First published as Letters to the Editor in Annals of Internal Medicine 151(58):587-88, October 20, 2009.

TO THE EDITOR:

The recently updated U.S. Preventive Services Task Force (USPSTF) guidelines¹ on aspirin for primary prevention of cardiovascular disease are an ideal example of evidence-based medicine in action. Although we cannot fully explain why aspirin effects differ by sex, the evidence strongly suggests that they do. Now our guidelines, and practices, can reflect that. The concept makes intuitive sense: Target aspirin use to those who are most at risk for cardiovascular disease. Because cigarette smoking is a major risk factor for stroke and myocardial infarction, the updated guidelines imply that smokers are more likely to be given aspirin.

However, if we accept differential effects of aspirin in men and women, we should also accept the possibility of differential effects in other physiologically distinct subgroups. The Women's Health Study² showed a significant interaction with smoking status (P <0.001), and somewhat unexpectedly, aspirin use was associated with increased harm in women who were currently smoking (relative risk for major cardiovascular event, 1.3).

Whether other studies have shown similar results is complicated by the established gender gap. The TPT Thrombosis Prevention Trial)³ studied high-risk patients, 44% of whom were smokers, and although a subgroup analysis was not published, benefits of aspirin were consistent with other trials. However, the TPT was performed solely in men. Other trials involving women had far fewer patients than the Women's Health Study, and subgroup analyses from those trials were not published.

Because smoking is also a risk factor for peptic ulcer⁴ and the bulk of the data available suggest harm rather than benefit in women who smoke, it may be prudent to take a more cautious approach to aspirin use in this group until further data can clarify the issue.

Vikram Budhraja, MD
Lincoln Medical and Mental Health Center
New York, NY 10451

References

1. U.S. Preventive Services Task Force. Aspirin for the prevention of cardiovascular disease: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2009;150:396-404. [PMID: 19293072]

2. Ridker PM, Cook NR, Lee IM, Gordon D, Gaziano JM, Manson JE, et al. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N Engl J Med 2005;352:1293-304. [PMID: 15753114]

3. The Medical Research Council's General Practice Research Framework. Thrombosis prevention trial: randomised trial of low-intensity oral anticoagulation with warfarin and low-dose aspirin in the primary prevention of ischaemic heart disease in men at increased risk. Lancet 1998;351:233-41. [PMID: 9457092]

4. Kurata JH, Nogawa AN. Meta-analysis of risk factors for peptic ulcer. Nonsteroidal antiinflammatory drugs, Helicobacter pylori, and smoking J Clin Gastroenterol 1997;24:2. [PMID: 9013343]

TO THE EDITOR:

Although we agree with the expanded role of aspirin for primary prevention of coronary heart disease recommended by the USPSTF¹, we are concerned that their designated risk-assessment tool²—which at this time is down for revision—may lead to inappropriate overuse of aspirin for primary prevention.

Using this calculator, based on Framingham data³, a 45-year-old man with low-risk (normotensive; nondiabetic; nonsmoker; total cholesterol level, 4.1 mmol/L [160 mg/dL]; high-density lipoprotein cholesterol level, 1.3 mmol/L [50 mg/dL]) is assigned a 10-year risk of 4%. This low-risk patient, according to the USPSTF would be prescribed preventive aspirin. However, calculating the same patient's risk by using the online tool⁴ accompanying the National Cholesterol Education Program guidelines, which is also based on Framingham data, generates a risk of 1%.

Clinicians often exhort patients to be wary of online information, yet medical professionals need to exercise the same caution. The USPSTF committee writes, "[a]vailable tools provide estimations of coronary heart disease risk," suggesting that all tools are equivalent and accurate. However, previous research^{5, 6} has shown wide variability in equation-based prediction tools. The Table demonstrates such inconsistency by using 3 readily available risk-assessment tools. Without more specific guidance from the USPSTF, use of any online tool may lead to inappropriate overuse or underuse of aspirin, depending on the tool chosen.

