Foreword

The Office of Health Technology Assessment (OHTA) evaluates the risks, benefits, and clinical effectiveness of new or unestablished medical technologies. In most instances, assessments address technologies that are being reviewed for purposes of coverage by Federally funded health programs.

OHTA's assessment process includes a comprehensive review of the medical literature and emphasizes broad and open participation from within and outside the Federal Government. A range of expert advice is obtained by widely publicizing the plans for conducting the assessment through publication of an announcement in the Federal Register and solicitation of input from Federal agencies, medical specialty societies, insurers, and manufacturers. The involvement of these experts helps assure inclusion of the experienced and varying viewpoints needed to round out the data derived from individual scientific studies in the medical literature.

After OHTA receives information from experts and the scientific literature, the results are analyzed and synthesized into an Assessment report. Each report represents a detailed analysis of the risks, clinical effectiveness, and uses of new or unestablished medical technologies. If an assessment has been prepared to form the basis for a coverage decision by a Federally financed health care program, it serves as the the Public Health Service's recommendation to that program and is disseminated widely.

OHTA is one component of the Agency for Health Care Policy and Research (AHCPR), Public Health Service, Department of Health and Human Services.

• Thomas V. Holohan, M.D.
• Director
• Office of Health Technology Assessment
• J. Jarrett Clinton, M.D.
• Administrator
• Questions regarding this assessment should be directed to:
• Office of Health Technology Assessment
• AHCPR
• Executive Office Center, Suite 400
• 2101 East Jefferson Street
• Rockville, MD 20852
• (301) 227-8337

[Introduction]

This reassessment is prompted by interest in determining whether the measurement of cardiac output (CO) by bioimpedance is accurate, safe, clinically useful, and equivalent to thermodilution (TD) techniques. In addition, the appropriateness of bioimpedence measurements of CO on an outpatient basis will be addressed.

The resistance to the flow of an electrical current is termed impedance, and tissue resistance to an applied current is known as electrical bioimpedance (EB).(1,2). The impedance within a body segment changes as the volume of blood within the segment changes. If the volume of blood within that segment is high relative to the volume of other body tissues, the impedance change will be high, primarily because the conductivity of blood is significantly higher than that of any other body tissues.(3,4) In addition, the resistivity of blood changes with its velocity and hematocrit variations.

Thoracic EB applied to monitoring the heart (impedance cardiography) utilizes a beat-to-beat analysis of the transthoracic impedance changes during the cardiac cycle as a means of calculating stroke volume (SV) and CO.(5-8) Electrical bioimpedance is based on the observation that pulsatile changes in flow between the aortic valve and the descending aorta during each systole create an impedance change, apparently as a consequence of the additional volume of blood entering the segment being monitored.(3,9) During sinus rhythm, EB reaches a maximum prior to the onset of ventricular ejection and then decreases rapidly, reaching a minimum approximately synchronous with the peak of aortic pressure.(4,9,10)

The ability to measure EB has been incorporated into automated, computerized monitoring devices that can obtain and display physiologic data continuously and noninvasively. The measurement of CO by EB devices involves the creation of an electrical field within the chest by the application and subsequent transmission of an alternating, sinusoidal, high-frequency (50-100 kHz), low-amplitude (1-4 mA) current from a constant-current oscillator through the thorax via an outer pair of electrodes placed at the upper neck and upper abdomen.(1,2,7,9,11,12) These pulsatile impedance changes (which depend on the fluid and electrolyte characteristics of the thoracic volume) are readily sensed and measured as voltage differences by means of an inner pair of external skin electrodes at the base of the neck and xiphoid process, and they are displayed by an EB device as an impedance waveform that peaks with the beginning of cardiac, systole.(1-3,9,12,13)

Although electrical currents passing through the body may produce burns, respiratory arrest, or ventricular fibrillation, it is only frequencies below 5 kHz that are regarded as hazardous.(9) Frequencies greater than 20 kHz do not produce appreciable electrode polarization, and they permit better signal-noise ratios.(9)

The relationship of changes in impedance to changes in blood volume can be derived mathematically, and a variety of formulas are used as mathematical models for calculating SV and CO from these waveforms (vide infra) (1,9,14-18)

Background

The measurement of CO offers a means of assessing global cardiac function in that, normally, CO is linearly related to cardiac work load.(19) As an important expression of cardiovascular performance, CO provides one index of determining the response of the cardiovascular system to demand, as well as a measure of the hemodynamic state of the patient.(20) It has been stated that CO should be routinely monitored in critically ill patients(21). or perioperatively in high-risk patients.(22) However, other physicians propose hemodynamic monitoring only for the small subset of critically ill patients who are unresponsive to conventional therapy.(23) Changes in CO may be used to identify a deterioration or improvement in the hemodynamic status of a patient or confirm the need for, or efficacy of, treatment.(24) However, CO is affected by factors other than cardiac performance, and the distribution of blood to the brain, kidneys, and other vascular beds is probably more important than the magnitude of the CO itself; thus in critically ill patients, CO measurements may add little to the data that can be derived from other hemodynamic monitoring and observations of renal output.(25) In addition, none of the techniques used to measure CO indicate right-or left-sided filling pressures, further limiting their usefulness in hemodynamic assessments.(26)

The measurement of CO, as a quantitative index of heart function, has generally required one of a number of invasive techniques --most frequently involving cardiac or arterial catheterization --which may be cumbersome, time consuming, and associated with some degree of risk.(23,27) Interest in noninvasive methods for measuring CO derives from a desire to extend this quantitative measurement to a wide variety of patients without the risks attending catheterization, which include pain, sepsis, pneumothorax, emboli, and arrhythmias; these complications are seen in up to 24 of patients and are regarded as serious in up to 4-5 of these patients.(28,29)

The first experimental application of electrical impedance as a plethysmography technique was reported in Germany in 1907, (30). and the first investigators to relate changes in thoracic impedance to the cardiac cycle were Atzler and Lehmann in 1932.(31) These investigators attempted to measure SV noninvasively by means of an electric current, and they also described its application to monitor the hemodynamic effects of drugs and for peripheral pulse detection. This early work was modified by Nyboer, who in 1939 applied this technology to monitoring impedance changes in the heart related to the electrocardiogram and heart sounds. He calculated SV based on the variations of impedance induced by fluctuations in the caliber of major vessels by using the following equation(18,32).

In 1966, Kubicek et al(14). used a four-electrode impedance plethysmography system and a modification of the Nyboer equation (to account for respiratory variations in EB) to calculate CO from changes in EB and introduced the concept of estimating SV by EB waveform analysis. This method is partly empirical and based on the unlikely assumptions of a uniform current density, a uniform configuration of the thorax, and a homogeneous conductor having a uniform cross section.(24,33) The Kubicek equation, based on modeling the thorax as a cylinder, is as follows(11,14.).

