1Institute of Anaesthesiology, 2Clinic for Cardiovascular Surgery and 3Department of Internal Medicine, University Hospital, CH-8091 Zurich, Switzerland*Corresponding author
Accepted for publication: March 20, 2000
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Br J Anaesth 2000; 85: 37988
Keywords: monitoring, transoesophageal echocardiography; heart, transoesophageal echocardiography; heart, ventricles; surgery, cardiovascular
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The aim of this study was to validate in humans intraoperative area measurements obtained by another type of automated echocardiographic system, using digital echo quantification (DEQ) against simultaneous volume measurements, obtained by an LV CC. Comparison of the two methods during cardiac surgery was undertaken to investigate the strengths and limitations of the intraoperative use of the DEQ method in patients with normal systolic LV function. We hypothesized that, during steady state and acute changes in loading, area changes measured with transoesophageal echocardiography at the mid- papillary short axis level would reflect accurately changes in CC volume.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Anaesthesia and operative techniques
Preoperative medication was continued until the morning of surgery. Patients were premedicated with oral flunitrazepam 0.020.03 mg kg1. In the operating room, a peripheral venous cannula and a fluid-filled femoral artery catheter (Seldicath, 1.3 mm; Plastimed, St-Leu-La-Forêt, France) were inserted under local anaesthesia. The arterial catheter was connected to a single-use transducer (Deltran II; Utah Medical Inc., Midvale, UT, USA). Continuous ECG (leads II and V5), arterial pressure and transcutaneous oxygen saturation monitoring were installed and connected to the monitoring system (Solar 8000; Marquette-Hellige GmbH, Freiburg, Germany). A continuous three-lead ECG was connected to the echo machine for DEQ signal-processing. Anaesthesia was induced with flunitrazepam (total dose 0.010.05 mg kg1) and fentanyl (total dose 2555 µg kg1). Pancuronium 0.1 mg kg1 was given to facilitate tracheal intubation, and patients lungs were mechanically ventilated with a Servo 900C ventilator (Siemens Elema AB, Upplands Väsby, Sweden). Anaesthesia was maintained with propofol by continuous infusion (1.84.5 mg kg1 h1). Central venous and pulmonary artery catheters (Baxter Intellicath 7.5 F; Baxter Healthcare Corp., Irvine, CA, USA) were placed via the right internal jugular vein, and the transoesophageal echo probe (see below) was inserted.
After sternotomy and pericardiotomy, epicardial pacemaker wires were fixed on the right atrium. After systemic heparinization, the aorta was cannulated, the CC was inserted via the right upper pulmonary vein and measurements were performed before right atrial cannulation.
Drug therapy for weaning from CPB consisted of nitroglycerine 0.6 (95% CI 0.260.94) µg kg1 min1 (six patients) or nifedipine 0.046 and 0.064 µg kg1 min1 (in two cases). In one case, norepinephrine 0.037 µg kg1 min1 was given for low systemic vascular resistance; no other catecholamine therapy was necessary in the study group. Propofol 3.3 (95% CI 2.64.1) mg kg1 h1 was used to maintain hypnosis during rewarming, chest closure and transfer to the intensive care unit.
Echocardiographic DEQ
Transoesophageal echocardiography with DEQ was done with an Omniplane 5 MHz echo probe and a Vingmed CFM 800 system (Vingmed Sound, Horten, Norway). We reduced overall gain, adjusted time gain, lateral gain compensations and compress levels and narrowed the image sector for optimal DEQ conditions and the highest possible frame rate (3050 frames s1).14 The DEQ uses high-quality two-dimensional (2D) images in a transgastric mid-papillary view; raw backscatter data are stored in memory and are subjected to a dynamic processing algorithm that applies statistically based edge-detection enhancement. A contour identifying the interface between blood and myocardium is superimposed on the real-time echocardiographic image and used for automated area calculation; it is displayed as a real-time trace in a manually tracked region of interest. The trace is exported as an analogue signal to an analoguedigital (A/D) converter and recorded with a sample rate of 1000 Hz on a workstation.
