To the Editor:
Although the article by Leesar and coworkers in the June 3, 1997, issue of Circulation1 concludes that “adenosine preconditions human myocardium,” in our opinion their investigation in 30 patients undergoing balloon angioplasty (PTCA) and pretreatment with normal saline or adenosine (2 mg/min for 10 minutes) merely allows them to reiterate their introductory statement that “to the best of our knowledge, evidence that adenosine can precondition human myocardium in vivo is still lacking.” First of all, the fact that there have been recent studies demonstrating reduced signs of myocardial ischemia during repeated balloon inflations does not necessarily suggest the existence of ischemic preconditioning in the conventional, ie, biochemical, sense. As long as the contribution of the gradual opening of collateral channels with successive balloon occlusions is not accounted for in the observed phenomenon of attenuated myocardial ischemia, ischemic “preconditioning” may be not a biochemical but a biophysical process of collateral recruitment due to the temporally increasing effect of a pressure gradient across the anastomoses between the nonstenotic donor vessel and the occluded recipient vessel. Hence, the finding that such a thing as ischemic preconditioning, in the normally used sense of the word, exists in humans is very controversial and is inversely associated with whether or not the authors accounted for collaterals.2 Second, and considering the effects of adenosine on the resistance and conductance vessels of the coronary circulation, it is similarly disputable whether there is such a thing as adenosine-induced preconditioning or whether the phenomenon of mitigated myocardial ischemia during coronary occlusion on adenosine administration is actually pharmacologically induced collateral recruitment. After all, exogenously administered adenosine is known to cause not only profound microvascular coronary dilatation3 but also increased flow indexes through collateral channels.4 Therefore, as long as Leesar et al do not treat coronary collaterals as a covariable in the outcome of reduced myocardial ischemia in response to adenosine administration before PTCA, they cannot conclude that “adenosine preconditions human myocardium,” or else the scientific community has to redefine preconditioning by extending its meaning to hemodynamic events such as enhanced flow along an increased (collateral) pressure gradient.
Finally, it is hard to imagine that the ST-segment shifts of the intracoronary ECGs shown in Figure 1 of the article by Leesar et al amount to 20 mm and more (up to 60 mm). The y axis of this graph must have been mixed up with that of Figure 2 illustrating the summed ST-segment shifts of all 11 surface ECG leads.
- Copyright © 1998 by American Heart Association
Leesar M, Stoddard M, Ahmed M, Broadbent J, Bolli R. Preconditioning of human myocardium with adenosine during coronary angioplasty. Circulation. 1997;95:2500–2507.
Wilson R, Wyche K, Christensen B, Zimmer S, Laxson D. Effects of adenosine on human coronary arterial circulation. Circulation. 1990;82:1595–1606.
Seiler C, Fleisch M, Meier B. Direct intracoronary evidence of collateral steal in humans. Circulation. 1997;96:4261–4267.
Contrary to what Seiler et al assert, several studies support the notion that alleviation of ischemia after the first balloon inflation during PTCA is a manifestation of ischemic preconditioning (PC). Although Seiler et al cite a paper by Dupouy et alR1 as evidence that ischemic PC does not exist in humans, the results of that study do not support the conclusions of the authors for reasons that have been detailed elsewhere.R2 Seiler and colleagues assert “the finding that such a thing as ischemic preconditioning … exists in humans is … inversely associated with whether or not the authors accounted for collaterals.” This assertion is incorrect. Although Cribier et alR3 found increased collateral perfusion during subsequent balloon inflations, they observed that in individual cases, myocardial ischemia decreased with repeated inflations despite the fact that there was no evidence of improved collateral circulation. Thus, the increased tolerance to ischemia did not correlate with indexes of collateral function.R3 In patients undergoing PTCA of the left anterior descending coronary artery, Tomai et alR4 demonstrated that the average peak flow velocity recorded in the right coronary artery during the first and second balloon inflation (an index of collateral recruitment) was not significantly different, yet there was increased tolerance to ischemia during the second inflation. In that study,R4 only ≈20% of the patients exhibited a modest increase in blood flow velocity at the end of the second inflation compared with the first inflation, and again, this increase did not correlate either with the changes in ST-segment shifts or with the severity of chest pain. Using a similar technique, Kyriakidis et alR5 found that only ≈30% of the patients exhibited progressive collateral recruitment after the first inflation. It should be noted that the blood flow velocity changes in the contralateral coronary artery have been shown to be a reliable index of collateral perfusion during PTCA and to be more accurate than other indexes, such as measurements of occlusion pressure through the balloon catheter or coronary angiography.
A recent study by Sakata et alR6 demonstrates that enhanced tolerance to ischemia occurred in patients undergoing PTCA in whom there was no recruitable collateral circulation during balloon inflation, as assessed by myocardial contrast echocardiography. Unlike coronary angiography, which only detects collateral vessels with a diameter >100 μm, contrast echocardiography reflects myocardial perfusion in the occluded bed and therefore is a more sensitive technique. These studiesR4 R6 indicate that ischemic PC develops during repetitive balloon inflations in the course of PTCA independently of changes in collateral perfusion.