The early Nyboer and Kubicek equations for calculating EB, which used the measured thoracic length (subject to potentially significant error) and the value of the specific resistivity of blood, led to unacceptable inaccuracies.(34)

In 1982, Sramek proposed an alternative SV equation, which modeled the thorax as a truncated cone and eliminated a term for the resistivity of blood (as a trivial factor in total resistively) (35,36) The Sramek equation employs an entity termed volume of electrically participating tissue (VEPT). The Sramek equation calculates SV and VEPT as follows:

The most recent modification of the Kubicek and Sramek equations was introduced by Bernstein in 1986.(15) It corrected for body habitus (height and weight rather than height alone) and was incorporated into a commercially available instrument for on-line measurement of CO by means of algorithms to calculate SV from thee electrical field size of the thorax, base impedance of the thorax, and SV changes related to ventricular contractions.(37).

Stroke volume and CO are not constant values, (38). and absolute methods for their direct clinical determination do not exist.(39). There are a number of indirect methods, but most are invasive and all are associated with limitations and significant degrees of error, so it cannot be said that a single universal standard method exists.(14,39,40)

Currently, the most widely accepted methods for the determination of CO are Fick oximetry, TD, and indicator dilution (dye or radioisotope) techniques, all of which depend on the principle of measuring the concentration of a known volume of indicator (such as dye, oxygen, or cold liquid) and generally estimate CO over a period of time rather than on a beat-by-beat basis.(4,24,25,34,41,42)

Although moderately difficult, tedious, and prone to error, the Frick method is generally regarded as the most accurate.(26,27) It requires the measurement of the partial pressures of oxygen and carbon dioxide in expired air and measurement of oxygen saturation in arterial and mixed venous blood, via right heart catheterization into the pulmonary artery.(6) The rate of error of this technique compared with electromagnetic flowmeters (feasible only in animal models) is approximately 10 but can exceed 19 in low CO states.(43,44). (The indirect Fick method is noninvasive and involves the analysis of breathed gases --carbon dioxide or nitrous oxide.(45-48) ) Both the indicator dilution and TD methods overcame the problems of the non-steady-state conditions of patient oxygenation and the specific breathing requirements of the Fick methods by injecting a known quantity of marker.(46,49) The indicator dilution technique, which supplanted the Fick methods, requires cardiac catheterization and 15-to 30-minute intervals for sampling; however, using the dye technique, only a limited number of samplings are possible owing to recirculation of the dye.(25,50) The radioisotope technique does not require arterial sampling.(51) Individual differences between indicator dilution measurements of up to 20 can be accounted for by the nonuniform distribution of the indicator and by instrumentation error.(52) Variations in measurements, overtime in the same patients, have been reported to range between 11 and 17.(50,53)

Because of its relative simplicity, precision, and accuracy, TD (also an indicator dilution method) is currently the most commonly used technique for measurement of CO.(25,54) However, it requires hospitalization, a skilled operator, and the absence of pneumothorax.(55) A thermistor mounted on the tip of a cardiac catheter, with radiographic confirmation of its position, is advanced into the pulmonary artery and records the temperature changes of cool saline of dextrose injected into the subclavian or internal jugular vein.(54,57) Determinations of CO by TD are based on an automatic computation of the area under a temperature-time curve displayed on a monitor and are highly correlated with values obtained by the Fick method (r = .89-.96) (38,54) Although CO measurement by TD is generally regarded as being accurate and precise and has achieved widespread application as the procedure of choice in critical care medicine, the variability of intra-subject TD measurements, depending on the clinical situation, is stated to be in the range of 15-20.(39,41,57) During mechanical ventilation, variations in measurements can be in the range of 10-50.(58,59)

The accuracy of these commonly accepted methods for determining CO has been questioned, especially in seriously ill patients.(9,48,50) Fick, indicator dilution, and TD methods all rely on the recording of distortion-free washout curves from the ventricle, and they assume rapid injection, complete mixing, repeated use of the same sampling site, and a steady-state condition of the patient, all of which are unlikely in a critically ill patient. Complete mixing rarely occurs in the ventricle, and the positions of the sampling catheters can also influence the result. In addition, washout curves are unreliable in low CO states. These methods, which use mean values of successive readings, have an inherent physiologic variability of 5-10 in most circumstances and a much greater error in nonsteady circulatory states.(9)

Despite these drawbacks, invasive monitoring for measuring physiologic functions including CO has been the dominant method of evaluating hemodynamics in the critical care setting for the purpose of making clinical decisions regarding fluid or pharmacologic support of the circulation.(11,27) More recent interest in simpler and commercially available noninvasive techniques to measure CO and SV has led to the use of both EB and Doppler ultrasonography.(60,61)

Continuous-wave suprasternal Doppler ultrasound and the Doppler pulmonary artery catheter are among the newer techniques that calculate CO by using the reflection of ultrasound from moving red blood cells to measure ascending aortic or pulmonic blood flow velocity.(25,62,63) Although relatively high correlations between Doppler and TD measurements have been reported, repeated measurements are required for continuous monitoring, and considerable technical skill is required for the accurate measurement of the arterial segment diameter.(21,64) The composite of problems associated with the Doppler technique has led to variations in CO estimates of +/-40.(65) The transtracheal and transesophageal Doppler techniques are regarded as invasive by some clinicians.(1,21)

Although use in critical care medicine and anesthesia appears to be its primary application, EB has been used to measure CO in normal subjects(66). ; it has also been used in variations conditions and procedures including dialysis, (67-69). cardiac pacing, (70). asthma, (71). sleep apnea, (72). pregnancy, (73). cesarean section, (74). and the neonatal period(75,76). ; in drug studies(77-85). ; and in simulated space flights.(86)

Other (less common) methods used to measure CO include meas spectrometry, radionuclide left ventriculography, electromagnetic flowmeter, acetylene absorption, pressure gradient and aortic pulse contour techniques, the thoracocardiograph, M-mode ultrasound cardiography of the left ventricle, and the mechanosphygmogram.(3,34,64,87-90)

Rationale

The basic rationale proposed for the use of EB for the measurement and monitoring of CO is that it is as simple to apply as an electrocardiogram and is noninvasive. The technique is rapid, unobtrusive, innocuous, automatic, and portable; it can be readily applied by nursing or technical staff with minimal and the patient can be left unattended for continuous and prolonged monitoring. Proponents claim that EB can measure CO with the same clinical accuracy as either the Fick or TD technique and that it offers the potential for sequential measurements of CO in patients for whom invasive measurements are impractical or contraindicated. In addition, EB can determine CO on a beat-to-beat basis or a predetermined intermittent frequency, which may, if required, permit a more rapid intervention than techniques using time-averaged data.

