A comparison of transoesophageal echocardiographic Doppler across ...

Anaesthesia, 1999, 54, pages 128–136 ................................................................................................................................................................................................................................................

A comparison of transoesophageal echocardiographic Doppler across the aortic valve and the thermodilution technique for estimating cardiac output J. Poelaert,1 C. Schmidt,2 H. Van Aken,3 F. Hinder,2 T. Mollhoff 2 and H. M. Loick 2 1 Clinical Director, Department of Intensive Care, University Hospital, De Pintelaan 185, B9000 Gent, Belgium 2 Dozent and 3 Professor and Head, Klinik und Poliklinik fu¨r Ana¨sthesiologie und operative Intensivmedizin, Westpha¨lische Wilhelmsuniversita¨t, Mu¨nster, Germany Summary

This study was undertaken in order to elucidate the differences between various planes of measurement and Doppler techniques (pulsed- vs. continuous-wave Doppler) across the aortic valve to estimate cardiac output. In 45 coronary artery bypass patients, cardiac output was measured each time using four different Doppler techniques (transverse and longitudinal plane, pulsed- and continuous-wave Doppler) and compared with the thermodilution technique. Measurements were performed after induction of anaesthesia and shortly after arrival in the intensive care unit. Optimal imaging was obtained in 91% of the patients, in whom a total of 82 measurements of cardiac output were performed. The respective mean (SD) areas of the aortic valve were 3.77 (0.71) cm2 in the transverse plane and 3.86 (0.89) cm2 in the longitudinal plane. A correlation of 0.87 was found between pulsed-wave Doppler cardiac output and the thermodilution technique in either transverse or longitudinal plane. Correlation coefficients of 0.82 and 0.84 were found between thermodilution cardiac output and transverse and longitudinal continuous-wave Doppler cardiac output, respectively. Although thermodilution cardiac output is a widely accepted clinical standard, transoesophageal Doppler echocardiography across the aortic valve offers adequate estimations of cardiac output. In particular, pulsed-wave Doppler cardiac output in both the transverse and longitudinal plane provides useful data. Keywords Measurement techniques; cardiac output, echocardiography, thermodilution. ...................................................................................... Correspondence to: Dr Jan Poelaert Accepted: 20 July 1998

Transoesophageal echocardiography (TOE) has been applied widely in ventilated patients with haemodynamic instability [1, 2]. For more than 25 years, pulmonary artery catheterisation has been the clinical standard in the management of the haemodynamically unstable patient and, in particular, for determining cardiac output. However, both methodological and patient-related theoretical and practical problems have been described with respect to estimation of cardiac output by the thermodilution technique [3]. In mechanically ventilated patients, left ventricular compliance may vary significantly throughout the ventilatory cycle [4]. The consequence is a simultaneously changing left ventricular preload, resulting in a condition of nonconstant blood flow during the thermodilution cardiac output measurement. 128

Non-invasive measurement of cardiac output by echocardiography has been evaluated by many investigators, using a two-dimensional technique or Doppler assessment [5–8]. Certain well-recognised requirements must be fulfilled: (i) the absence of turbulence; (ii) parallel orientation of the Doppler beam to the direction of the blood flow; and (iii) measurement of cross-sectional area at the level of the sampling site. Sampling the spectral Doppler signal at a certain level of the heart or in the great vessels can be used to display the instantaneous blood-flow velocities over time. Former approaches to Doppler estimation of cardiac output employed the pulmonary artery and the mitral valve as sampling sites, but showed a high failure rate [8]. Recently, a transgastric long axis view has been described [9–11], Q 1999 Blackwell Science Ltd

Anaesthesia, 1999, 54, pages 128–136 J. Poelaert et al. • Thermodilution vs. Doppler cardiac output determination ................................................................................................................................................................................................................................................

