In Reply:—
We thank Drs. Djalali and Sadovnikoff for their interest in our study on the influence of modalities of administration of sevoflurane on its cardioprotective properties in patients undergoing coronary surgery with cardiopulmonary bypass.1We regret that their review of the cardioprotective properties of volatile anesthetic agents was limited to the article of Conzen et al. 2During the past years, there has been increasing evidence that these cardioprotective effects that had been demonstrated in experimental studies could also been observed in clinical practice.2–6
Demonstration of these cardioprotective effects was not solely based on analysis of troponin I concentrations but also on the evaluation of systemic hemodynamics parameters and variables of left ventricular function. When analyzing the literature, it was striking to observe that the clearcut clinical beneficial effects were only apparent in protocols where the volatile anesthetic was administered throughout the entire procedure and not in protocols where the volatile anesthetic was given solely before aortic cross clamping (preconditioning protocols) (references 12–16 from De Hert et al. 1). This suggested that the beneficial effects on myocardial function of anesthetic preconditioning protocols observed in the experimental setting do not necessarily hold for the clinical setting. This was what our most recent study confirmed.1In this study, the cardioprotective effects of an anesthetic regimen were clinically most apparent when sevoflurane was administered throughout the surgical procedure. This was evident from a lower postoperative release of troponin I but also from better early postoperative hemodynamic and left ventricular function. When sevoflurane was administered only before aortic cross clamping or after completion of the coronary anastomoses, the postoperative troponin I release was not different from that observed with the total intravenous anesthetic regimen. However, postoperative recovery of stroke volume occurred earlier, suggesting that some cardioprotection may also be present with these pre– and post–cardiopulmonary bypass protocols.
The question on the effects of sevoflurane administered only during the period of cardiopulmonary bypass deserves some attention. During total cardiopulmonary bypass, the heart is “bypassed” from the circulation, and as such, virtually no sevoflurane could be administered to the myocardial tissue. The alternative approach to address this question is to add sevoflurane to the cardioplegic solution.
It should be noted that in our study, coronary anastomoses were performed using intermittent aortic cross clamping. This implies that during cardiopulmonary bypass, periods of ischemia are alternated with periods of reperfusion. As such, two protective mechanisms may be involved, which are ischemic and anesthetic preconditioning, and potential beneficial effects cannot unequivocally be attributed to sevoflurane. Therefore, this specific clinical situation does not allow one to address the question of whether sevoflurane administered only during cardiopulmonary bypass would be cardioprotective.
Substantial efforts were made by Drs. Djalali and Sadovnikoff to comment on the interpretation of troponin I as a marker of myocardial damage. Troponin is currently the routine marker used to determine extent of myocardial damage. As clearly stated in the methodology of the current study—as in the others—careful attention was paid to the random assignment of the patients to groups. There were no differences in patient characteristics or in perioperative variables such as number of grafts, duration of aortic cross clamping, and cardiopulmonary bypass time. Random assignment of the patients to the different groups was such that each individual surgeon operated a similar number of patients in each group in each study. The consequence is that the only variable between the groups in all our studies was the anesthetic protocol used, and any difference in any of the variables could therefore be essentially related to this difference in anesthetic protocol. It should again be emphasized that the conclusions on the cardioprotective properties of the volatile agent were not based solely on postoperative concentrations of troponin I but also on different variables of myocardial function.
Finally, isoflurane has indeed been shown in many experimental studies to exhibit cardioprotective properties. As such, it can be expected that these properties should also be apparent in the clinical setting. There are many reasons to choose one particular anesthetic agent, among which economical concerns certainly are to be considered. However, the real economical impact relies not on the individual cost of the specific agent but rather on the fact that cardioprotective properties of some anesthetic techniques may result in a reduction of intensive care unit and hospital durations of stay.6
* University Hospital Antwerp, Edegem, Belgium. stefan.dehert@ua.ac.be