Response to Letter by Zaugg et al: STAT5 Activation and Cardioprotection By Remote Ischemic Preconditioning in Patients Undergoing Coronary Revascularization During Isoflurane Anesthesia
Response: We appreciate the interest of Zaugg et al 1 in our studies on remote ischemic preconditioning (RIPC) in humans and the opportunity to respond to the issues raised. Both our recent studies on signal transducer and activator of transcription 5 (STAT5) activation in human left ventricular myocardium with RIPC 2 and on the comparison of isoflurane and propofol anesthesia with respect to the efficacy of RIPC 3 were derived from our ongoing clinical trial (NCT 01406678). Obviously, the
... Obviously, the STAT5 data are only from that subset of patients who consented to left ventricular biopsies. We caution to overinterpret modest differences in peak troponin I concentrations between this subset and those of the larger cohort, given the wide interindividual variation of troponin I concentrations and the time course of its release. In fact, we therefore compare the entire time course of troponin I concentrations to assess protection offered by RIPC versus control. We identified STAT5 tyrosine 694 phosphorylation as the only signal that differed among more than 20 proteins and their phosphorylation state that have been suggested previously to play a role in cardioprotection. 4 As detailed in our original article, 2 each (phospho-) protein was blotted on four separate gels and subjected to four separate comparisons, ie, two withingroup paired comparisons from baseline to early reperfusion, separately for RIPC and controls, and two between-group unpaired comparisons between RIPC and controls, separately at baseline and early reperfusion. As for STAT5 tyrosine 694 phosphorylation, this entire analysis was repeated and confirmed with a different antibody. Therefore, a type I error is highly unlikely to account for our positive finding on STAT5 activation by RIPC in human left ventricular myocardium. Considering the modest differences in just peak troponin I concentrations, the extrapolation by Zaugg et al to better protection by cold blood cardioplegia in a 1999 study from Paris 5 than by crystalloid cardioplegia in our study 3 is far-fetched. Such conclusion on better protection is also somewhat naive, without consideration of other patient characteristics and surgical variables, notably of cross-clamp time, which was markedly longer in our study. We also do not share the enthusiasm of Zaugg et al for microarray gene expression analysis of right atrial biopsies. Apart from the type I error to which such microarray analyses are known to be susceptible, it is the protein and its activity, but not the mRNA, that will eventually mediate the acute protection by RIPC, if any. Also, whereas the right atrium is easily amenable for tissue harvesting, it is probably the cardiac chamber with the least need for protection by RIPC during cardioplegic cardiac arrest. Therefore, our STAT5 tyrosine 694 protein phosphorylation data from human left ventricular myocardium are certainly much more relevant for cardioprotection. Two recent, randomized, double-blind, clinical studies found no protection by RIPC. 6,7 Many confounders may interfere with cardioprotection, including age, 8 comorbidities, and comedications. 9 -11 In fact, the use of propofol 3 may have offset the protection in the Rahman et al study, 6 and the inclusion of diabetic individuals may have offset the protection in the study by Karuppasamy et al. 7 Exactly for these reasons, we have carefully excluded diabetic individuals, used isoflurane, and avoided propofol, even for induction of anesthesia in our mechanistic studies of RIPC in humans so far. In conclusion, we have no evidence that isoflurane anesthesia interferes with cardioprotection by RIPC in humans or that it makes it redundant.