Appropriateness of aspirin prescribing for primary and secondary prevention of cardiovascular disease in type 2 diabetes in different care settings

Type 2 diabetes mellitus is a chronic condition associated with multiple complications and comorbidities. If poorly managed, diabetes can lead to functional limitations, disability, reduced quality of life, and life expectancy. According to the Irish CODEIRE study (2006), approximately 10.0% of the Ireland’s national healthcare budget was spent on diabetes with nearly half that amount due to hospitalizations attributable to the microvascular and macrovascular complications associated with the disease. A large proportion of these admissions could be prevented with adequate blood glucose control and mitigating overall vascular risk. [1].

In 2012, diabetes alone claimed 1.5 million lives worldwide [2] with the greatest mortality due to cardiovascular disease (CVD). [3] Persons with type 2 diabetes have a two to fourfold increased risk of developing CVD, [4] mainly due to increased arachidonic acid metabolism and thromboxane A₂ synthesis, resulting in augmented platelet aggregation and a pro-inflammatory substrate. Low-dose aspirin inhibits platelet cyclooxygenase 1, the key enzyme responsible in thromboxane A₂ synthesis, subsequently preventing thrombus formation. [5] .

Multiple studies support aspirin prescription for secondary prevention of cardiovascular events as it reduces mortality, and morbidity from non-fatal vascular events (i.e., stroke or myocardial infarction), and total vascular events, respectively. [6, 7] With regard to primary prevention, historically persons with type 2 diabetes without a prior history of CVD were recommended to commence on aspirin, as diabetes is considered as a CVD risk equivalent. [8] However, the benefits of aspirin are ambiguous and recent studies including the ARRIVE, ASPREE, and in particular ASCEND trials have resulted in changes to guidelines. [9‐11].

Generally, the guidelines for aspirin use in primary prevention are contradictory. The American Diabetes Association (ADA) has recommended aspirin use as primary and secondary prevention of CVD in diabetes since 1997 [12] and has updated their guidelines several times with minor modifications. In the recent 2020 guidelines, the advice given is that low-dose aspirin (75–162 mg/day) should:

The American Diabetes Association stated that individuals should not be on aspirin therapy if they are younger than 21 years old, or have any specific contraindications, which include aspirin allergy, bleeding tendency, recent gastrointestinal bleeding, and clinically active hepatic disease. Clinical judgment on a case-to-case basis is required for low-risk patients, such as those aged < 50 years with one or more risk factors, older patients with no risk factors, or those with 5–10% 10-year cardiovascular risk. [13].

As recent evidence demonstrates that there are limited benefits to aspirin use in the primary prevention setting, both the 2016 European Guidelines on Cardiovascular disease prevention in clinical practice and the latest practical guide to integrated type 2 diabetes mellitus by Irish College of General Practitioners (ICGP) in 2016 [14] have altered their guidelines and do not support routine aspirin use in type 2 diabetes without CVD.

This study aims to investigate the appropriateness of aspirin prescribing in type 2 diabetes by (a) comparing aspirin prescribing in a primary and secondary care setting and (b) determining whether aspirin is being used for primary or secondary prevention of CVD.

In this study, 49.0% of the population was prescribed aspirin, among whom 86.2% were attending shared care. However, despite such a high prevalence of CVD and the international consensus for aspirin use as secondary prevention for CVD, [13, 14, 16] 10.8% of individuals with CVD, who were attending shared care, were not on aspirin or any other antiplatelet/anticoagulant. This may be due to potential contraindications to aspirin use, such as previous gastrointestinal bleeding, history of peptic ulcer disease, or possible medication interactions.

Primary prevention with aspirin has always been controversial. Multiple randomized controlled trials (RCTs) were conducted on aspirin use in primary prevention of CVD over the past 30 years, notably the British Male Doctors Study (1988), [17] Physicians’ Health Study (1989), [18] Early Treatment Diabetic Retinopathy Study (1992), [19] Hypertension Optimal Treatment randomised trial (1998), [20] Thrombosis Prevention Trial (1998), [21] Primary Prevention Project (2001), [22] and Women's Health Study (2005), [23] among others. A meta-analysis of four of the mentioned RCTs reported a significant reduction of all cardiovascular events by 15.0%, and myocardial infarctions by 30.0%. However, this was balanced with an increased risk of bleeding by 69.0%. [24] Two other meta-analyses done in 2011 reported similar findings—a decrease in total cardiovascular events, and non-fatal MI, but no statistical reduction in stroke, or all-cause mortality. [25, 26] Berger et al. [26] also concluded that over a mean follow-up of 6.9 years, the number needed to harm was 1 major bleed in 261, which counterbalanced the number needed to treat (253 patients), in order to prevent 1 major cardiovascular event.

