Abstract

Background: In this study we compared the myocardial protective effects of fentanyl-based total intravenous anesthesia (TIVA), sevoflurane and isoflurane.

Methods: After ethic committee approval, 57 patients who were scheduled to undergo open heart surgery using cardiopulmonary bypass (CPB) were randomized into three groups. Following standard induction, group 1 (TIVA group; n=17) received fentanyl-midazolam-pancuronium, group 2 (Sevo group; n=20) received 1 MAC sevoflurane and group 3 (Iso group; n=20) received 1 MAC isoflurane for maintenance. During preoperative period and on postoperative third day, left ventricle functions and cardiac scores were evaluated by transthoracic echocardiography. Cardiac troponin I (cTnI) and creatine kinase MB fraction (CKMB) were measured before CPB and at sixth hour, day one, two and three following CPB. For lactate measurement, blood samples were collected from arterial line and retrograde cannula before and after CPB and following declamping.

Results: Demographic parameters were similar among three groups. Cardiac index and output increased in all three groups after CPB. Creatine kinase MB fraction and cTnI levels were similar among groups before CPB, those levels increased in all groups at sixth hour after CPB. Although there was not a statistically significant difference among the groups, this increase was the highest in TIVA group. After declamping, the lactate levels collected from retrograde cannula increased in all groups.

Conclusion: As a result, as all the groups have preconditioning potential, although we did not find a significant difference between fentanyl-based TIVA group and volatile anesthetics in terms of myocardial protection, we could say that volatile anesthetics are more cardioprotective, compared to fentanylbased TIVA group.

Volatile anesthetics have been shown to protect the myocardium from ischemic injury, and the related clinical process has been termed anesthetic-induced preconditioning. The mechanism for anesthetic preconditioning is not clearly understood, but protein kinase, adenosine triphosphate (ATP)-sensitive potassium channels, and adenosine A1 receptors play key roles in this process. The volatile anesthetic, sevoflurane, also increases basal and stimulated coronary flow along with nitric oxide (NO) production.[1] Sevoflurane selectively increases large and small coronary blood flow via mechanisms independent of ATP-sensitive potassium (KATP) channels.[2] Although clinical studies have demonstrated the translocation of the protein kinase C-d and € isoform in the human myocardium in response to sevoflurane,[3] conflicting results were found for perioperative ST segment changes, arrhythmias, troponin I (TnI), and creatine kinase-MB isoenzyme (CK-MB) release.[4,5]

Morphine, an opioid, also has preconditioning effects which have been proven by many studies.[6,7] Its protective effect is exercised by way of d receptor intermediation, and some studies have shown that pure d opioid agonists may also behave in this way.[7,8] Morphine acts by activating d r eceptors, t hereby a ctivating KATP channels and providing cardioprotection. In other studies using remifentanil,[9] fentanyl,[10] and sufentanil as anesthetic agents, it was suggested that besides d receptors, k receptors may also contribute to cardioprotection. Although fentanyl has a preferential affinity for m opioid receptors, it also interacts with d and k receptors.[11,12] Kato et al.[12,13] demonstrated that fentanyl has a cardioprotective effect which is mediated by both m receptors and protein kinase C (PKC) in a model of myocardial ischemia reperfusion (I/R) injury in vitro. Although there are conflicting results regarding the cardioprotective effect of fentanyl,[6] it is one of the most preferred opioid agents due to its reduced side effects. Cardiac anesthesia frequently has a preference for a combination of balanced anesthetic techniques with opioid and volatile anesthetics. This balanced anesthetic technique may enhance protection against myocardial ischemia more than either agent alone. In experimental studies, the combination of isoflurane and morphine provides increased cardiac protection but their clinical effects have not been clearly demonstrated.[14]

Although there are a lot of studies which use fentanyl and volatile anesthetics in balanced anesthesia protocols, fentanyl doses were given in equivalent doses to groups; therefore, only the effects of volatile anesthetics were evaluated.[3,4,15] This study a imed to compare the cardioprotective effects of anesthesia groups using a combination of low-dose fentanyl, sevoflurane, and isoflurane with a fentanyl-midazolam anesthesia group by means of clinical, biochemical, and hemodynamic parameters.