which is a prerequisite for measurement of cardiac output at the level of the left ventricular outflow tract (LVOT). With this view, the LVOT, the aortic valve and the ascending aorta are situated in a two-dimensional image in one straight line from either the transverse or longitudinal plane, permitting exact alignment of the Doppler beam to the direction of blood flow. The current study was designed to evaluate the feasibility and accuracy of determining Doppler cardiac output at the level of the LVOT using either pulsed- or continuous-wave Doppler in both a transverse and longitudinal plane. The Doppler data were obtained in various haemodynamically different situations in the peri-operative setting of coronary artery bypass surgery and compared with thermodilution cardiac output measurements. Patients and methods

Patients In 45 consecutive patients scheduled for elective coronary artery bypass surgery, both a pulmonary artery catheter and a transoesophageal echocardiographic probe were introduced for haemodynamic monitoring. All patients were ventilated mechanically. After an initial TOE study, patients with significant aortic, mitral or tricuspid valve disease were excluded. Patients who were not in sinus rhythm at the time of investigation were disregarded. Written, informed consent was obtained from all patients, and the study was approved by the ethics committee of the University Hospital of Mu¨nster. Patients received oral premedication with flunitrazepam (2 mg) 2 h prior to induction of anaesthesia. The anaesthetic technique was standardised. An infusion of propofol (1–2 mg.kg¹1) and sufentanil (1–3 mg.kg¹1) was administered, until disappearance of the eyelash reflex. Subsequently, pancuronium bromide (0.1 mg.kg¹1) was injected, facilitating tracheal intubation. Anaesthesia was maintained with a continuous infusion of propofol (1.5– 5.0 mg.kg¹1.h¹1) and sufentanil (3–5 mg.kg¹1.h¹1). A pulmonary artery catheter (Swan Ganz Catheter, BaxterEdwards Lab., Santa Ana, CA, USA) was inserted via an internal jugular vein. The appearance of a typical wedge curve on the pressure monitor indicated proper placement of the tip of the catheter in the pulmonary circulation. A 5.0/3.7 MHz multiplane transoesophageal probe (Omniplane I, Hewlett-Packard Inc., Andover, MA, USA) was introduced and connected to an Hewlett Packard Sonos 1500 echocardiograph (Hewlett Packard Inc., Andover, MA, USA). Methods Simultaneous cardiac output measurements were performed using both the thermodilution technique and Doppler Q 1999 Blackwell Science Ltd

echocardiography during apnoea. Thermodilution cardiac output measurements were performed in triplicate by the anaesthetist in charge of the patient. The anaesthetist was blinded to the results of the echocardiographic examination, as was the echocardiographer to the results of the thermodilution method. At different time points throughout the peri-operative course, a total of 90 comparative measurements in 45 patients were performed under different haemodynamic situations and at different sample times. Cardiac output was measured first under clinical steady state 15 min after the induction of anaesthesia. Second, cardiac output was determined 30 min after arrival in the ICU. Thermodilution cardiac output data were derived by injecting 10 ml of cold (4 8C) dextrose 5% solution during apnoea. The bolus was injected within 3 s. Dilution curves were analysed automatically by a computerised monitoring system (Siemens Sirecust). TOE was performed simultaneously. Blood-flow velocities throughout the LVOT were first measured in a transverse (Fig. 1) and then in a longitudinal (Fig. 2) plane. The LVOT, the aortic valve and the aortic root were visualised in a transgastric long axis view (Fig. 1), as described recently [9]. To obtain these images, the probe has to be advanced deep into the stomach. By flexing the tip of the probe anteriorly and to the left, the transducer is positioned close to the apex of the left ventricle. This image is well suited to acquisition of blood-flow velocities across the aortic valve, because the intercept angle between the ultrasound beam and orientation of the blood flow remains consistently below 208. Both pulsed- and continuous-wave Doppler signals were first obtained in the transverse plane. Care was taken to maximise the velocity signal and to reduce the wall filter to obtain complete signals aligned to the baseline. Following this measurement, the multiplane probe was switched to scan an analogous image in the longitudinal plane of the left ventricle (Fig. 2). This view, which is equivalent to the transthoracic longitudinal parasternal image, permits perpendicular alignment of the ultrasound beam to the LVOT and ascending aorta, and utilisation of the axial resolution of the imaging system. Doppler signals were attained again using the procedure described above. The aortic diameter was determined in a longitudinal plane on a two-dimensional image of the LVOT at the bases of the aortic cusps (Fig. 3). Pulsed- and continuouswave Doppler signals were recorded on standard VHS videotapes for off-line analysis. Off-line analysis For every cardiac output determination, high-quality spectral Doppler signals of instantaneous flow velocities from three consecutive beats were analysed. Each analysis comprised pulsed- and continuous-Doppler waves, as 129

J. Poelaert et al. • Thermodilution vs. Doppler cardiac output determination Anaesthesia, 1999, 54, pages 128–136 ................................................................................................................................................................................................................................................

