Q: Professor Heusch, it is a true pleasure to discuss a fascinating article of your group on the vasoconstrictor potential of saphenous vein graft (SVG) lesions, just published in Circulation Research. This is truly a terrific study and a prime example of translational research – congratulations!
The no-reflow phenomenon continues to be a dreaded event in the cath lab and is seen in about a third of AMI patients undergoing primary percutaneous coronary intervention (PCI) and up to nearly a half of all SVG interventions. Even with PCI on native coronary arteries lesions characterized by prominent plaque burden, one can encounter the no-reflow phenomenon with a chance of about 1 in 20. Your study now points into the direction that it is not all about the particulate matter but that vasoconstricting substances are expressed in and released from the atherosclerotic plaque more clearly than any other study before. Moreover, for the first time your study addresses the question of the “no-reflow antidote” very elegantly. Would you mind summarizing your key findings for us?
A: Thank you for your kind comments! Indeed, we found that during PCI of SVG there is not only release of particulate debris from the stented atherosclerotic lesion which might, if not retrieved by aspiration, physically obstruct the coronary microcirculation, but there is also a significant release of soluble vasoconstrictor substances, notably serotonin and thromboxane, along with the cytokine TNFα. TNFα sensitizes the vasculature for constrictor responses to serotonin and thromboxane. Verapamil and nitroprusside potently antagonize the aspirate-induced vasoconstriction, but adenosine does not.
Q: In essence then, there were two components to your work. The bedside component was the cath lab, where 22 patients underwent PCI of a SVG lesion with a distal protection device. Aspirates were taken distal to the lesion before inflation as a pre-PCI reference. Then, the protection device was inflated to guarantee complete distal occlusion and PCI was performed, after which another aspirate was taken from the “column of blood” above the protection device point. Is this correct? Why did you decide not to take an aspirate right after inflation of the protection device and just before the intervention?
A: Your description of our aspiration procedure is correct. We did take a control aspirate at the site of the lesion prior to PCI. Then, however, we wanted to have the aspirate with a biochemical fingerprint of the stented lesion. Also, we wanted to have a realistic estimate of physical debris and soluble substances which might contribute to microvascular obstruction, if not retrieved by aspiration.
Q: You filtered the aspirate and obtained particulate debris. Did you examine the debris further? Was there any correlation with the amount or composition of the debris and the plasma analysis results or any other angiographic or clinical parameter?
A: We do regular histology on the particulate debris and regularly identify parts of the atherosclerotic lesion (foam cells, cholesterol crystals, lipid material) along with superimposed platelet/fibrin-thrombi. So far, we have not found an obvious correlation of particulate debris with soluble substances or with angiographic/clinical parameters. However, the magnitude of vasoconstriction correlates to angiographic diameter stenosis, plaque volume on IVUS and CCS classification (JACC 47,2006,981-6). Also, there is a correlation of the amount of TNFα release to the angiographic degree of restenosis 6 months later (Am J Physiol 292, 2007, H2295-9).
Q: Subsequent processing of the aspirate was essentially generation of plasma samples, which were then immediately frozen. Did the freezing process influence the results in any way?
A: No, in preliminary experiments the responses to fresh aspirate plasma were not different from those to quickly frozen plasma.
Q: Looking at the plasma analyses that you performed, which vasoconstrictors were released from the SVG lesions upon catheter-based intervention?
A: First and foremost, it is serotonin. Thromboxane adds to and potentiates the serotonin-induced vasoconstriction, and TNFα sensitizes the vasculature to vasoconstrictor responses.
Q: Did you restrict your analysis to one SVG lesion per patient?
A: Yes, in the present cohort of patients only one SVG lesion per patient was analyzed.
Q: As one would expect the SVGs in your study population were at the “age” for PCI, about 11 years on average. Were there some additional morphological characteristics that your noted to correlate with the vasonconstrictor potential, e.g. ulceration or thrombus? Did your group happen to perform IVUS or OCT to further characterize the lesions?
A: Indeed, we have performed IVUS in a number of patients. Notably, the larger the necrotic core volume on virtual histology, the larger was the troponin release as evidence for myonecrosis (Basic Res Cardiol 103, 2008, 587-97).
Q: Were there any particular biomarkers at baseline that would correlate with very high vasoconstrictor release such as cTnI and CRP elevation?
A: No, so far we could not predict the release of soluble vasoconstrictors during PCI with any biomarker before PCI.
Q: What about those patients in which TIMI flow was not III before and/or after PCI – did their aspirate analysis results differ from those with TIMI III flow?