Review of Available Literature

There are no published data that relate CO measurements by EB to patient outcomes. Virtually all studies of EB report comparative results of EB measurements of CO with those obtained by other techniques. Thus, the accuracy and limitations of EB for determining CO have been compared with more established techniques in animals, (39,91-93). in normal volunteers (Table 1), and in a variety of patients in various clinical settings (Table 2). The tables summarize published reports (selected to include a minimum of 10 subjects) comparing EB with TD. Results of these comparisons were commonly reported as correlation coefficients (r) with subsequent conclusions by the investigators that EB provided satisfactory estimates of CO if correlations were approximately in the range of .80-.90 and that EB did not provide reliable estimates of CO if r as less than .70. However, the rationale for selecting these levels of correlation was not made explicit. A recent study by Spahn et al(26). reported significant underestimation of CO by EB despite relatively high correlation coefficients (r = . 78-.82). The considerable scatter of EB results compared with TD results in this study could have misled therapeutic interventions, and in a considerable number of situations (4.7-12.3) the CO changes measured by EB were in the opposite direction of the CO measurements by TD. In addition, the precision and accuracy of EB is significantly reduced in patients with cardiac valve lesions and in the presence of intracardiac shunts, pacemakers, sepsis, mechanical ventilation, dysrhythmias, or other conditions associated with extremely high or low levels of CO.(30,41,42,95,108,111,112).

Table

Table 1. EB (using the Minnesota model 304 A impedance and the Kubicek equation) versus TD.

Table

Table 2. EB (using a BoMed impedance device and the Sramek-Bernstein equation) versus TD.

Examples of other factors that may potentially affect the accuracy of EB include fluid in the lungs or pleural space, assisted ventilation, the nonuniform changing volume of blood along the length of the thorax, changes in blood resistivity, movement artifacts, electrode type and placement, and the method used to determine signal sampling.(99,101,113)

Discussion

There has been a growing interest in recent years in EB as a noninvasive method for determining CO.(114) However, the application of EB has been and continues to be controversial, with conflicting reports regarding its precision and accuracy.(16,115) The variety of techniques used to measure CO would seem to indicate that no one method is entirely satisfactory, and the conflicting results and wide ranges in correlations of EB seen in these comparative studies may be explained, in part, by the heterogeneity of the subjects tested and the methodologic diversity in the application of EB, including equipment, equations, and signal sampling and processing strategies.(9,17,66,106,116) Advocates of EB have, for the most part, based their conclusions of clinical effectiveness on correlational studies comparing EB with TD (vide supra). As seen in the tables, while some studies exhibited relatively high correlations, the range of r was from. 17 to 1.0. This very wide range of correlations serves to cast doubt on the reliability of EB as a measure of CO. It should be noted that there r values in these studies were derived from multiple measurements in the same subjects, which itself leads to a source of error. Also, investigators did not attempt to compare the measurement variation within subjects with the variation between subjects. Categorizing patients into broad subsets, such as "critically ill," "intensive care," "postsurgery," "coronary artery disease," and "primary trauma," rather than more precisely described categories, stratified according to severity of illness, suggests weak study design and makes suspect the extrapolation of the reported data.

A 1989 editorial by Sun and Van De Water(117). emphasized the need to standardize EB methodology, including the type and placement of electrodes, blood resistivity, etc. In addition, the equations used to compute CO from EB measurements are based on oversimplified and empirically derived models of thoracic hydraulic and electrical events, and impedance changes due to volume and conductivity in each major thoracic structure have not been calculated.(4,7,17,95,118)

Serious problems arise when comparisons are made between different methods of measuring CO. With standard invasive techniques, such as Fick and TD, measurements in individuals are subject to sources of error such that the precision and accuracy of a single measure has been deemed insufficient for routine clinical use, so a series of measurements are usually performed and a mean then calculated. These standard methods do not exhibit completely satisfactory intercorrelations.(101). Therefore, an evaluation of EB versus a standard technique should take into account the presence of systematic errors affecting both techniques.(48,119) In addition, the commonly reported correlation coefficients is, at best, only partially descriptive of data obtained by theoretically different methods.(48,119) La Mania et al(119). have addressed the shortcoming of using correlation coefficients as a sole means of determining whether to accept or reject a new methodology.

Calculating a correlation coefficient is a very common method of comparing two variables. However, this represents only a method for exhibiting an association, not a measure of degree of agreement (equivalence) or accuracy, and it may be worthless when used to measure two equally inaccurate methods.(21,33) A measure of equivalence would require bias plot analysis (performed only in relatively few, and mostly recent, studies) (13,49,55,100,101,113). and calculation of the mean of the difference between the two measurements.(16). In addition, the discrepancies between the different methods of measuring CO may be related to the physical principles used by the techniques, so that a high correlation under one set physiologic conditions may not be seen under another.(113)

Regression analysis, which was more frequently performed to compare different methods of measurement, has its own limitations and does not provide a direct answer to the questions of equivalence.(119,120)

The fact that such a wide range of correlation coefficients (e.g., .17-.98) have been reported in a substantial number of studies fuels the debate on whether EB can reliably estimate CO, especially under certian conditions. In many cases the lowest correlation coefficients were seen in hemodynamically unstable patients (e.g., preeclampsia, .17; cardiomyopathy, .51; sepsis or dysrhythmias, .55).

Few of these reported studies quantified in detail either the precision or the accuracy of the measurements on which r was based, and none of the commonly used standard methods is considered a true "gold standard."(15,65,121,122). Methods have become standard because their accuracy has been widely deemed sufficient to make decisions regarding patient care.(16) As noted above, all methods of measuring CO have differing sources of inaccuracy.(23) All available methods, including the more recently developed radiopharmaceutical techniques, produce errors of at least +/-10, and the overall clinical error of most methods of measuring CO is in the range of 15-20.(41,47,122) Measurements of CO in the same individual may vary considerably, even under controlled conditions, and single measurements in different individuals yield the expected wide spectrum of values.(123)

A 1978 review of EB by Miller and Horvath(124). concluded that the technology is "reliable and practical" when its limitations are fully understood. They agreed with observations that EB cannot produce absolute estimates of CO and is not reliable for intersubject comparisons. However, they concluded that EB is useful as a noninvasive monitor of intrasubject variations in CO because the sources of error remain relatively constant within subjects.

