Use of low-dose acetylsalicylic acid for cardiovascular disease prevention: A practical, stepwise approach for pharmacists

Abstract

Low-dose acetylsalicylic acid (ASA) is recommended in patients with established cardiovascular disease. However, the role of ASA in those without cardiovascular disease (i.e., primary prevention) is less clear, which has led to discordance among Canadian guidelines. In 2018, 3 double-blind, randomized controlled trials were published that evaluated ASA 100 mg daily versus placebo in patients without established cardiovascular disease. In the ASPREE trial, ASA did not reduce the risk of all-cause death, dementia, or persistent physical disability in patients ≥70 years of age but increased the risk of major bleeding. In the ARRIVE trial, ASA failed to lower the risk of a composite of cardiovascular events but increased any gastrointestinal bleeding in patients at intermediate risk of cardiovascular disease. In the ASCEND trial, ASA significantly reduced the primary composite cardiovascular outcome in patients with diabetes for a number needed to treat of 91 over approximately 7.4 years. Yet major bleeding was increased with ASA for a number needed to harm of 112. Therefore, in most situations, ASA should not be recommended for primary cardiovascular prevention. However, there are additional indications for ASA beyond cardiovascular disease. Thus, a sequential algorithm was developed based on contemporary evidence to help pharmacists determine the suitability of ASA in their patients and play an active role in educating their patients about the potential benefits (or lack thereof) and risks of ASA. Can Pharm J (Ott) 2020;153:xx-xx.

Introduction

Case Vignette

Mr. Singh is a 55-year-old man of Indian descent who has been a long-standing patient at your pharmacy. He has hypertension, which is well controlled with the combination of an angiotensin-converting enzyme (ACE) inhibitor and thiazide-like diuretic, but no other cardiovascular history or risk factors. His medical history is otherwise noncontributory. He inquires whether he should be taking low-dose acetylsalicylic acid (ASA) to lower his risk of a heart attack.

Knowledge Into Practice.

• The role of low-dose acetylsalicylic acid (ASA) in the primary prevention of cardiovascular disease is ambiguous.

• Previous meta-analyses have shown that ASA reduced the risk of cardiovascular events in patients without cardiovascular disease but also increased the risk of major bleeding.

• Three recently published randomized, placebo-controlled trials investigated the use of ASA to prevent cardiovascular events in patients without cardiovascular disease. In patients ≥70 years of age or at intermediate cardiovascular risk, ASA did not reduce cardiovascular events but increased the risk of bleeding. In patients with diabetes, ASA reduced cardiovascular events but also increased the risk of major bleeding.

• A stepwise approach was developed to aid pharmacists in assessing the appropriateness of ASA in their patients.

Mise En Pratique Des Connaissances.

• Le rôle de la faible dose d’AAS dans la prévention primaire des MCV est ambigu.

• Des méta-analyses antérieures ont montré que l’AAS réduit le risque d’événements cardiovasculaires chez les patients sans MCV, mais accroît également le risque d’hémorragie majeure.

• Trois études à répartition aléatoire et contrôlées par placebo dont les résultats ont récemment été publiés examinaient l’utilisation de l’AAS pour prévenir les événements cardiovasculaires chez les patients sans MCV. Chez les patients âgés de 70 ans et plus ou chez les patients présentant un risque cardiovasculaire intermédiaire, l’AAS n’a pas réduit les événements cardiovasculaires, mais a accru le risque d’hémorragie. Chez les patients diabétiques, l’AAS a réduit les événements cardiovasculaires, mais a également accru le risque d’hémorragie majeure.

• Une approche graduelle a été conçue pour aider les pharmaciens à évaluer l’avantage de l’AAS pour leurs patients.

Background

Pharmacists are often asked about the role of ASA in the prevention of cardiovascular disease (CVD). A cross-sectional survey conducted in 2009-2010 of patients at family practices in Alberta found that 40% of respondents took ASA.¹ Of those, roughly 41% took ASA for primary CVD prevention, most of whom stated it was based on advice from their health care provider. These findings are similar to the results of a national survey in the United States in 2012, where 52% of respondents aged 45 to 75 years reported taking ASA.²

ASA is recommended in patients with established CVD (i.e., secondary prevention).³ However, the role of ASA in those without CVD (i.e., primary prevention) is less clear. Previous meta-analyses have shown that ASA reduces the risk of cardiovascular events in patients without CVD but also increases the risk of major bleeding.^4-8 The included studies have a number of limitations, including variable ASA dosing (ranging from 100 mg every other day to 500 mg daily) and treatment duration. Furthermore, most participants were at relatively low cardiovascular risk, as patients with moderate or high cardiovascular risk constituted <10% of the study populations. These trials do not represent contemporary practice, as few patients were on other therapies for cardiovascular risk reduction (e.g., statins).