Methods

This study was designed as a double-blinded, randomized study and was performed after obtaining the permission of the ethics committee of Türkiye Yüksek İhtisas Training and Research Hospital, Ankara, Turkey. A total of 57 patients scheduled for elective coronary artery bypass graft surgery (CABG) were enrolled, and informed written consent was obtained from all patients.

Patients under the age of 30 and those older than 75 years old were excluded from the study. Other exclusion criteria were as follows: diabetes mellitus, renal failure, a history of myocardial infarction for at least one month before surgery, preoperative unstable angina pectoris, and patients using theophylline. All patients taken to the operating room were monitored with electrocardiography (ECG), end-tidal carbon dioxide concentration (ETCO2) by capnography, and oxygen saturation (SO2) by pulse oxymetry during the operation. Two peripheral venous catheterizations (16 or 18G needle) and radial artery catheterization with an 18G needle were performed. After local anesthesia, an 8F catheter was placed in the internal jugular vein. A 7F 4-lumen Swan-Ganz catheter (Abbott, TORQUE-LINE 7 F, 110 cm, 1.5 thermo dilution catheter, 4 lumen, RA) was inserted via the jugular catheter.

All patients were premedicated with diazepam (0.1 mg/kg) per os (P.O.) the night before and morphine sulphate (0.1 mg/kg) intramuscular (I.M.) 30 minutes before the operation. An intravenous bolus infusion of midazolam (0.1 mg/kg), fentanyl (15-20 m/kg), and intravenous pancuronium bromide (0.1 mg/kg) were used in the induction of anesthesia. A tidal volume of 8-10 ml/kg with 50% oxygen-air mixture was achieved with controlled mechanical ventilation after endotracheal intubation.

Hemodynamic parameters were scaled at five different times: T1: Before induction, T2: After induction, T3: Before cardiopulmonary bypass (CPB), T4: After CPB, and T5: Six hours after the operation.

The mean arterial blood pressure (MAP), heart rate (HR), mean pulmonary artery pressure (MPAP), pulmonary capillary wedge pressure (PCWP), cardiac output (CO), cardiac index (CI), stroke volume (SV), systemic vascular resistance (SVR), pulmonary vascular resistance (PVR), left ventricular stroke volume index (LVSVI), and right ventricular stroke volume index (RVSVI) were the hemodynamic parameters tested. Cardiac output was measured with a specific device (3300 Cardiac Output Computer Abbott Laboratories, North Chicago, Illinois, USA).

Patients were randomized into three groups
Group 1 total intravenous anesthesia (TIVA group; n=17): This group was also accepted as a control group. In anesthesia maintenance, fentanyl (0.25-0.5 μg/kg/min) and midazolam (0.3 mg/kg/min) were given by intravenous infusion, and pancuronium was given in a dose of 2 mg as required.

Group 2 sevoflurane (Sevo group; n=20): After the same anesthetic induction procedure as the TIVA group, maintenance doses of fentanyl 5 μg/kg and pancuronium bromide 2 mg were applied repeatedly as required in this group. Sevoflurane was administrated in 1 MAC (minimal alveolar concentration), causing a decrease of approximately 20-25% in the arterial blood pressure.

Group 3 isoflurane (Iso group; n= 20): The procedure was the same as the Sevo group except that the volatile anesthetic agent isoflurane was used instead of sevoflurane.

In groups 2 and 3, the volatile anesthetics were given until CPB was started. After unclamping the aortic cross clamp, the gases were administered again until the end of the operation. In group 1, the agents were infused over the same periods.

A standard CPB technique was performed in all groups. After anticoagulation with routine doses of heparin (400 IU/kg), CPB was started, and a membrane oxygenator (MEDOS Hilite 7000, Vingmed Danmark A/S, Taastrup, Denmark) and a non-pulsatile pump (mean f low= 2 .4 L/min/m2 body surface area) were used. The body temperature was between 32-34 °C during CPB. The heart was arrested by a hyperkalemic crystalloid cardioplegic solution (Plegisol 15 ml/kg) given through anterograde cannula under a pressure of 120 mmHg. During the CPB period, 400 cc of diluted cold blood cardioplegia was applied every 20 minutes, and warm blood cardioplegia was given before unclamping the aortic cross clamp in order to obtain myocardial protection. After unclamping the aorta, CPB was terminated when a rectal temperature of 35-36 ºC and adequate cardiac contractions were achieved. At the level of 2.0 L/min/m² of CI, inotropic agents were administrated. The inotropic agent requirement, incidence, type, and treatment of postoperative reperfusion arrhythmia were recorded.

