The cardioprotective effects of propofol, which is commonly used in patients undergoing anaesthesia for cardiac surgery, are less well established than those of the volatile anaesthetics [1,2]. However, there is indirect evidence that propofol may also possess protective properties in the setting of ischaemia and reperfusion-induced injury. This evidence is based partly on the finding that propofol has free radical-scavenging properties and protects the heart against H2O2-induced damage in isolated rat hearts [3-5]. In contrast, several in vivo studies performed in different laboratories could not demonstrate any beneficial effect of propofol at clinically relevant doses on systolic functional recovery from myocardial stunning, and it was concluded that there was certainly less protection than with the volatile anaesthetics [6-8]. While it is generally accepted that myocardial stunning is at least partially induced by oxidative stress, the lack of conclusive evidence about the cardioprotective effect of propofol is difficult to reconcile with its purported free radical-scavenging properties.
Until now, no studies have specifically focused on the effects of propofol on diastolic function in ischaemia- and reperfusion-induced myocardial damage. Previous studies [4,9,10] and our own preliminary findings have demonstrated an attenuation of the increase in left ventricular end-diastolic pressure (LVEDP) by propofol during ischaemia. This evidence, and the fact that diastolic function is disturbed in ischaemia and reperfusion , has led us to examine the effects of propofol on both systolic and diastolic function during ischaemia and reperfusion. We studied both components of diastole: the early, active phase of relaxation and the late, passive component, which depends mainly on chamber compliance. The experimental setting was an isolated, parabiotic, blood perfused rabbit heart model.
Sixty-eight adult New Zealand White rabbits (2.5-3.3 kg) were included in the study, half of which acted as parabiotic 'support' animals. All animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals prepared by the US National Academy of Sciences (US National Institute of Health, revd 1978). The study was approved by the Ethics Committee of the Katholieke Universiteit Leuven.
A modified Langendorff set-up (Fig. 1) was used as described [12,13], primed with Ringer's solution 30 mL, Geloplasma® (Institut Mérieux Benelux, Brussels, Belgium) 20 mL, sodium bicarbonate 0.8 molar in 1.5 mL, and heparin 7500 IU.
The rabbits were premedicated with pirinitramide (piritramide) 20 mg intramuscularly (Dipidolor®; Janssen Pharmaceutica, Beerse, Belgium). Anaesthesia was induced by intravenous injection of sodium pentobarbital 25 mg kg−1 and maintained with pirinitramide 2 mg kg−1 h−1 and sodium pentobarbital 4 mg kg−1 h−1. The trachea was intubated with a 3.5 mm tube via a tracheotomy and the lungs ventilated with a Mark 7® ventilator (Bird Products Corp., Palm Springs, CA, USA) using a 50% mixture of oxygen and air. In the 'support' animal, arterial blood-gas status was kept within the physiological range by adjustment of lung ventilation and the administration of sodium bicarbonate when necessary. Heparin was administered continuously at 300 IU kg−1 h−1. The left carotid artery and right jugular vein of the support animal were cannulated for withdrawal and the return of blood to and from the isolated heart preparation. The heart was removed from the donor animal via a median sternotomy, briefly submerged in cold Ringer's solution (0-4°C) and mounted onto the perfusing system within 30 s. An 18-G catheter was inserted through the apex of the left ventricle for venting. Arterial blood was withdrawn from the support animal by a roller pump into an overflow blood reservoir located 85 cm above the isolated heart, providing a constant perfusion pressure of 70 mmHg. A thermistor needle was inserted into the ventricular septum to monitor the myocardial temperature, which was maintained at 37 (±0.5)°C by means of heating tubing around all vascular catheters and by immersion of the heart in warm water during the ischaemic phase. The heart was paced continuously at 180 beats min−1 by means of a pair of pacing needles inserted in the free wall of the right ventricle. A fluid-filled latex balloon, connected to a pressure transducer (Model 1280C®; Hewlett Packard, Waltham, MA, USA), was placed in the left ventricle via an incision in the left atrial appendage. Saline solution was injected into the balloon until LVEDP reached 12 mmHg. The perfusion pressure was recorded through a side arm of the aortic cannula and an inline flow probe (T206 Small Animal Flowmeter®; Transonic Systems, Inc., Ithaca, NY, USA) immediately above the aortic cannula measured the coronary blood flow. All variables were recorded on a heat-writing chart recorder (WT-655G®; Nihon Kohden Corp., Tokyo, Japan).