This has led to discordance among Canadian guidelines. The Canadian Cardiovascular Society states that ASA, at any dose, should not be recommended for the routine prevention of cardiovascular events in patients without CVD but also notes that ASA 75 to 162 mg daily could be considered in patients with high cardiovascular and low bleeding risk, which is based on consensus opinion.⁹ Similarly, Diabetes Canada recommends that low-dose ASA should not be used routinely in patients with diabetes but may be considered in patients with additional cardiovascular risk factors based on consensus opinion.¹⁰ Hypertension Canada recommends that low-dose ASA should be considered in hypertensive patients aged ≥50 years but that caution should be exercised if their blood pressure is uncontrolled.¹¹

In 2018, 3 double-blind, randomized controlled trials were published that evaluated ASA 100 mg daily versus placebo in patients without established CVD.^12-16 A detailed summary of the trials is included in Table 1. These contemporary trials informed the development of the following stepwise approach for pharmacists to assess the appropriateness of ASA in their patients (Figure 1).

Table 1.

Detailed summary of recent trials comparing ASA to placebo in patients without cardiovascular disease

				Primary outcome*				Composite cardiovascular outcome†				Major bleeding*
	n	Population	Follow-up	ASA	Placebo	HR/RR (95% CI)	NNT	ASA	Placebo	HR/RR (95% CI)	NNT	ASA	Placebo	HR/RR (95% CI)	NNH
ASPREE12-14	19,114	Mean age 74 years, 44% male, 11% DM	4.7 years (median)	21.5 events per 1000 person-years	21.2 events per 1000 person-years	1.01 (0.92-1.11)	NS	7.8 events per 1000 person-years	8.8 events per 1000 person-years	0.89 (0.77-1.03)	NS	8.6 events per 1000 person-years	6.2 events per 1000 person- years	1.38 (1.18-1.62)	89
ASCEND15	15,480	Mean age 63 years, 63% male, 94% DM	7.4 years (mean)	8.5%	9.6%	0.88 (0.79-0.97)	91	7.0%	7.6%	0.92 (0.82-1.03)	NS	4.1%	3.2%	1.29 (1.09-1.52)	112
ARRIVE16	12,546	Mean age 64 years, 70% male (DM excluded)	5 years (median)	4.3%	4.5%	0.96 (0.81-1.13)	NS	3.3%	3.5%	0.95 (0.79-1.15)	NS	1.1%	0.6%	1.73 (1.17-2.53)‡	200

ASA, acetylsalicylic acid; CI, confidence interval; DM, diabetes mellitus; HR, hazard ratio; NNH, number needed to harm; NNT, number needed to treat; NS, not significant; RR, rate ratio.

^*Refer to text for specific definition.

^†Includes cardiovascular/coronary heart disease death, fatal/nonfatal myocardial infarction and fatal/nonfatal stroke.

^‡Calculated by the authors using a χ² test.

Figure 1.

A stepwise algorithm to assess the appropriateness of ASA

Stepwise approach

• Step 1: Does the patient have CVD? If yes, recommend ASA (but consider caveats). If no, go to step 2.

Patients with established CVD can broadly be categorized as being either symptomatic or asymptomatic. Patients with symptomatic CVD include those with a previous acute coronary syndrome (myocardial infarction [MI] or unstable angina), stable angina, stroke, transient ischemic attack (TIA) and peripheral artery disease. Asymptomatic CVD includes patients with atherosclerosis identified on imaging, such as a coronary angiogram or carotid ultrasound. Thus, it is important to inquire of patients or review their medical record to identify all types of CVD. Patients with established CVD are generally recommended low-dose ASA (80-81 mg) daily as secondary prevention. In patients with CVD, ASA has been shown to reduce serious vascular events, major coronary events, stroke and vascular death but also to increase the risk of major extracranial bleeding.^5,17 However, the general consensus is that the benefits of therapy exceed the risks. Of note, most of these trials enrolled patients with symptomatic CVD.

