methods (COICCOFP) were calculated and compared as Zb decreased at 1 intervals.
Results. During the experiment mean Zb decreased from 35.9 (SD 5.2) to 27.8 (6.5) (P=0.0037). This occurred over a period of 225 (range 112338) min and Zb decreased by 1
every 51 (2268) min. The presence of excessive lung fluid was confirmed at post-mortem. Before lung oedema was induced, COIC was 1.5 (0.6) litre min1 and the corresponding value of COFP was1.5 (0.7) litre min1 (data from eight dogs). As Zb decreased, and lung fluid accumulated, the error between COIC and COFP widened (P<0.0001, ANOVA for repeated measures). Eventually, COIC decreased to 0.7 (0.3) litre min1 and the corresponding value of COFP was1.2 (0.3) litre min1 (
Zb=5
, data from six dogs). Mean arterial pressure, central venous pressure and systemic vascular resistance were kept constant.
Conclusion. The presence of lung fluid attenuates COIC measurements with respect to COFP.
Keywords: lung, oedema ; measurement techniques, cardiac output ; model, dog
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In the present study we investigated the effects of lung oedema on impedance cardiography measurements. Both impedance and aortic flow probe cardiac output were measured in anaesthetized dogs. Lung oedema was induced by oleic acid. Bias and percentage errors between the two modes of cardiac output measurement were compared as lung oedema developed. Basal transthoracic impedance (Zb) was used to assess the degree of lung fluid accumulation.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Placement of flow probe
A thoracotomy was performed at the left fourth intercostal space. The pericardium was incised longitudinally to expose the aortic root. The ascending aorta was separated from the pulmonary artery for 23 cm by blunt dissection using the finger. Free fat surrounding the aorta was carefully removed. A snug-fitting flow probe, either 16 or 20 mm A-series ultrasonic probe (Transonic Systems Inc., Ithaca, NY, USA), was placed around the ascending aorta and ultrasonic gel was applied. The probe size was chosen such that it neither compressed the vessel wall nor was loose enough to allow kinking. The probe cable was brought out of the thorax posteriorly. The pericardium was closed with sutures, a chest drain to an underwater seal was inserted, the collapsed lung was re-expanded and the chest wall was closed with sutures.
Blood flow and pressure measurements
Aortic blood flow was measured by the flow probe which used a high-precision four-crystal array. It was connected to a T106 single-channel flowmeter (Transonic Systems Inc., Ithaca, NY, USA) that also processed the transduced arterial blood pressure wave. CVP was measured intermittently using the same transducer. The arterial pressure system was kept patent with a 10 ml h1 saline infusion. The data were collected on a laptop computer and displayed using the software WinDaq (DataQ Instruments, OH, USA).
Impedance measurements
The impedance method of cardiac output measurement involves detecting the transthoracic impedance to a high-frequency (50100 MHz) low-amperage (15 mA) electric current passed between the neck and lower thorax. The impedance waveform of aortic blood flow during each cardiac cycle is thus detected and impedance variables are measured from the waveform, notably the initial upslope (dZ/dtmax) which coincides with aortic blood flow and the left ventricular ejection time (LVET). Stroke volume and cardiac output are derived from these variables using the Kubicek equation.1
In the present study, cardiac output (COIC) and Zb were measured using a RheoCardioMonitor (RheoCardioTechnologies, Singapore). This impedance system consists of six leads which are connected to the subject, a cardiograph module, a visual display unit and a dedicated computer that analyses the signal and stores data. An alternating current (100 kHz, 2 mA) was applied to the thorax of the dogs by two electrodes placed on the head and lower limb. The impedance to this current was detected by two pairs of opposing electrodes placed laterally on the mid-neck and the lower thorax at the level of the diaphragm. Normally surface electrodes are used, but we used subcutaneous needle electrodes to improve skin contact.
Stroke volume (SV) was calculated by the RheoCardioMonitor using a modified Kubicek equation:16
![]() |
Preparation of oleic acid
Oleic acid (Sigma, St Louis, MO, USA) was stored as frozen oil at 20°C. The frozen oil was weighed and dissolved by vortexing (gentle shaking) for 20 s in 100% ethanol to produce a 10 mg ml1 solution.
Induction of pulmonary oedema
After baseline measurements, oleic acid 0.1 mg kg1 was slowly injected (over 3 min) into the right atrium via the central line to induce pulmonary oedema. Lung oedema developed over 12 h, although the rate was variable.