Conductance catheter
A 7 French, 12 electrode LV CC (Cordis Europa NV, LJ Roden, The Netherlands) with an integrated tip manometer (Sentron BV, AC Roden, The Netherlands) was introduced via the right upper pulmonary vein through the mitral valve with the help of an LV vent introducer and fixed with a purse-string suture. The catheter was advanced until four correct volume tracings were displayed on the screen of the CC monitoring system; its position within the left ventricle was controlled for centred radial position by transoesophageal echocardiography in a longitudinal view of the left ventricle. Tip manometer catheters were immersed in saline, then calibrated relative to atmospheric pressure; specific blood resistivity (), which is defined as 1/blood conductance (
), was measured before and after CPB with the help of a blood sample and the measuring cuvette supplied with the CC analysis system (Leycom Sigma 5 DF; CardioDynamics BV, Zoetermeer, The Netherlands). Injections of 12.5% saline into the distal port of the pulmonary artery catheter for determination of parallel conductance2 3 were performed before and after CPB, immediately after determination of cardiac output with thermodilution (three to five injections of iced saline).
Data acquisition and registration
The ECG and femoral arterial pressure signals were amplified by the Solar 8000 monitoring system with a direct output to the A/D system. All analogue signals of interest (DEQ area, five CC volumes, micromanometer LV pressure, femoral arterial pressure and ECG were digitized with a sample rate of 1000 Hz for display and stored with the help of an A/D card and customized software under the environment of Superscope II (MacAdios; GW Instruments, Somerville, MA, USA) on an Apple Macintosh computer.
Compensation of the time used by the echocardiography system for calculation and display of DEQ was achieved by time-advancing the DEQ area curve until the first three to five beats of one 15 beat run were congruent to the CC volume signal. Compensation of the general, system-related, and the individual physiological time delay of a fluid-filled catheter against a rapid response micromanometer was performed by advancing the femoral arterial pressure curve until its ejection phases exactly matched the ejection phases of the LV pressure curve of the micromanometer.
To reduce electrical noise, all signals were filtered with a low-pass HAM filter with a 50 Hz frequency cut-off, signals >50 Hz being attenuated and those 50 Hz left unchanged.
Protocol
Simultaneous CC volume, DEQ area, LV and femoral arterial pressure signals were obtained during apnoea at end-expiration with zero positive airway pressure for 30 s. After a baseline measurement, one or two runs with injection of hypertonic saline into the distal port of the pulmonary artery catheter were then recorded. Acute alterations in preload were induced by partially clamping the inferior vena cava and afterload was increased by a partial aortic cross-clamping manoeuvre (aortic occlusion). Before CPB and after weaning from CPB and haemodynamic stabilization, we obtained duplicate measurements 5 min apart (i) during caval (Fig. 1A) and aortic occlusion (Fig. 1B) while the patient was in sinus rhythm, and (ii) with atrial pacing at a heart rate of 90 bpm. The registered LV pressurevolume and LV pressurearea loops were analysed off-line.
|
DEQ area and CC volume were normalized based on the first end-diastolic DEQ area (100% area) and CC volume (100% volume) values (Figures 3A and 4); values measured subsequently were expressed as a percentage of the initial end-diastolic values.
|
|
For visual analysis, the DEQ signal (in mV) was calibrated for amplitude, baseline offset and time delay; the femoral arterial pressure curve was adjusted in time as described above. The LV pressurearea and LV pressurevolume loops were displayed for the selected beats. End-systolic and end-diastolic points of each loop were analysed. End-systole was measured at aortic valve closure, as determined by the femoral arterial pressure curve (Fig. 1), and at end-diastole at the end of the diastolic pressure plateau.
End-systolic and end-diastolic slopes of LV pressure plotted against DEQ area and CC volume, respectively, were calculated by linear regression analysis. All values of interest were exported to a data sheet for statistical analysis.
Statistics
Raw data are expressed as median and 95% CI; BlandAltman analyses in the text are reported as mean (SD). BlandAltman analyses and one linear regression analysis were used to present correlation of DEQ area and CC volume data. The significance of differences between measurements was determined using the Wilcoxon ranked sign test for paired comparisons and the MannWhitney U-test for unpaired comparisons. P values <0.05 were considered significant. Calculations were made with Statview 4.5 (Abacus Concepts Inc, Berkeley, CA, USA).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
A total of 112 measurement runs of 15 beats each (median 14, 95% CI 1214 runs per patient) was analysed during manoeuvres for acute loading changes (Fig. 1). For visual comparison with the CC volume data, amplification of the area signal was 9.5 (95% CI 8.410.6) times, baseline offset 7.2 (95% CI 4.89.6) mV and the area signal time offset compared with the CC volume signal was 194 (95% CI 202 to 168) ms.