However, not all patients with CVD should be on ASA. Caveats include patients with a contraindication (e.g., hypersensitivity reaction) or intolerance (e.g., severe dyspepsia). As well, patients may be on an oral anticoagulant (direct-acting oral anticoagulant or warfarin), which may preclude the need for ASA depending on their individual factors. Also, patients may have an unacceptable bleeding risk due to a previous bleed (or be at high risk of bleeding) or potential interactions between ASA and drugs that increase the risk of bleeding (e.g., nonsteroidal anti-inflammatory drugs [NSAIDs], corticosteroids, selective serotonin reuptake inhibitors).

• Step 2: Does the patient have another indication for ASA? If yes, recommend ASA (but consider caveats). If no, go to step 3.

Patients may have additional indications for ASA beyond CVD. These may be broadly categorized as cardiac or noncardiac indications. Cardiac indications include patients with atrial fibrillation (if they are unable or unwilling to take an oral anticoagulant)¹⁸ or a previous valve replacement (e.g., bioprosthetic aortic or mitral valve).¹⁹ Some noncardiac indications include prevention of colorectal cancer,²⁰ venous thromboembolism prophylaxis (if they are unable or unwilling to take an oral anticoagulant),²¹ preeclampsia (a condition in pregnancy characterized by hypertension and proteinuria)²² or polycythemia vera or essential thrombocythemia (myeloproliferative disorders characterized by an elevated red blood cell or platelet count, respectively).²³

• Step 3: Is the patient ≥70 years of age? If yes, do not recommend ASA. If no, go to step 4.

The ASPREE trial enrolled >19,000 “healthy elderly” patients (i.e., no history of CVD, dementia or physical disability) who were ≥70 years of age (or ≥65 years of age among blacks and Hispanics in the United States).^12-14 Roughly 70% of patients had 2 or more cardiovascular risk factors, although only 11% had diabetes. The trial was terminated early after 4.7 years of follow-up based on the determination that ASA was unlikely to show a benefit. The primary composite outcome of all-cause death, dementia and persistent physical disability was not significantly different between groups. A secondary cardiovascular composite outcome of coronary heart disease death, nonfatal MI and nonfatal stroke was also not significantly different between groups. However, major bleeding (defined as hemorrhagic stroke, symptomatic intracranial bleeding or extracranial bleeding that led to transfusion, hospitalization, surgery or death) was higher with ASA for a number needed to harm (NNH) of 89. Upper and lower gastrointestinal bleeding were also higher with ASA for a NNH of 213 and 532, respectively. All-cause death was actually higher with ASA for a NNH of 143, but this finding was considered to be hypothesis generating. Based on the results of this trial, ASA should generally not be recommended for primary disease prevention in patients ≥70 years of age.

• Step 4: Does the patient have diabetes mellitus and is ≥40 years of age? If yes, discuss the potential benefits and risks of ASA. In general, do not recommend ASA, but select patients may make an informed decision to take ASA for primary cardiovascular prevention. If no, go to step 5.

The ASCEND trial included >15,000 patients ≥40 years of age, of whom 94% had type 2 diabetes.¹⁵ Approximately 75% were on a statin at baseline, and mean baseline systolic blood pressure was 136 mmHg. Eighty-two percent were considered to be at low or moderate cardiovascular risk. After a mean of 7.4 years, ASA significantly reduced the primary composite outcome of vascular death, nonfatal MI and nonfatal stroke or TIA for a number needed to treat (NNT) of 91. However, the secondary composite outcome of nonfatal MI, nonfatal stroke and vascular death (i.e., without TIA) was not significantly different between groups. Of note, in the subgroup of patients ≥70 years of age (24% of the overall population), ASA did not reduce the risk of cardiovascular events or revascularization. Both major bleeding (defined as intracranial hemorrhage, sight-threatening eye bleeding that required hospitalization or any bleeding that resulted in transfusion or was fatal or disabling) and serious gastrointestinal bleeding (defined as any gastrointestinal bleeding that resulted in hospitalization, transfusion or was fatal or disabling) were increased with ASA for a NNH of 112 and 200, respectively. The incidence of any gastrointestinal bleeding was not reported. There was no significant difference in all-cause or vascular death. These results are consistent with a preceding meta-analysis of patients with diabetes.⁸

Based on these data, ASA may have a role in the prevention of cardiovascular events in patients with diabetes; however, the reduction in cardiovascular events is approximately offset by an increase in the risk of bleeding. It is important to note that patients with risk factors for bleeding (e.g., previous bleeding event, peptic ulcer disease, severe kidney or liver disease, concurrent NSAID use) may be at higher risk of bleeding compared with the study population. Therefore, it is recommended to engage in shared decision making with patients to discuss the potential benefits and risks of therapy, thereby allowing them to make an informed decision based on their individual factors as well as their values and preferences.