Figure 1 Deep transgastric view obtained

by maximal flexion anteriorly and laterally of the tip of the TOE probe (AA, ascending aorta; LA, left atrium; LV, left ventricle).

explained above. Each flow-velocity spectrum was analysed by manually enveloping along the brightest border of the flow-velocity pattern to sum the instantaneous blood flow velocities (Fig. 4). The time velocity integral (TVI) is the calculated area under the Doppler curve over a specified period [12]. TVI represents the distance (cm) that one red blood cell travels with each cardiac beat. Volumetric flow as stroke volume was derived as a product of TVI and cross-sectional area (CSA). A circular shape was assumed and CSA was obtained from the formula: p(D2/4) ¼ 0.785 D, where D is the diameter of the LVOT (cm). Multiplying stroke volume and heart rate (HR) provides cardiac output: Doppler cardiac output (ml.min¹1) ¼ TVI (cm) × CSA (cm2) × HR (beat.min¹1).

All measurements were performed by off-line analysis using the software of the echocardiograph. All components after systole were ignored, as they most probably originated from reflection waves against the walls of the aorta. Heart rate was derived from the ECG on the echocardiograph and incorporated into the above-mentioned formula. Thus, cardiac output was calculated in four different ways: 1 transverse long axis transgastric view with pulsed spectral Doppler and left ventricular outflow tract diameter from the midoesophageal view; 2 longitudinal long axis transgastric view with pulsed spectral Doppler and left ventricular outflow tract diameter from the same imaging plane; 3 transverse long axis transgastric view with continuouswave Doppler and left ventricular outflow tract diameter from the midoesophageal view;

Figure 2 Longitudinal view from the

gastro-oesophageal junction obtained by flexion anteriorly of the tip of the TOE probe (AA, ascending aorta; AV, aortic valve; LA, left atrium; LV, left ventricle). 130

Q 1999 Blackwell Science Ltd


Figure 3 Longitudinal view at the upper level of the mediastinum, showing the outflow tract of the left ventricle (LVOT), the ascending aorta (AO) and two cusps of the aortic valve (LA, left atrium; RVOT, right ventricular outflow tract).

(A)

(B)

Figure 4 Example of pulsed- (A) and continuous-wave (B) Doppler pattern.


131


4 longitudinal long axis transgastric view with continuouswave Doppler and left ventricular outflow tract diameter from the same imaging plane. To calculate Doppler cardiac output in the longitudinal plane, the TVI was linked with the respective aortic valve diameter from the same scanning plane. Data from the transverse plane were also combined. In order to determine the reproducibility and accuracy of the methods used, an independent observer (CS) reviewed the videotaped images of 40 Doppler cardiac output measurements of 12 randomly selected patients twice at different times. Statistical analysis Linear relationship between the different methods used to determine cardiac output was assessed using Pearson correlation analysis. To investigate the agreement between the different methods of cardiac output measurement and to exclude systematic errors, bias was evaluated by pairing the values. The difference between two measurements obtained from two techniques was related to their mean value according the method of Bland & Altman [13]. The mean bias and the 95% CI were calculated and displayed graphically. The variability coefficient of repeated measures was calculated as the quotient of the largest difference between repeated measurements and their average. Mean values, slopes, intercepts and the null hypothesis were tested using the Student’s t-test. Fisher’s Z-transformation was applied to compare correlation coefficients. Results of data variance were compared with the variance ratio F-test. Intra-observer and interobserver variability was assessed statistically in the same manner. A t-test was applied to compare correlation coefficients. Significance was assumed when p-values were < 0.05. Results