Experimental protocol
Haemodynamic data, including cardiac output measurements comparing the impedance cardiography with the flow probe, were recorded as lung oedema developed. The dogs were kept anaesthetized throughout the experiment. The haemodynamic status was kept as constant as practical throughout the experiment by maintaining a constant CVP of 810 mmHg with i.v. boluses of normal saline (100 ml). The use of drugs that maintain blood pressure and cardiac output were avoided. After several hours the degree of lung oedema became so severe that the dogs could no longer tolerate the circulatory effects and died, which concluded the experiment.
Data collection and statistical analysis
Fluid intake and output were recorded. The volume of saline given over the course of the experiment was divided by the duration to give the hourly fluid intake. The dog's bladder was catheterized and urine collected in a beaker. The urine collected over the course of the experiment was measured and divided by the duration to yield the hourly fluid output.
The following impedance and haemodynamic variables were recorded at 1 min intervals: COIC, Zb, flow probe cardiac output (COFP), mean arterial pressure (MAP) and CVP. The systemic vascular resistance (SVR) was calculated from the equation
![]() |
|
![]() |
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The average duration of data collection after injection of oleic acid and the development of lung oedema was 225 (112338) min. Zb decreased by 36 during the experiment, i.e. a decrease of 1
every 51 (2268) min. The absolute value of Zb decreased from 35.9 (5.2) to 27.8 (6.5)
, i.e. a decrease of 22% (P=0.0037).
The haemodynamic status of each dog was kept reasonably constant by giving i.v. boluses of fluids during the experiment. There was little change in COFP, MAP, CVP and SVR until the dog's condition started to deteriorate and Zb had decreased by 46 (Table 1 and Fig. 1). However, COIC decreased with respect to COFP as Zb decreased (P<0.0001). The effects on bias are shown for each dog experiment in Figure 1, where Zb and bias are seen to decrease. The wide variation in the value of bias between sequential bias measurements, indicating a lack of reproducibility, should be noted. The effect of changes in COFP (upper plots) on bias measurements, particularly in dog 4, should also be noted. The increasing negative percentage error between the two cardiac output measurements as Zb decreases is summarized for all eight experiments in the box plot in Figure 2.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
A number of explanations have been proposed for the repeatedly poor results found in critically ill patients. Over 30 years ago, Van de Water and colleagues11 suggested that excessive lung fluid has a detrimental effect on dZ/dtmax (the flow factor in the Kubicek equation). Bernstein3 was the first to report that the BoMed, a predecessor of the current generation of impedance cardiographs, underestimated impedance in patients with sepsis, which he attributed to impedance currents traversing the luxuriantly perfused skin and thoracic musculature which, being highly conductive, diverted them away from the aorta. Later, Shoemaker and colleagues12 suggested that the excessive lung fluid, or pulmonary oedema, affected impedance measurements and recommended a lower limit of 20 for Zb, below which the use of impedance cardiography is contraindicated. We hypothesized that the presence of lung fluid resulted in a short circuit of the heart and major blood vessels, including the aorta which is the origin of the impedance signal.13 14 These blood-filled organs are highly conductive compared with normal lung tissue which consists of mainly air-filled alveoli. Hence, during impedance cardiography, most of the impedance-detecting current normally flows through the blood-filled structures. However, the presence of lung fluid causes the mediastinum to be partly bypassed and consequently the impedance signal is attenuated, resulting in cardiac output measurements which are lower than expected.14 15 However, the evidence for this explanation is mainly anecdotal and it had never been rigorously tested in vivo using experimental models of lung oedema. Thus the present study provides in vivo evidence of the wet-lung effect.
Although animal models do not fully replicate the clinical situation, they allow data to be collected and pathological conditions to be produced that could not be achieved ethically in human studies. Thus, by creating a dog model of pulmonary oedema and directly measuring aortic blood flow we were able to show the effects of increasing lung fluid on impedance cardiography. When extrapolated to the clinical setting our data provide a further explanation of why impedance cardiography has been found to be unreliable in critically ill patients in whom pulmonary oedema and lung pathology are very common. Thus lung fluid can be added to the list of factors, such as changes in peripheral resistance due to sepsis and vasopressor therapy17 and cardiac arrhythmias, anomalies and valve defects,2 that have been reported to have an adverse effect on impedance cardiography measurements in critically ill patients.
Oleic acid is the most common agent used experimentally to induce pulmonary oedema; hence it was used in the present study.18 19 The alternative method of overloading the circulation with fluids was found to be unreliable and produced large changes in haemodynamic parameters, which would have complicated our data analysis. However, liberal volumes of i.v. saline were still given throughout the study to maintain CVP and encourage the accumulation of lung fluid. This is reflected in the high urine outputs of 120150 ml h1 observed in this study. A number of techniques are available for assessing lung fluid content, such as double-indicator dilution (the gold standard method), chest radiography and thoracic impedance.1822 Only the latter was available in the present study. We also confirmed at post-mortem that lung fluid had indeed accumulated.
