Diffuse Coronary Disease and Atherothrombosis: A Rationale for Long-Term Therapy to Prevent Recurrent Ischemic Events

When examining the diffuse nature of coronary disease, it is important to recognize the fact that acute coronary syndromes are just the tip of the atherothrombotic “iceberg” (Figure 1). That is, they are the events which are clinically manifested as unstable angina, non-ST segment elevation myocardial infarction, and ST-segment elevation myocardial infarction. What is often not obvious to the clinician initially, however, is that the patients who present with acute coronary syndromes also have multiple coronary plaques present elsewhere, as well as hyperreactive platelets, and ongoing vascular inflammation. Diagnostic techniques. There are a number of diagnostic techniques available to evaluate unstable plaque, some of which include established invasive techniques such as angiography, intravascular ultrasound (IVUS), and angioscopy (Figure 2). Angiography is considered the gold standard for detecting coronary artery disease, but is not particularly good at detecting vulnerable plaque. Intravascular ultrasound is slightly more invasive than routine angiography, but is better at detecting vulnerable plaque. Angioscopy is even more invasive, and also more sensitive in delineating the presence of unstable plaques. There are a number of clinically promising noninvasive diagnostic techniques currently being studied which include magnetic resonance imaging (MRI) and electron beam computed tomography (EBCT). Some of the emerging invasive diagnostic technologies include thermography, intravascular tomography (OCT), transesophageal MRI, intravascular tissue Doppler, and intravascular spectroscopy. Evidence of multiple plaques. Data from several sources show that most myocardial infarctions arise from smaller, non-flow limiting stenoses (Figure 3).¹ Indeed, the majority of myocardial infarctions are due to plaques that are 2–6 As the data from these studies show, the prevalence of multiple plaques found depends on the sensitivity of the diagnostic technique used. For example, with angiography, the prevalence of multiple plaques ranges from 14–50%. However, if intravascular ultrasound is used, 79% of patients are found to have multiple plaques. If angioscopy is performed, 100% of patients appear to have multiple plaques. Thus, the more sensitive the diagnostic technique used to delineate the presence of unstable plaque and multiple coronary plaques, the greater the prevalence appears to be. An excellent study by Rioufol et al. demonstrated the number and distribution of multiple plaques in patients with acute coronary syndromes as assessed by intravascular ultrasound (Figure 5).5 Over 80% of patients had at least one or more unstable plaques that were visualized based on the presence of one or more plaques in the same artery that had the culprit lesion, unstable plaque in a non-culprit artery, or unstable plaque indeed found in all three coronary arteries. Intravascular ultrasound also revealed that most patients had multiple unstable plaques in the culprit artery, while a sizeable number of patients had unstable plaques visualized in all three coronary arteries. Using angioscopy, investigators have found multiple vulnerable plaques throughout the coronary arterial tree. Hyperreactive platelets. The concept of hyperreactive platelets also ties in to the broader concept of multiple plaques and multiple plaque rupture. Data from the TIMI 12 study show just how long platelet activation persists after a patient presents with an acute coronary syndrome (Figure 6).⁷ It is interesting to see just how prevalent platelet activation is initially, regardless of whether the treatment modality is mechanical or chemical reperfusion. Also interesting is the fact that platelet activation persists not only out to one week, but out to one month post-event as well. The TIMI 12 data emphasize the importance of the platelet, not only in the acute ischemic event, but in the chronic phase of therapy as well. Data also illustrate the point that platelet aggregability is detrimental (Figure 7). In five-year outcomes following acute coronary syndromes, the death rates (34.6%) and the overall cardiac event rates (46.2%) are much higher in patients with heightened platelet aggregability.⁹ Thus, the more reactive the platelets are, the higher the rate of adverse cardiac events, and indeed, the higher the rate of death.⁸The role of inflammation. Inflammation, in addition to platelet activation, is the other key player in acute coronary syndromes and multiple plaque rupture. Data published by Rioufol, et al. look at the incremental value and the relative value of different prognosticators of risk.