Steady-state correlation of DES area and CC volume measurements
During steady state, curves of area changes as registered by DEQ showed a close agreement with curves obtained by the CC technique (Fig. 2): Figure 2A shows an example of parallel registration and Fig. 2B a BlandAltman analysis of 32 steady-state recordings at 5 ms intervals; the mean (SD) difference between normalized DEQ area and normalized CC volume was 0.6 (2.5)%.
|
Correlation during interventions
DEQ area versus CC volume
Calibrated absolute DEQ area and CC volume values are presented as box plots (Fig. 3A,B) and as linear regression analyses (Fig. 3C,D). Numeric values are shown in Table 1.
|
Figure 4 shows the same analysis for end-systole and end-diastole separately (panels A and B, respectively). Agreement was significantly less in the cava occlusion manoeuvre. The end-systolic difference was 0.5 (3.7)% and 3.9 (4.4)% during aortic and cava occlusion, respectively (P<0.0001 for absolute differences); end-diastolic difference was 1.3 (4)% and 0.2 (5.7)% during aortic and cava occlusion, respectively (P<0.0001 for absolute differences).
Differences between pressurearea and pressurevolume slopes
Differences between slopes of elastance (end-systolic LV pressure in mm Hg)/(DEQ area in (cm2x10)) and (end-systolic LV pressure in mm Hg)/(CC volume in ml) during the two types of intervention are depicted in Fig. 5. These differences were significantly but not uniformly influenced by the type of clamping (aortic or cava occlusion). Elastance slope differences were significantly smaller during aortic occlusion (mean (SD) 0.2 (3.9) than during cava occlusion (2.1 (8.4); P=0.03), whereas the slope differences for end-diastolic LV pressure plotted against DEQ area and CC volume, respectively, were higher during aortic occlusion (0.11 (0.85)) than during cava occlusion (0.26 (0.65); P=0.0013). The latter significance is only statistically important because the SDs of differences were very small with both manoeuvres.
|
Effects of CPB and atrial pacing
Differences between preoperative and postoperative results are presented in Table 1. Pooled end-systolic and end-diastolic DEQ area and CC volume values were significantly larger after CPB than before CPB, and end-diastolic LV pressure was slightly, but significantly, higher.
The end-systolic LV pressure/DEQ area and LV pressure/CC volume slopes were the same before and after CPB; the end-diastolic LV pressure/CC volume slope increase following CPB was minimal, but statistically significant. DEQ LV fractional area change, CC LV ejection fraction and thermodilution measured stroke volume all decreased significantly after CPB.
The effects of atrial pacing are summarized in Table 2. Before CPB period, a heart rate of 90 beats min1, induced by atrial pacing, led to end-systolic and end-diastolic LV pressures being lower than those at spontaneous, lower frequency. All these differences were statistically significant, but very small. After CPB, only end-diastolic LV pressure decreased significantly when atria were stimulated at 90 beats min1.
|
End-systolic pressurearea and pressurevolume slopes were not influenced by atrial pacing, either before or after CPB. End-diastolic pressurevolume slopes decreased slightly during atrial pacing before but not after CPB.
Early postoperative outcome
No early mortality, no new regional wall motion abnormalities (as assessed by intraoperative transoesophageal echocardiography) and no perioperative myocardial infarction (as assessed by ECG and cardiac enzymes) occurred. In one patient, transoesophageal echocardiographic examination led to a non-penetrating muco-serosal lesion of the oesophagogastric transition, as diagnosed by endoscopy in a previously unknown situs inversus abdominalis. The patient was discharged from the hospital on day 11 without sequelae. One patient developed postoperative pneumonia necessitating prolonged mechanical ventilation; his hospital stay was 18 days and his condition was good at discharge. Median postoperative stay in the intensive care unit was 1.0 day (95% CI 0.14.4 days); median hospital stay was 9.5 days (95% CI 7.211.8 days).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
This is in accordance with the study of Chen and colleagues,7 who used an identical echocardiography system but used a transthoracic approach for comparison with the conductance technique. In the apical four-chamber view, these authors based volume determination on automated digital echo quantification. We acquired the 2D images and, hence, the DEQ signal with a transoesophageal echo probe in a fixed position; patients were anaesthetized and measurements were taken at end-expiration. Stable images and good DEQ quality were, therefore, easier to obtain than with a hand-held transthoracic echo probe in spontaneously breathing patients. True cardiac volumes cannot be measured intraoperatively for validation of an echocardiographic volume or area determination. The CC measurements used in our study, however, have been shown to correlate with electromagnetic flow, thermodilution, contrast ventriculography, magnetic resonance imaging and ultrafast computed tomography in animals and in humans.3 15 17 Although automatic border detection systems slightly underestimate fractional area changes by underestimating end-diastolic and overestimating end-systolic areas, compared with ultrafast computed tomography,19 several studies have shown good correlation between the echocardiographic automated border detection and (i) real volume changes in an isolated canine heart preparation18 and (ii) CC measurements in humans.6 7
Limitations of methodology and critical evaluation of results
Conductance catheter method
Theoretical and animal studies have shown that the accuracy of the CC method depends on the distance between the catheter and the wall of the ventricle, relative to the distance between the catheter injecting electrodes.15 2023 With larger ventricular volumes and, hence, in diastole, the method thus becomes less precise. We did not find this relevant in our study; on the contrary, differences between end-systolic volume and area measurements were significantly greater than differences at end-diastole.
