Aspirin and Secondary Prevention in Peripheral Artery Disease: Title and subTitle BreakA Perspective for the Early 21st Century

Lower extremity peripheral artery disease (PAD) is common among older men and women in developed countries and will become increasingly prevalent as populations survive longer with chronic disease. Recent estimates in the United States indicate that 1 in 16 men and women 40 years or older have PAD.¹ In primary care medical practices in the United States, nearly 30% of men and women who are either 70 years or older or aged 50 to 69 years with a history of diabetes or smoking have an ankle-brachial index (ABI) of less than 0.90, consistent with PAD.²

Multiple epidemiological studies completed during the past 20 years consistently demonstrate increased rates of cardiovascular morbidity and mortality among men and women with PAD compared with those without PAD even after adjusting for cardiovascular disease risk factors and history of cardiovascular events.^{3 - 4} Therefore, preventing cardiovascular events is a major therapeutic goal for clinicians treating patients with PAD. Current clinical practice guidelines from the American Heart Association and the American College of Cardiology recommend low-dose aspirin (75 to 325 mg/d) to reduce the high risk of cardiovascular morbidity and mortality in patients with PAD.⁵ Clopidogrel is recommended as an alternative antiplatelet medication for secondary prevention in PAD.⁵

In this issue of JAMA, Berger et al⁶ report results of a meta-analysis of 18 randomized controlled clinical trials assessing the effect of aspirin therapy, alone or with dipyridamole, on the risk of cardiovascular events in 5269 patients with PAD.⁶ Overall, compared with a control group, aspirin therapy was associated with a 12% relative risk reduction in the primary outcome measure of cardiovascular events (nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death). However, this 12% risk reduction did not achieve statistical significance. For secondary outcomes, aspirin therapy was associated with a statistically significant reduction in nonfatal stroke but was not associated with significant reductions in nonfatal myocardial infarction, cardiovascular mortality, or all-cause mortality.

Several characteristics of this meta-analysis must be considered when interpreting the results. First, the sample size is small. The authors report that a sample size of 5000 participants provided 88% power to detect a 25% difference and 70% power to detect a 20% difference between the aspirin and control groups for the primary outcome measure. Therefore, the meta-analysis lacks statistical power for demonstrating clinically meaningful differences of 20% or less between the aspirin and control groups. Second, of the 5269 participants included in the meta-analysis, more than 25% were from 2 studies that included only PAD participants with diabetes mellitus, which is a major risk factor for and is common among those with PAD. However, patients with diabetes mellitus may derive less benefit from aspirin than those without diabetes mellitus, as suggested by results of the Antithrombotic Trialists' Collaboration meta-analysis and other recent clinical trial data.^{7 - 8}

Third, of the 18 trials included in the meta-analysis, 15 were published prior to 1995; however, the diagnosis and recognition of PAD has changed dramatically in the past 10 years. Although PAD was previously considered synonymous with intermittent claudication, it is now known that most patients with PAD are asymptomatic or present with leg symptoms other than classic intermittent claudication.^{2 ,9} The results of aspirin clinical trials that enrolled patients with PAD during earlier decades may have limited generalizability to patients with PAD treated by clinicians in the 21st century. Fourth, given the heterogeneity of results, subset analyses stratified by patient characteristics such as age, sex, and diabetes would be informative. However, individual data were not available for these comparisons.

The current guideline-recommended dose of 75 to 325 mg of aspirin for patients with PAD is derived from the Antithrombotic Trialists' Collaboration meta-analysis of 287 studies involving 135 000 patients at high risk of vascular events.⁷ Among all patients, the Antithrombotic Trialists' Collaboration meta-analysis demonstrated that proportional reductions in vascular events were 19% for daily aspirin doses of 500 to 1500 mg, 26% for aspirin doses of 160 to 325 mg, 32% for aspirin doses of 75 to 150 mg, and 13% for aspirin doses less than 75 mg.⁷ Importantly, of the 18 clinical trials included in the meta-analysis by Berger et al,⁶ only 2 trials (including 1742 participants) evaluated once-daily aspirin therapy in currently recommended doses of 75 to 325 mg. Therefore, data evaluating aspirin monotherapy in doses currently recommended for patients with PAD are lacking. The ongoing Aspirin in Asymptomatic Atherosclerosis clinical trial¹⁰ with a sample size that is nearly two-thirds the size of the meta-analysis by Berger et al is expected to provide important information about the benefits of low-dose aspirin therapy in PAD.

Clopidogrel offers a small but statistically significant advantage over aspirin for secondary prevention among all patients with cardiovascular disease.¹¹ However, data from the Clopidogrel vs Aspirin in Patients at Risk of Ischemic Events (CAPRIE) trial suggesting that clopidogrel may be more beneficial for patients with PAD than for those with clinically evident heart disease or stroke were not supported by results of the more recent Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management, and Avoidance (CHARISMA) trial.^{11 - 12}

The lack of statistical power in the meta-analysis by Berger et al underscores a significant obstacle to the development of evidence-based clinical practice guidelines for patients with PAD: relatively few clinical trial data are available to inform practice guidelines on PAD management. In the Antithrombotic Trialists' Collaboration's most recent meta-analysis of all antiplatelet therapies, 15 828 patients with coronary artery disease (CAD) vs 9731 patients with PAD were available for analyses in clinical trials of antiplatelet therapy.⁷

Between 1990 and 1997, there were 20 new antiplatelet therapy trials for patients with CAD vs 11 for patients with PAD.⁷ Unfortunately, the relative deficiency of PAD clinical trials does not appear to be improving. A 2009 search of ClinicalTrials.gov, the registry of federally and privately funded clinical trials conducted around the world, yielded 546 open clinical trials testing therapeutic interventions for patients with CAD vs 186 for patients with PAD. If most of the PAD trials are focused on preventing adverse lower extremity outcomes, the discrepancy in numbers of clinical trials designed to prevent heart disease or stroke events among these 2 groups may be even greater than these numbers suggest.

The paucity of clinical trial data to inform clinical practice for patients with PAD is likely driven by multiple factors, including insufficient numbers of trained clinical investigators dedicated to the study of PAD; difficulty accruing patients¹³ ; and perhaps failure on the part of federal agencies, private industry, and leaders in cardiovascular medicine to recognize the importance of studying therapies for patients with PAD. Although CAD is highly prevalent in patients with PAD, clinical trial data of patients with CAD are not necessarily generalizable to those with PAD. Many patients with PAD do not have symptomatic CAD. The high burden of inflammation, extremely low levels of physical activity, and high prevalence of chronic kidney disease are also characteristics that distinguish patients with PAD^{14 - 15} and may influence their response to therapeutic interventions.

The meta-analysis by Berger et al enriches current understanding of the association of aspirin with cardiovascular outcomes in patients with PAD. However, based on the limitations of data available, the findings should not alter recommendations for aspirin as an important therapeutic tool for secondary prevention in patients with PAD. To best inform evidence-based clinical practice guidelines, more high-quality clinical trials are needed. Achieving this will require greater resources for research and a larger critical mass of clinical investigators dedicated to the study of PAD.