ADVERTISEMENT
Feature
Antiplatelet Therapy: Aspirin
March 2003
History of aspirin. Although aspirin’s active ingredient has been used for medicinal purposes for centuries, there is still much we don’t know about it. In fact, the first published research is from 1763 entitled “An Account of the Success of the Bark of the Willow in the Cure of Agues.”1 Obviously, our thoughts about aspirin have changed considerably over time. For instance, in an attempt to defend aspirin, Bayer launched an advertisement in the late 1930s stating that aspirin did not affect the heart because there had been concerns about direct cardiac toxicity which we now know was due to the use of very large doses that resulted in salicylism.
Individual propensity for thrombosis. Before discussing aspirin resistance, one needs to consider the inter-individual variability in thrombosis. Patients who present with a thrombotic event are typically given the same weight-adjusted antithrombotic therapy: heparin, aspirin, clopidogrel, and a IIb/IIIa inhibitor. Yet each patient’s thrombotic system is different. A group from the Mayo Clinic has conducted a couple of very forward-thinking, yet simple studies on animal models that consistently showed — at least in the porcine model — the individualization in effects of arterial thrombosis.2 In the most recent study3 the blood from 25 pigs was perfused over an arterial substrate from a single aorta and the amount of platelet deposition was measured. The measurement revealed that the strongest predictor of the amount of platelet deposition was the animal, not the thrombogenic stimulus of the aortic strip. The variability in this platelet deposition was over 24-fold for all 25 animals. When the experiment was repeated using different aortic strips, there was only a 3-fold difference. A strong correlation (0.86) is shown between the inter-individual or inter-pig response to these aortic strips. Several genetic studies have looked at the coagulation system in twin female pairs. Investigators in one of these studies4 found that the genetic factors in these twins contributed 41–75% of the variation in concentrations of hemostatic proteins and approximately 45–65% of the variability in active markers of coagulation and fibrinolysis.5 A recent study from the Framingham group showed that the heritability in platelet aggregation response among 2,413 participants explained 21–30% of the variance. To put this in perspective, there was greater sibling correlation with platelet aggregability than there was between more standard risk factors such as HDL levels and systolic blood pressure.6 Thus, each of us has a different tendency to clot that is at least partly related to a variability in coagulation factors and platelet activity.
“Aspirin resistance?” Although there is no conclusive information available, it appears that aspirin resistance is a reality. One can examine this question by first analyzing the data from the Antithrombotic Trialists’ Collaboration a little differently than usual (Figure 1).7 Though we typically focus on the 22% relative risk reduction in virtually every group with aspirin therapy, we can also look at it from the angle of the 80% of patients at risk who had an event despite aspirin therapy. Unfortunately, at this time we can only speculate as to why this is the case. A study by Buchanan was one of the first to explore variability in bleeding times.8 The investigators studied 40 CABG patients treated with 325 mg of aspirin daily, compared them on aspirin and off aspirin, and found that approximately 58% of the patients were responders, with a mean change in bleeding time of approximately 58 ± 10%. On the other hand, 42% of the patients were considered non-responders and had no significant response or no change in bleeding time. Other studies have looked at a number of different markers of platelet inhibition with aspirin, and every one of them has shown a wide variability in response with “aspirin resistance” being identified in 5–50% of patients.
A small but important eight-person volunteer study9 measured increased sensitivity to collagen and aspirin resistance at baseline and then at 30 months. The five aspirin responders were still responders at 30 months, while the three non-responders were also still non-responders at 30 months. This obviously does not prove that aspirin responsiveness is an inherited quality or an individual characteristic, but it does imply that in patients with no known atherosclerotic disease, responsiveness remains constant over a 30-month period. Other trials also provide data on some measure of aspirin responsiveness and at the same time correlate them with clinical outcomes (Figure 2).10–13 A substudy from the HOPE group, a case-control study involving 488 patients with five years of follow-up, measured thromboxane biosynthesis on aspirin and cardiovascular events. Uninary 11-dehydroxy thromboxane B2 was measured at baseline in these patients and then broken into quartiles, revealing a gradually increased risk of the combined endpoint of myocardial infarction, stroke, or cardiovascular death. Although both the lay and academic literature attribute this to aspirin resistance, it actually only shows that the patients who experienced more events had more active platelets. This may therefore just be a measure for atherosclerotic burden and stability rather than aspirin responsiveness.