Another source of error was found by Szwarc and colleagues.27 Parallel conductance changes within cardiac cycles led to volume measurements being significantly different from those obtained by magnetic resonance imaging in an animal model. This was not considered to be of importance by Lankford and colleagues24 or White and colleagues25
The metallic retractor in the open chest situation seems to be another source of artefacts for CC signals. Cabreriza and colleagues26 found that isolation of the heart by latex wrapping could improve the reliability of volume signals. This isolation procedure, however, was introduced primarily to decrease the changes in parallel conductance, which is of importance only for absolute volume measurements. Another potential source of artefacts is the transoesophageal echo probe near the CC. We did not find that conductance signals were affected by whether the echo probe was emitting sound or not and this problem has not been reported by other investigators.
Echocardiography
During intraoperative monitoring, the mid-papillary short-axis view is widely accepted for the evaluation of LV filling state and contractility. Our aim was to validate intraoperative DEQ measurements in this standard plane against a well established method of LV volume determination. We found a small difference between the two methods within cardiac cycles. Following normalization of the DEQ area and correction for time delay, we found a cycle-specific pattern of under- and overestimation by DEQ (Fig. 3A), which supports the observations of Chen and colleagues.7
An important source of error in DEQ detection is the presence of the metal-containing CC in the ventricle. The echo probe had to be positioned carefully to avoid refraction and shadowing artefacts; the probe was then fixed using a specially designed apparatus. Nevertheless, DEQ area measurements were often blurred by the shadow of the CC, requiring exclusion of the catheter artefact area; particularly during the cava occlusion manoeuvre, lateral displacement of the hypovolaemic heart further impaired proper area detection. The measured LV area was therefore very small at the end of caval occlusion in a considerable number of patients, especially in those patients with small left ventricles at baseline. These artefacts probably led to spreading of the end-systolic area values during caval occlusion (Fig. 4) and the consequent spreading of elastance slope values.
Another difficulty of repeated 2D measurements is the reproducibility of the transgastric mid-papillary short-axis view. Although the probe was not moved during the measurement sequence, it cannot be guaranteed that its position was identical before and after CPB. However, each loading manoeuvre was analysed individually with regard to the areavolume relationship, and differences in area parameters before and after CPB are mirrored in CC volume measurements (Table 1).
Study patients had normal LV function without akinetic or dyskinetic wall segments and were therefore ideal for DEQ measurements. It is evident that the more regional wall motion abnormalities there are, the more bias there will be when any 2D measurements are compared with measurements of cardiac volumes.
Data analysis
Two methods were used to analyse differences between DEQ area and CC volume curves. During steady state, differences between filling and emptying patterns of LV were evaluated, the difference between each data pair being determined every 5 ms throughout the cardiac cycle. Although it would have been of interest to examine entire cardiac cycles during clamping manoeuvres too, this was not possible because of large quantities of data, so we restricted analysis of clamping manoeuvres to the clinically important time points of end-systole and end-diastole.
Calibration of the CC data for absolute volume was performed using the stroke volume derived from thermodilution measurements. This method is widely accepted; additional calibration with the saline dilution technique for parallel conductance did not seem to improve precision in our study. Variation of differences between normalized DEQ area and normalized CC volume data, however, was independent of calibration method.