Porter and Swain, (24). in a 1987 review of 19 studies performed between 1968 and 1985, reported that r values ranged between .26 and .98. They concluded that although EB measurements were generally higher than the values obtained by other methods, a series of consecutive CO measurements can give valuable information if a trend in CO can be identified, which may give warning of hemodynamic deterioration or confirm the efficiency of treatment. According to Patterson et al, (21). the theoretical basis for EB is very weak, and in high-risk patients the EB waveform may change very significantly and fail to give an accurate prediction of CO. Indeed, most investigators have concluded that EB cannot be regarded as a reliable method for providing absolute values of CO, but that changes in CO can be adequately monitored by EB compared with invasive methods.(9,60,66,90,95,116,118,125-127)

Computerized ensemble averaging has been recently proposed for improving the accuracy of EB measurement.(128) Although the results were promising, this technology has not been validated by other investigators.

In addition to questions concerning the precision and accuracy of CO measurements, the general clinical utility of such measurements in many types of ill patients has been questioned, and controlled clinical trials establishing improved patients outcome from their use have not been performed.(10,65,129) Moreover, the criteria for the appropriate selection of subsets of patients for whom EB determinations of CO may be most useful have not been delineated or validated. In 1989 editorial. Nishimura(130). stated that the clinical utility of monitoring immediate changes in CO in the critically ill patient needs to be further assessed. A 1987 editorial in Chest addressed the risk and lack of proven benefit of the use of all hemodynamic monitoring in the intensive care setting.(131)

Although the majority of clinical studies of EB were of hospitalized patients who had previously been subjected to right heart catheterization for CO determinations by TD or other techniques, the simplicity of applying EB technology, which is comparable with that of electrocardiography, also lends itself to numerous potential applications in outpatient settings.(9) A number of laboratory studies of EB involving normal volunteers and/or ambulatory nonhospitalized patients included measurements of CO at rest, during exercise, or during tilt table studies.(10,71,87) However, no published studies have specifically addressed the use of EB in the outpatient setting, and its potential use in this context remains conjectural.

In response to the Federal Register notice of this assessment(132). and the solicitation of information and opinions from organizations and institutions with interest and/or experience with this technology, the Office of Health Technology Assessment (OHTA) has received the following input:

The University of Minnesota Medical School (February 1991) believes that measurement of CO by EB is, at this time, not a valid technique for use in critically ill patients. The reasons for this opinion include the observation that all the available models used to calculate CO are poor representations of the true electrical configuration of the thorax; in addition, the validity of impedance measurements relies on their agreement with other accepted methods for determining CO, which have a better physical basis in theory and model.

The American College of Cardiology (March 1991) stated that EB offers a very promising method for the noninvasive evaluation of cardiac and critically ill patients, provided the limitations of the system are fully understood. They believe that EB is most useful in research and clinical settings where serial measurements of CO and hemodynamic trends are required. However, EB may be suboptimal where accurate estimation of CO is critical. In addition, random measurement of CO in an outpatient setting would be the least useful application of this technology and can be recommended only for clinical investigation.

The Medical College of Wisconsin at Milwaukee (April 1991) believes there is considerable evidence that, in selected patients, EB can provide reliable CO measurements in a number of clinical situations.

The University of Illinois at Chicago (April 1991) uses EB to follow changes in CO during hemodialysis and believes that EB gives very consistent CO measurements.

Texas Tech University Health Sciences Center (April 1991), which has approximately 4 years of experience in the evaluation and application of ensemble-averaged EB, stated that EB determinations of CO are "quite valid" compared with simultaneous TD measurements in a large and diverse population of seriously ill patients. They consider Swan-Ganz, TD, and EB techniques to be "roughly equivalent," and they use EB after the first 24 hours of intensive care monitoring in place of invasive techniques.

The Mayo Clinic (April 1991), on the basis of somewhat limited experience with EB, has compared it with TD and Doppler ultrasound measurements of CO and concluded that EB offers a means of providing simple, reliable, and reproducible measurements for monitoring sequential changes in CO within individual patients in most clinical situations. Theoretical limitations of using EB to measure CO include assumptions regarding the volume of the thorax, the relationship of SV to basal impedance, and the processing of the gated impendance waveform. Clinical limitations of EB massive ascites or thoracic fluid accumulation resulting in extremely low thoracic impedance and extreme ranges of thoracic configuration such as marked obesity or cachexia.

The Food and Drug Administration (FDA) (January 1992) has informed OHTA that it has no new information to offer concerning EB. It has previously stated that EB devices are Class II devices and were approved under the 510(k) (This process permitted the device manufacturers to market their product on the basis of its being equivalent to devices already categorized as Class II, without having to submit clinical data.) The FDA believes that EB devices are safe in regard to patient exposure to electrical hazard and appear to be reasonably effective.

The National Institutes of Health (February 1992) has conclude that EB as currently used is not clinically acceptable. It appears to provide reasonable results most often in normal adults. However, in specific abnormal populations, the discrepancies between EB and direct methods of CO determinations occur frequently and unpredictably. Although there is more support for the use of EB for monitoring changes in CO rather than for determining absolute values, the method fails often in this mode as well.

Although adverse consequences of injecting a 4-mA current briefly into the torso have not been shown, prolonged exposure to these high-frequency current levels should be discouraged until appropriate data are collected on EB used for long-term monitoring.

Summary

Thoracic bioimpedance relates change in throcic electrical conductivity to changes in thoracic aotic blood volume and blood flow. Electrical bioimpedance has been proposed as a simple and readily reproducible noninvasive technique for the determination of CO. The absolute values of CO by EB have frequently differed from values obtained by other methods regarded as "standard." It modest gain in popularity as a clinical technique appears to be related to its suggested usefulness as a monitor to detect changes in CO within individual subject as an alternative to invasive techniques, especially when serial measurements are required. There continues to be lack of persuasive data derived from rigorous clinical trials supporting the use of EB determinations of CO for the clinical management of any subset of patients.

Conflicting reports concerning the accuracy of EB continue to be published. Although many investigators have concluded that EB yields satisfactory results with a probability of error similar to that of other accepted methods, in general their reliance on correlation coefficients as the main evidence supporting their stance appears to provide necessary but insufficient evidence of clinical utility in either hospital or outpatient settings. The precision and accuracy of EB has not been quantified, and its equivalence to other methods has not been proven. In addition, the methodologic diversity in the application of EB and the heterogeneity of the subjects tested have posed serious problems in attempting to compare results between different method of measuring CO. Although proponents favoring EB for CO measurements are more supportive of its application for monitoring changes in CO rather than for determination of absolute values of CO, the method often fails in this mode as well. Electrical bioimpedance continues to suffer aa lack of widespread clinical acceptance, and the subsets of patients for whom EB measurement might be clinically effective have been neither delineated nor validated.

References

1.

Gayes JM Transthoracic electrical bioimpedance: A noninvasive measurement of cardiac output J Post Anesth Nurs 1989 4:300–305. [PubMed: 2477531]

2.

Buell JC A practical, cost-effective, noninvasive system for cardiac output and hemodynamic analysis Am Heart J 1988 116:657–664. [PubMed: 3134811]

3.