Another study conducted by Pat Gum employed two different methods to determine if patients were aspirin responders or non-responders.14 In 325 patients undergoing cardiac catheterization, aggregometry was performed using two different agonists: ADP and arachidonic acid (Figure 3). Based on the aggregometry data, 70% of these patients were labeled aspirin responders (aspirin-sensitive); 24% of the patients were considered partial responders; and 5.5% were considered resistant to aspirin. The investigators used the PFA–100®, a point-of-care assay that revealed a 90% aspirin sensitivity rate and an approximate 10% resistance rate. When Dr. Gum looked at clinical outcomes of aspirin responsiveness as determined by aggregometry studies, a significant increase in events in the patients who were considered aspirin-resistant was found (Figure 4).15 Obviously, with only 17 patients, these numbers are quite small, but the data were nonetheless significant. Again, much like the previous study just mentioned, platelet aggregation response may or may not be a more appropriate marker of platelet activity rather than a pure measure of aspirin responsiveness.
An older study from 199716 involved 100 patients with peripheral vascular disease who were undergoing percutaneous intervention and receiving 100 mg of aspirin daily. The investigators looked at restenosis or total limb occlusion rates at one year as the endpoint for aspirin responders. The results showed a very strong correlation between the patients’ measured aspirin resistance. Those patients considered either partially or completely resistant had a 13% limb occlusion rate at one year, compared to the normal responders whose limb occlusion rate was 0% — a significant difference. One of the best studies conducted, which used single platelet counting as a marker for platelet responsiveness to aspirin, involved 181 patients who presented with a cerebrovascular event and were started on aspirin 500 mg T.I.D., then followed up for 24 months.17 Surprisingly, the two-thirds of the study’s patients who were considered aspirin responders had a 5% rate of death, myocardial infarction, or recurrent stroke, whereas the one-third of the study’s patients who were considered non-responders to aspirin had a 40% event rate. Thus, there was a very strong correlation between the measurement of a true responsiveness to aspirin — not just baseline platelet function in patients on aspirin — and long-term outcomes.
In more recent trials, data on the measure of potential aspirin resistance come from virtually all of the acute coronary syndrome trials in which patients are broken down into non-prior aspirin users and prior aspirin users, with prior aspirin users consistently experiencing significantly worse outcomes. Interestingly, in most of these trials — PURSUIT, TIMI 11B, PRISM-PLUS and ESSENCE — approximately two-thirds of the patients are typically in the prior aspirin-users group and one-third are in the non-prior aspirin users group (Figure 5). Prior aspirin use was also a predictor of response to therapy, so whether it was the addition of low-molecular weight heparin or a GP IIb/IIIa receptor antagonist, only the prior aspirin users showed significant benefit with the additional therapies. Numerous potential mechanisms for aspirin responsiveness or resistance have been postulated, but none confirmed. When patients present with an acute coronary syndrome while on aspirin and have worse outcomes, it may be due to the simple fact that they have a more thrombogenic stimulus that overwhelms the protective effects of aspirin. This may explain why patients have significantly worse outcomes if they are already on aspirin therapy.
Another potential mechanism again involves the arachidonic pathway (Figure 6). When we consider the metabolism of arachodonic acid we typically only focus on the COX-1–dependent production of thromboxane A2. However, arachadonic acid has two other important pathways that have platelet aggregating effects: the non-enzymatic lipid peroxidation catalyzed by free radicals that leads to the formation of isoprostanes which amplify platelet response to other agonists and which are vasoconstrictors. There is also a lipoxygenase pathway, which acts upon arachidonic acid forming by-products that increase platelet adhesivity. PLA polymorphism is yet another potential mechanism for aspirin responsiveness or resistance.19,20 PlA2 patients comprise about one-quarter of the Northern European ancestry population, and several studies have found a significant difference in responsiveness to aspirin compared to patients who do not have PlA2 polymorphism. A final potential mechanism for aspirin resistance is COX-2-based thromboxane production in platelets, of which aspirin is only a weak inhibitor.21 Studies of post-surgery patients suggest that platelets regulate COX-2, while other studies have refuted this.
What is the Appropriate Aspirin Dose? At the start of the 20th century, Bayer aspirin was originally sold in the U.S as a 5 grain tablet (approximately 325 mg) because that particular size happened to make a convenient pill. This 325 mg dose is still primarily used today. A 1985 study presented some early data on the aspirin dose necessary to inhibit platelet thromboxane production.22,23 The data show that approximately 12 mg was the daily dose required for chronic therapy to completely inhibit thromboxane metabolism. As a single dose, approximately 100 mg were required. Obviously, there was a variability among patients, but in these small studies of volunteers, the data suggest a considerably lower dose than is currently utilized. The large Dutch TIA trial involving 3,131 patients who had a recent TIA or minor stroke, made head-to-head comparisons between 283 mg of aspirin and a very low 30 mg dose of aspirin.24 If anything, the trend showed a benefit with the 30 mg daily dose of aspirin, but there was essentially no difference over time. A substudy of 31 patients showed identical levels of suppression (99%) of thromboxane metabolite.