A systematic review including all of the 11 RCTs aforementioned noted a 14.0% reduction in nonfatal stroke with the use of ≤ 100 mg aspirin, on top of the well-established nonfatal MI decline. However, despite the addition of 18,369 participants in two other RCTs, the decrease in all-cause, and cardiovascular mortality remains insignificant. [27] The most recent ASCEND trial comparing 15,480 randomised patients with diabetes on either 100 mg aspirin daily or placebo, suggested similar findings. The benefits of CVD risk reduction (8.5% vs 9.6%; p = 0.01) was counterbalanced with major bleeding risks (3.2% vs 4.1%, p = 0.003). [11] These results have been further borne out in the ASPREE study and the ARRIVE trial. Both of these studies excluded diabetic patients but still demonstrated risk: benefit neutrality for aspirin therapy without confirmed vascular disease^. [9, 10].

As of 2020, aspirin use for primary prevention is no longer recommended by most guidelines, [14,16] except for the ADA which advise consideration of use for individuals with > 10% of 10-year cardiovascular risk. [13] Our study documented that 40.0% and 70.0% of primary prevention persons under GP care only had less than 10% risk of 10-year CHD risk and 10-year stroke risk, respectively. Similarly, 15.5% and 36.9% of the shared care participants had less than 10% risk of 10-year CHD, and 10-year stroke risks, respectively (Fig. 3). According to the ADA guideline, [12] these individuals should not be prescribed aspirin. It was suggested that doctors tend to miscalculate individuals’ CVD risk due to the lack of universal risk assessment tools or calculators. [28] Discrepancies in guidelines may result in confusion for physicians, and physicians may also be unlikely to discontinue a medication prescribed by other healthcare professionals.

Upon review of the available literature to date, it is imperative to note that an increased risk for CVD in individuals with type 2 diabetes should not mandate aspirin use. Healthcare professionals are required to perform a thorough risk assessment tailored to each individual before prescribing aspirin therapy. Type 2 diabetes mellitus is associated with a greater number of comorbidities and complications, which inevitably, increase the cost of management. An Irish study done in 2010 indicated that persons with type 2 diabetes ≥ 65 years who were covered by the General Medical Services scheme had a higher average annual pharmaceutical cost for management of comorbidities (€1238.67) in contrast to those without diabetes (€799.28). [29] Therefore, patient education and strict adherence to therapeutic guidelines are recommended to prevent complications, and thereby also reduce costs to the healthcare system. Evidence-based indications should be implemented to improve appropriate drug usage, including aspirin, in individuals with diabetes. In addition, these individuals should have their medications reviewed thoroughly every year.

Some limitations of this research include the retrospective design of the study, incomplete data from clinical case notes, and measurement of the most recent, single clinical and laboratory data, which could result in over- or under-diagnoses. Aetiology of stroke was important as haemorrhagic stroke would preclude the need for aspirin, but it was not differentiated in this study. In addition, most studies adopted the Framingham risk score for 10-year CVD risk or the QRISK2 score, while this study used the UKPDS risk engine, as the latter is tailored for diabetes patients. A larger sample size and a longer study period would be recommended to provide better statistical results. A larger geographical catchment area also should be considered for future studies.

In conclusion, type 2 diabetes and CVD account for the most non-communicable diseases’ deaths globally. Therefore, achievement of therapeutic targets for individuals with type 2 diabetes should be monitored closely and improved in order to reduce cardiovascular risk. A significant proportion of persons with diabetes should be, but were not, prescribed aspirin for secondary prevention. This is in contrast to the substantial population of those who should not be, but were on aspirin for primary prevention in this study. Both primary care and secondary care health professionals should re-evaluate their aspirin prescription among individuals with type 2 diabetes to ensure the utmost benefits are achieved.