Throughout the experiment we aimed to maintain the dog's circulation in a steady state. This was of particular importance as other haemodynamic factors, such as changes in peripheral resistance, have also been shown to affect impedance measurements.17 Unwanted haemodynamic effects may still have influenced some of our plots in Figure 1. There was also occasional loss of impedance data because the RheoCardioMonitor failed to function at times as lung oedema developed (Fig. 1). Drugs that support blood pressure and cardiac output were also omitted, and the circulating blood volume, measured by CVP, was maintained at a constant level by giving aliquots of i.v. saline. Changes in COFP and SVR were seen only when Zb decreased by >4 or COFP changed (Table 1 and Fig. 1; dogs 3, 5, 7 and 8).
Thoracic impedance has been used for many years to assess extravascular lung fluid.11 20 21 Spinale and colleagues20 showed in a pig model, where lung oedema was also induced by oleic acid, that decreases in Zb were related to increases in extravascular lung fluid.20 These authors used the gold standard method of double-indicator dilution to assess lung fluid. Zb is not a very consistent measurement of lung fluid and varies greatly between individuals. The normal value for Zb in humans is quoted to range between 20 and 35 owing to differences in chest size and morphology.11 23 Later, Spinale and colleagues24 improved their agreement between Zb and lung fluid measurements in pigs by adjusting for factors such as thoracic shape, haematocrit and right ventricular end-diastolic volume. Roos and colleagues25 tried to overcome this problem of varying body habitus in critically ill patients by indexing Zb to thoracic length. This variability in the baseline value makes it difficult to use Zb as a bedside measure of lung fluid and it is the relative change that is more important. This point was evident in the present study where baseline Zb ranged from 29 to 43
, but the change in Zb with the development of pulmonary oedema was only 36
. It is also noteworthy that the values of Zb in the present dog study (2943
) were higher than that in human subjects (2035
), reflecting differences in the size and shape of the thorax between the two species. Despite these topographical differences between canine and human morphology, the underlying electrophysiological principles are the same. Thus it would be reasonable to apply our findings in animals to the clinical setting. Zb values <20
in humans are generally regarded as indicative of lung oedema,12 14 and most impedance cardiograph manufacturers warn that their systems become unreliable and should not be used below the 20
threshold.
Our data also demonstrated a wide variation from the mean (or SD) of individual impedance cardiac output readings as measurements were repeated (Fig. 1). Assuming that the flow probe provided relatively consistent readings, and the quoted error is 12%,26 27 most of the variability in the error arises from the impedance measurements. For patient safety reasons, the RheoCardioMonitor uses a relatively small transthoracic current (2 mA) to detect the thoracic impedance. This results in a high signal-to-noise ratio with other bioelectric signals, such as respiratory movement and electrical activity of the heart and muscle movement, as well as electrical artifacts arising from the electrode system and device circuitry. The RheoCardioMonitor incorporates a number of signal-processing and filter technologies aimed at reducing these artifacts.16 28 However, the system still provides an impedance signal that contains a significant element of noise, making the measurement of impedance variables such as dZ/dtmax and LVET unreliable. Whether this problem exists in other impedance cardiography devices is not known. Therefore impedance measurements should be averaged over a number of readings to provide reliable measurements of impedance cardiac output. This problem potentially detracts from the use of the method as a real-time monitor of acute changes in cardiac output. To overcome the problem in the experimental setting and permit statistical analysis, we grouped the data at 1 intervals and used median values (Fig. 2).
In conclusion, we have demonstrated an association between lung oedema induced by oleic acid and changes in impedance cardiac output estimates. There appears to be an attenuation of the COIC value as lung fluid accumulates. This finding helps to explain why the impedance method fails to provide reliable data in critically ill patients.