Bache RJ, Harley A, Greenfield JC Jr Evaluation of thoracic impedance plethysmography as an indicator of stroke volume in man Am J Med Sci 1969 258:100–113. [PubMed: 4240999]

4.

Harley A, Greenfield JC Jr Determination of cardiac output in man by means of impedance plethysmography Aerospace Med 1968 39:248–252. [PubMed: 4865723]

5.

Masaki D, Greenspoon JS, Ouyounian JG Measurement of cardiac output in pregnancy by thoracic electrical bioimpedance and thermodilution Am J Obster Gynecol 1989 161:680–684. [PubMed: 2782352]

6.

Guyton AC, Jones CE, Coleman TG Circulatory Physiology: Cardiac Output and Its Regulation, Philadelphia: WB Saunders, 1973.

7.

Penny Penney BC Theory and cardiac applications of electrical impedance measurements CRC Crit Rev Bioeng 1984 1986 13: 227–281. [PubMed: 3516573]

8.

Braden DS, Leatherburg Leatherbury L, Treiber FA, et al Noninvasive assessment of cardiac output in children using impedance cardiography Am Heart J 1990 120:1166–1172. [PubMed: 2239669]

9.

Mohapatra SN Non-invasive Cardiovascular Monitoring by Electrical Impedence Technique, London: Pitman Medical, 1981.

10.

Allison RD, Lewis CE, Rezek TW. Vascular dynamics: Impedance plethysmography study during a standardized tilt table procedure Space Life Sci 1970; 2:261–393. [PubMed: 4940655]

11.

Kubicek WG, Patterson RP, Witoe DA. Impedance cardiograph as a noninvasive method of monitoring cardiac function and other parameters of the cardiovascular system Ann NY acad Sci 1970 170:724–732.

12.

Wong DH, Tremper KK, Stemmer EA, at al Noninvasive cardiac output: Simultaneous comparison of two different methods with thermodilution Anesthesiology 1990 72: 4–12. [PubMed: 2187376]

13.

Colley Cooley WL The calculation of cardiac stroke volume from variationvariations in transthoracic electrical impedance Biomed Eng 1972 19:316–319. [PubMed: 5049477]

14.

Kubicek WG, Karnegis JN, Patterson RP, et al Development and evaluation of an impedance cardiac output system Aetospace Aerospace Med 1966 37:1208–1212. [PubMed: 5339656]

15.

Bernstein DP Continuous noninvasive real-time monitoring of stroke volume and cardiac output by thoracic electrical bioimpedance Crit Care Med 1986 14:898–901. [PubMed: 3757531]

16.

Temper Tremper K Transthoracic electrical bioimpedance versus thermodilution technique for cardiac output measurement during mechanical ventilation Intensive Care Med 1989 15:219–220. [PubMed: 2745863]

17.

Bernstein DP. A new stroke volume equation for thoracic electrical bioimpedance: Theory and rationale Crit Care Med 1986; 14:904–909. [PubMed: 3757533]

18.

Nyboer J, Bagno S, Barnett A, et al Radiocardiograms: Electrical impedance changes of the heart in relation to electrocardiograms and heart sounds. Proceedings of the Thirty-Second Annual Meeting of the American Society of Clinical Investigation J Clin Invest 1940 19:

19.

Pepke-Zaba J, Higenbottam TW, Dinh Xuan AT, et al Validation of impedance cardiography measurements of cardiac output during limited exercise in heart transplant recipients Transplant Int 1990 3:108–112. [PubMed: 2206215]

20.

Sramek BB The impact on diagnosis and therapy of computerized integration, processing and display of noninvasive hemodynamic and cardiodynamic parameters Intensive Care World 1989 6: 205–211.

21.

World MJ Estimation of cardiac output by bioimpedance cardiography J R Army Med Corps 1990; 136:92–99. [PubMed: 2388195]

22.

Rao TKL, Jacobs KH, El-Etr AA Reinfarction following anesthesia in patients with myocardial infarction Anesthesiology 1983 59:493–505. [PubMed: 6650905]

23.

Patel C, Laboy V, Venus B, et al Acute complications of pulmonary artery catheter insertion in critically ill patients Crit Care Med 1986 14:195–197. [PubMed: 3943335]

24.

Porter JM, Swain ID Measurement of cardiac output by electrical impedance plethysmography J Biomed Eng 1987 9:222–231. [PubMed: 3613546]

25.

[no authors listed] Measurement of cardiac output (editorial) Lancet 1988 2:257–258. [PubMed: 2899240]

26.

Spahn DR, Schmid ER, Tornio M, et al Noninvasive versus invasive assessment of cardiac output after cardiac surgery: Clinical validation J Cardiothorac Anesth 1990 4:46–59. [PubMed: 2131856]

27.

Weisel RD, Berger RL, Hechtman HB. Measurement of cardiac output by thermodilution N Engl J Med 1975 292:682–684. [PubMed: 1089881]

28.

Boyd KD, Thomas SJ, Gold J, et al A prospective study of complications of pulmonary artery catheterizations in 500 consecutive patients Chest 1983 84:245–249. [PubMed: 6884097]

29.

Woo MA, Hamilton M, Stevenson LA, et al Comparison of thermodilution and transthoracic electrical bioimpedance cardiac outputs Heart Lung 1991 20:357–362. [PubMed: 2071427]

30.

Cremer M Uber die registrierung mechanischer voraumlrange elektrishem wege. Speziell mit hilfe des saitengal vanometers und saitenilektrometers Muench Med Wochenschr 1907 54:1629–1637.

31.

Atzler E, Lehmann G Uber ein neues Verfahren zur Darstellung de Herztatigkeit (Dielektrographie) Arbeitsphysiologie 1932 5:636–680.

32.

Nyboer J Electrical Impedance Plethysmography, Springfield, IL:Charles C Thomas,1959 .

33.

Ebert TH, Eckberg DL, Vetrovec GM, et al Impedance cardiograms reliably estimate beat-by-beat changes of left ventricular stroke volume in humans Cardiovasc Res 1984 18:354–360. [PubMed: 6744356]

34.

Donovan KD, Dobb GJ, Woods WPD, et al Comparison of transthoracic electrical impedance and thermodilution methods for measuring cardiac output Crit Care Med 1986 14: 1038–1044. [PubMed: 3780246]

35.

Sramek BB Cardiac output by electrical impedance Med Electronics 1982 4:93–97. [PubMed: 10257596]

36.

Sramek BB Hemodynamic and pump-performance monitoring by electrical bioimpedance: New concepts Probl Respir Care 1989 2:274–290.

37.

Huang KC, Stoddard M, Tsueda K, et al Stroke volume measurements by electrical bioimpedance and echocardiography in healthy volunteers Crit Care Med 1990 18:1274–1279. [PubMed: 2225898]

38.