⁴ LDL cholesterol is, of course, a powerful risk prognosticator, but when pitted head-to-head with LDL cholesterol, high-sensitivity C-reactive protein (HSCRP) actually emerges as a better prognosticator of risk. Both of these forms of measuring patient risk appear to be incremental. How in fact does C-reactive protein (CRP) tie in with multiple plaque rupture? The CRP level predicts the prevalence of multiple coronary plaques (Figure 8). The data on CRP levels show the CRP, stratified by low, medium and high tertiles, and the prevalence of multiple plaques increase quite dramatically, especially in the highest tertile of C-reactive protein.⁴ How then are the two concepts of inflammation and thrombosis inter-related? The platelet seems to be the key player that connects these two concepts. An excellent recent review article shows how activated platelets, by generating a number of inflammatory modulators can lead to plaque rupture (Table).⁹ Data published on the detrimental role of platelet-derived sCD40L in cardiovascular disease expands on the concept that the platelet is actually not the inactive, minor player in thrombosis that we were taught in medical school (Figure 9).^9,10 In addition to its major role in thrombosis, the platelet is also an essential mediator in the inflammatory pathways in acute coronary syndromes, and potentially in other pathological states as well. The importance of CD40 and CD40 ligand cannot be over-emphasized. The principle repository of these elements lies within the platelet. The role of CD40 is only just now being given its due — not only in inflammation and thrombosis, but in restenosis as well. This perhaps ties together some of the older data showing that antiplatelet therapy does indeed decrease restenosis. What therapies do we have that might affect CD40 and CD40 ligand? One might think that aspirin would have an effect on CD40 release, but unfortunately, it does not. On the other hand, clopidogrel, or the combination of aspirin and clopidogrel, does appear to decrease CD40L release, as evidenced by the results of a small study by Hermann, et al. (Figure 10).¹¹ How does C-reactive protein fit into the interventional cardiology realm? Data from Stefanadis et al., show a relatively strong correlation between serum markers of inflammation such as C-reactive protein, and the temperature at the coronary plaque (Figure 11).¹² This indirect surrogate marker of inflammation that is occurring at the coronary arteries led us, in our own interventional database, to examine the impact of baseline C-reactive protein on 30-day outcomes, namely rates of death or myocardial infarction, including peri-procedural myocardial infarction.¹³ When patients are stratified by C-reactive protein quartiles, ranging from the lowest to the highest quartiles, the rates of death or myocardial infarction increase (Figure 12). And indeed, the deaths appear to be clustered only in the patients in the highest two C-reactive protein quartiles. This corroborates the work that has been done in other settings regarding the value of C-reactive protein in risk prognostication. Antiplatelet therapies and inflammation. What antiplatelet therapies are available that might affect this? Data on the 30-day death or myocardial infarction rates in patients undergoing percutaneous coronary intervention in the highest CRP quartile revealed a > 50% risk reduction in this high-risk cohort of patients with just a few extra tablets of clopidogrel given prior to the time of balloon or stent injury (Figure 13).¹⁴ These results are, if anything, further amplified in the diabetic subgroup from the patient population just referred to. That is, patients with diabetes in the highest quartile of C-reactive protein who were not pretreated with clopidogrel had dramatically higher event rates (28.6% death or MI) than those in that subgroup who did receive pre-treatment with the drug (10.9%, p = 0.028) (Figure 14).¹⁵ Our group has submitted data for review that show patients who underwent percutaneous coronary intervention, stratified by clopidogrel pre-treatment versus no clopidogrel pre-treatment, had no difference in their baseline pre-procedural high sensitivity C-reactive protein level. But what happens to the change in high sensitivity C-reactive protein after PCI, which increases inflammatory markers and the level of C-reactive protein? It appears that clopidogrel pre-treatment blunts the increase in C-reactive protein that occurs post-intervention — a statistically significant difference that also holds up on multivariate analysis. Long-term risk. Patients who have undergone percutaneous coronary intervention face not only a short-term risk of stent thrombosis, or an intermediate-term risk of subacute stent thrombosis, but also later risks such as in-stent restenosis and the composite of major cardiac events that, if one includes death, myocardial infarction, and target vessel revascularization, are approximately 20% at one year (Figure 15). What about post-myocardial infarction patients who have a single coronary plaque versus multiple coronary plaques? Data from Goldstein et al. clearly show that patients with multiple plaques have much higher rates of repeat catheterization, recurrent acute coronary syndromes, repeat PTCA, PTCA of the non-infarct related artery, and CABG (Figure 16).