Selective high thoracic epidural anesthesia (HTEA) of the upper five thoracic sympathetic segments using local anesthetics offers good pain relief in patients with acute myocardial infarction or unstable angina pectoris. In addition to an antianginal action, which is attributable to the blockade of sympathetic afferent nerves, antiischemic effects have also been demonstrated (1). The antiischemic effects of the inhibition of sympathetic nervous outflow to the heart are supposed to arise from changes in the major determinants of myocardial oxygen demand because it reduces heart rate (HR) and preload and afterload of the left ventricle (LV) without affecting coronary perfusion pressure (CPP) (2). Furthermore, HTEA attenuates the paradoxical vasoconstrictor response that has been observed at the site of atherosclerotic lesions (3) and increases the luminal diameter of dynamic stenoses of epicardial coronary arteries (4). Thus, HTEA is assumed to alleviate myocardial ischemia by improving global myocardial oxygen balance and by redistributing myocardial blood flow to vulnerable regions. Both effects of HTEA may result in an improvement of overall systolic and diastolic LV function.
Despite several previous clinical and experimental studies, questions remain about the effect of HTEA on systolic LV function, which has variably been reported to be unchanged (5), impaired (6), or even improved (4) in healthy individuals and in patients with coronary artery disease (CAD). Moreover, there have been no studies published on the effect of HTEA on diastolic function in CAD patients, although it is well appreciated that a change in diastolic function is the first hemodynamic manifestation of myocardial ischemia and that subclinical ischemia can alter LV relaxation, filling, and distensibility in the presence of normal systolic function (7).
The objective of this study was to investigate the effects of sympathetic blockade by HTEA on systolic and diastolic LV function. LV function was quantified using a combined systolic/diastolic Doppler echocardiographic index (myocardial performance index [MPI]) (8), and additional echocardiographic and hemodynamic variables, which specifically address systolic (e.g., fractional area change [FAC]) (9) or diastolic (e.g., intraventricular flow propagation velocity [Vp]) (10) function of the LV.
The study was approved by the IRB of the University Hospital of Münster, Germany. Thirty-seven consecutive patients who were scheduled for coronary artery bypass surgery were prospectively enrolled in the investigation after written informed consent had been obtained. Patients were not considered for inclusion if they had (a) supraventricular or ventricular rhythm disturbances, (b) a history of significant valvular disease, (c) clinical instability or typical chest pain occurring at rest upon arrival in the operating room, or (d) any coagulation disorder or medication interfering with the insertion of an epidural catheter. All patients received their ordinary cardiac medication at 6:30 am on the day of the investigation. All but two patients were taking β-adrenoceptor-blocking drugs; 9 were receiving calcium entry blockers, 17 long-acting nitrates, and 12 angiotensin-converting enzyme inhibitors.
The epidural catheter (19-gauge; Arrow International Inc., Reading, PA) was inserted on the day before surgery. Either the T1-2 or the T2-3 interspace was chosen, as described elsewhere (11). No neurologic sequelae were observed, nor did multiple needle passes with bloody taps occur. After insertion of the epidural catheter, the patients were transferred back to the ward.
On the day of the investigation, IV access and invasive arterial blood pressure monitoring by catheterization of the radial artery were established. A continuous IV infusion of NaCl 0.9% (body weight, 5 mL/kg) was started. Automated ST segment analysis at J + 60 ms was instituted (Hellige Marquette Solar 8000 Patient Monitor; Marquette Medical Systems, Milwaukee, WI). Under local anesthesia, a 7.5F quadruple-lumen, balloon-tipped, flow-directed pulmonary artery catheter (PAC; Baxter Edwards, Irvine, CA) was advanced via the right or left internal jugular vein. Thereafter, the patients were allowed to rest for at least 15 min.
Central neuraxial blockade was instituted after the simultaneous recording of echocardiographic and hemodynamic variables (baseline). Bupivacaine 0.075 mL/kg 0.5% was given as an initial dose to achieve sensory blockade from T1-5. The level of the blockade was tested 30 min after the epidural injection by assessing both temperature and pinprick discrimination. The upper and lower sensory levels were recorded. Additional doses of epidural anesthetic solution were administered as a bolus if the spread of sensory blockade did not completely include segments T1-5. The level of motor blockade was verified by checking finger grip (C8/T1), hand flexion (C7/C8), and elbow flexion (C5/C6) (ESSAM score) (12). Echocardiographic and hemodynamic measurements were again recorded after achievement of sensory blockade at the T1-5 level (HTEA).
