Primary and secondary prevention of cardiovascular disease in diabetes with aspirin

Despite advances in clinical management, cardiovascular disease remains the main cause of mortality in individuals with diabetes.^1,2 The thrombotic milieu associated with diabetes, characterised by enhanced platelet activation and elevated levels/activity of coagulation factors,^3,4 contributes to the higher risk of vascular ischaemia in this population. Until recently, clinical guidelines advocated the use of aspirin for both primary and secondary cardiovascular protection in diabetes, a practice that has changed in the past 3–4 years after newer studies questioned the clinical benefit of such an approach.

The review by Schnell et al. analyses the evidence for cardiovascular protection by aspirin in diabetes. Aspirin is known for its platelet inhibitory activity, which is thoroughly discussed in the review. The authors also describe the effects of aspirin on coagulation factors including thrombin, fibrinogen and factor XIII, which studies have generally failed to address when investigating the biochemical efficacy of this agent. Despite the wealth of data, the exact pathways involved in the reduced clinical efficacy of aspirin in diabetes are unclear, and the authors discuss possible mechanisms with special emphasis on increased platelet turnover. This poses the question as to whether higher and/or multiple doses of aspirin are more effective in diabetes, which has been recently suggested. ⁵

The original guidelines recommending aspirin for primary cardiovascular prevention in diabetes were, surprisingly, based on limited and inconclusive data, conducted before the era of statins and other established cardiovascular preventative strategies, with some studies employing unconventional doses of aspirin.^6–8 More recent clinical trials such as the prevention of progression of arterial disease and diabetes (POPADAD) and Japanese primary prevention of atherosclerosis with aspirin for diabetes (JPAD) failed to show a benefit for aspirin therapy when used for primary cardiovascular protection,^9,10 although both studies were underpowered, and concrete conclusions cannot be made. Nevertheless, these newer data resulted in a paradigm shift and guidelines were modified to limit aspirin therapy in primary prevention to high-risk individuals (detailed below). It should be noted, however, that these recent data suggest that subgroups of diabetes patients may benefit from aspirin treatment, such as older patients and individuals with moderate impairment in renal function. This is an important point, as studies have generally regarded diabetes as a single clinical entity, when it is in reality a clinical continuum. An individual with diabetes of 20 years duration, on insulin therapy and with established nephropathy, has a different vascular risk compared to an individual who is diet controlled with no apparent complications. Given the absence of macrovascular disease, using aspirin therapy in either of these individuals would be classified as ‘primary cardiovascular prevention’, implying both belong to a group with similar cardiovascular risk, when these two individuals cannot be more different. These are two extremes of the same condition with numerous possibilities in between, each with a different degree of vascular risk. Therefore, future work assessing response to aspirin therapy should take into account the subgroups of diabetes, demanding large trials and long follow-ups. There are ongoing studies [a study of cardiovascular events in diabetes (ASCEND), aspirin and simvastatin combination for cardiovascular events prevention trial in diabetes (ACCEPT-D) and aspirin in reducing events in the elderly (ASPREE)], which may be able to provide some answers on subgroups, although none has been adequately powered to address this specific issue.

In contrast to primary prevention, aspirin continues to be used for secondary cardiovascular protection in diabetes, with compelling evidence for such an approach.^11,12 It is somewhat perplexing as to why aspirin is effective for secondary but not primary cardiovascular protection in diabetes. One simple explanation may be related to fewer events and reduced power of primary prevention studies. However, the meta-analyses of primary prevention studies in diabetes, discussed in detail by the authors, have failed to demonstrate a clinical benefit for aspirin therapy. Therefore, we can conclude that primary cardiovascular prevention by aspirin in diabetes is at best reduced and at worst nonexistent. A key question emerging from primary and secondary prevention studies is ‘why would a vascular event result in increased aspirin efficacy in diabetes’? Is this simply a marker of more extensive vascular disease, thereby representing a ‘high-risk group’ who would benefit? This concept offers an attractive explanation for the differential effect of aspirin in diabetes, further supported by fragments of the data from primary prevention trials (which are by no means conclusive). Therefore, the American Diabetes Association guidelines limit aspirin use in diabetes for primary prevention in high-risk individuals, which includes subjects with a 10-year risk of cardiovascular events exceeding 10%. Aspirin is not recommended for individuals with less than 5% risk, whereas for those having a 5%–10% risk, aspirin may be considered with the decision left at the discretion of the attending physician. ¹³ On the other hand, aspirin is recommended for all diabetes patients for secondary cardiovascular prevention provided there are no contraindications to this treatment. Given our current knowledge, these guidelines are a reasonable compromise but more work is needed to better characterise diabetes subjects who would benefit from this therapy.

It should be stressed that aspirin is not without side effects, with special concerns over gastrointestinal and intracerebral bleeding. ¹⁴ The risk of complications related to aspirin treatment varies between individuals and is affected by age and associated clinical conditions. Consequently, the benefit/harm ratio can vary wildly from highly advantageous to seriously unfavourable, necessitating careful assessment of patients before prescription of this agent.

A number of conclusions can be drawn from the review by Schnell et al. in relation to aspirin use in diabetes: