Author + information
- ↵∗Address for correspondence:
Dr. Michael Pignone, DMS Health Discovery Building, Room 7.704, 1701 Trinity Street, Stop Z0900, Austin, Texas 78712-1876.
With the publication of 3 new, large trials in the past year (1–3), the use of aspirin for primary prevention has once again sparked extensive debate and deliberation, including updated preventive care guidelines from the American Heart Association and American College of Cardiology (4). Although the new trials have brought the debate to the forefront, the preventive care community has had significant uncertainty about the role of aspirin for many years. The most recent U.S. Preventive Services Task Force recommendations from 2017 (created prior to the new trials) downgraded the strength of recommendation from the “A” level for adults age 50 to 75 years at increased cardiovascular risk to a “B” recommendation for adults age 50 to 60 years at increased cardiovascular risk and “C” recommendation for shared decision making for ages 60 to 70 years at increased cardiovascular risk (5). The Task Force based their recommendation on information from multiple systematic reviews (on cardiovascular benefits, colorectal cancer benefits, and bleeding risks) and a decision analytic model that attempted to synthesize these different potential outcomes (6,7). The differing recommendations by age reflected increasing bleeding risk with age, as well as finding that multiple years of aspirin use were needed to observe the colorectal cancer benefits (5).
The new 3 trials each provide helpful additional information for policymakers and guideline developers (1–3). However, these new trials have to be considered within the totality of the evidence available. As such, we are fortunate that, in the few months since the new trials were published, 2 high-quality systematic reviews and meta-analyses have been performed, each incorporating these new trials along with the many previous trials that have addressed aspirin’s role for primary prevention (8,9). The systematic review in this issue of the Journal by Abdelaziz et al. (8) included 15 trials with >165,000 total participants (mean age 61.5 years). Similar to previous meta-analyses, they found that aspirin reduced the risk of nonfatal myocardial infarction and ischemic strokes. The relative risks for all-cause mortality (0.97) and cardiovascular mortality (0.93) with aspirin were <1.00, but did not reach statistical significance. No overall effects were observed for cancer during the combined periods of observation (mean of 6.4 years). As with previous reviews, aspirin was associated with a clear increased risk of major bleeding (risk ratio [RR]: 1.50), including mostly major gastrointestinal bleeding (RR: 1.52), but also the less common but more debilitating intracranial hemorrhage (RR: 1.32) (8).
The investigators used meta-regression to examine whether several factors influenced the effect of aspirin. They found differences in stroke outcome by sex (women had greater reductions) and for lower doses of aspirin (<100 mg), perhaps through less risk of hemorrhagic stroke. Limiting to trials with longer follow-up produced somewhat more favorable estimates of effect on all-cause mortality. Other factors examined, including diabetes, smoking, and underlying cardiovascular risk, did not produce clear differences in effect. Zheng and Roddick (9) used somewhat different methods, but reached essentially the same conclusions.
Although both systematic reviews reached similar conclusions, several underlying limitations to the evidence base remain, despite the availability of multiple large, high-quality trials. First, the duration of the trials (mean 6.4 years) is insufficient to understand the full effects of aspirin when used for primary prevention. Longer observational follow-up of trial populations (even if the intervention and control conditions are not maintained) is essential for better understanding the impact of aspirin for all-cause mortality and for cancer incidence and mortality. Second, individual patient-level meta-analysis should be conducted to better understand any potential differences in effect of aspirin for key patient groups, including women, people with diabetes, and those using other cardiovascular risk-reducing therapies, especially statins. Finally, we need additional studies focusing on the effect of aspirin in women, to separate differences in effect by dose from differences in effect by sex.
While additional research is being undertaken, we must still cope with the question of what to do currently. Foremost, we must recognize that the decision about aspirin for primary prevention is hard, because, in addition to the uncertainty about the evidence, it involves weighing the modest potential benefits of prevention of first cardiovascular events versus the modest risk of adverse effects of increased bleeding. It is not sufficient to simply count up cardiovascular events prevented versus (mostly gastrointestinal) bleeding events caused, as they do not have the same immediate or eventual health effect. Modeling studies, including cost-effectiveness analyses, can be helpful in this regard, if one agrees with the assumptions used to build the models (10,11).
Such analyses suggest that the greatest potential net benefit would accrue to adults with increased cardiovascular risk who are not at increased risk for bleeding. Operationally, this group includes those ages 50 to 59 years and perhaps 60 to 69 years with 10-year cardiovascular risk over 10%. Because other forms of risk-reducing therapy are similarly or more effective and have fewer nonreversible adverse effects, it is prudent to first employ smoking cessation, blood pressure control, and statins in eligible patients, reserving a discussion about low-dose aspirin (including both its benefits and potential risks) for those with residual risk over 10% who are not at increased bleeding risk. Because bleeding risk increases with age, and in light of newer trial evidence (1), it seems prudent to avoid initiating aspirin for primary prevention in adults over age 70 years. Those who are taking aspirin and reach age 70 years without any adverse effects should discuss with their clinician the question of whether to continue aspirin or stop it, based on their overall risk and personal preferences.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Pignone is a former member of the U.S. Preventive Services Task Force. The views presented here are his and not necessarily those of the Task Force.
- 2019 American College of Cardiology Foundation
- McNeil J.J.,
- Wolfe R.,
- Woods R.L.,
- et al.
- Bowman L.,
- Mafham M.,
- Wallendszus K.,
- et al.
- Gaziano J.M.,
- Brotons C.,
- Coppolecchia R.,
- et al.
- Arnett D.K.,
- Blumenthal R.S.,
- Albert M.A.,
- et al.
- Zheng S.L.,
- Roddick A.J.