Standard transthoracic two-dimensional, pulsed, color-flow, and color M-mode Doppler echocardiographic examinations were performed with a System FiVe or Vivid 7 ultrasound machine (GE Medical Systems, Milwaukee, WI) equipped with a multifrequency phased-array transducer. Recordings were stored digitally and on super-VHS videotape for subsequent off-line analysis with EchoPac software. Three consecutive beats were measured and averaged for each two-dimensional and Doppler variable by two independent investigators.
LV dimensions were measured at the mid-ventricular level from two-dimensional images obtained in the parasternal short-axis view. LV end-diastolic area (EDA) and FAC were determined per published criteria (9). Afterload was assessed by an index of wall stress (end-systolic meridional wall stress [σm(es)], as described by Reichek et al. (13), according to the formula:
where BPsyst is systolic arterial blood pressure (mm Hg), ESD is LV end-systolic diameter (cm), and ESWT is end-systolic wall thickness (cm).
Mitral inflow was recorded at the mitral valve tips from the apical four-chamber view. LV outflow velocity was recorded from the apical long-axis view, with the sample volume positioned just below the aortic ring. The mitral inflow and LV outflow velocity curves were analyzed with respect to Doppler time intervals “a” and “b,” as described by Tei et al. (8) (Fig. 1). The interval “a” from cessation to onset of mitral inflow equals the sum of LV isovolumic contraction time (ICT), ejection time (ET), and isovolumic relaxation time (IRT). Ejection time “b” was measured as the duration of LV outflow. The sum of ICT and IRT was obtained by subtracting “b” from “a”. IRT was measured by subtracting the interval between the R wave of the electrocardiograph and the cessation of LV outflow from the interval between the R wave and the onset of mitral inflow. ICT was calculated by subtracting IRT from a–b. MPI was calculated as shown in Figure 1: (a–b)/b. Conventional Doppler variables of transmitral inflow velocity curve were also measured. On the basis of mitral E/A ratio (peak early filling velocity to peak velocity at atrial contraction) and deceleration time (DT) of mitral E velocity (14), patients were divided into four groups according to the grade of diastolic dysfunction: normal, impaired relaxation (E/A < 1 and DT > 220 ms), pseudonormal (1.5 > E/A > 1 and 220 ms > DT > 160 ms), and restrictive filling (E/A > 1.5 and DT > 160 ms). These different patterns correspond to increasing LV stiffness.
From the apical four-chamber view, the color Doppler sector map of the mitral inflow was adjusted to obtain the longest column of color flow from the mitral annulus to apex. The M-mode cursor was placed through the center of this flow, avoiding boundary regions. The color M-mode Vp was measured as the slope of the first aliasing velocity during early filling, from the mitral valve plane to 4 cm distally into the LV cavity (15). Doppler velocity curves were recorded during end-expiratory apnea, with a sweep speed of 100 mm/s.
To test the interobserver variability, the measurements were performed off-line from digital recordings by a second observer who was unaware of the results of the first examination. Variability was calculated as the mean percent error, derived from the difference between the two sets of measurements, divided by the mean of the observations.
The systemic and pulmonary artery pressures, including the pulmonary occlusion (POP) and central venous (CVP) pressures were registered. Waveforms were digitally processed via an analog-to-digital converter with 12-bit resolution (Dataq Instruments, Akron, OH). Cardiac output (CO) was measured using the thermodilution technique and a standardized patient monitoring system (Tram-Modul 250SL, Hellige Marquette Solar 8000 Patient Monitor; Marquette Medical Systems, Milwaukee, WI). Ten milliliters of sterile, ice-cold, isotonic (0.9%) saline solution was injected in triplicate through the right atrial lumen of the catheter, and the decrease in temperature at the distal thermistor was recorded and analyzed. Systemic vascular resistance (SVR) was calculated with standard formula. CPP was calculated as the difference between radial diastolic pressure and POP.
Results are expressed as mean ± sd. The paired Student’s t-test was used to compare mean scores of continuous variables before and after HTEA, and the Wilcoxon test was used for comparison of categories of Doppler transmitral inflow pattern. Pearson correlation analysis was performed to assess the association between the change in MPI and the change in SVR. P values of <0.05 were considered statistically significant. All calculations were performed using statistical software (SPSS 6.1; SPSS Inc, Chicago, IL).