![]() |
Acknowledgments |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Bernstein DP. Continuous noninvasive real-time monitoring of stroke volume and cardiac output by thoracic electrical bioimpedance. Crit Care Med 1986; 14: 898901[ISI][Medline]
3 Bernstein DP. A new stroke volume equation for thoracic electrical bioimpedance: theory and rationale. Crit Care Med 1986; 14: 9049[ISI][Medline]
4 Critchley LA. Impedance cardiography: impact of new technology. Anaesthesia 1998; 53: 67785[CrossRef][ISI][Medline]
5 Critchley LA, Critchley JA. A meta-analysis of studies using bias and precision statistics to compare cardiac output measurement techniques. J Clin Monit Comput 1999; 15: 8591[CrossRef][ISI][Medline]
6 Young JD, McQuillan P. Related comparison of thoracic electrical bioimpedance and thermodilution for the measurement of cardiac index in patients with severe sepsis. Br J Anaesth 1993; 70: 5862[Abstract]
7 Sageman WS, Amundson DE. Thoracic electrical bioimpedance measurement of cardiac output in postaortocoronary bypass patients. Crit Care Med 1993; 21: 113942[ISI][Medline]
8 Wong DH, Tremper KK, Stemmer EA, et al. Noninvasive cardiac output: simultaneous comparison of two different methods with thermodilution. Anesthesiology 1990; 2: 78492
9 Clarke DE, Raffin TA. Thoracic electrical bioimpedance measurement of cardiac output: not ready for prime time. Crit Care Med 1993; 21: 111112[ISI][Medline]
10 Tremper KK. Continuous noninvasive cardiac output: are we getting there? Crit Care Med 1987; 15: 2789[ISI][Medline]
11 Van de Water JM, Miller IT, Milne EN, et al. Impedance plethysmography: a noninvasive means of monitoring the thoracic surgery patient. J Thorac Cardiovasc Surg 1970; 60: 6417[ISI][Medline]
12 Shoemaker WC, Wo CC, Bishop MH, et al. Multicenter trial of a new thoracic electrical bioimpedance device for cardiac output estimation. Crit Care Med 1994; 22: 190712[ISI][Medline]
13 Critchley LA, Critchley JA. Wet lungs and impedance cardiography. Anaesth Intensive Care 1997; 19: 31415
14 Critchley LA, Critchley JA. Lung fluid and impedance cardiography. Anaesthesia 1998; 53: 36972[CrossRef][ISI][Medline]
15 Critchley LA, Calcroft RM, Tan PY, Kew J, Critchley JA. The effect of lung injury and excessive lung fluid, on impedance cardiac output measurements, in the critically ill. Intensive Care Med 2000; 26: 67985[CrossRef][ISI][Medline]
16 Haryadi DG, Westenskow DR, Critchley LA, et al. Evaluation of a new advanced thoracic bioimpedance device for estimation of cardiac output. J Clin Monit Comput 1999; 15: 1318[CrossRef][ISI][Medline]
17 Critchley LA, Peng ZY, Fok BS, James AE. The effect of peripheral resistance on impedance cardiography measurements in the anaesthetized dog. Anesth Analg 2005; 100: 170812
18 Scillia P, Kafi SA, Melot C, Keyzer C, Naeije R, Gevenois PA. Oleic acid-induced lung injury: thin-section CT evaluation in dogs. Radiology 2001; 219: 72431
19 Katzenelson R, Perel A, Berkenstadt H, et al. Accuracy of transpulmonary thermodilution versus gravimetric measurement of extravascular lung water. Crit Care Med 2004; 32: 15504[CrossRef][ISI][Medline]
20 Spinale FG, Reines HD, Cook MC, Crawford FA. Noninvasive estimation of extravascular lung water using bioimpedance. J Surg Res 1989; 47: 53540[ISI][Medline]
21 Zellner JL, Spinale FG, Crawford FA. Bioimpedance: a novel method for the determination of extravascular lung water. J Surg Res 1990; 48: 4549[CrossRef][ISI][Medline]
22 Raaijmakers E, Faes TJ, Kunst PW, et al. The influence of extravascular lung water on cardiac output measurements using thoracic impedance cardiography. Physiol Meas 1998; 19: 4919[CrossRef][ISI][Medline]
23 Fein A, Grossman RF, Jones JG, Goodman PC, Murray JF. Evaluation of transthoracic electrical impedance in the diagnosis of pulmonary oedema. Circulation 1979; 60: 115660
24 Spinale FG, Hendrick DA, Crawford FA, Carabello BA. Relationship between bioimpedance, thermodilution, and ventriculographic measurements in experimental congestive heart failure. Cardiovasc Res 1990; 24: 4239[ISI][Medline]
25 Roos JC, Koomans HA, Boer P, Dorhout Mees EJ. Transthoracic electrical impedance as an index of extracellular fluid volume in man. Intensive Care Med 1985; 11: 3942[ISI][Medline]
26 Bednarik JA, May CN. Evaluation of a transit-time system for the chronic measurement of blood flow in conscious sheep. J Appl Physiol 1995; 78: 52430
27 Dean DA, Jia CX, Cabreriza SE, et al. Validation study of a new transit time ultrasonic flow probe for continuous great vessel measurements. ASAIO J 1996; 42: M6716[ISI][Medline]
28 Barin E, Haryadi DG, Schookin SI, et al. Evaluation of a thoracic bioimpedance cardiac output monitor during cardiac catheterization. Crit Care Med 2000; 28: 698702[CrossRef][ISI][Medline]
|