Table. Predicted 10-Year Risk for 3 Hypothetical Patients, Using 3 Online Risk Calculators

^a Man aged 45 years, nonsmoker, untreated blood pressure of 120/80 mm Hg, total cholesterol of 4.1 mmol/L (160 mg/dL), low-density lipoprotein cholesterol of 3.6 mmol/L (140 mg/dL), high-density lipoprotein cholesterol of 1.3 mmol/L (50 mg/dL), triglyceride of 1.7 mmol/L (150 mg/dL), no family history of coronary heart disease.
^b Man aged 55 years, nonsmoker, untreated blood pressure of 130/90 mm Hg, total cholesterol of 5.2 mmol/L (200 mg/dL), low-density lipoprotein cholesterol of 3.6 mmol/L (140 mg/dL), high-density lipoprotein cholesterol of 1.3 mmol/L (50 mg/dL), triglyceride of 1.7 mmol/L (150 mg/dL), no family history of coronary heart disease.
^c Man aged 65 years, smoker, treated blood pressure of 120/80 mm Hg, total cholesterol of 6.2 mmol/L (240 mg/dL), low-density lipoprotein cholesterol of 5.2 mmol/L (200 mg/dL), high-density lipoprotein cholesterol of 0.8 mmol/L (30 mg/dL), triglyceride of 2.26 mmol/L (200 mg/dL), no family history of coronary heart disease.

We encourage the USPSTF to reissue their most recent recommendations with a specific risk-assessment tool that has been thoroughly studied to ensure the clinically appropriate application of these important guidelines.

Arun V. Mohan, MD, MBA
Carmen Patrick Mohan, MD
Richard Balaban, MD
Harvard Medical School and Cambridge Health Alliance
Cambridge, MA 02139

References

1. U.S. Preventive Services Task Force. Aspirin for the prevention of cardiovascular disease: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2009;150:396-404. [PMID: 19293072]

2. Medical College of Wisconsin. Risk Assessment Tool. Accessed 20 March 2009.

3. Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837-47. [PMID: 9603539]

4. National Cholesterol Education Program. 10-year CVD Risk Calculator. 19 March 2009. National Heart Lung and Blood Institute. Accessed on 11 September 2009.

5. Lenz M, Mühlhauser I. [Cardiovascular risk assessment for informed decision making. Validity of prediction tools]. Med Klin (Munich). 2004;99:651-61.

6. Brindle P, Beswick A, Fahey T, Ebrahim S. Accuracy and impact of risk assessment in the primary prevention of cardiovascular disease: a systematic review. Heart 2006;92:1752-9. [PMID: 16621883]

7. PROCAM risk calculator. Accessed 20 March 2009.

We appreciate the thoughtful letters from Dr. Budhraja and Dr. Mohan and colleagues regarding the USPSTF recommendation on aspirin prophylaxis for the prevention of cardiovascular disease. Dr. Budhraja calls attention to subgroups of women in whom the effect of aspirin in preventing cardiovascular disease may differ from that of the general population. In general, the USPSTF is cautious when considering unplanned subgroup analyses of randomized trials, which are the basis for Dr. Budhraja's comment. The authors of the original report from the Women's Health Study¹ mention multiple comparisons as an additional caution in interpreting this subgroup analysis. All subgroup analyses should be considered hypothesis-generating rather than independently persuasive.

The possibility suggested by Dr. Budhraja that the higher risk for peptic ulcer disease in smokers might place them at higher risk for hemorrhage when taking aspirin merits further research.

Dr. Mohan and colleagues raise many valuable points. The inadequate and contradictory information derived from Web-enabled coronary and cardiovascular disease risk calculators has been a matter of great concern for the USPSTF. The calculator referenced in the recommendation was selected primarily because it is easy to use and does not require information about high-density lipoprotein cholesterol concentration. As of this writing, the calculator has been removed from the Medical College of Wisconsin's Web site and reportedly is being revised.*

The USPSTF felt that making a recommendation meant to be tailored to estimation of cardiovascular disease risk without making any suggestions to help clinicians use the recommendation would be worse than mentioning an imperfect calculator. Research in this field is sorely needed. The development of a "gold standard" cardiovascular disease risk calculator to aid in predicting contemporary rates of cardiovascular disease in the United States should be a pressing priority for analysis of data derived from large cohort studies done in the past decade, perhaps pooling individual-level data across studies. The Agency for Healthcare Research and Quality has funded a project to evaluate the models currently available for risk calculation for cardiovascular disease. The results will be available soon. The use of risk-prediction models, and the tools based on them, to guide decisions about use of preventive and therapeutic medications, as well as decisions about screening, will become increasingly important in the emerging era of personalized medicine.

Ned Calonge, MD, MPH
Diana Petitti, MD, MPH
Mary Barton, MD, MPP
U.S. Preventive Services Task Force
Rockville, MD 20850

Reference

1. Ridker PM, Cook NR, Lee IM, Gordon D, Gaziano JM, Manson JE, et al. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N Engl J Med 2005;352:1293-304. [PMID: 15753114]

*At the time of this posting (October 2009), the calculator has been reinstated on the Medical College of Wisconsin's Web site (https://www.mcw.edu/calculators/coronary-heart-disease-risk). Accessed May 8, 2019.