Thirty-seven patients satisfied the admission criteria for the current investigation. After initial instrumentation, 4 patients were excluded because of unilateral spread of the epidural blockade (Patient 1), incomplete blockade of the upper 2 thoracic segments (Patient 3), new-onset atrial fibrillation (Patient 23), and failure to insert a PAC (Patient 34). Thus, 33 patients completed the protocol and composed the final study group. The age, height, and body weight of the remaining patients (26 men and 7 women) were 67 ± 7 yr, 171 ± 9 cm, and 79 ± 11 kg, respectively. From the 33 patients, 27 had 3-vessel disease and 6 had 2-vessel disease, 17 had previous myocardial infarction, 8 had previous coronary angioplasty, and 6 had previous coronary artery bypass surgery.
An average amount of 6.0 ± 1.0 mL of bupivacaine induced a sensory blockade of 9 ± 3 of the upper thoracic segments, with a mean rostral spread to C7 ± 1.5 vertebral level and a mean caudal spread to T7 ± 1.5 vertebral level. In 10 patients, handgrip was missing after 30 min, indicating motor block of the C8/T1 segment; another 5 patients had loss of handgrip and wrist flexion (C7/C8). Elbow flexion (C6/C7) was unimpaired in all patients.
The results for HR, mean arterial blood pressure (MAP), CO, and CPP are summarized in Table 1. HTEA produced significant decreases in HR (P < 0.001), MAP (P < 0.001), and CPP (P < 0.001). Although significant (P = 0.003), the decrease in CO (5.9 ± 1.3 to 5.6 ± 1.3 L/min) was small in terms of absolute values.
Preload was indicated as CVP, POP, and EDA (Table 1). None of these variables changed significantly after HTEA. Individual values for EDA at baseline and after HTEA are given in Figure 2.
Afterload was quantified as SVR and σm(es). SVR, which is a measure of vasomotor tone, changed significantly (Table 1). σm(es), which is an indication of wall stress that combines data on LV geometry with pressure data, was not significantly changed after HTEA (Table 1; Fig. 2).
No changes were observed after HTEA in variables of LV contractile function, regardless of whether systolic time intervals (ICT or ICT/ET) or a global index of LV fiber shortening (FAC) was used (Table 1; Fig. 2).
IRT and the ratio of IRT versus ejection time (IRT/ET) decreased significantly. Also Vp showed a significant change (P < 0.001). Vp data at baseline and after HTEA are illustrated in Figure 2. The Doppler transmitral flow velocity profile (Dopplertrmit) at baseline exhibited different grades of diastolic dysfunction in 20 patients (60.6%). Dopplertrmit was categorized as normal in 13 patients, as abnormal relaxation in 18 patients, and as a pseudonormalized pattern in 2 patients for the baseline study. Of the patients with an abnormal relaxation pattern, the transmitral flow curve changed to normal in 10 patients, remained an abnormal relaxation in 7 patients, and changed to a pseudonormalized pattern in 1 patient. In the 2 patients with a pseudonormalized pattern at baseline, the transmitral velocity profile changed to an abnormal relaxation pattern in 1 patient. Changes after HTEA were statistically significant (Wilcoxon test; P = 0.005) (Table 1).
MPI is the sum of two ratios, namely ICT/ET and IRT/ET, of which the former is a reflection of systolic function and the latter of diastolic function. The results for MPI, ICT/ET, and IRT/ET are illustrated in Figure 3. MPI was easily obtained in all patients. HTEA induced a significant decrease in MPI. Compared with baseline, MPI changed from 0.51 ± 0.13 to 0.35 ± 0.13. There was no correlation between the change in MPI after HTEA and the change in SVR (Pearson correlation analysis; r = −0.031) (Fig. 4).
Interobserver variability for measurement of MPI, Vp, EDA, and FAC was 7.1% ± 2.9%, 7.7% ± 6.1%, 8.1% ± 6.3%, and 9.1% ± 7.1%, respectively.
The key finding of this study was that diastolic, but not systolic, LV function improved in awake patients with CAD after HTEA. The observed improvement in diastolic function was reflected by a change of MPI, which is a combined Doppler echocardiographic index of global systolic/diastolic myocardial performance.