³ Thus, rates for all modalities of revascularization are much higher in patients with multiple plaques than those with a single plaque. To support this 20% figure, three contemporary trials — EPISTENT, ESPRIT, and TARGET — involving percutaneous coronary intervention, including stenting and IV GP IIb/IIIa use, also show approximately a 20% rate at one year. Treatment. There are a number of ways to stabilize vulnerable plaque in a patient who is at risk for a plaque rupture (Figure 17). Focal treatment to stabilize a ruptured plaque is something clinicians are very skilled at, and percutaneous coronary intervention with stenting is probably the best way to mechanically seal an unstable plaque. However, the problem arises from the fact that coronary artery disease is diffuse, and multiple vulnerable plaques are present. Medical therapy, therefore, is key in stabilizing vulnerable plaque(s) because it can reduce thrombogenicity in the blood via antiplatelet therapy; reduce plaque lipid content via statins, primarily; decrease inflammation via statins, aspirin, and potentially clopidogrel; improve endothelial function via statins and ACE inhibitors; and potentially increase plaque collagen content through MMP inhibition or other novel modalities.¹⁶CAPRIE. The data from the CAPRIE trial highlight the important fact that patients with coronary artery disease often have concomitant cerebrovascular disease or peripheral arterial disease (Figure 18). The same applies to patients who present with cerebrovascular disease or peripheral arterial disease. While this discussion is focused on coronary artery disease and multiple plaques in the coronary artery circulation, it applies to the panvascular system as well.¹⁷ The CAPRIE trial results showed an 8.7% relative risk reduction favoring clopidogrel versus active control with aspirin. There are also some important subset analyses that have been conducted subsequent to the publication of the main CAPRIE paper in the Lancet. A subgroup analysis has been performed on patients who underwent prior revascularization which was specifically coded as“prior cardiac surgery” in the CAPRIE database.¹⁸ Patients who were randomized to clopidogrel had a significant risk reduction in the primary endpoint of myocardial infarction, stroke, and cardiovascular death compared to those who were randomized to aspirin. These patients also had a relative risk reduction far larger than seen in the overall CAPRIE patient cohort. The Kaplan-Meier curves from that same subgroup analysis of patients who underwent prior cardiac surgery again show the benefit of clopidogrel (Figure 19). In terms of the endpoint of cardiovascular death, myocardial infarction, stroke, or rehospitalization for ischemia or bleeding, and follow-up out to three years, it is important to note how prevalent adverse ischemic events are in this patient population with atherosclerotic disease. That is, when the different ischemic endpoints such as myocardial infarction that would present to the cardiologist, stroke, that would present to the neurologist, and rehospitalization for peripheral limb ischemia that would present to the vascular surgeon, are combined, and when these patients are followed up over a long enough period, they indeed experience a high cumulative rate of recurrent ischemic events. Thus, patients with atherosclerosis are at much higher risk than any of us may appreciate as individual practitioners or individual subspecialists. Data published recently feature a subset analysis from the CAPRIE study specifically looking at patients with diabetes mellitus (Figure 20).¹⁹ As would be expected, the aspirin-treated patient group, which included non-diabetics, all diabetics, and those diabetics specifically treated with insulin, had a higher event rate with the presence of diabetes, particularly insulin-requiring diabetes. But what is the impact of clopidogrel therapy on these patients? The benefit of clopidogrel is larger in the diabetic patients and even larger still in the insulin-requiring diabetics, such that in terms of the adverse events prevented per 1,000 patients treated per year for clopidogrel versus aspirin, nine events were prevented in the non-diabetics, but a much larger benefit of 21 events were prevented in the diabetic patient group, and 38 events were prevented in the insulin-requiring diabetic group. Thus, much like there is a gradient of risk from the non-diabetic to the insulin-requiring diabetic patient, there also appears to be a gradient of benefit with more potent antiplatelet therapy. CHARISMA. The CHARISMA trial’s primary