to the editor: Dr. Solenski’s review¹ of treatment options for transient ischemic attack (TIA) discusses the additive benefits of combination therapy with aspirin and extended-release dipyridamole (Persantine), compared with placebo or aspirin alone, as noted in the European Stroke Prevention Study 2 (ESPS-2).² However, Dr. Solenski raises a concern that the “expected similar benefits for reducing myocardial infarction and vascular death were not observed.”¹ This concern is misplaced because a trial limited to stroke patients would not be expected to show such benefits.

ESPS-2 only enrolled patients with recent ischemic stroke or TIA². Such patients are overwhelmingly more likely to have a recurrent stroke than a myocardial infarction during the two-year follow-up period of the trial.³ Although a nonsignificant trend toward reduction of myocardial infarction in favor of aspirin and combination therapy was seen in ESPS-2,² there were simply too few myocardial infarction endpoints to draw any firm conclusion. This paucity of myocardial infarction endpoints in stroke patients has occurred in other antiplatelet trials. In the Clopidogrel versus Aspirin in Patients at Risk of Ischemic Events (CAPRIE) trial,⁴ patients enrolled after a stroke were seven times more likely to have a stroke than a myocardial infarction during the follow-up period. More strikingly, the recent Management of Atherothrombosis with Clopidogrel in High-risk Patients (MATCH) with Recent Transient Ischemic Attack or Ischemic Stroke trial enrolled patients with recent stroke and TIA, 73 percent of whom would be presumed to have a high risk of cardiac events caused by diabetes or previous myocardial infarction. Even in this population, ischemic stroke was five times more likely as an endpoint than myocardial infarction.⁵

Patients with a recent stroke or TIA are at exceedingly high risk for a recurrent ischemic stroke. Preventing myocardial infarction in these patients is important, but recurrent stroke prevention is paramount, at least for the first two years. In this context, the concern about combined aspirin/extended-release dipyridamole raised by Dr. Solenski should not prevent the use of this therapy in patients with recent stroke or TIA.

in reply: I appreciate the opportunity to address Dr. Bernstein’s letter and thank him for his interest in our article.¹ The European Stroke Prevention Study 2 (ESPS-2) consisted of a large cohort of 6,602 patients randomized into four treatment groups with more than 1,600 patients enrolled in each limb. Primary endpoints were stroke, death, and stroke and death together. Transient ischemic attack (TIA) and other vascular events were secondary endpoints. More than 33 percent of the patients in each limb had known ischemic heart disease, and an additional 8 percent in each limb had known cardiac failure at the time of enrollment. This calculates to more than 500 patients with known ischemic heart disease enrolled in each limb of the study. An unknown number of patients likely had undiagnosed ischemic heart disease, potentially raising the true number of patients with cardiac ischemia. Patients were followed on treatment with the study drug for two years. A total of 167 patients experienced myocardial infarction during this time with no statistically significant difference between the study limbs, particularly between patients receiving aspirin and those receiving placebo. There are strong data showing that even low-dose aspirin given over short periods of time is cardioprotective.² For example, in a cohort of patients with silent ischemia, there is significant benefit from taking low-dose aspirin (75 mg per day) as seen in a randomized, double-blind, placebo-controlled trial.³ The group randomized to aspirin had significantly less myocardial infarctions or death at three months compared with the placebo group (4 versus 21 percent, respectively) and also at 12 months (9 versus 28 percent, respectively). Therefore, the effect of acetylsalicylic acid on preventing myocardial infarction in this relatively large cohort of patients would be expected. However, I agree that myocardial infarction was a secondary outcome point and that the trial was not designed to evaluate this. Secondary outcome points still should be analyzed critically because they may provide important data in understanding the representative nature of the patient population being tested. I strongly agree with Dr. Bernstein that outcomes for the evaluation of stroke therapies are most accurately determined if stroke alone is chosen as the primary endpoint, as was eloquently argued by Dr. Albers.⁴ As pointed out in Dr. Albers’ article, long-term follow-up studies have found that after a stroke or TIA, patients have a greater risk of dying of a cardiac event than of a stroke.^5,6 This suggests that data on myocardial infarction in the context of the medication being tested for secondary stroke prophylaxis should continue to be critically analyzed as a secondary outcome point