Mackenzie JD, Haites NE, Rawles JM Method of assessing the reproducibility of blood flow measurement: Factors influencing the performance of thermodilution cardiac output computers Br Heart J 1986 55:14–24. [PMC free article: PMC1232063] [PubMed: 3947477]

39.

Jivegard L, Frid I, Haljamae H, et al Cardiac output determination determinations in the pig: Thoracic electrical bioimpedance versus thermodulition thermodilution Crit Care Med 1990 18:995–998. [PubMed: 2394124]

40.

Levett JM, Replogle RL Thermodilution cardiac output: A critical analysis and review of the literature J Surg Res 1979 27:392–404. [PubMed: 529796]

41.

Spinale FG, Reines HD, Crawford FA Jr Comparison of bioimpedance and thermodilution methods of determining cardiac output: Experimental and clinical studies Ann Thorac Surg 1988 45:421–425. [PubMed: 3355284]

42.

Naggar CZ, Dobnik DB, Flessas AP, et al Accuracy of the stroke index as determined by the transthoracic electrical impedance method Anethesiology 1975 45:201–205. [PubMed: 1115371]

43.

Braunwald E (Ed) Heart Disease: A textbook of Cardiovascular Medicine, Philadelphia: WB Saunders, 1980.

44.

Selzer A, Sudrann RB Reliability of cardiac output in man by means of Fick principle Circ Res 1958 6: 485–490. [PubMed: 13547408]

45.

Edmunds AT, Godfrey S, Tolley M Cardiac output measured by transthoracic impedance cardiography at rest, during exercise and at various lung volumes Clin Sci 1982 63:107–113. [PubMed: 7083772]

46.

Miles DS, Sawka MN, Wilde SW, et al Estimation of cardiac output by electrical impedance during arm exercise in women J Appl Physiol Respir Environ Exerc Physiol 1981; 51:1488–1492. [PubMed: 7319881]

47.

Hatcher DD, Srb OB Comparison of two noninvasive techniques for estimating cardiac output during exercise J Appl Physiol 1986 61: 155–159. [PubMed: 3090011]

48.

Smith SA, Russell AE, West MJ, et al Automated non-invasive measurement of cardiac output: Comparison of electrical bioimpedance and carbon dioxide re-breathing techniques Br Heart J 1988 59:292–298. [PMC free article: PMC1216462] [PubMed: 3128316]

49.

Teo KK, Hetherington MD, Haennel RG, et al Cardiac output measured by impedance cardiography during maximal exercise tests Cardiovasc Res 1985 19:737–743. [PubMed: 4084931]

50.

Richardson DW, Wyso EM, Hecht AM, et al Value of continuous photoelectric recording of dye curves in the estimation of cardiac output Circulation 1959 20:1111–1117.

51.

Goddard J, Pullan BR, Rowlands DJ, et al Radioisotope determination of cardiac output: Relative methods of recording over the precordium, the lung and the head Eur J Intensive Care Med 1975 1: 31–36.

52.

Sleeper JC, Thompson HK Jr, McIntosh HD, et al Reproducibility of results obtained with indicator-dilution technique for estimating cardiac output in man Circ Res 1962; 11:712&–720. [PubMed: 13989212]

53.

Fox IJ, Swan HJC, Wood EH Clinical and physiologic applications of a new dye for continuous recording of dilution curves in whole blood independent of variations in oxygen saturation in Zimmerman HA(Ed): Intravascular Catheterization, Springfield, IL:Charles C Thomas,1959,

54.

Branthwaite MA, Bradley RD Measurement of cardiac output by thermal dilution in man J Appl Physiol 1968 24:434–438. [PubMed: 5640735]

55.

Mattar JA, Baruzzi ACA, Diament D, et al A clinical comparison between cardiac output measurement by thermodilution versus noninvasive thoracic electrical bioimpedance Acute Care 1986 12:58–60. [PubMed: 3434160]

56.

Betz RC, Bastanier CK, Mocellin R Impedance cardiography: A method to evaluate quantitatively cardiac output? Comparisons with the Fick principle. Basis Res Cardiol 1977; 72:46–56. [PubMed: 843321]

57.

Secher NJ, Arnsbo P, Heslet L, et al Measurements of cardiac stroke volume in various body positions in pregnancy and during ceasarean section: A comparison between thermodilution and impedance cardiography Scand J Clin Lab Invest 1979 39:569–576. [PubMed: 531485]

58.

Synder JV, Powner DJ. Effects of mechanical ventilation on the measurement of cardiac output by thermodilution Crit Care Med 1982; 10:677–682. [PubMed: 7116891]

59.

Van Grondelle A, Ditchey RV, Groves BM, et al Thermodilution method overestimates low cardiac output in humans Am J Physiol 1983 245: J690–H692. [PubMed: 6624939]

60.

Gotshall RW, Wood VC, Miles DS Comparison of two impedance cardiographic techniques for measuring cardiac output Ann Biomed Eng 1989 17:495–505. [PubMed: 2610421]

61.

Silke B, Zezulka AV, Verma SP, et al Hemodynamic assessment of nicardipine alone and with atenolol in coronary artery disease using a modified echo-Doppler device Am J Cardiol 1989 64:28H–34H. [PubMed: 2801572]

62.

Hanna L, Lopez-Majano V, Ward J Non-invasive ejection fraction monitoring: A comparison of the impedance method to the radionuclide cardiography Anesthesiology 1988 69(3A):A308.

63.

Segal EL, Pearl RG, Ford AJ Jr, et al Instantaneous and continuous cardiac output using a Doppler pulmonary artery catheter J Am Coll Cardiol 1989 13: 1382–1392. [PubMed: 2703619]

64.

Belz GG, Breithaupt K, Erb KA, et al Non-invasive assessment of stroke volume in clinical pharmacology Br J Clin Pharmacol 1990 24:640P.

65.

Taylor SH, Silke B Is the measurement of cardiac output useful in clinical practice? Br J Anaesth 1988 60: 90S–98S. [PubMed: 3284572]

66.

deMey de Mey C, Enterling D Noninvasive assessment of cardiac performance by impedance cardiography: Disagreement between two equations to estimate stroke volume Aviat Space Environ Med 1988 59:57–62. [PubMed: 3355468]

67.

Spence JA, Baliga R, Nyboer J, et al Changes during hemodialysis in total body water, cardiac output and chest fluid as detected by biolectrical impedance analysis Trans Am Soc Artif Intern Organs 1979 25:51–56. [PubMed: 524631]

68.

Thayssen P, Andersen KH, Pindborg T Non-invasive monitoring of cardiac function during haemodialysis Scand J Urol Nephrol 1981 15: 13–317. [PubMed: 7323755]

69.