Changes in the ST segment on the DII and V5 derivations were recorded electronically on the monitor (Datex Ohmeda S/5, GE Healthcare, Madison, USA) simultaneously with the hemodynamic parameters. Transthoracic echocardiography was done preoperatively and on the third day postoperatively in order to assess the left ventricular functions, cardiac scores, and ejection fraction (EF).

The biochemical parameters, cardiac troponin I (cTnI) level, and CK-MB fraction were determined at six different times: B1: Before CPB, B2: After CPB, B3: Six hours after surgery, B4: The first day postoperatively, B5: The second day postoperatively, and B6: The third day postoperatively.

For determining the CK-MB levels, blood samples were taken from the patients who had undergone CABG before CPB, after CPB, at the sixth hour postoperatively, and on the first, second, and third days after surgery. Measurements were performed to determine the CK-MB activity with immunologic method by a Hitachi 911 Chemistry Analyzer (Boehringer Ingelheim, Germany).

Cardiac troponin I blood samples were taken from the patients who had undergone CABG before CPB, after CPB, at the sixth hour postoperatively, and on the first, second, and third days after surgery. Cardiac TnI measurements were performed by the Microparticle Enzyme Immunoassay (MEIA) method with an Abbott AxSYM immunoassay diagnostic system in ng/ml (Abbott, Turkey).

Blood samples collected for L-lactate level simultaneously obtained from arterial line and retrograde cannula were examined for lactate levels at two different period: L1: before CPB after the retrograde cannula was placed and L2: after the unclamping of the aorta. For the lactate measurement, 5 ml blood was centrifuged and cooled at –60 °C for eight months. The measurement was done by the enzymatic calorimetric method using the Sigma diagnosis reagent (Procedure No 735) on the Hitachi 911 chemical analyzer (Boehringer Ingelheim, Germany).

Data analysis was performed by using the Statistical Package for the Social Sciences (SPSS Inc., Chicago, Illinois, USA) for Windows version 11.5 software. Data was shown as mean ± standard deviation for continuous variables and number of patients and percentage (%) for categorical variables. Analysis of variance (ANOVA) or the Friedman test was used to evaluate the differences among repeated measures where applicable. When the p value from the repeated measures of ANOVA or Friedman test statistics was statistically significant, Bonferroni adjusted t-tests or Friedman multiple comparison tests were used to know which measurement differed from the others. When the number of repeated measures was two, a paired t test was applied to determine whether the differences between the preoperative and postoperative measurements were statistically significant or not. A p v alue of l ess than 0.05 was considered statistically significant. For all possibilities within group comparisons, the Bonferroni adjustment was applied to control the type I error rate.

Results

Demographic data were given in Table 1. The patients in all three groups were not statistically different in terms of sex, age, body surface area (BSA), American Society of Anesthesiologists (ASA) physical classification, cross-clamp time, and the number of times the patients had undergone CABG and CPB (Table 1). None of the patients developed myocardial ischemia intraoperatively as judged by the automated ST segment analysis, and no patient required a second operation.

Table 1: Demographics of patients

There were no significant differences between the groups regarding preoperative cardiac hemodynamic parameters. Neither were the CI values different among the three groups preoperatively. After induction, all patients in the three groups showed similar responses. At the T3 period, there was a significant decrease in the Sevo and Iso group in the SVR measurements compared with the patients in the TIVA group. The CI and CO values were not different between the groups at the T3 p eriod; h owever, a t t he T4 period (after CPB measurement), all group values increased significantly over the T3 v alues, b ut t here w ere n o s ignificant differences between the groups (Table 2).

Table 2: Haemodynamic parameters

The incidence of perioperative myocardial ischemic and arrythmic events was assessed with the help of ST segment analysis.