The experimental protocols are shown in Figure 2. The viability of the isolated heart preparation was assessed by means of coronary blood flow of at least 5 mL min−1 and the left ventricle developed a pressure of at least 100 mmHg after the initial stabilization. If this was not achieved, the experimental protocol was not initiated.
Systolic function (Fig. 2, Protocol 1)
To achieve an intracoronary propofol concentration of 168 μmol L−1 at an average coronary blood flow of 10 mL min−1, the propofol suspension (Diprivan® 1%; AstraZeneca n.v., Destelbergen, Belgium) was diluted to 16.8 mmol L−1 with glucose 5%. A suspension of the vehicle, Intralipid® 10% (Fresenius Kabi, Schelle, Belgium) was diluted to the same extent. Following a 30 min stabilization period, infusion of propofol (n = 10) or its vehicle, Intralipid® (n = 10), was started at 0.1 mL min−1 via a side arm of the aortic cannula. After 15 min of infusion of the test drug, a 30 min period of total global normothermic ischaemia was imposed by bypassing the inflow cannula of the isolated heart in the perfusion circuit. During ischaemia, pacing was continued, but the administration of the test drug was stopped. Upon reperfusion, the propofol or Intralipid® infusion was restarted and maintained during 120 min of reperfusion. After the experiment, the support rabbit was killed with an overdose of pentobarbital and potassium chloride.
Diastolic function (Fig. 2, Protocol 2)
In a second series of experiments, the diastolic function was specifically examined by means of LVEDP-volume curves  constructed after a 30 min stabilization ('baseline') period, after 15 min perfusion with propofol (n = 7) or Intralipid® (n = 7), immediately following 30 min ischaemia, and after 30 min reperfusion with the test drug. To achieve this, the intraventricular balloon was emptied, then refilled by 0.2 mL increments while recording the systolic and diastolic pressure, until a LVEDP = 40 mmHg was reached. The balloon was then emptied to the volume corresponding with an LVEDP = 12 mmHg and the protocol was continued. The diastolic function was studied in a separate series of experiments and a limited number of times during each experiment to avoid influencing the results by the repeated filling of the balloon.
Coronary blood flow and left ventricular pressure (LVP) and its first derivative (LVdP/dt) were recorded continuously. In the first experimental protocol, maximal positive LVdP/dt (dP/dtmax) was registered as an index of contractility. Maximal negative LVdP/dt (dP/dtmin) was used as an initial parameter of diastolic function. LVEDP was measured as a parameter of ischaemic contracture. In the second experimental protocol, the time constant of ventricular pressure decline τ was calculated as a variable for early isovolumic relaxation using the non-zero asymptote method described by Raff and Glantz . Chamber stiffness was calculated from the LVEDP-volume curves by comparing VdP/dV-pressure relationships at a common pressure level of 12 mmHg according to Mirsky .
Changes in global haemodynamics, τ and VdP/dV within groups were analysed using ANOVA for repeated measurements, with Fisher's least significant difference as the post hoc test where appropriate. Differences between groups were analysed using unpaired t-tests. The analyses were preformed using the Statview® 5.0 software package (SAS Institute, Cary, NC, USA). P < 0.05 was considered as significant. Data are the mean ± 95% confidence interval.
Under baseline conditions, propofol decreased dP/dtmax and dP/dtmin while Intralipid® had no effect. Both groups displayed significant decreases in dP/dtmax during ischaemia and reperfusion. Recovery of dP/dtmax and dP/dtmin during reperfusion was similar in the two treatment groups: expressed as percentage of values measured during drug infusion at baseline, mean dP/dtmax at 120 min was 84% in the propofol group and 88% in the Intralipid® group; mean dP/dtmin at 120 min was 68% of the drug control value in the propofol group and 65% in the Intralipid® group. Propofol, but not Intralipid®, caused a reduction of left ventricular systolic pressure (LVSP) at baseline. LVSP was reduced during reperfusion in both groups (mean LVSP at 120 min was 74% of the drug control value in the propofol group and 78% in the IL group). The mean LVEDP increased during ischaemia in both groups, but less in the propofol group than in the Intralipid® group. While LVEDP returned rapidly to control values in the propofol group, it remained significantly elevated for 30 min in the Intralipid® group. The changes in coronary blood flow were comparable in the two groups throughout the experiment, although reactive hyperaemia was more pronounced in the Intralipid® group.