The effects of HTEA on LV function are thought to be produced by blockade of cardiac sympathetic efferent nerve fibers that have their origin in segments T1-5 (16). Activation of these fibers results in stimulation of both α- and β-adrenergic receptors. Stimulation of β-receptors leads to increased inotropy, chronotropy, and blood pressure, whereas stimulation of α-adrenergic receptors induces vasoconstriction of epicardial coronary arteries (17). In healthy humans, metabolic autoregulation through different pathways and stimulation of vascular β-receptors override the α-adrenergic response. However, in patients with CAD, cardiac sympathetic stimulation induces (because of endothelial dysfunction, exhaustion of autoregulation, or severe coronary stenosis) unrestrained α-adrenergic vasoconstriction that is powerful enough to reduce coronary blood flow and initiate myocardial ischemia (18). Nabel et al. (3) demonstrated a progressive reversal from vasodilation to vasoconstriction during the cold pressor test in both epicardial and resistive vessels with increasing severity of atherosclerosis. Baumgart et al. (19) reported a decrease in coronary blood flow in the presence of atherosclerosis sufficient to induce net myocardial lactate production and to precipitate myocardial ischemia upon intracoronary infusion of selective α1- and α2-receptor agonists in CAD patients. Moreover, it was established in a study using quantitative coronary angiography that HTEA significantly increased the diameter of stenotic coronary artery segments without causing any changes in the tone of the coronary resistance vessels (4).
HTEA-induced loss of sympathetic drive to the myocardium, the epicardial coronary arteries, and the small resistance vessels may thus influence LV function. LV function can be analyzed according to systole and diastole. With respect to HTEA, much more attention has been devoted to the assessment of systolic function, but the results remain controversial. In healthy subjects (6) and in experimental animals (20), a decrease in contractility has been reported when load-independent measures were applied. HTEA reduced the slope of the linear approximation of the LV end-systolic pressure-volume relationship by 50%, and the severe alteration in contractility was attributed to the loss of sympathetic innervation. In CAD patients, HTEA preserved (21) or even improved (4,22) LV systolic function. Interestingly, these studies revealed a reduced incidence of regional wall motion abnormalities, which was ascribed to antiischemic effects of HTEA. In the current report, global systolic LV function was not altered by HTEA. Systolic function was evaluated by a classical ejection phase index (FAC) and by systolic time intervals (ICT and ICT/ET), all of which depend on ventricular loading conditions and HR. Nevertheless, preload, quantified by EDA and POP, and afterload, when indicated in terms of wall stress (σm(es)), remained stable throughout the study. SVR however, which is more an indication of vasomotor tone than of afterload, showed a significant decrease. Against the background of virtually unchanged CO, MAP and CPP declined as a consequence of reduced vasomotor tone.
The effect of HTEA on diastolic function in awake CAD patients has not been previously evaluated, although the prevalence of diastolic dysfunction in asymptomatic patients is significantly more frequent than the prevalence of systolic dysfunction. Furthermore, a change in diastolic function is the earliest hemodynamic manifestation in CAD (23) because diastolic function can be modified reversibly by subclinical myocardial ischemia, even in asymptomatic patients (7). For that reason, diastolic function is considered to be the most sensitive variable of ischemic injury (24). In this study, analysis of the transmitral flow velocity profile (Dopplertrmit) disclosed relaxation abnormality as the most common form of diastolic dysfunction in surgical CAD patients. At baseline, a delayed relaxation pattern was present in 18 patients and a pseudonormalized Dopplertrmit in 2 more patients. Restrictive filling was not encountered. Thus, 60.6% of the patients exhibited diastolic dysfunction. After HTEA, 12 patients converted to a lower grade of diastolic dysfunction or to normal and 1 patient to a higher grade. With the advent of recent echocardiographic techniques, such as color M-mode echocardiography, which is a relatively load-independent method, the ability to accurately detect diastolic dysfunction has been significantly improved. An age-related cutoff value of <0.50 m/s of the Vp determined from color M-mode recordings is considered as a marker of abnormal diastolic LV function (25). In the current investigation, baseline values for Vp were in the range previously established in patients with ischemic heart disease. After HTEA, Vp increased significantly. This observation is consistent with the known relationship between CAD and diastolic dysfunction, in which relaxation is affected before contractility. Vp has been shown to have a very strong inverse correlation with the time constant of isovolumic relaxation (tau) (10). Because relaxation of the myocytes in diastole is a process that is energy dependent and thus sensitive to improved perfusion, this may be why Vp changed after HTEA.