objective is to assess whether clopidogrel 75 mg is superior to placebo in preventing the occurrence of ischemic complications, namely stroke, myocardial infarction, or cardiovascular death in high-risk patients who are already receiving aspirin. The secondary goal of this trial is to evaluate the safety of clopidogrel in this patient population, specifically the risks of primary intracranial hemorrhage and life-threatening bleeding. CHARISMA is designed as a Phase III multicenter, multinational, multidisciplinary, randomized, parallel group, double-blind trial. It is worth highlighting the multidisciplinary nature of this trial in which the executive committee consists not only of interventional cardiologists, but also preventive cardiologists, neurologists, vascular medicine specialists, and diabetologists. The goal of this trial is to approach atherosclerosis from a truly comprehensive standpoint. About 15,200 patients are expected to be enrolled in the trial over one to two years; follow-up will be anywhere from three months to 3.5 years; and enrollment will end after 1,040 primary events have occurred. CHARISMA is thus an event-driven trial. The inclusion criteria, broadly speaking, consist of the following: 1) documented coronary artery disease; 2) documented cerebrovascular disease; 3) documented peripheral arterial disease; or 4) a combination of 2 major or 3 minor or 1 major and 2 minor atherothrombotic risk factors — which translates into “high-risk primary prevention”. Conclusions. Coronary artery disease, as well as other arterial diseases involving the cerebral vasculature and the peripheral arterial system, are diffuse by their nature and often involve multiple unstable plaques. Medical therapy is complementary to endovascular therapy and these should no longer be viewed as mutually exclusive treatments. Instead, both treatment approaches ought to be harmonized. The future will very likely lie both in long-term antiplatelet therapy and anti-inflammatory therapy. Further research may also reveal that long-term antiplatelet and anti-inflammatory therapies are interconnected and that the same drugs are effective in both pathways. DISCUSSION Chris Cannon: Thank you, Deepak. How do you all view this? Do we think about inflammation from the cath lab perspective? Is that on the radar screen in general? Spencer King: What is your speculation about why the C-reactive protein is reduced by clopidogrel? Is it blocking a platelet thrombotic event, or is some other mechanism driving it? Deepak Bhatt: That is a terrific question. The honest answer is that I don’t know. I suspect that part of it has to do with a primary antithrombotic effect. It is conceivable that any effective antithrombotic therapy may secondarily have some affect on inflammatory markers. I do think that it is more than that, however. The only study I discussed regarding CD40, interestingly, did not show a benefit with aspirin. Even in terms of C-reactive protein reduction, it is a mixed bag whether aspirin conclusively decreases inflammatory markers. The Physicians Health Study showed an incremental benefit of aspirin in patients in the highest quartile of CRP. Aspirin has not been conclusively shown to decrease inflammatory markers; there are excellent studies showing that aspirin does decrease inflammatory markers, and other excellent studies showing that it does not. On the other hand, the CD40 study did show that clopidogrel decreases CD40 ligand. Thus, I think that something else occurs with clopidogrel in terms of an anti-inflammatory effect separate from its direct antithrombotic effects. And perhaps the same mechanism is at play in terms of how it influences CD40 and CD40L and CRP. C-reactive protein, of course, is a fairly broad measure of inflammation. On the other hand, a good deal of new data suggest that CRP is a direct pathogen as well. There was a recent paper published showing how CRP and oxidized LDL interact directly in terms of promoting atherogenesis. While we viewed CRP largely as a surrogate marker of risk — and it is certainly an excellent, relatively cheap, and easily available risk marker — it appears that it is also a key player in the direct pathogenesis of plaque progression and disruption. Thus, I think that clopidogrel probably exerts both antiplatelet and, indirectly, anti-inflammatory pathway benefits. Elliot Rapaport: In your comments regarding the benefit of pre-treatment with clopidogrel in the diabetic patient going to the catheterization laboratory, I presume this was a comparison of no clopidogrel to subsequent post-PTCA or post-stent clopidogrel. What was the time difference that showed the benefit? How long beforehand was pre-treatment required in order to observe the benefit you described? Deepak Bhatt: In the data I presented here, all of the patients received clopidogrel afterwards because they were just the stent