Optimal imaging allowing complete analysis in both transverse and longitudinal imaging planes was obtained in 41 of 45 patients (91%). In 41 patients (six female and 39 male, mean age 63 years), 82 measurements were performed. In one patient, the transverse transgastric long axis view could not be obtained due to the presence of a pronounced kyphoscoliosis. The other three failures comprised patients in whom the transverse or longitudinal image, respectively, could not be obtained properly, owing either to air or excessive fluid in the oesophagus and stomach. The mean (SD) diameter of the left ventricular outflow tract measured in the transverse and longitudinal plane was 2.18 (0.20) cm and 2.20 (0.25) cm, with respective areas of 3.77 (0.71) cm2 and 3.86 (0.89) cm2. The mean (SD) angle of deviation between blood flow and Doppler beam was 132

12 (7)8 in the transverse plane and 9 (5)8 in the longitudinal plane. A summary of the correlation graphs for the four methods of Doppler cardiac output in comparison with the thermodilution method is given in Figs 5 and 6 (pulsed-wave Doppler and continuous-wave Doppler, respectively). Bland and Altman confidence diagrams are shown simultaneously. Table 1 demonstrates the cardiac output estimated by the different methods and the respective correlation coefficients between thermodilution cardiac output, spectral pulsed-wave Doppler cardiac output and continuouswave Doppler cardiac output determinations. A correlation coefficient of 0.87 was found between spectral pulsed-wave Doppler in either transverse or longitudinal plane and thermodilution cardiac output. Correlation coefficients of 0.82 and 0.84 were found between thermodilution cardiac output and transverse and longitudinal continuous-wave Doppler cardiac output determinations, respectively. The inter- and intra-observer variability of the Doppler cardiac output measurements is are shown in Table 2.

Discussion

The current study primarily demonstrates that cardiac output data, estimated in the transverse plane either with pulsed- or continuous-wave Doppler of the LVOT, obtained from a deep transgastric view, correlate well with those derived from thermodilution cardiac output. We also showed the feasibility of determining cardiac output in an accurate manner in 91% of a series of cardiac surgical patients using this deep transgastric view [9, 14]. To the best of our knowledge, the present findings represent the first clinical demonstration using different methods of cardiac output determination by TOE in the peri-operative setting from this particular view. Thermodilution is widely and reliably used as a clinical standard to measure cardiac Table 1 Cardiac output estimation and correlation between thermodilution technique and pulsed- or continuous-wave Doppler in transverse and longitudinal plane. The diameter of the left ventricular outflow tract was measured in the longitudinal image. Cardiac output (l . min¹1)

Mean (SD)

TD

PW t

PW l

CW t

CW l

TD PW t PW l CW t CW l

5.41 (1.97) 4.87 (2.05) 5.10 (2.01) 5.62 (2.45) 5.80 (2.28)

1.00 0.87 0.87 0.82 0.84

1.00 0.77 0.85 0.48

1.00 0.74 0.70

1.00 0.55

1.00

CW, continuous wave Doppler; TD, thermodilution technique; PW, pulsed wave Doppler; t, transverse plane; l, longitudinal plane.



Figure 5 (A) Correlation graph of the comparison between pulsed-wave Doppler echocardiography in transverse plane vs. the

thermodilution technique. (B) Correlation graph of the comparison between pulsed-wave Doppler echocardiography in longitudinal plane vs. the thermodilution technique. (C) Difference between cardiac output measured by pulsed-wave Doppler in transverse plane and by thermodilution vs. mean of cardiac output estimated by Doppler and thermodilution. (D) Difference between cardiac output measured by pulsed wave Doppler in longitudinal plane and by thermodilution vs. mean of cardiac output estimated by Doppler and thermodilution.