We used a visually based, manually performed adjustment of the DEQ area curve. The time delay between registration of DEQ area and CC volume was variable and longer than reported with other automated border systems5 9 11 25 (Fig. 1). Technically, the time delay depends on the time the system requires for calculating DEQ; physiologically, it depends on the position of the echo probe. The more apical data are acquired, the later the area curve is displayed, as compared with the CC volume curve. End-systole and end-diastole were determined manually based on the femoral arterial pressure curve (aortic valve closure) and on LV pressure. ECG was not taken into consideration as future analyses will be performed in patients with possible bundle-branch block or ventricular pacing.
Clamping manoeuvres
Partial occlusion for acute reduction in preload is widely accepted4 5 7 9 10 and, although a partial aortic occlusion manoeuvre has been performed in animals,22 it is not to our knowledge a standard procedure in humans. By rapidly increasing LV afterload, it assesses the capacity of the left ventricle to maintain stroke volume against an increased aortic resistance without ventricular dilation. The so-called homeometric autoregulation capability28 29 should leave end-diastolic volume and pressure unchanged.
Reproducibility
It is difficult to repeat partial clamping of the inferior vena cava and the ascending aorta for quantitative comparison. We controlled the clamping manoeuvres on line and the surgeon tried to get a similar decrease or increase in LV pressure. Nevertheless, the resulting pressurearea and pressurevolume slopes at end-systole yielded considerable variations, those of the pressurearea loops being significantly greater.
Findings
One patient presented insufficient correlation during the caval occlusion manoeuvre despite excellent DEQ quality. One of the reasons was insufficient clamping of the vena cava inferior, which led to a relatively small decrease in LV volume, correctly measured by CC, but not detected by the DEQ method in the 2D short-axis view. The consecutive DEQ elastance revealed an opposite slope to the corresponding CC elastance slope. During aortic occlusion, only very small volume changes but important LV pressure changes are induced. Slopes are therefore determined more by the increase in LV pressure than by (the virtually unchanged) LV area and volume; the correlation of the pressurearea with the pressurevolume is therefore superior than during the cava occlusion manoeuvre.
Conclusions
We found that intraoperative changes in LV volume, assessed by CC, are reliably reflected by transoesophageal echocardiographic DEQ area changes of the LV short axis during steady-state conditions and during acute increase of afterload by partial occlusion of the ascending aorta.
When very small systolic area values resulted from partial occlusion of the vena cava, the decreased precision of the DEQ method made it difficult to perform reproducible end-systolic LV pressureDEQ area elastance slopes.
Patients with enlarged left ventricles and reduced ejection fraction might present more favourable conditions for LV function evaluation with vena cava occlusion manoeuvres and DEQ area measurements.
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Baan J, Jong TT, Kerkhof PL et al. Continuous stroke volume and cardiac output from intra-ventricular dimensions obtained with impedance catheter. Cardiovasc Res 1981; 15: 32834[ISI][Medline]
3 Baan J, van der Velde ET, de Bruin HG et al. Continuous measurement of left ventricular volume in animals and humans by conductance catheter. Circulation 1984; 70: 81223[Abstract]
4 Schreuder JJ, Biervliet JD, van der Velde ET et al. Systolic and diastolic pressurevolume relationships during cardiac surgery. J Cardiothorac Vasc Anesth 1991; 5: 53945[Medline]
5 Gorcsan J III, Denault A, Mandarino WA, Pinsky MR. Left ventricular pressurevolume relations with transesophageal echocardiographic automated border detection: comparison with conductance-catheter technique. Am Heart J 1996; 131: 54452[ISI][Medline]
6 Nomura K, Kurosawa H, Morita K et al. Comparison of left ventricular pressurevolume loops measured by two-dimensional echocardiography and a conductance catheter. Nippon Kyobu Geka Gakkai Zasshi 1997; 45: 3741 (English abstract)[Medline]
7 Chen CH, Nevo E, Fetics B et al. Comparison of continuous left ventricular volumes by transthoracic two-dimensional digital echo quantification with simultaneous conductance catheter measurements in patients with cardiac diseases. Am J Cardiol 1997; 80: 75661[ISI][Medline]
8 Deneault LG, Kancel MJ, Denault A et al. A system for the on-line acquisition, visualization, and analysis of pressurearea loops. Comput Biomed Res 1994; 27: 617[ISI][Medline]
9 Gorcsan J III, Romand JA, Mandarino WA, Deneault LG, Pinsky MR. Assessment of left ventricular performance by on-line pressure-area relations using echocardiographic automated border detection. J Am Coll Cardiol 1994; 23: 24252[ISI][Medline]
10 Gorcsan J III, Gasior TA, Mandarino WA, Deneault LG, Hattler BG, Pinsky MR. Assessment of the immediate effects of cardiopulmonary bypass on left ventricular performance by on-line pressure-area relations. Circulation 1994; 89: 18090[Abstract]
11 Gorcsan J III, Denault A, Gasior TA et al. Rapid estimation of left ventricular contractility from end-systolic relations by echocardiographic automated border detection and femoral arterial pressure. Anesthesiology 1994; 81: 55362[ISI][Medline]
12 Denault AY, Gorcsan J III, Mandarino WA, Kancel MJ, Pinsky MR. Left ventricular performance assessed by echocardiographic automated border detection and arterial pressure. Am J Physiol 1997; 272: H13847
13 Mandarino WA, Pinsky MR, Gorcsan J III. Assessment of left ventricular contractile state by preload-adjusted maximal power using echocardiographic automated border detection. J Am Coll Cardiol 1998; 31: 8618[ISI][Medline]
14 Bednarz JE, Marcus RH, Lang RM. Technical guidelines for performing automated border detection studies. J Am Soc Echocardiogr 1995; 8: 293305[Medline]
15 Szwarc RS, Laurent D, Allegrini PR, Ball HA. Conductance catheter measurement of left ventricular volume: evidence for nonlinearity within cardiac cycle. Am J Physiol 1995; 268: H14908.