Several meta-analyses have been conducted analyzing the relationship between aspirin dose and efficacy.25 A meta-analysis of 10 randomized trials showed a consistent benefit of aspirin with no significant difference based on low, medium, or high doses (Figure 7). Although there was no clear efficacy dose-response — at least over 30 mg — there was a clear dose response in terms of side effects. In the UK-TIA trial,26 patients either received placebo, high-dose aspirin (600 mg b.i.d.), or low-dose aspirin (300 mg q.d.). Each increase in the dose of aspirin was associated with an increased occurrence of side effects, i.e., gastrointestinal symptoms, hemorrhage, upper gastrointestinal symptoms, and constipation (Figure 8).
Finally, if the patient is an aspirin non-responder, can the dose simply be increased? Two very small trials have suggested this to be the case. One of these trials involved 107 patients with prior stroke, all of whom received Risk of bleeding with aspirin during CABG. More than 20 studies have focused on the risk of bleeding with aspirin use. Most of the early studies suggest that there is an increased risk, while a majority of the later studies show no increased risk of bleeding with aspirin in the peri-operative period. A large, single-center study published in the Annals of Surgery in 2002 involved nearly 2,000 aspirin users at the time of surgery.28 It compared outcomes in over 700 patients who had received no peri-operative aspirin. The results showed a significant increase in the need for blood transfusions, other blood products besides red blood cells, and re-operation for bleeding in aspirin users compared with non-users (Figure 9). The absolute numbers were small, but with multivariate analysis, prior aspirin use was a significant predictor of the need for blood transfusions in the peri-operative period.
The CREDO trial. The 28-day CREDO data will be presented as late-breaking trial data at this year’s TCT meeting, with the one-year data to be presented at the upcoming AHA meeting. CREDO was designed several years ago when ticlopidine was still the standard of care in stenting (Figure 10). The goal was to look at a 300 mg clopidogrel loading dose 3 to 24 hours prior to the time of stenting compared to a matching placebo. At the time of stenting, both groups received 75 mg of clopidogrel and 325 mg of aspirin, and both groups continued to receive these agents out to 28 days. The results being presented tomorrow essentially compare 28.5 days versus 28 days of clopidogrel in a stent population. The group that received the loading dose of clopidogrel will continue on clopidogrel and aspirin after day 28 to one year, in order to determine the degree of protection from the long-term risk of multiple unstable plaques.
DISCUSSION
Elliot Rapaport: For the patients who present with with an acute coronary syndrome and who are on aspirin therapy, should we in fact be performing platelet aggregability studies in response to factors such as ADP stimulation? If so, if the patients are non-responders, is there any point in continuing aspirin therapy in these patients? Should we discontinue the aspirin and use only clopidogrel or just add clopidogrel?
Steve Steinhubl: The easiest answer, based on the clinical data, would be that you should add clopidogrel. Unfortunately, we currently lack a good measuring device. I mentioned the PFA–100® (Dade Behring, Inc., Deerfield, Illinois) but other point-of-care devices have been developed to measure response to aspirin. We need a prospective study in order to measure this response. Aggregometry would be very difficult to perform, especially because it would be confounded by the clinical scenario that the patient presented with. The consistency of the clinical data regarding the increased risk of prior aspirin users has led me to become more aggressive in treating these patients from an antiplatelet standpoint — I use a IIb/IIIa inhibitor for these cases. In the CURE study, the benefit was consistent regardless of whether the patients were prior aspirin users, which has made clopidogrel stand out from all of the other antithrombotic agents. Our institution is involved in a study of the Accumetrics point-of-care device for measuring aspirin responsiveness. The Accumetrics Ultegra® (San Diego, Calif.) point-of-care device clearly notes a wide range of responses. Whether it is clinically relevant has yet to be demonstrated.
Shamir Mehta: Roughly two-thirds of the CURE patients were on prior aspirin. Similar to what the other trials showed, their absolute event rates were very high compared to the non-aspirin users. But interestingly, as Steve mentioned, the relative benefit of clopidogrel was approximately 20% in both groups.
Elliot Rapaport: What if you performed tests and found that the patient was an aspirin responder? Would you only administer aspirin to that patient?