During infusion of propofol and Intralipid® in preischaemic conditions, there was no significant change in τ. During ischaemia, τ could not be analysed owing to the cessation of contractile function. Upon reperfusion, τ increased significantly in both groups and there was no significant difference between the two treatment groups throughout the protocol (Fig. 4a). Chamber stiffness (VdP/dV at 12 mmHg) did not change with propofol or with Intralipid® in preischaemic conditions, but it increased significantly during ischaemia in the Intralipid® group (from 34 ± 9 to 54 ± 8 mmHg). This persisted during reperfusion and remained significantly higher in the Intralipid® group than in the propofol-treated animals (43 ± 5 mmHg in the Intralipid® group versus 30 ± 8 mmHg in the propofol group after 30 min reperfusion) (Fig. 4b).
The results suggest that in the present model propofol has no significant cardioprotective effect during ischaemia and reperfusion except for a reduction in chamber stiffness. We observed no beneficial effect of propofol on the early, active phase of relaxation, nor on the recovery of systolic function after 30 min of global ischaemia. Although the absolute values of dP/dtmax were lower in the propofol group than in the control group during reperfusion, the recovery of dP/dtmax to drug control values was similar in the two groups.
The finding that propofol has no protective effect on systolic function is in agreement with two previous studies in dogs [6,7] and one in regionally ischaemic isolated rat hearts . In contrast, in a number of studies performed on isolated rat hearts perfused with Krebs-Henseleit bicarbonate, an improvement in systolic function after global ischaemia was observed [4,5,9,10]. The interest in the potential for propofol to protect the heart was raised by Kokita and Hara  who demonstrated that propofol attenuates mechanical and metabolic changes induced by exogenously applied hydrogen peroxide. This led to the hypothesis that propofol would have a cardioprotective effect against ischaemia and reperfusion injury via free radical scavenging, and it generated a number of studies in Krebs-Henseleit bicarbonate perfused rat hearts which showed improved systolic function. As far as we know, only one study in a different animal model (intracoronary propofol in open chest dogs) has reproduced this effect . An important difference between the present study and the aforementioned studies is the perfusion with blood rather than Krebs-Henseleit bicarbonate, which offers a closer resemblance to the clinical situation where the free fraction of propofol is <2% (in contrast to Krebs-Henseleit bicarbonate perfusion where the free fraction is far higher) . The use of a different animal model may also partly explain the different findings to do with systolic function, since it has recently been shown that the effect of propofol on myocardial contractility varies between different rodent species . The inconsistency of the findings so far reported suggests that the effect of propofol on systolic function in the ischaemic and reperfused myocardium is at best discrete and, in contrast to the volatile anaesthetics [5-8], unlikely to be significant at clinical concentrations.
In agreement with several other studies [4,9,10], we found that propofol protects the heart from the LVEDP rise characteristic of ischaemic contracture. For these reasons, our hypothesis was that propofol could have a protective effect on active or passive relaxation, or both. To our knowledge, this is the first study specifically to examine the effects of propofol on these two variables, which, compared with dP/dtmin, are less load-dependent.
In agreement with previous studies [20,21], under baseline conditions, propofol had no effect on τ or chamber stiffness. As described previously, the index of active relaxation τ was prolonged after myocardial ischaemia and reperfusion  and propofol did not influence this response. In contrast, the most important finding in our study is the increase of chamber stiffness that occurred during ischaemia (in particular) and reperfusion in the Intralipid® group but not in the propofol group. This variable has rarely been studied in the stunned myocardium. Data from one previous study in dogs suggest that disturbed chamber stiffness is a more important variable than active relaxation in the diastolic dysfunction secondary to myocardial stunning . We chose to analyse chamber stiffness using VdP/dV, as opposed to the stiffness constant α (which is derived by fitting the curves into the formula: EQUATION
where Ped and Ved are LVEDP and volume, respectively, and β is the extrapolated LVEDP (at zero volume)), because in contrast to α, the VdP/dV-pressure relationship takes into account variations in chamber size . Ischaemic preconditioning has been shown similarly to prevent the increase in dP/dV at common pressures associated with myocardial stunning in rabbit hearts . This, of course, does not imply a common mechanism but suggests that disturbed compliance is a relevant factor in ischaemia-reperfusion induced dysfunction.
The mechanisms behind the effects of propofol on myocardial function following ischaemia and reperfusion remain unclear. Myocardial stunning is a far more complex process than free radical formation alone . Since we observed no protective effect on systolic function, it seems unlikely that propofol should exert its effect via free-radical scavenging. In this respect, note that Intralipid® might attenuate the scavenging properties of propofol . Propofol also has calcium channel-blocking properties [25,26] - another mechanism that has been implicated in cardioprotection. This could possibly play a role in the effect of propofol on diastolic function in ischaemia-reperfusion damage.