A: In our 22 patients, none had TIMI III before PCI. After PCI, 3 patients had TIMI II and 19 had TIMI III; this asymmetrical distribution of TIMI flow does not permit to look for TIMI- related differences in the aspirate at this point.
Q: It is interesting that all of the “no-reflow” potential was seen in patients who were deemed to be on optimal medial therapy. Is there anything that could have been optimized? You noted that thromboxane was not suppressed as much as expected with aspirin. How many were pretreated with Plavix? How many received GPIIb/IIIa inhibitor treatment? How important to you view platelet activation suppression?
A: Almost all (20 out of 22) patients were pretreated with aspirin+clopidogrel. None of them had GPIIb/IIIa inhibition. I would speculate that the release of serotonin and thromboxane into the aspirate during PCI is not secondary to insufficient platelet inhibition, but from other cellular sources, such as macrophages in the atherosclerotic plaque.
Q: The second, bench-side component of your study was the “vasomotor bioassay”. Would you mind telling us more about the experimental setup and protocol that you used? Why did you decide on rat mesenteric arteries? How comparable are they to human myocardial microvessels, which are the site of action of the no-reflow phenomenon?
A: Our current bioassay was chosen because it is easy and feasible. In fact, rat mesenteric arteries have a similar diameter as isolated human coronary microvessels in studies by David Gutterman, and they also have a similar receptor arrangement for adenosine, serotonin and thromboxane.
Q: What happened when you exposed these arteries to aspirate taken before and after the intervention?
A: Whereas the aspirate taken before PCI induced only 20-30% of maximal KCl-induced vasoconstriction, the aspirate taken after PCI induced almost 100% vasoconstriction.
Q: It is quite impressive to observe the vasoconstriction response of the volume of blood that would usually spread downstream in unprotected SVG intervention. You also performed experiments with blood from healthy volunteers to reproduce and validate the vasoconstrictor cocktail responsible for the PCI-related findings – is this correct? What did you find?
A: Plasma from healthy volunteers was taken as a negative control and induced only minor vasoconstriction.
Q: Finally, and one of the most important findings, you examined the “relief potency” of substances commonly used in the cath lab to counteract the no-reflow phenomenon. Please tell us more about your exciting observations.
A: Indeed, contrary to the common notion that adenosine causes maximum coronary dilation, adenosine was almost ineffective in attenuating the aspirate-induced vasoconstriction, whereas both verapamil and nitroprusside dose-dependently, but ultimately completely abrogated vasoconstriction. So, clearly adenosine is not useful against the no-reflow phenomenon (Basic Res Cardiol 105, 2010, 1-5), whereas verapamil and nitroprusside are.
Q: Obviously, the important implication of your study is that we should use more nitroprusside and verapamil rather than adenosine. Should we do proactively before SVG-PCI or would it be sufficient enough to treat as needed? Should we look at this in a randomized clinical trial translating your findings further?
A: From a theoretical point of view, verapamil has the disadvantage of being a negative chronotrope and negative inotrope, but it would also attenuate stunning and reperfusion injury. Also, verapamil in contrast to nitroprusside would not require continuous infusion. Apart from the general rule that prevention is better than cure, I would not know whether or not pro-active administration of verapamil or nitroprusside is better. Certainly, I would welcome a clinical trial which attempts to translate our findings further.
Q: Along these lines, which protection device is to be used? Should we use balloon occlusion rather than filter devices? Should these be used in native vessel primary PCI at the time of AMI and perhaps even with PCI of lesions with extensive plaque burden in the native arteries?
A: Filter devices have a cut-off at about 80µm diameter, so would not retrieve any smaller particulate debris. However, particles with only 40µm diameter cause already arrhythmias, microinfarcts with an inflammatory reaction, contractile dysfunction, and reduction of coronary reserve (Circulation 120, 2009, 1822-36). Also, all soluble substances are not prevented from their exposure to the microcirculation by filter devices. Of course, balloon aspiration devices can only be used in vessels with a sufficient diameter to be introduced.
Q: Is there anything we missed to point out from you study or anything that you would like to add?
A: No, I think we have addressed all relevant issues.
A: Thank you for the opportunity to further highlight our translational study. We firmly believe, that not only particulate debris but also soluble vasoconstrictor and inflammatory substances are decisive for microvascular flow impairment and the no-reflow phenomenon.
Our mission: To reduce the burden of cardiovascular disease
© 2017 European Society of Cardiology. All rights reserved