Dalal S, Yu AW, Gupta DK, et al L-lactate high-efficiency hemodialysis: Hemodynamics, blood gas exchanges, potassium/phosphorus, and symptoms Kidney Int 1990; 38:896–903. [PubMed: 2266673]

70.

Goldstein DS, Cannon RO III, Zimlichman R, et al Clinical evaluation of impedance cardiography Clin Physiol 1986 6:235–251. [PubMed: 3087687]

71.

Edmunds AT, Godfrey S Cardiovacular Cardiovascular response during severe acute asthma and its treatment in children Thorax 1981 36:534–540. [PMC free article: PMC1020437] [PubMed: 7314025]

72.

Tolle FA, Judy WV, Yu PL, et al Reduced stroke volume related to pleural pressure in obstructive sleep apnea J Appl Physiol Respir Environ Exerc Physiol 1983 55:1718–1724. [PubMed: 6662762]

73.

Myhrman P, Granerus G, Karlsson K, et al Cardiac output in normal pregnancy measured by impedance cardiography Scand J Clin Lab Invest 1982 42:513–518. [PubMed: 7156865]

74.

Milsom I, Forssman L, Sivertsson R, et al Measurement of cardiac stroke volume by impedance cardiography in the last trimester of pregnancy Acta Obstet Gynecol Scand 1983 62:473–477. [PubMed: 6364687]

75.

Freyschuss U, Noack G, Zetterstrom R Serial measurements of thoracic impedance and cardiac output in healthy neonates after normal delivery and caesarean section Acta Paediatr Scand 1978 1979 68:357–362. [PubMed: 443037]

76.

FreychussFreyschuss U, Gentz J, Noack G, et al Circulatory adaptation in newborn infants of strictly controlled diabetic mothers Acta Paediatr Scand 1982; 71:209–214. [PubMed: 7136629]

77.

Scott DHT, Arthur GR, Scott DB. Haemodynamic changes following buprenorphine and morphine Anaesthesia 1980 35:957–961. [PubMed: 7004250]

78.

Giuliani G, Salandin V, Michielon P, et al Noninvasive hemodynamic evaluation (by bioimpedance) during brief anesthesia using propofol Minerva Anestesiol 1989 555: 21–23. [PubMed: 2789349]

79.

Kirch W, Halabi A, Linde M, et al Negative effects of famotidine on cardiac performance assessed by noninvasive hemodynamic measurements Gastroenterology 1989 96:1388–1392. [PubMed: 2565841]

80.

Giovannitti JA, Henteleff HB, Bennett CR Cardiorespiratory effects of meperidine, diazepam, and methohexital conscious sedation J Oral Maxillofac Surg 1982 40:92–95. [PubMed: 6950054]

81.

Hinrichsen H, Halabi A, Kirch W Hemodynamic effects of different H2-receptor antagonists Clin Pharmacol Ther 1990 48: 302–308. [PubMed: 1976052]

82.

Thomas SHL, Molyneux P, Kelley J, et al The cardiovascular effects of oral nifedipine and nicardipine: A double-blind comparison in healthy volunteers using transthoracic bioimpedance cardiography Eur J Clin Pharmacol 1990 39:233–240. [PubMed: 2257858]

83.

Orko R, Malmivuo J Anaesthesia for cardioversion: Immediate haemodynamics in patients with thiopental or althesin Ann Clin Res 1976 8: 248–253. [PubMed: 999210]

84.

Dionne RA, Driscoll EJ, Butler DP, et al Evaluation by thoracic impedance cardiography of diazepam, placebo, and two drug combinations for intravenous sedation of dental outpatients J Oral Maxillofac Surg 1983 41:782–788. [PubMed: 6581281]

85.

Keim HJ, WallanceWallace JM, Thurston H, et al Impedance cardiography for determination of stroke index J Appl Physiol 1976 41:797–799. [PubMed: 791918]

86.

Lathers CM, Diamandis PH, Riddle JM, et al Acute and intermediate cardiovascular responses to zero gravity and to fractional gravity levels induced by head-down or head-up tilt J Clin Pharmacol 1990 30:494–523. [PubMed: 2355102]

87.

Wilson MF, Sung BH, Pincomb GA, et al Simultaneous measurement of stroke volume by impedance cardiography and nuclear ventriculography: Comparisons at rest and exercise Ann Biomed Eng 1989 17:475–482. [PubMed: 2610419]

88.

Wilkins SB, Van De Water JM, Horowitz J, et al Noninvasive cardiac output measurement Surg Forum 1989 401: 52–54.

89.

Klausen K Comparison of CO2 rebreathing and acetylene methods for cardiac output J Appl Physiol 1965 20:763–766. [PubMed: 5838728]

90.

Sackner MA, Hoffman RA, Kreiger BP, et al Thoracocardiography, part 2: Noninvasive measurement of changes in stroke volume; comparisons to impedance cardiograph Chest 1991 99:896–903. [PubMed: 2009792]

91.

White SW, Quail AW, de Leeuw PW, et al Impedance cardiography for cardiac output measurement: An evaluation of accuracy and limitations Eur Heart J 1990 11:79–92. [PubMed: 2092994]

92.

Baker LE, Judy WV, Geddes LE, et al The measurement of cardiac output by means of electrical impedance Cardiovasc Res Center Bull 1971 9:135–145.

93.

Pate TD, Baker LE, Rosborough JP The simultaneous comparison of the electrical impedance method for measuring stroke volume and cardiac output with four other methods Cardiovasc Res Center Bull 1975 14:39–52. [PubMed: 764970]

94.

Mehlsen J, Bonde J, Stadeager C, et al Reliability of impedance cardiography in measuring central hemodynamics Clin Physiol 1991 11:579–588. [PubMed: 1769191]

95.

Muzi M, Ebert TJ, Tristani FE, et al Determination of cardiac output using ensemble-averaged impedance cardiograms J Appl Physiol 1985 58: 200–205. [PubMed: 3968011]

96.

Boer P, Roos JC, Geyskes GG, et al Measurement of cardiac output by impedance cardiography under various conditions Am J Physiol 1979 6:H481–H496. [PubMed: 495735]

97.

Denniston JC, Maher JT, Reeves JT, et al Measurement of cardiac output by electrical impedance at rest and during exercise J Appl Physiol 1976 40:91–95. [PubMed: 765317]

98.

Nechwatal W, Bier P, Eversmann A, et al The noninvasive determination of cardiac output by means of impedance cardiography: Comparative evaluation with a thermal dilution technique Basic Res Cardiol 1976; 71:542–552. [PubMed: 1016191]

99.

Balestra B, Malacreda Malacrida R, Leonardi L, et al Esophageal electrodes allow precise assessment of cardiac output by bioimpedance Crit Care Med 1992; 20:62–67. [PubMed: 1729048]

100.