Table 2 Intra-observer and interobserver variability of cardiac

output measurements. Cardiac output (l . min¹1)

Mean (SD)

1

2 vs. 1

2 vs. 2

1 2 vs. 1 2 vs. 2

4.03 (1.47) 3.93 (1.47) 4.02 (1.54)

1.00 0.91 0.92

1.00 0.93

1.00

1, observer 1; 2, observer 2.


output in the critical-care setting [15, 16]. The thermodilution technique has been shown to be accurate when a stationary flow is present and a series of precautions are taken into account [3, 17]. With respect to the recently discussed association between thermodilution determination of cardiac output in the initial care of critically ill patients and increase in mortality, less invasive techniques may be preferable. It is possible for TOE to provide an instantaneous diagnosis of hypovolaemia, qualitative assessment of contractility and valvular function in a very short time at the bedside [14, 18–20], permitting differentiation between cardiac and noncardiac causes of an acute haemodynamic derangement with a very low probability of morbidity [21]. 133


Figure 6 (A) Correlation graph of the comparison between continuous-wave Doppler echocardiography in transverse plane vs. the

thermodilution technique. (B) Correlation graph of the comparison between continuous-wave Doppler echocardiography in longitudinal plane vs. the thermodilution technique. (C) Difference between cardiac output measured by continuous-wave Doppler in transverse plane and by thermodilution vs. mean of cardiac output estimated by Doppler and thermodilution. (D) Difference between cardiac output measured by continuous-wave Doppler in longitudinal plane and by thermodilution vs. mean of cardiac output estimated by Doppler and thermodilution.

Both pulsed- and continuous-wave Doppler have been used to estimate cardiac output. The pulsed-wave modality relies on single-beam technology. The ultrasound pulses are transmitted and the back-scattered ultrasound is received by the same crystal with some delay. Continuous-wave Doppler employs different pulse transmission and receiver crystals and, hence, permits measurement of higher velocities. The comparison of the frequency shift is analysed and displayed after reception of the back-scattered signal. The major advantage consists of the ability to measure the highest velocities. These recordings contain flow information of all particles along the entire ultrasound beam and, hence, no selective sampling (range resolution) exists [12]. Many authors have assessed the accuracy of pulsed- and 134

continuous-wave Doppler cardiac output at various sampling sites (Table 3). Sampling at the level of the mitral valve appears to offer less accurate measurements of cardiac output [22, 23] owing both to the changing shape of the mitral annulus and the variation (6 12%) of the mitral annulus area throughout the cardiac cycle [24]. The pulmonary artery has been the preferred sampling site of other investigators [7, 8, 25–27] The failure rate of 10–29% (Table 3) is explained both by problems in measurement of the diameter of the vessel and by interference from the left main bronchus. It follows that Doppler cardiac output measured at the level of the aortic orifice area is an appealing technique, being relatively easily accessible [9, 10] and highly reproducible [28]. Q 1999 Blackwell Science Ltd


Table 3 Comparative analysis of the litera-

ture on human studies.

Site

First author

n

Failure (%)

Doppler approach

r

Reference

PA PA PA PA MV LVOT LVOT LVOT LVOT AA RVOT

Savino Muhuideen Gorcsan III Izzat Shimamoto Katz Darmon Feinberg Owen* Descorps-Decle`re Maslow

45 99 15 19 65 57 109 33 64 28 45

24 29 13 10 NA 12 2 12 9 7 16

CW short axis PW short axis CW short axis PW short axis PW four chamber CW transverse long axis CW transverse long axis PW longitudinal long axis PW transverse long axis PW transverse long axis PW longitudinal long axis

0.93 0.65 0.83 0.95 NA 0.91 0.94 0.91 0.95 0.98 0.98

[8] [7] [27] [26] [22] [9] [10] [11] [31] [32] [33]

CW, continuous-wave Doppler; PW, pulsed-wave Doppler; LVOT, left ventricular outflow tract; MV, mitral valve; AA, ascending aorta; PA, pulmonary artery; RVOT, right ventricular outflow tract; NA, not available; *limited to comparison with transthoracic echocardiography.

