Monday, June 27, 2011

Aspirin for primary CVD prevention: the continuing debate

In 2002, the U.S. Preventive Services Task Force (USPSTF) strongly recommended that primary care clinicians discuss preventive aspirin use with adults at increased risk of cardiovascular events. Four years later, the National Commission on Prevention Priorities (NCPP) ranked counseling for aspirin use the number one priority on its list of the most effective clinical preventive services. According to the NCPP, if the percentage of eligible patients using aspirin (then estimated to be about 50 percent) increased to 90 percent, 45,000 additional lives could be extended each year.

At that time, the benefits of aspirin use in men and women were assumed to be the same. However, an updated USPSTF recommendation statement published in the June 15th issue of American Family Physician indicates that aspirin use actually prevents heart attacks in men, but ischemic strokes in women. In addition, physicians and patients must weigh the benefits of reduced cardiovascular risk with the risk of gastrointestinal bleeding events, and use shared decision making when these risks are closely balanced.

To further complicate matters, a 2009 meta-analysis published in the journal The Lancet questioned the value of aspirin for primary prevention, concluding that for patients who without a history of cardiovascular disease, "aspirin is of uncertain net value." In response, family physicians and USPSTF members Ned Calonge and Michael LeFevre wrote an editorial that concluded, "There is not a simple message for aspirin prophylaxis as a primary preventive strategy, and we need to consider gender, age, and the associated balance of potential risks and benefits to provide the best advice and preventive care for our patients."

The debate continues with two thought-provoking editorials in the June 15th issue of AFP. Alison L. Bailey and colleagues caution that routine aspirin use is not justified for primary prevention in adults at low risk of CVD. On the other hand, W. Fred Miser asserts that the main issue regarding aspirin for primary prevention continues to be underuse in appropriate-risk patients.

So which is the bigger problem, overuse or underuse?


The above is a slightly edited version of a post that was originally published on the AFP Community Blog.


  1. once again, medicne is not simple and there is not one answer for everyone. The data is not inconsistent. Men >40 with 2 or more CV risk factors benefit. no risk factors don't bother. If I were a woman I would want to decrease my risk for stroke. Having seen strokes, they Suck.
    So again, women probaly over 55 wiht risk factors for stroke should take aspirin. As i tell my patients If you get side effects the risk starts to out weight the benefit and so you stop.

  2. Kenny
    I think you should take a stand and answer the question: Do you think that the greater problem is overuse or underuse?
    Myself, I think underuse in patients for whom it is indicated.
    Of course, there is a population of people who will (because they read and care about their health) assume that adopting a preventive strategy (esp cheap, like aspirin) is good even if they don't meet the criteria. Similarly, there is a cohort who will resist any treatment even if they do. point is that we need effectiveness and fidelity studies to find out why people who are, for example, not taking aspirin when the indications are there are not. This is meaningful research; if we do not know the reasons, we cannnot design effective interventions.

  3. Josh, I agree with you and the NCPP that underuse is probably the bigger problem. Part of it may be a misperception of one's personal risk for a CV event - although there are Web calculators that do this easily, few physicians use them, and fewer patients know about them.

    And then there is the tolerance for adverse effects of aspirin. To us, there may be an obvious threshold where averting an MI or ischemic stroke would outweigh the risk of a GI bleed or hemorrhagic stroke (and the USPSTF actually suggests such thresholds in its statement), but the asymptomatic patient may not readily agree.

    To take another example in chemoprevention, most of the (relatively few) women who take tamoxifen or raloxifene to prevent primary breast cancer are not those who would benefit most from doing so, statistically speaking. (Of course you could take the reasonable position that compared to aspirin, breast ca chemoprevention is a very questionable strategy, as it's been shown to reduce breast cancers but not lower breast cancer mortality, and may simply be targeting overdiagnosed cancers. Ditto for prostate cancer chemoprevention.)