Clancy TV, Norman K, Reynolds R, et al Cardiac output measurements measurement in critical care patients: Thoracic electrical bioimpedance versus thermodilution J Trauma 1991 31:1116–1119. [PubMed: 1875438]

101.

Jewkes C, Sear JW, Verhoeff F, et al Non-invasive measurement of cardiac output by thoracic electrical bioimpedance: A study of reproducibility and comparison with thermodilution Br J Anaesth 1991; 67:788–794. [PubMed: 1768552]

102.

Thomas AN, Ryan J, Doran BR, et al Bioimpedance versus thermodilution cardiac output measurement: The Bomed NCCOM3 after coronary bypass surgery Intensive Care Med 1991 17:383–386. [PubMed: 1774390]

103.

Northridge DB, Findlay IN, Wilson J Non-invasive determination of cardiac output by Doppler echocardiography and electrical bioimpedance Br Heart J 1990 63:93–97. [PMC free article: PMC1024333] [PubMed: 2317415]

104.

Sullivan PJ, Martineau RJ, Hull KA, et al Comparison of bioimpedance and thermodilution measurement measurements of cardiac output during aortic surgery Can J Anaesth 1990 37: [PubMed: 2361316]

105.

Silke B, Pernenkil R, Maitra S, et al A new transthoracic bioimpedance (TEI) method of cardiac output determination Br J Clin Pharmacol 1989 28:

106.

Easterling TR, Benedetti TJ, Carlson KL, et al Measurement of cardiac output in pregnancy by thermodilution and impedance techniques Br J Obstet Gynaecol 1989 96:67–69. [PubMed: 2923843]

107.

Boldt J, Kling D, Thiel A, et al Noninvasive versus invasive hemodynamic monitoring Anaesthesist 1988 37:218–223. [PubMed: 3261552]

108.

Salandin V, Zussa C, Risica G, et al Comparison of cardiac output estimation by thoracic electrical bioimpedance, thermodilution, and Fick methods Crit Care Med 1988 16:1157–1158. [PubMed: 3168510]

109.

Shoemaker WC, Appel PL, Kram HB, et al Multicomponent noninvasive physiologic monitoring of circulation function Crit Care Med 1988 16:482–490. [PubMed: 2966035]

110.

Appel PL, Kram HB, Mackabee J, et al Comparison of measurements of cardiac output by bioimpedance and thermodilution in severely ill surgical patients Crit Care Med 1986 14:933–935. [PubMed: 3769504]

111.

LabadidiLababidi Z, Ehmke DA, Durnin RE, et al Evaluation of impedance cardiac output in children Pediatrics 1971 47:870–879. [PubMed: 4929902]

112.

Preiser JC, Doper A, Parquier JN, et al Transthoracic electrical bioimpedance versus thermodilution technique for cardiac measurement during mechanical ventilation Intensive Care Med 1989 15: 221–223. [PubMed: 2745864]

113.

Kinnen E Cardiac output from transthoracic impedance variations Ann NY Acad Sci 1970 170:747–756.

114.

Baker LE, Geddes LE The measurement of respiratory volumes in animals and man with use of electrical impedance Ann NY Acad Sci 1970 170:667–668.

115.

Patterson RP, Wang L, Raza B, et al Mapping the cardiogenic impedance signal on the thoracic surface Med Biol Eng Comput 1990 28: 212–216. [PubMed: 2377002]

116.

Sherwood A, Carter LS Jr, Murphy CA Cardiac output by impedance cardiography: Two alternative methodologies compared with thermodilution Aviat Space Environ Med 1991 62:116–122. [PubMed: 2001207]

117.

Sun HH, Van De Water JM Editorial Ann Biomed Eng 1989 17:457–458. [PubMed: 2610424]

118.

Veigl VL, Judy WV Reproducibility of hemodynamic haemodynamic measurements by impedance cardiography Cardiovasc Res 1983 17:728–734. [PubMed: 6661744]

119.

La Mantia KR, O'Connor T, Barash PG Comparing methods of measurement: An alternative approach Anesthesiology 1990 72:1–3. [PubMed: 2339794]

120.

Judy WV, Langley Fm, McCowen KD, et al Comparative evaluation of the thoracic impedance and isotope di lution methods for measuring cardiac output Aerospace Med 1969 40:532–536. [PubMed: 4890370]

121.

Patterson RP, Wang L, Raza SB Impedance cardiography using band and regional electrodes in supine, sitting, and during exercise IEEE Trans Biomed Eng 1991 38:393–400. [PubMed: 1874520]

122.

Haites NE, McLennan FM, Mowat DHR, Rawles JM How far is the cardiac output? Lancet 1984 2: 1025–1027. [PubMed: 6149404]

123.

Porter JM, Shakespeare PG Cardiac output after burn injury Ann R Coll Surg Engl 1984 66: 33–35. [PMC free article: PMC2493656] [PubMed: 6691694]

124.

Miller JC, Horvath SM Impedance cardiography Psychophysiology 1978 15: 80–91. [PubMed: 625525]

125.

Musumeci V, Cardillo C, Guerrera G, et al Heterogeneity of individual changes in total peripheral resistance after monotherapy with nicardipine, enalapril or ketanserin in hypertensives at rest and during exercise J Drug Dev 1990 3: 77–85.

126.

Mickell JJ, Lucking SE, Chaten FC, et al Trending of impedance-monitored cardiac variables: Method and statistical power analysis of 100 control studies in a pediatric intensive care unit Crit Care Med 1990 18:645–650. [PubMed: 2344756]

127.

Miles DS, Gotshall RW Impedance cardiography: Non-invasive assessment of human central hemodynamics at rest and during exercise Exerc Sport Sci Rev 1989 17: 231–263. [PubMed: 2676548]

128.

Pickett BR, Buell JC Validity of cardiac output measurement by computer-averaged impedance cardiography, and comparison with simultaneous thermodilution determinations Am J Cardiol 1992 69:1354–1358. [PubMed: 1585872]

129.

Eckman LG, Milsom I, Arvidson S, et al Clinical evaluation of an ensemble-averaging impedance cardiograph for monitoring stroke volume during spontaneous breathing Acta Anaesthesiol Scand 1990; 34;190–196. [PubMed: 1693029]

130.

Nishimura RA Another measure measurement of cardiac output: Is it truly needed? J Am Coll Cardiol 1989 13:1393–1394. [PubMed: 2703620]

131.

Robin ED. Death by pulmonary artery flow-directed catheter: Time for a moratorium? Chest 1987 92: 727–731. [PubMed: 3652759]

132.

[no authors listed] Federal Register. January 17,1991 56:

AHCPR Publication No. 92-0073