The incidence of arrythmias after cross-clamp release was did not differ among the three groups. In the TIVA group, ventricular fibrillation (VF) was seen in 10 patients (58.8%), and seven patients (41.1%) were defibrillated. In the Sevo group, VF was seen in 12 patients (60%) and eight patients (40%) were defibrillated. Finally, in the Iso group, 11 patients (55%) were diagnosed with VF, and seven patients (35%) needed defibrillation.

Troponin I concentrations were similar in all groups before CPB. After revascularization, the cTnI levels increased and reached peak levels at the sixth hour postoperatively. This increase was seen more significantly in the TIVA group (p=0.001) than in the Iso and Sevo groups (p=0.005) in the post-CPB period, but there was no statistically significant difference between the groups. The cTnI levels on the first, second, and third day postoperatively were similar in the Iso and TIVA groups but were lower in the Sevo group. No statistical significance between groups was evident (Figure 1).

Figure 1: The change in troponin I (as ng/ml). + p<0.001; * p<0.005 compared to before-CPB: Before cardiopulmonary bypass; After-CPB: After cardiopulmonary bypass.

The creatinine kinase-myocardial fraction measurement results were also comparable at the beginning of surgery and all groups’ data were similar to each other. These levels were significantly different in the postoperative periods in all groups compared with the basal (preoperative) levels. At the sixth hour, this significance was very high (p=0.001) compared with the baseline in all groups, but CK-MB results at the sixth hour and first day postoperatively were not significantly different between the groups (Figure 2).

Figure 2: The change in CK-MB (as U/L). CK-MB: Creatinine kinase myocardial fraction; * p<0.001 for all groups compared to before-CPB: Before cardiopulmonary bypass; After-CPB: After cardiopulmonary bypass.

The pre-CPB arterial and retrograde cannula lactate levels were similar in all the groups. Arterial and retrograde cannula lactate levels after the unclamping of the aorta were higher than before CPB in all the groups (Figure 3). Compared with the other groups, the retrograde cannula lactate levels were found to be higher in the TIVA group.

Figure 3: Lactate levels (as mg/dl). * p<0.001 as compared to before-CPB: Before cardiopulmonary bypass; After-CPB: After cardiopulmonary bypass; + p<0.05 as compared between groups.

The fentanyl consumption of the TIVA group was significantly higher than in the volatile anesthetic groups (TIVA 6643.82 micgr ±1184.92; Sevo group 2485.00 micgr ±184.85; Iso group 2441.25 micgr ±174.76).

Preoperative echocardiography was maintained, and the ejection fractions (EF1) were similar in the three groups. Also, postoperative echocardiography showed that the ejection fractions at the third postoperative day (EF2) were not different within the groups. On the other hand, according to the preoperative and postoperative EF change, the groups showed statistical differences (p=0.044). There was an increase in EF in the Sevo group while it decreased in the TIVA and Iso groups (Figure 4). This difference in the Sevo group was statistically significant (p=0.003 compared with the TIVA group; p=0.001 compared with the Iso group). When preoperative and postoperative end-diastolic size and ventricular scores were evaluated, there was no difference either within the groups or between the groups. The preoperative and postoperative changes in end-diastolic size and ventricular scores were statistically similar (p=0.26; p=0.66, respectively).

Figure 4: Ejection fraction change between groups. * p=0.003 compare to TIVA group; + p≤0.001 compared to Iso group.

The need for an inotropic agent was seen in two patients in the TIVA group, three patients in the Sevo group, and two patients in the Iso group, and there was no difference between the groups (p=0.05). Mechanical ventilation periods (MVP) were similar in the TIVA, Sevo, and Iso groups postoperatively (8.04±0.2 hours; 7.8±0.1 hours; 8.02±0.2 hours, respectively). Intensive care unit (ICU) stays were calculated and found to be similar in all the groups (14.4±0.2 hours; 13.9±0.3 hours; 14.1±0.3 hours, respectively). Hospital stays (HS) were also similar (4.9±0.1 days; 5.2±0.1 days; 5.4±0.3 days, respectively).

No patient developed a cerebrovascular event, and no hemofiltration or dialysis was indicated in the groups. Postoperatively, no death or transmural infarct was seen in the three groups.