MPI is a new Doppler-derived index of combined LV systolic and diastolic performance, which has been found to correlate well with noninvasive and invasive measures of both systolic (ejection fraction, peak +dP/dt) and diastolic (time constant of relaxation, peak −dP/dt) LV function (8). Thus, the index incorporates phases of active LV contraction and relaxation. In experimental settings, MPI can distinguish a wide variety of functional states (26) and follow acute changes in ventricular function (27). In clinical settings, MPI was reported to provide relevant information on overall LV function in various heart diseases, and increasing values of MPI were closely related to worsening LV function (28). MPI is the sum of 2 ratios, namely ICT/ET and IRT/ET. ICT/ET correlated to +dP/dt, whereas IRT/ET correlated with tau and -dP/dt. Both systolic and diastolic dysfunctions are thus reflected by prolongation of the isovolumic times relative to a shortened ET, resulting in an increased value of the index. However, the duration of IRT parallels in the course of evolving diastolic dysfunction the changes in Doppler transmitral inflow pattern. IRT increases with delayed relaxation, returns to normal values with pseudonormalization, and finally decreases with restrictive pattern. Considering the current results, pseudonormalization of IRT was seen in only two patients at baseline, and no patient exhibited restrictive filling. Thus, an improvement in diastolic function should have been mirrored by a shortening of IRT. Consequently, a decline in baseline values of IRT was documented after HTEA indicating a shift of delayed relaxation pattern towards normal filling characteristics. Because ICT/ET did not change, a decrease in IRT/ET accounted for the improvement in MPI observed in this study.
The present study contains certain limitations. Most importantly, the variables used to quantitate systolic and diastolic LV function are variably sensitive to changes in preload and afterload. However, in view of unchanged EDA and σm(es), the influence of such changes appeared to be small. Second, it is conceivable that the results are obscured by the observed significant decline in HR, although, for example, Vp (10) and MPI (29) are reported to have only weak correlations. Finally, the spread of HTEA in this study was extensive and not exclusively confined to the segments T1-5. The resultant loss of vasomotor tone induced a significant depression of MAP and CPP not encountered by others (2,4). Possibly an infusion technique would have been more appropriate to accurately restrict the block to the upper five thoracic segments (30).
In conclusion, several lines of evidence indicate that LV diastolic function is improved in resting patients with CAD after HTEA. Further studies are warranted to identify the mechanisms of action and to define the clinical impact of HTEA.
1. Olausson K, Magnusdottir H, Lurje L, et al. Anti-ischemic and anti-anginal effects of thoracic epidural anesthesia versus those of conventional therapy in the treatment of severe refractory unstable angina pectoris. Circulation 1997;96:2178–82.
2. Blomberg S, Emanuelsson H, Ricksten SE. Thoracic epidural anesthesia and central hemodynamics in patients with unstable angina pectoris. Anesth Analg 1989;69:558–62.
3. Nabel E, Ganz P, Gordon JB, et al. Dilation of normal and constriction of atherosclerotic coronary arteries caused by the cold pressor test. Circulation 1988;77:43–52.
4. Blomberg S, Emanuelsson H, Kvist H, et al. Effects of thoracic epidural anesthesia on coronary arteries and arterioles in patients with coronary artery disease. Anesthesiology 1990;73:840–7.
5. Saada M, Catoire P, Bonnet F, et al. Effect of thoracic epidural anesthesia combined with general anesthesia on segmental wall motion assessed by transesophageal echocardiography. Anesth Analg 1992;75:329–35.
6. Goertz AW, Seeling W, Heinrich H, et al. Influence of high thoracic epidural anesthesia on left ventricular contractility assessed using the end-systolic pressure-length relationship. Acta Anaesthesiol Scand 1993;37:38–44.
7. Bonow RO, Vitale DF, Bacharach SL, et al. Asynchronous left ventricular regional function and impaired global diastolic filling in patients with coronary artery disease: reversal after coronary angioplasty. Circulation 1985;71:297–307.
8. Tei C, Nishimura RA, Seward JB, Tajik AJ. Noninvasive Doppler-derived myocardial performance index: correlation with simultaneous measurements of cardiac catheterization measurements. J Am Soc Echocardiogr 1997;10:169–78.
9. Clements FM, Harpole DH, Quill T, et al. Estimation of left ventricular volumes and ejection fraction by two-dimensional echocardiography: comparison of short axis imaging and simultaneous radionuclide angiography. Br J Anaesth 1990;64:331–6.