Although the noninvasive and safe character of TOEDoppler is attractive, certain hypotheses must be made. With this technique, the sampling site is assumed to be a fixed cylinder with a constant diameter [29]. The Doppler method inherently implies that the Doppler beam is parallel to the blood flow through the vessel. The deep transgastric view in the transverse plane allows visualisation of the complete LVOT and the aortic valve in one straight line, parallel to the Doppler beam [9]. Angles of incidence greater than 208 between the Doppler beam and the blood flow may lead to underestimation of the velocity and flow measurement [30]. It is essential that a low signalto-noise ratio is present and that the measured velocity is maximised. Finally, the orifice used as outflow is accepted as remaining constantly circular. In patients undergoing coronary artery bypass grafting, Darmon et al. [10] compared the data obtained from continuous-wave Doppler across the aortic valve with the thermodilution technique. They noted that cardiac output measured with continuous-wave Doppler echocardiography was related more closely to thermodilution cardiac output when a triangular measured, rather than a circular calculated, aortic valve opening was used. This technique appeals from a scientific point of view. Nevertheless, with this study we propose a clinically and easily available technique for estimating cardiac output through the aortic valve using either pulsedor continuous-wave Doppler. In conclusion, this study demonstrates the feasibility of measuring cardiac output reliably and offers a valuable alternative to the thermodilution technique, in particular in view of the additional value of TOE. Either pulsed- or continuous-wave Doppler can be used to measure cardiac output in patients after cardiac surgery without tricuspid regurgitation or aortic valvular pathology from a deep transgastric image in a transverse plane with a low failure rate. Q 1999 Blackwell Science Ltd

References 1 Heidenreich PA, Stainback RF, Redberg RF, Schiller NB, Cohen NH, Foster E. Transesophageal echocardiography predicts mortality in critically ill patients with. Journal of the American College of Cardiology 1995; 26: 152–8. 2 Poelaert JI, Trouerbach J, De Buyzere M, Everaert J, Colardyn FA. Evaluation of transesophageal echocardiography as a diagnostic and therapeutic aid in a critical care setting. Chest 1995; 107: 774–9. 3 Jansen JRC. Thermodilution method for the clinical assessment of cardiac output. Intensive Care Medicine 1995; 21: 691–7. 4 Hansen RM, Viquerat CE, Matthay MA et al. Poor correlation between pulmonary arterial wedge pressure and left end-diastolic volume after coronary artery bypass graft surgery. Anesthesiology 1986; 64: 764–70. 5 Dittman H, Voelker W, Karsch KR, Seipel L. Influence of sample site and flow area on cardiac output measurements by Doppler echocardiography. Journal of the American College of Cardiology 1987; 10: 818–23. 6 Hozumi T, Shakado M, Applegate A, Shah PM. Accuracy of cardiac output estimation with biplane transesophageal echocardiography. Journal of the American Society of Echocardiography 1993; 6: 62–8. 7 Muhiudeen IA, Kuecherer HF, Lee E, Cahalan MK, Schiller NB. Intraoperative estimation of cardiac output by trasnesophageal pulsed Doppler echocardiography. Anesthesiology 1991; 74: 9–14. 8 Savino JS, Troinaos CA, Aukburg S, Weiss R, Riechek N. Measurement of pulmonary blood flow with transesophageal two-dimensional and Doppler echocardiography. Anesthesiology 1991; 75: 445–51. 9 Katz WE, Gasior TA, Quinlan JJ, Gorcsan III J. Transgastric continuous-wave Doppler to determine cardiac output. American Journal of Cardiology 1993; 71: 853–7. 10 Darmon PL, Hillel Z, Mogtader A, Mindich B, Thys D. Cardiac output by transesophageal echocardiography using 135