Discussion

This study compared the cardioprotective effects of the Sevo and Iso groups with the fentanyl-based TIVA group, and there was no difference between the groups in terms of protection. However, in the Sevo group, there was a significant increase in EF on the postoperative third day. The increments of the lactate levels following the unclamping of the aorta were less in the volatile groups when compared with the TIVA group. Although they were not statistically significant, the increases in the cTnI and CK-MB levels at the postoperative sixth hour in the Sevo group were less than in the TIVA group. Therefore, we concluded that the volatile anesthetic groups, especially the Sevo group, seemed to protect the myocardium better.

In the past, high-dose opioid anesthetics were preferred during cardiac surgery anesthesia because of their ability to provide hemodynamic stability, but the most recent studies have shown that halogenated volatile anesthetics are more effective in protecting the myocardial structure, integrity, and function.[15-17] Two of these recent studies used propofol, which is known to have no preconditioning effect, in their TIVA groups, However, in our study, opioid anesthesia (fentanyl-midazolam) was choosen for the TIVA group. Opioids have been suggested to have preconditioning effects, and several investigators have reported that morphine may protect the myocardium from ischemic injury during cardiac surgery.[6,7] Although it is frequently used clinically, the cardioprotective effect of fentanyl still remains controversial. Kato et al.[10] demonstrated that fentanyl alleviated postischemic ventricular dysfunction. Opioid receptors and KATP channels mediated this cardioprotective effect. Again, Kato et al.[12] showed that in a rat model, fentanyl limited the infarct area by the mediation of delta opioid receptors and PKC. However, Murphy et al.[6] compared morphine and fentanyl in their clinical study and suggested that the morphine-isoflurane combination protects the myocardium better than the fentanyl-isoflurane combination against ischemia. Our study compared the protective effects of fentanyl in clinical practical doses with volatile anesthetics, and it was concluded that the volatile anesthetics seemed to be slightly more protective than the fentanyl. In addition, sevoflurane and isoflurane used after ischemia until the end of the operation were shown to have protective effects against reperfusion injury and stunning; therefore, cardiac function was improved, disarrythmias were degraded in the postoperative period, myocardial energy was saved, and the infarct size was also reported to have shrunk.[18-21] This effect may further contribute to preconditioning. For these reasons, we prefer to use sevoflurane and isoflurane until the end of the operation.

In our study, there was no difference between the groups with regard to demographic values and hemodynamic parameters. In other studies which focused on anesthetic preconditioning and ischemic preconditioning, the CO and CI were increased compared with the control groups after CPB.[15,17] In our study, although the CI and CO increased in all three groups after bypass, there were no significant differences. This may be due to using opioids in the control group instead of propofol. We performed an echocardiography preoperatively and on the postoperative third day. The variation in EF in the Sevo group was meaningful, but there were no differences between the preoperative and postoperative echocardiographic ventricular score evaluations.

Cardiac TnI levels are the most important indicator of cardiac damage. In a study by Etievent at al.[22] there was a correlation between postoperative high cTnI values and the aortic cross-clamping period. Also, cTnI values have been found to be useful in early acute myocardial infarction (MI) and postoperative MI diagnosis and for comparisons between different cardioprotective procedures. In our study, the crossclamping period was similar between the three groups. Our results showed that even though they were statistically insignificant, the cTnI levels were lower in the volatile anesthetics groups than in the TIVA group. This insignificant difference may be due to the preconditioning effect of fentanyl. In studies in which sevoflurane and propofol were compared, the cTnI levels were lower in the sevoflurane group.[17,23] In another study, there were no differences in the volatile anesthetic groups with regard to the CK-MB and cTnI levels; however, the N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels were different, and the PKC delta and epsilon translocation was shown. Therefore, the sevoflurane preconditioning effect was clinically accepted.[3]

Biochemical markers may be an indicator of early graft occlusion. There were no ECG changes in the ST segment analysis in the three groups in our study, so we did not consider graft occlusion or any other adverse effect of surgery.