10. Garcia MJ, Thomas JD, Klein AL. New Doppler echocardiographic applications for the study of diastolic function. J Am Coll Cardiol 1998;32:865–75.
11. Loick HM, Schmidt C, Van Aken H, et al. High thoracic epidural anesthesia, but not clonidine, attenuates the perioperative stress response via sympatholysis and reduces the release of troponin T in patients undergoing coronary artery bypass grafting. Anesth Analg 1999;88:701–9.
12. Abd Elrazek E, Scott NB, Vohra A. An epidural scoring scale for arm movements (ESSAM) in patients receiving high thoracic epidural analgesia for coronary artery bypass grafting. Anaesthesia 1999;54:1104–9.
13. Reichek N, Wilson J, St John Sutton M, et al. Noninvasive determination of left ventricular end-systolic stress: validation of the method and initial application. Circulation 1982;65:99–108.
14. Nishimura RA, Tajik AJ. Evaluation of diastolic filling of left ventricular in health and disease: Doppler echocardiography is the clinician’s Rosetta stone. J Am Coll Cardiol 1997;30:8–18.
15. Garcia MJ, Smedira NG, Greenberg NL, et al. Color M-mode Doppler flow propagation velocity is a preload insensitive index of left ventricular relaxation: animal and human validation. J Am Coll Cardiol 2000;35:201–8.
16. Magnusdottir H, Kirnö K, Ricksten SE, Elam M. High thoracic epidural anesthesia does not inhibit sympathetic nerve activity in the lower extremities. Anesthesiology 1999;91:1299–304.
17. Young MA, Vatner SF. Regulation of large coronary arteries. Circ Res 1986;59:579–96.
18. Heusch G, Baumgart D, Camici P, et al. α-adrenergic vasoconstriction and myocardial ischemia in humans. Circulation 2000;101:689–94.
19. Baumgart D, Haude M, Goerge G, et al. Augmented α-adrenergic constriction of atherosclerotic human coronary arteries. Circulation 1999;99:2090–7.
20. Mizukoshi Y, Shibata K, Yoshida M. Left ventricular contractility is reduced by hypercapnic acidosis and thoracolumbar epidural anesthesia in rabbits. Can J Anaesth 2001;48:557–62.
21. Berendes E, Schmidt C, Van Aken H, et al. Reversible cardiac sympathectomy by high thoracic epidural anesthesia improves regional left ventricular function in patients undergoing coronary artery bypass grafting. Arch Surg 2003;138:1283–90.
22. Kock M, Blomberg S, Emanuelsson H, et al. Thoracic epidural anesthesia improves global and regional left ventricular function during stress-induced myocardial ischemia in patients with coronary artery disease. Anesth Analg 1990;71:625–30.
23. Castello R, Pearson AC, Kern MJ, Labovitz AJ. Diastolic function in patients undergoing coronary angioplasty: influence of degree of revascularization. J Am Coll Cardiol 1990;15:1564–9.
24. Apstein CS, Grossman W. Opposite initial effects of supply and demand ischemia on left ventricular diastolic compliance: the ischemia-diastolic paradox. J Mol Cell Cardiol 1987;19:119–29.
25. Palecek T, Linhart A, Bultas J, Aschermann M. Comparison of early diastolic mitral annular velocity and flow propagation velocity in detection of mild to moderate left ventricular diastolic dysfunction. Eur J Echocardiogr 2004;5:196–204.
26. Broberg CS, Pantely GA, Barber BJ, et al. Validation of the myocardial performance index by echocardiography in mice: a noninvasive measure of left ventricular function. J Am Soc Echocardiogr 2003;16:814–23.
27. LaCorte JC, Cabreriza SE, Rabkin DG, et al. Correlation of the Tei index with invasive measurements of ventricular function in a porcine model. J Am Soc Echocardiogr 2003;16:442–7.
28. Dujardin K, Tei C, Yeo TC, et al. Prognostic value of a Doppler index combining systolic and diastolic performance in idiopathic-dilated cardiomyopathy. Am J Cardiol 1998;82:1071–6.
29. Poulsen SH, Nielsen JC, Andersen HR. The influence of heart rate on the Doppler-derived myocardial performance index. J Am Soc Echocardiogr 2000;13:379–84.
© 2005 International Anesthesia Research Society
30. Kessler P, Neidhart G, Bremerich DH, et al. High thoracic epidural anesthesia for coronary artery bypass grafting using two different surgical approaches in conscious patients. Anesth Analg 2002;95:791–7.