16 Odake M, Takeuchi M, Takaoka H, Hata K, Hayashi Y, Yokoyama M. Determination of left ventricular volume using a conductance catheter in the diseased human heart. Eur Heart J 1992; 13 (Suppl E): 227[Abstract]
17 Akaishi M, Matsubara T, Abe S, Goto S, Yokozuka H, Handa SC. Conductance catheter for determining left ventricular volume and end-systolic pressurevolume relations: its clinical application. J Cardiol 1992; 22: 53947[Medline]
18 Gorcsan J III, Morita S, Mandarino WA et al. Two-dimensional echocardiographic automated border detection accurately reflects changes in left ventricular volume. J Am Soc Echocardiogr 1993; 6: 4828[Medline]
19 Marcus RH, Bednarz J, Coulden R, Shroff S, Lipton M, Lang RM. Ultrasonic backscatter system for automated on-line endocardial boundary detection: evaluation by ultra fast computed tomography. J Am Coll Cardiol 1993; 3: 83947
20 Woodard JC, Bertram CD. Effect of radial position on volume measurements using the conductance catheter. Med Biol Eng Comput 1989; 27: 2532[ISI][Medline]
21 Steendijk P, Van der Velde ET, Baan J. Left ventricular stroke volume by single and dual excitation of conductance catheter in dogs. Am J Physiol 1993; 264: H2198207
22 Boltwood CM, Appleyard RF, Glantz SA. Left ventricular volume measurement by conductance catheter in intact dogs. Circulation 1989; 80: 136077.[Abstract]
23 Hettrick DA, Battocletti JH, Ackmann JJ, Linehan, Warltier DC. Effects of physical parameters on the cylindrical model for volume measurements by conductance. Ann Biomed Eng 1997; 25: 12634[ISI][Medline]
24 Lankford EB, Kass DA, Maughan WL, Shoukas AA. Does volume catheter parallel conductance vary during a cardiac cycle? Am J Physiol 1990; 258: H193342
25 White PA, Chaturvedi RR, Shore D, Lincoln C, Szwarc RS, Bishop AJ et al. Left ventricular parallel conductance during cardiac cycle in children with congenital heart disease. Am J Physiol 1997; 273: H295302
26 Cabreriza SE, Dean DA, Jia CX, Dickstein ML, Spotnitz HM. Electrical isolation of the heart. Stabilizing parallel conductance for left ventricular volume measurement. ASAIO J 1997; 43: M50914[ISI][Medline]
27 Szwarc RS, Mickleborough LL, Mizuno SI, Wilson GJ, Liu P, Mohamed S. Conductance catheter measurements of left ventricular volume in the intact dog: parallel conductance is independent of left ventricular size. Cardiovasc Res 1994; 28: 2528[ISI][Medline]
28 Sarnoff SJ, Mitchell JH, Gilmore JP, Remensnyder JP. Homeometric autoregulation in the heart. Circ Res 1960; 8: 107791[ISI]
29 Baan J, Van Der Velde ET, Steendijk P. Ventricular pressurevolume relations in vivo. Eur Heart J 1992; 13 (Suppl E): 26.[ISI][Medline]