In the present study, dP/dtmax decreased slightly after the start of the propofol infusion and remained lower than in the control group throughout the experiment. A mild negative inotropic effect of propofol has been described in a number of studies [19,25,27], but, in general, propofol is considered to have little direct effect on contractility [20,21]. There is evidence that propofol has multiple sites of action in the cardiomyocyte, and, interestingly, recent evidence suggests that these include mechanisms with both positive and negative inotropic effects . A propofol-induced decrease in force of contraction during ischaemia could possibly explain the slight protective effect of propofol on diastolic function, since the degree of myocardial stunning is strongly influenced by the extent and duration of ischaemia . The protection against the rise in LVEDP during ischaemia indeed suggests that propofol acts before reperfusion. One plausible mechanism is that propofol attenuates the rise in left ventricular tension during ischaemia, reducing the mechanical damage to the myofilaments or non-contractile elements, hence improving diastolic function.
Limitations of the present work include the fact that biochemical and molecular alterations were not studied. As already mentioned, a large number of possible mechanisms of action upon the cardiomyocyte have been proposed, but it remains far from clear which are relevant in the physiological setting of ischaemia and reperfusion. However, this study could act as a stimulus for the further exploration of these mechanisms. The maximal blood propofol concentration occurring during induction of anaesthesia has been reported to be 112 μmol L−1. It has been demonstrated in vitro that a concentration of 168 μmol L−1 is required to produce radical-scavenging activity ; we therefore decided to use this concentration in our study to maximize the likelihood of observing any effect. The concentration of propofol used in the present study must therefore be considered as supraclinical. An extensive investigation of dose-response characteristics was beyond the scope of the present study; further investigation is therefore required. We acknowledge that our data concerning diastolic function only encompass a limited number of moments in time. We chose to limit the number of pressure-volume curve constructions because we believe that repetitive filling of the balloon may influence the passive characteristics we were examining, thus influencing the results. We chose 30 min reperfusion for the post-ischaemic measurement because at this time there is still a clear systolic dysfunction, though the initial changes associated with reperfusion (such as the characteristic reactive hyperaemia) have at least partly returned to normal. For a further description of the effects of propofol over time during reperfusion, a more extensive study is required. Finally, one cannot exclude an effect of the basal anaesthetic of the support rabbit on our results. While there is no evidence that barbiturates influence the pathophysiology of ischaemia-reperfusion injury , opioids may have a preconditioning effect . However, both the drug and the vehicle groups received identical background anaesthesia, making it unlikely that background anaesthesia influenced the results.
In summary, our results suggest that propofol at supraclinical concentrations slightly improves passive diastolic function but has no protective effect on active relaxation or systolic function in globally ischaemic and reperfused isolated rabbit hearts. Further studies are required to address the dose-response relationship and explain the mechanisms involved.
The authors thank Veerle Leunens and Dr Frank Weekers for excellent technical assistance.
1. Warltier DC, al-Withiqui MH, Kampine JP, Schmeling WT. Recovery of contractile function of stunned myocardium in chronically instrumented dogs is enhanced by halothane or isoflurane. Anesthesiology
2. Toller WG, Gross ER, Kersten JR, et al.
Sarcolemmal and mitochondrial adenosine triphosphate-dependent potassium channels: mechanism of desflurane-induced cardioprotection. Anesthesiology
3. Kokita N, Hara A. Propofol attenuates hydrogen peroxide-induced mechanical and metabolic derangements in the isolated rat heart. Anesthesiology
4. Ko SH, Yu CW, Lee SK, et al.
Propofol attenuates ischemia-reperfusion injury in the isolated rat heart. Anesth Analg
5. Kokita N, Hara A, Abiko Y, Arakawa J, Hashizume H, Namiki A. Propofol improves functional and metabolic recovery in ischemic reperfused isolated rat hearts. Anesth Analg
6. Ross S, Munoz H, Piriou V, Ryder WA, Foëx P. A comparison of the effects of fentanyl and propofol on left ventricular contractility during myocardial stunning. Acta Anaesthesiol Scand
7. Meissner A, Weber TP, Van Aken H, Zbieranek K, Rolf N. Recovery from myocardial stunning is faster with desflurane compared with propofol in chronically instrumented dogs. Anesth Analg
8. Ebel D, Schlack W, Comfere T, Preckel B, Thamer V. Effect of propofol on reperfusion injury after regional ischaemia in the isolated rat heart. Br J Anaesth
9. Mathur S, Farhangkhgoee P, Karmazyn M. Cardioprotective effect of propofol and sevoflurane in ischemic and reperfused rat hearts: role of Katp
channels and interaction with the sodium-hydrogen exchange inhibitor HOE 642 (Cariporide). Anesthesiology