Myocardial lactate secretion can be used as a marker of myocardial protection because in the case of hypoxia and ischemia, lactate is released into the blood to compensate for the ATP level in the tissue. Hence, the lactate level shows anaerobic glycolysis and gives information about the metabolic mode.[24] In this study, the arterial and retrograde cannula lactate levels after the declamping of the aorta were higher than in the period before CPB in all groups, and this elevation was higher in the TIVA group than in the volatile anesthetic groups. Lactate metabolism is affected by many factors which results in challenging results. Some other studies detected that lactate production is only related to the CPB period, but in our study, the CPB duration was not different statistically The increase in the arterial lactate level may be related to surgical stress.[25] In a study which compared off-pump and on-pump bypass, the lactate levels were higher in the on-pump group. Accordingly, the increase in the retrograde cannula lactate levels in all groups in our study may be related to CPB.[26]

Unlike in other laboratory studies, we could not find the clinically protective effect of fentanyl. We also found no distinct difference in the cardioprotective effect between the TIVA (fentanyl-based) and volatile anesthetic groups, which is in contrast to findings in other studies comparing propofol and volatile anesthetics. Fentanyl concentrations used in experimental studies is outside the clinical range. In other words, as fentanyl binds to plasma protein at a high rate (82%), the fentanyl concentrations used in clinical studies might be insufficient.[13] Otherwise, the possibility exists that there may be some difference between species. We did not test whether or not fentanyl may potentiate the effects of volatile anesthetics, but it is known that morphine and volatile anesthetics potentiate each other.[14] Moreover, we have no clinical practice which shows that only volatile anesthetics are used in cardiac surgery. The fentanyl dose was much lower in the volatile anesthetic group than in the TIVA group. Also, it has been noted that midazolam has no preconditioning effect. In a rat study, midazolam had no effect on KATP channels or ischemic myocyte survival.[27]

As a result, since all the groups have preconditioning potential, we could not find a distinctive difference between the volatile anesthetics and the fentanylmidazolam- based TIVA group. Even if no statistical significance was shown, we can conclude that volatile anesthetics seem to be more cardioprotective than fentanyl based TIVA regimen.

Declaration of conflicting interests
The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding
This study was supported by ABBOTT Laboratory, İstanbul, Turkey Group.

References

1) Novalija E, Fujita S, Kampine JP, Stowe DF. Sevoflurane mimics ischemic preconditioning effects on coronary flow and nitric oxide release in isolated hearts. Anesthesiology 1999;91:701-12.

2) Kersten JR, Schmeling T, Tessmer J, Hettrick DA, Pagel PS, Warltier DC. Sevoflurane selectively increases coronary collateral blood flow independent of KATP channels in vivo. Anesthesiology 1999;90:246-56.

3) Julier K, da Silva R, Garcia C, Bestmann L, Frascarolo P, Zollinger A, et al. Preconditioning by sevoflurane decreases biochemical markers for myocardial and renal dysfunction in coronary artery bypass graft surgery: a double-blinded, placebo-controlled, multicenter study. Anesthesiology 2003;98:1315-27.

4) Belhomme D, Peynet J, Louzy M, Launay JM, Kitakaze M, Menasché P. Evidence for preconditioning by isoflurane in coronary artery bypass graft surgery. Circulation 1999;100(19 Suppl):II340-4.

5) Fellahi JL, Gue X, Philippe E, Riou B, Gerard JL. Isoflurane may not influence postoperative cardiac troponin I release and clinical outcome in adult cardiac surgery. Eur J Anaesthesiol 2004;21:688-93.

6) Murphy GS, Szokol JW, Marymont JH, Avram MJ, Vender JS. Opioids and cardioprotection: the impact of morphine and fentanyl on recovery of ventricular function after cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2006;20:493-502.

7) McPherson BC, Yao Z. Morphine mimics preconditioning via free radical signals and mitochondrial K(ATP) channels in myocytes. Circulation 2001;103:290-5.

8) Schultz J el-J, Hsu AK, Nagase H, Gross GJ. TAN-67, a delta 1-opioid receptor agonist, reduces infarct size via activation of Gi/o proteins and KATP channels. Am J Physiol 1998;274:H909-14.

9) Zhang Y, Irwin MG, Wong TM, Chen M, Cao CM. Remifentanil preconditioning confers cardioprotection via cardiac kappa- and delta-opioid receptors. Anesthesiology 2005;102:371-8.

10) Kato R, Ross S, Foëx P. Fentanyl protects the heart against ischaemic injury via opioid receptors, adenosine A1 receptors and KATP channel linked mechanisms in rats. Br J Anaesth 2000;84:204-14.