It should also be pointed out that evidence of collateral recruitment during PTCA has been found in only ≈20% to 50% of the patients.R3 R4 R5 R6 If the increased tolerance to ischemia observed during subsequent balloon inflations were due solely to collateral recruitment, then one would expect this phenomenon to occur only in a minority of the patients. Instead, our study,R7 as well as many other studies (reviewed in Reference 2), have demonstrated increased tolerance to ischemia in most or even all of the patients. In addition, this increased tolerance to ischemia during subsequent inflations can be abolished by pharmacological manipulations, such as glibenclamideR8 and adenosine receptor antagonists.R9 R10 If the mechanism responsible for this increased tolerance to ischemia were solely collateral recruitment, it seems unlikely that glibenclamide and adenosine antagonists would block it.
Seiler et al speculate that in our study,R7 administration of adenosine caused enhanced tolerance to ischemia during the first inflation simply by inducing collateral recruitment. This speculation is unsupportable for several reasons. First, we allowed a 10-minute interval between the end of the adenosine infusion and the first balloon inflation. The plasma half-life of adenosine in humans is on the order of seconds. We have recently studied 3 patients in whom we infused intracoronary adenosine at the same dose used in our previous studyR7 (2 mg/min for 10 minutes) and measured coronary flow with a 0.014-in Doppler guidewire and quantitative coronary angiography at baseline, at the end of adenosine infusion, and 5 and 10 minutes later. In all 3 patients, adenosine-induced hyperemia subsided completely within 10 minutes of the end of the adenosine infusion. Therefore, any vasodilation induced by adenosine in our studyR7 should have resolved before the first balloon inflation. Second, Seiler et al seem to confuse the effects of intravenous adenosine with those of intracoronary adenosine. In order for collateral flow to increase in the presence of a complete coronary occlusion, collateral vessels need to be dilated in their entire length, not just within the collateral-dependent vascular bed. Because in our study adenosine was administered by the intracoronary route and produced no changes in systemic hemodynamics, it seems unlikely that adenosine dilated the portion of collateral vessels that was outside of the collateral-dependent vascular bed. (In contrast, intravenous adenosine may dilate the entire length of collateral vessels.) Taken together, the above considerations strongly support the conclusion that the enhanced tolerance to ischemia after pretreatment with adenosineR7 was unrelated to increased collateral perfusion.
With regard to the measurements of ST-segment shifts on the intracoronary ECG, it would appear that Seiler and coworkers are not familiar with the pertinent literature. Numerous previous studies have reported ST-segment shifts >1 mV (10 mm) during PTCAR4 R8 R9 R10 (reviewed in Reference 2). In these studies, ST-segment elevation has been found to be much lower in the surface ECG than in the intracoronary ECG. Obviously, the precise magnitude of the ST-segment shifts can vary depending on a number of factors, including, among others, the size of the ischemic region, the exact position of the intracoronary wire, and the duration of the inflation.
Dupouy P, Geschwind H, Pelle G, Aptecar E, Hittinger L, El Ghalid A, Dubois-Rande JL. Repeated coronary artery occlusions during routine balloon angioplasty do not induce myocardial preconditioning in humans. J Am Coll Cardiol. 1996;27:1374–1380.
Tomai F, Crea F, Gaspardone A, Versaci F, Ghini AS, De Paulis R, Chiariello L, Gioffre PA. Phentolamine prevents adaptation to ischemia during coronary angioplasty: role of α-adrenergic receptors in ischemia preconditioning. Circulation. 1997;96:2171–2177.
Kyriakidis MK, Petropoulakis PN, Tentolouris CA, Marakas SA, Antonopoulos AG, Kourouclis CV, Toutouzas PK. Relation between changes in blood flow of the contralateral coronary artery and the angiographic extent and function of recruitable collateral vessels arising from this artery during balloon coronary occlusion. J Am Coll Cardiol. 1994;23:869–878.
Sakata Y, Kodama K, Kitakaze M, Masuyama T, Hirayama A, Lim YJ, Ishikura F, Sakai A, Adachi T, Hori M. Different mechanisms of ischemic adaptation to repeated coronary occlusion in patients with and without recruitable collateral circulation. J Am Coll Cardiol. 1997;30:1679–1686.
Leesar MA, Stoddard M, Ahmed M, Broadbent J, Bolli R. Preconditioning of human myocardium with adenosine during coronary angioplasty. Circulation. 1997;95:2500–2507.
Tomai F, Crea F, Gaspardone A, Versaci F, De Paulis R, Penta de Peppo A, Chiariello L, Gioffre PA. Ischemic preconditioning during coronary angioplasty is prevented by glibenclamide, a selective ATP-sensitive K+ channel blocker. Circulation. 1994;90:700–705.
Claeys MJ, Vrints CJ, Bosmans JM, Conraads VM, Snoeck JP. Aminophylline inhibits adaptation to ischaemia during angioplasty: role of adenosine in ischaemic preconditioning. Eur Heart J. 1996;17:539–544.
Tomai F, Crea F, Gaspardone A, Versaci F, De Paulis R, Polisca P, Chiariello L, Gioffre PA. Effects of A1 adenosine receptor blockade by bamiphylline on ischaemic preconditioning during coronary angioplasty. Eur Heart J. 1996;17:846–853.