patients. Thus, the only difference was that the clopidogrel pre-treatment group received some dosing of clopidogrel prior to PCI. We defined it as “any sort of clopidogrel pre-treatment,” so if the patient received clopidogrel on the table after the diagnostic catheterization, before the attending interventional physician walked in, that counted as clopidogrel pre-treatment. If anything, that definition would bias the analysis against clopidogrel pre-treatment. That is, if anything, we are biasing the analysis to appear worse than it actually is. In fact, if we had only studied those patients who had received an adequate course of clopidogrel pre-treatment, the data would have been even stronger, but the way these data were captured in our interventional registry, we could only demarcate clopidogrel pre-treatment with a “yes” or a “no.” Thus, I think these data are probably more robust than they appear at first glance. Mercedes Dullum: As a cardiac surgeon, I see multiple plaques all the time. The angiogram may appear totally smooth except for the one lesion that we are theoretically going after, but when we actually perform the anastamosis, these plaques are found throughout the vessel — not just lumpy, bumpy stuff found in diabetics. It would be interesting to study these patients to determine if they have earlier vein graft closure downstream. In any case, from what you have described, this would be an ideal group of patients to treat with clopidogrel. Have you considered this at all? Deepak Bhatt: No, we have not, but it certainly would be a terrific avenue of research to pursue. You are right, perhaps it does tie in to some of the reasons, at least as shown in the randomized trials, that diabetic patients with multi-vessel disease — probably as a surrogate of diffuse coronary artery disease — appear to do better with coronary artery bypass graft surgery than multi-vessel intervention. The importance of multiple plaque rupture in the CABG setting has certainly been under-researched. Mercedes Dullum: Yes. We see diffuse disease not only in diabetics, but also in younger patients who undergo bypass surgery. Mitch Driesman: Can you tell whether the plaque is active or inflamed? It bothers me that the most recent studies suggesting that we see multiple active plaques simultaneously are based on IVUS imaging. These studies assume that IVUS is the gold standard. I just wish we had more objective techniques. In surgery, it seems that diffuse disease can be identified, but we would have difficulty determining whether the plaque is active or old. Mercedes Dullum: Sometimes you can see the active inflammation. I suppose we could measure the temperature differences with an electrode, but it is not easy. We see diffuse plaques throughout. Deepak Bhatt: I agree that IVUS can be a bit tricky when trying to delineate plaque rupture. The more sensitive the measure used, such as angioscopy, the more the IVUS findings are corroborated. In fact, angioscopy shows that multiple plaque rupture is even more prevalent than IVUS would suggest. Therefore, I think that the composite of all the data: the MRI studies, the IVUS studies, and the angioscopy studies, point to the reality that multiple plaques are indeed present. Ron Waksman: How long must clopidogrel be given in order to achieve the effect, and what happens if the drug is stopped at 30 days? Most operators are giving clopidogrel for stent procedures now. Does stopping it affect the other plaques and the inflammation phenomenon? When clopidogrel is stopped, is there a rebound effect in which CRP rises again? Do you have any data on that? Deepak Bhatt: Those are excellent questions. I personally do not have any data in that regard. I only have the CRP data from our center that I showed you earlier. As for the clinical side of things, we will find out from Steve Steinhubl when he presents the 30-day CREDO results tomorrow and then the one-year data at the AHA meeting, whether all patients who undergo a stent procedure should receive long-term treatment with clopidogrel, regardless of CRP, inflammation, or multiple plaque rupture. I am betting that the answer to that question is Yes, because I think that this concept of multiple plaque rupture is more real than we interventionalists believe it to be. Ron Waksman: It would be relatively easy to follow because you have already shown that CRP is reduced with clopidogrel. Thus, you could stop the drug and see if CRP rises again. Deepak Bhatt: Yes, it would actually be very easy to observe. The data I have just presented are preliminary. They have been submitted and have not yet undergone peer review. I think they open up an avenue for further research. Chris Cannon: Thank you all for your participation in this discussion. Let’s now move on to the topic of aspirin and aspirin resistance which Steve Steinhubl from U.N.C. Chapel Hill, will discuss.