10. Javadov SA, Lim KH, Kerr PM, et al.
Protection of hearts from reperfusion injury by propofol is associated with inhibition of the mitochondrial permeability transition. Cardiovasc Res
11. Moreyra AE, Gelpi RJ, Mosca SM, Cingolani HE. Chronic administration of nicardipine attenuates myocardial stunning in isolated rabbit hearts. J Mol Cell Cardiol
12. Chen CC, Morishige N, Masuda M et al.
R56865, a Na+
- and Ca2+
-overload inhibitor, reduces myocardial ischemia-reperfusion injury in blood-perfused rabbit hearts. J Mol Cell Cardiol
13. Van de Velde M, DeWolff M, Leather HA, Wouters PF. Effects of lipids on the functional and metabolic recovery from global myocardial stunning in isolated rabbit hearts. Cardiovasc Res
14. Mosca SM, Gelpi RJ, Cingolani HE. Dissociation between myocardial relaxation and diastolic stiffness in the stunned heart: its prevention by ischemic preconditioning. Mol Cell Biochem
15. Raff GL, Glantz SA. Volume loading slows left ventricular isovolumic relaxation rate. Evidence of load-dependent relaxation in the intact dog heart. Circ Res
16. Mirsky I. Assessment of diastolic function: suggested methods and future considerations. Circulation
17. Yoo KY, Yang SY, Lee J, et al.
Intracoronary propofol attenuates myocardial but not coronary endothelial dysfunction after brief ischaemia and reperfusion in dogs. Br J Anaesth
18. Mazoit JX, Samii K. Binding of propofol to blood components: implications for pharmacokinetics and for pharmacodynamics. Br J Clin Pharmacol
19. Van Klarenbosch J, Stienen GJ, de Ruijter W, Scheffer GJ, de Lange JJ. The differential effect of propofol on contractility of isolated myocardial trabeculae of rat and guinea-pig. Br J Pharmacol
20. Graham MR, Thiessen DB, Mutch WC. Left ventricular systolic and diastolic function is unaltered during propofol infusion in newborn swine. Anesth Analg
21. Pagel PS, Schmeling WT, Kampine JP, Warltier DC. Alteration of canine left ventricular diastolic function by intravenous anesthetics in vivo.
Ketamine and propofol. Anesthesiology
22. Kunichika H, Katayama K, Sakai H, Yonezawa T, Matsuzaki M. The effect of left ventricular chamber compliance on early diastolic filling during coronary reperfusion. Jpn Circ J
23. Bolli R, Marban E. Molecular and cellular mechanisms of myocardial stunning. Physiol Rev
24. Green TR, Bennet SR, Nelson VM. Specificity and properties of propofol as an antioxidant free radical scavenger. Toxicol Appl Pharmacol
25. Cook DJ, Housmans PR. Mechanism of the negative inotropic effect of propofol in isolated ferret ventricular myocardium. Anesthesiology
26. Yang CY, Wong CS, Yu CC, Luk HN, Lin CI. Propofol inhibits cardiac L-type calcium current in guinea pig ventricular myocytes. Anesthesiology
27. Kanaya N, Murray PA, Damron DS. Propofol and ketamine only inhibit intracellular Ca2+
transients and contraction in rat ventricular myocytes at supraclinical concentrations. Anesthesiology
28. Kanaya N, Murray PA, Damron DS. Propofol increases myofilament Ca2+
sensitivity and intracellular pH via activation of Na+
exchange in rat ventricular myocytes. Anesthesiology
29. Wouters PF, Van de Velde M, Van Aken H, Flameng W. Ischemic event characteristics determine the extent of myocardial stunning in conscious dogs. Basic Res Cardiol
30. Zhou W, Fontenot HJ, Liu S, Kennedy RH. Modulation of cardiac calcium channels by propofol. Anesthesiology
31. Mullenheim J, Molojavyi A, Preckel B, Thamer V, Schlack W. Thiopentone does not block ischemic preconditioning in the isolated rat heart. Can J Anaesth
32. 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