11

12

13

14

15

16

17

18

19

20

21

22

136

continuous-wave Doppler across the aortic valve. Anesthesiology 1994; 80: 796–805. Feinberg MS, Hopkins WE, Davila-Roman VG, Barzilai B. Multiplane transesophageal echocardiographic Doppler imaging accurately determines cardiac output measurements in critically ill patients. Chest 1995; 107: 769–73. Nishimura RA, Miller FA Jr, Callahan MJ. Doppler echocardiography: theory, instrumentation, technique and application. Mayo Clinic Proceedings 1985; 60: 321–43. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–10. Poelaert J, Schmidt C, Colardyn F. Transoesophageal echocardiography in the critically ill. A review of the literature. Anaesthesia 1998; 53: 55–68. Ganz W, Donoso R, Marcus H, Forrester J, Swan H. A new technique for measurement of cardiac output by thermodilution in man. American Journal of Cardiology 1971; 27: 392–6. Boldt J, Menges T, Wollbruck M, Hammerman H, Hempelmann G. Is continuous cardiac output measurement using thermodilution reliable in the critically ill patient? Critical Care Medicine 1994; 22: 1913–18. Latson TW, Whitten CW, O’Flaherty D. Ventilation, thermal noise, and errors in cardiac output measurements after cardiopulmonary bypass. Anesthesiology 1993; 79: 1233–43. Berqvuist BD, Bellows WH, Leung JM. Transesophageal echocardiography in myocardial revascularisation: II. Influence on intraoperative decision making. Anesthesia and Analgesia 1995; 82: 1139–45. Sheikh KH, de Bruijn NP, Rankin JS, Stanley T, Wolfe WG, Kisslo J. The utility of transesophageal echocardiography and Doppler color flow imaging in patients undergoing cardiac valve surgery. Journal of the American College of Cardiology 1990; 15: 363–72. Poelaert J, Schmidt C, Van Aken H, Colardyn F. Transoesophageal echocardiography in critically ill patients. A comprehensive approach. European Journal of Anesthesiology 1997; 14: 350–8. Daniel WG, Erbel R, Kasper W et al. Safety of transesophageal echocardiography. A multicenter survey of 10,419 examinations. Circulation 1991; 83: 817–21. Shimamoto H, Kito H, Kawazoe K, Fujita T, Shimamoto Y. Transoesophageal Doppler echocardiographic

23

24

25

26

27

28

29

30

31

32

33

measurement of cardiac output by the mitral annulus method. British Heart Journal 1992; 68: 510–15. Estagnasi P, Djedaini K, Mier L, Coste E, Dreyfus D. Measurement of cardiac output by transesophageal echocardiography in mechanically ventilated patients. Comparison with thermodilution. Intensive Care Medicine 1997; 23: 753–9. Reynolds T. Measure the gap! A proposed simplified approach for measuring the descent of the base of the left ventricle. Journal of the American Society of Echocardiography 1997; 10: 818–21. Roewer N, Bednardz F, Schulte am Esch J. Continuous measurement of intracardiac and pulmonary blood flow velocities with transesophageal pulsed Doppler echocardiography: technique and initial clinical experience. Journal of Cardiothoracic Anesthesia 1987; 1: 418–28. Izzat MB, Regragui IA, Wilde P, Angelini GD, Bryan AJ. Transesophageal echocardiographic measurements of cardiac output in cardiac surgical patients. Annals of Thoracic Surgery 1994; 58: 1486–9. Gorcsan III J, Diana P, Ball BA, Hattler BG. Intraoperative determination of cardiac output by transesophageal continuous wave Doppler. American Heart Journal 1992; 123: 171–6. Ihlen H, Amlie J, Dale J et al. Determination of cardiac output by Doppler echocardiography. British Heart Journal 1984; 51: 54–60. Greenfield JC, Patel DC. Relation between pressure and diameter in ascending aorta of man. Circulation Research 1962; 10: 778–81. Bolger AF, Eidenvall L, Ask P, Loyd D, Wranne B. Understanding continuous-wave Doppler signal intensity as a measure of regurgitant severity. Journal of the American Society of Echocardiography 1997; 10: 613–22. Owen AN, Simon P, Moidl R et al. Measurement of aortic flow velocity during transesophageal echocardiography in the transgastric five-chamber view. Journal of the American Society of Echocardiography 1995; 8: 874–8. Descorps-Decle`re A, Smail N, Vigue B et al. Transgastric pulsed Doppler echocardiographic determination of cardiac output. Intensive Care Medicine 1996; 22: 34–8. Maslow A, Comunale ME, Haering JM, Watkins J. Pulsed wave Doppler measurement of cardiac output from the right ventricular outflow tract. Anesthesia and Analgesia 1996; 83: 466–71.