Steve Steinhubl: If the patient presented already on aspirin and had an event while on it, I would not give only aspirin. But if the patient presented without prior aspirin use and had an event, then I would be more comfortable giving aspirin alone as the cornerstone of therapy. Actually, it is premature to say that because we need to see study data showing that aspirin response can be measured, that it correlates with clinical outcomes, and that the therapy is appropriate. Perhaps some day we will even determine that some patients would do better on clopidogrel, or maybe even warfarin. The first step, however, is to be able to measure the response and then correlate that measurement with outcomes.
Ron Waksman: Do you have any data showing that the event rate curves match those of aspirin responders for patients who are resistant to aspirin and are thus given clopidogrel?
Steve Steinhubl: That is a terrific question, but unfortunately we currently have no data. However, I am aware of several studies in the planning stages that may hopefully help answer this question.
Chairman’s Note: The CREDO trial (Clopidogrel for Recurrent of Events During Observation trial) was presented at the American Heart Association 75th Scientific Sessions by Steve Steinhubl the day after this roundtable meeting. It was simultaneously published in the Journal of the American Medical Association (JAMA) by Steinhubl, Topol and colleagues. This is a landmark trial that will lead to dramatic benefits for PCI patients. In this study, patients with planned or likely PCI (which included approximately one third of patients with ACS) were randomized to receive a loading dose of clopidogrel (300 mg) or placebo between 3 and 24 hours prior to PCI. Following stenting, all patients received open label clopidogrel for 28 days post stent; after 28 days, patients in the pretreatment group continued on clopidogrel for 1 year, whereas the non-pretreatment group was treated with matching placebo. The results at 28 days found that loading of clopidogrel at least 6 hours prior to PCI leads to a 38% relative risk reduction in 30 day POST PCI events! It is clear that for ACS patients we need to get the clopidogrel started as soon as possible. For elective PCI patients, it means that we need to consider starting clopidogrel as an outpatient procedure prior to taking them to the cath lab. Overall, longterm treatment (1 year) with clopidogrel plus aspirin led to 26.9% relative reduction in death, MI, or stroke compared to post-PCI clopidogrel therapy for one month (8.5% vs 11.5% [placebo], p=0.02). This included an impressive further 37.4% relative reduction in events, ve further 37.4% relative reduction in major events from day 29 to 1 year with clopidogrel (p=0.04). The results of PCI-CURE and CREDO support pre-procedural loading and long-term therapy with clopidogrel in those scheduled or expected to undergo PCI. The significant benefits were seen with or without the concomitant use of glycoprotein IIb/IIIa inhibitors. – Christopher Cannon, MD
Chairman’s Note: The CREDO trial (Clopidogrel for Recurrent of Events During Observation trial) was presented at the American Heart Association 75th Scientific Sessions by Steve Steinhubl the day after this roundtable meeting. It was simultaneously published in the Journal of the American Medical Association (JAMA) by Steinhubl, Topol and colleagues. This is a landmark trial that will lead to dramatic benefits for PCI patients. In this study, patients with planned or likely PCI (which included approximately one third of patients with ACS) were randomized to receive a loading dose of clopidogrel (300 mg) or placebo between 3 and 24 hours prior to PCI. Following stenting, all patients received open label clopidogrel for 28 days post stent; after 28 days, patients in the pretreatment group continued on clopidogrel for 1 year, whereas the non-pretreatment group was treated with matching placebo. The results at 28 days found that loading of clopidogrel at least 6 hours prior to PCI leads to a 38% relative risk reduction in 30 day POST PCI events! It is clear that for ACS patients we need to get the clopidogrel started as soon as possible. For elective PCI patients, it means that we need to consider starting clopidogrel as an outpatient procedure prior to taking them to the cath lab. Overall, longterm treatment (1 year) with clopidogrel plus aspirin led to 26.9% relative reduction in death, MI, or stroke compared to post-PCI clopidogrel therapy for one month (8.5% vs 11.5% [placebo], p=0.02). This included an impressive further 37.4% relative reduction in events, ve further 37.4% relative reduction in major events from day 29 to 1 year with clopidogrel (p=0.04). The results of PCI-CURE and CREDO support pre-procedural loading and long-term therapy with clopidogrel in those scheduled or expected to undergo PCI. The significant benefits were seen with or without the concomitant use of glycoprotein IIb/IIIa inhibitors. – Christopher Cannon, MD
1. Stone E. An account of the success of the bark of the willow in the cure of agues. Philos Trans R Soc Lond [Biol]. 1763;53:195–200.
2. Wysokinski WE, McBane II RD, Owen WG. Individual propensity for arterial thrombosis. Arterioscler Thromb Vasc Biol 1999;19:883–886.