• Step 5: Discuss the potential benefits and risks of ASA. In general, do not recommend ASA, but select patients may make an informed decision to take ASA for primary cardiovascular prevention.

The ARRIVE trial enrolled >12,000 men aged ≥55 years and women aged ≥60 years at intermediate cardiovascular risk (defined as a 10-year risk of coronary heart disease of 10%-20% based on various American and European risk calculators).¹⁶ The mean baseline Framingham Risk Score was 14%. Patients with diabetes were excluded. After 5 years of follow-up, the primary composite outcome of cardiovascular death, MI, unstable angina, stroke and TIA was not significantly different between groups, nor was the composite of cardiovascular death, MI and stroke. Overall major bleeding was not reported. However, any gastrointestinal bleeding was increased with ASA for an NNH of 197. In addition, the combination of any gastrointestinal bleeding and hemorrhagic stroke was higher with ASA. There was no significant difference in all-cause or cardiovascular death. Of note, this trial had several limitations that may have influenced the results. The actual cardiovascular risk of the included population was lower than the predicted event rate, as the calculated 10-year risk of cardiovascular death, MI and stroke was approximately 7% in the placebo group compared with a Framingham Risk Score of 14%. Therefore, the trial may have been underpowered. Furthermore, roughly 30% of patients in each group prematurely withdrew from the study.

A subsequent meta-analysis included data from 13 randomized controlled trials including the ASPREE, ARRIVE and ASCEND trials (>164,000 patients) that compared ASA to either placebo or no treatment.²⁴ ASA reduced the composite endpoint of cardiovascular death, nonfatal MI and nonfatal stroke by an absolute 0.38% (NNT of 265) over 5 years but also increased the risk of major bleeding with ASA by an absolute 0.47% (NNH of 210). There was no significant difference in all-cause death between groups.

Because the relative benefits and risks of ASA are similar, they could be considered to counterbalance each other in terms of net clinical effect. However, it is recommended to engage in shared decision making with patients to discuss the potential benefits and risks of therapy, thereby allowing them to make an informed decision based on their individual values and preferences. For example, some patients may want to do everything possible to prevent a cardiovascular event and are willing to accept the increased risk of bleeding. Other patients may feel as though the absolute benefit is too small to warrant taking ASA daily considering the higher risk of bleeding. The aforementioned evidence may prompt some patients who currently take ASA to consider discontinuing therapy. Some clinicians have raised concerns that this may precipitate cardiovascular events.²⁵ However, there are no substantive data to support this concern in those taking ASA for primary cardiovascular prevention. Furthermore, one must note that any intervention aimed at reducing cardiovascular events does not eliminate the risk (e.g., ASA in secondary prevention only decreased the risk of subsequent cardiovascular events by approximately 20%). Use of ASA, or lack thereof, should be reassessed on an annual basis, as patients’ overall cardiovascular and bleeding risk will change over time.

Finally, it is important to recommend that patients engage in a multifaceted approach to lower their cardiovascular risk, such as healthy eating, regular physical activity, smoking cessation, healthy weight goals, stress reduction and appropriate blood pressure and lipid control.³ It is recommended that any patient with cardiovascular risk factors undergo an objective assessment of their cardiovascular risk (e.g., Framingham Risk Score), which can be discussed with the patient to help inform and motivate their approach to lowering their cardiovascular risk.²⁶

Case vignettes

To follow up the introductory case vignette using the stepwise approach (Figure 1), Mr. Singh does not have a history of CVD and has no other indications for ASA. Further, he is <70 years of age and does not have a history of diabetes. He would be similar to the population that was studied in the ARRIVE trial, which showed ASA was not beneficial at reducing cardiovascular events but increased the risk of bleeding. However, it should be noted that only approximately 2% of the trial population was nonwhite. Overall, it is reasonable to recommend that he not start taking ASA.

The following additional case vignettes represent other common clinical scenarios to help illustrate some key concepts regarding the stepwise approach to the use of low-dose ASA for CVD prevention.