11) Maguire P, Tsai N, Kamal J, Cometta-Morini C, Upton C, Loew G. Pharmacological profiles of fentanyl analogs at mu, delta and kappa opiate receptors. Eur J Pharmacol 1992;213:219-25.

12) Kato R, Foëx P. Fentanyl reduces infarction but not stunning via delta-opioid receptors and protein kinase C in rats. Br J Anaesth 2000;84:608-14.

13) Kato R, Foëx P. Myocardial protection by anesthetic agents against ischemia-reperfusion injury: an update for anesthesiologists. Can J Anaesth 2002;49:777-91.

14) Ludwig LM, Patel HH, Gross GJ, Kersten JR, Pagel PS, Warltier DC. Morphine enhances pharmacological preconditioning by isoflurane: role of mitochondrial K(ATP) channels and opioid receptors. Anesthesiology 2003;98:705-11.

15) Haroun-Bizri S, Khoury SS, Chehab IR, Kassas CM, Baraka A. Does isoflurane optimize myocardial protection during cardiopulmonary bypass? J Cardiothorac Vasc Anesth 2001;15:418-21.

16) De Hert SG, Cromheecke S, ten Broecke PW, Mertens E, De Blier IG, Stockman BA, et al. Effects of propofol, desflurane, and sevoflurane on recovery of myocardial function after coronary surgery in elderly high-risk patients. Anesthesiology 2003;99:314-23.

17) De Hert SG, ten Broecke PW, Mertens E, Van Sommeren EW, De Blier IG, Stockman BA, et al. Sevoflurane but not propofol preserves myocardial function in coronary surgery patients. Anesthesiology 2002;97:42-9.

18) Buljubasic N, Stowe DF, Marijic J, Roerig DL, Kampine JP, Bosnjak ZJ. Halothane reduces release of adenosine, inosine, and lactate with ischemia and reperfusion in isolated hearts. Anesth Analg 1993;76:54-62.

19) Kanaya N, Kobayashi I, Nakayama M, Fujita S, Namiki A. ATP sparing effect of isoflurane during ischaemia and reperfusion of the canine heart. Br J Anaesth 1995; 74:563-8.

20) Heindl B, Reichle FM, Zahler S, Conzen PF, Becker BF. Sevoflurane and isoflurane protect the reperfused guinea pig heart by reducing postischemic adhesion of polymorphonuclear neutrophils. Anesthesiology 1999;91:521-30.

21) Marijic J, Stowe DF, Turner LA, Kampine JP, Bosnjak ZJ. Differential protective effects of halothane and isoflurane against hypoxic and reoxygenation injury in the isolated guinea pig heart. Anesthesiology 1990;73:976-83.

22) Etievent JP, Chocron S, Toubin G, Taberlet C, Alwan K, Clement F, et al. Use of cardiac troponin I as a marker of perioperative myocardial ischemia. Ann Thorac Sur 1995;59:1192-4.

23) Conzen PF, Fischer S, Detter C, Peter K. Sevoflurane provides greater protection of the myocardium than propofol in patients undergoing off-pump coronary artery bypass surgery. Anesthesiology 2003;99:826-33.

24) Pietersen HG, Langenberg CJ, Geskes G, Kester A, de Lange S, Van der Vusse GJ, et al. Myocardial substrate uptake and oxidation during and after routine cardiac surgery. J Thorac Cardiovasc Surg 1999;118:71-80.

25) Bortone F, Mazzoni M, Repossini A, Campolo J, Ceriani R, Devoto E, et al. Myocardial lactate metabolism in relation to preoperative regional wall motion and to early functional recovery after coronary revascularization. J Cardiothorac Vasc Anesth 2003;17:478-85.

26) Koh TW, Carr-White GS, DeSouza AC, Ferdinand FD, Hooper J, Kemp M, et al. Intraoperative cardiac troponin T release and lactate metabolism during coronary artery surgery: comparison of beating heart with conventional coronary artery surgery with cardiopulmonary bypass. Heart 1999;81:495-500.

27) Zaugg M, Lucchinetti E, Spahn DR, Pasch T, Garcia C, Schaub MC. Differential effects of anesthetics on mitochondrial K(ATP) channel activity and cardiomyocyte protection. Anesthesiology 2002;97:15-23.