3. Karnicki K, Owen WG, Miller RS, McBane RD II. Factors contributing to individual propensity for arterial thrombosis. Arterioscler Thromb Vasc Biol 2002;22:1495–1499.
4. de Lange M, Snieder H, Ariens RA, et al. The gentics of hemostasis: A twin study. Lancet 2001;357:101–105.
5. Ariens RA, de Lange M, Snieder H, et al. Activation markers of coagulatuion and fibrinolysis in twins: Heritability of the prethrombotic state. Lancet 2002;359:667–671.
6. O'Donnell CJ, Larson MG, Feng D, et al. Genetic and environmental contributions to platelet aggregation: The Framingham Heart Study. Circulation 2001;103:3051–3056.
7. Antithrombotic Trialists. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients [erratum appears in Br Med J 2002;324:141]. Br Med J 2002;324:71–86.
8. Buchanan MR, Brister SJ. Individual variation in the effects of ASA on platelet function: Implications for the use of ASA clinically. Can J Cardiol 1995;11:221–227.
9. Kawasaki T, Ozeki Y, Igawa T, Kambayashi J. Increased platelet sensitivity to collagen in individuals resistant to low-dose aspirin. Stroke 2000;31:591–595.
10. Pappas JM, Westengard JC, Bull BS. Population variability in the effect of aspirin on platelet function. Implications for clinical trials and therapy. Arch Pathol Lab Med 1994;118:801–804.
11. Valettas N, Morgan CD, Reis M. Aspirin resistance using flow cytometry (Abstr). Blood 1997;10(Suppl 1):124b.
12. Valles J, Santos MT, Aznar J, et al. Erythrocyte promotion of platelet reactivity decreases the effectiveness of aspirin as an antithrombotic therapeutic modality. The effect of low-dose aspirin is less than optimal in patients with vascular disease due to prothrombotic effects of erythrocytes on platelet reactivity. Circulation 1998;97:350–355.
13. Eikelboom JW, Hirsh J, Weitz JI, et al. Aspirin-resistant thromboxane biosynthesis and the risk of myocardial infarction, stroke, or cardiovascular death in patients at high risk for cardiovascular events. Circulation 2002;105:1650–1655.
14. Gum P, Kottke-Marchant K, Poggio ED, et al. Profile and prevalence of aspirin resistance in patients with cardiovascular disease. Am J Cardiol 2001;88:230–235.
15. Gum P. Abstract. ACC 2002.
16. Mueller MR, Salat A, Stangl P, et al. Variable platelet response to low-dose ASA and the risk of limb deterioration in patients submitted to peripheral arterial angioplasty. Thromb Haemost 1997;78:1003–1007.
17. Grotemeyer KH, Scharafinski HW, Husstedt IW. Two-year follow-up of aspirin responder and aspirin non-responder. A pilot study including 180 post-stroke patients. Thromb Research 1993;71:397–403.
18. Braun M. Platelets 1994;5:325–331.
19. Cooke GE, Bray PF, Hamlington JD, et al. PlA2 polymorphism and efficacy of aspirin. Lancet 1998;351:1253.
20. Cooke GE. Abstract. J Am Coll Cardiol 2001;276A.
21. Weber AA, Zimmermann KC, Meyer-Kirchrath J, Schror K. Cyclooxygenase-2 in human platelets as a possible factor in aspirin resistance. Lancet 1999;353:900.
22. Patrono C, Ciabattoni G, Patrignani P, et al. Clinical pharmacology of platelet cyclooxygenase inhibition. Circulation 1985;72:1177–1184.
23. Lee M, Cryer B, Feldman M. Dose effects of aspirin on gastric prostaglandins and stomach mucosal injury. Annals Int Med 1994;120:184–189.
24. The Dutch TIA Study Group. A comparison of two doses of aspirin (30 mg vs. 283 mg a day) in patients after transient ischemic attack or minor ischemic stroke. N Engl J Med 1991;325:1261–1266.
25. Algra A, van Gijn J. Aspirin at any dose above 30 mg offers only modest protection after cerebral ischaemia (comment). J Neurology, Neurosurgery & Psychiatry 1996;60:197–199.
26. UK-TIA study Group. The United Kingdom transient ischemic attack (UK-TIA) aspirin trial: Final results. J Neurology, Neurosurgery & Psychiatry 1991;54:1044–1054.
27. Helgason CM, Tortorice KL, Winkler SR, et al. Aspirin response and failure in cerebral infarction. Stroke 1993;24:345–350.
28. Ferraris VA, Ferraris SP, Joseph O, et al. Aspirin and postoperative bleeding after coronary artery bypass grafting. Annals of Surgery 2002;235:820–827.