Case 1

A 72-year-old woman approaches your pharmacy counter to ask whether she should continue taking ASA, as she recently saw a news report that suggested she should not be taking it. She has a history of hypertension (you measure her blood pressure, which is slightly above target) and hypothyroidism. Upon questioning, she denies any history of CVD. She reports taking ASA 81 mg daily of her own accord for about 12 years with no evidence of adverse effects. As per her pharmacy medication profile, her other medications include an ACE inhibitor and levothyroxine. She does not like taking medications and expressed a preference to be on as few medications as possible.

She is similar to the population that was studied in the ASPREE trial, which showed ASA was not beneficial at reducing cardiovascular events but increased the risk of bleeding. Although she has been taking ASA for years without incident, it does not preclude her from future harm. Her cardiovascular risk, as well as her bleeding risk, will continue to increase with age. Based on her preference to be on as few medications as possible, it is reasonable to recommend that she stop ASA. Conversely, she is likely going to derive more cardiovascular benefit by optimizing her antihypertensive therapy to achieve her blood pressure target.

Case 2

A 67-year-old man approaches you to request a medication review, as he feels as though he is taking “too many” medications. You perform a best possible medication history—he is currently taking ASA 81 mg daily in addition to an ACE inhibitor, β-blocker, dihydropyridine calcium channel blocker and a statin, all of which are “for his heart.” He also takes metformin and a sulfonylurea for type 2 diabetes mellitus, a proton pump inhibitor for gastroesophageal reflux disease, gabapentin for diabetic neuropathy, tamsulosin for benign prostatic hypertrophy and acetaminophen for lower back pain. He denies ever having an MI or stroke. However, you check his electronic medical record and discover he had a percutaneous coronary intervention with a drug-eluting stent placed in his left anterior descending artery about 2 years ago. When asked, he states he forgot to mention this procedure and that it was because he was having chest pains while golfing.

Although he denies any history of a symptomatic cardiovascular event, he is technically secondary prevention because he underwent percutaneous coronary intervention with stenting for what sounds to be stable angina. Therefore, he does have a valid indication for ASA and should be recommended to continue therapy. Despite having gastroesophageal reflux disease, the benefits of ASA likely outweigh the risks, and his use of a proton pump inhibitor may lower his risk of gastrointestinal bleeding.

Case 3

A 56-year-old man attends an appointment with you at your family health team based on a referral by his primary care provider for a cardiovascular risk assessment. His medical history is noncontributory save for a 40 pack-year history of smoking. He also has a positive family history of premature CVD, as his father passed away from an MI at age 55. Using a tool in your electronic medical record, you calculate his Framingham Risk Score to be >20%, indicating he is at high cardiovascular risk. He is quite concerned about his family history and is interested in taking ASA to reduce his risk of an MI.

The role of ASA in this specific patient population (i.e., high primary cardiovascular risk) has not been studied. Therefore, it is difficult to make a recommendation for him based on recent evidence. Overall, his cardiovascular risk is quite elevated, and he does not have any obvious risk factors for bleeding. Further, he is interested in taking a medication to reduce his risk. Therefore, it would be prudent to engage in shared decision making with him, acknowledging that the cardiovascular benefit may outweigh the risk of bleeding (although both are relatively small), thereby allowing him to make an informed decision whether or not to start ASA. In general, it is advisable to recommend that he not take ASA because of the lack of evidence for his specific circumstance but to support him if he chooses to start taking ASA. More importantly, he should be educated that both smoking cessation and statin therapy would likely provide a greater relative cardiovascular benefit than ASA. He should also be educated about ways to mitigate any other modifiable cardiovascular risk factors, if applicable (e.g., obesity, sedentary lifestyle, poor eating habits, stress).

Conclusion

Recently published randomized controlled trials informed the creation of a sequential algorithm for pharmacists to assess the appropriateness of ASA in their patients. It is recommended that clinicians follow this stepwise approach in practice when assessing the use of ASA. Many patients take ASA for a variety of reasons, either self-prescribed or on the advice of a health care professional. The role of ASA in the secondary prevention of CVD is well established. However, recent clinical trials have shown that, in general, ASA is not beneficial for the prevention of cardiovascular events in patients without CVD but increases the risk of bleeding. In patients with diabetes, ASA was shown to reduce cardiovascular events but also to increase the risk of bleeding. Future guideline recommendations will likely reflect these new data to provide more consistency for frontline practitioners. Notwithstanding, it is important to consider patients’ values and preferences when discussing the potential benefits and risks of ASA. ■