Author + information
- William Wijns, MD, PhD∗ ( and )
- Dan Rusinaru, MD, PhD
- ↵∗Reprint requests and correspondence:
Dr. William Wijns, Cardiovascular Center Aalst, Moorselbaan 164, Aalst, Belgium 9300.
The cardiovascular disease (CVD) epidemic remains the leading cause of death worldwide (1). Widespread adoption of the Western diet and lifestyle by populations in emergent countries with low or middle income has resulted in dramatic increases in the incidence of coronary heart disease and stroke. An overwhelming proportion of the cardiovascular risk is explained by the cumulative presence of classic and potentially modifiable risk factors, and this effect does not seem to be influenced by sex, ethnicity, or geography (2). Therefore, to effectively reduce risk and improve outcomes, prevention strategies should be conducted and implemented at the population level. Yet in 2003, Wald and Law (3) introduced the concept of preventive pharmacotherapy as an attractive, but controversial, weapon to beat the CVD epidemic. According to most optimistic calculations, the implementation of a strategy based on preventive pharmacotherapy in subjects age ≥55 years, without known but at moderate risk of CVD, would result in 60% to 90% reduction in coronary heart disease events and 48% to 80% reduction in stroke rates (4,5). Unfortunately, hard data on the actual risk reduction that is associated with preventive pharmacotherapy using the polypill are presently still lacking.
In this issue of the Journal, Bittencourt et al. (6) propose an interesting novel approach to preventive pharmacotherapy to potentially improve its effectiveness. Using the coronary artery calcium (CAC) score, the authors enrich the population amenable to polypill treatment by selecting a higher-risk subset with subclinical disease present. Applying “what-if” calculations, the authors report the impact of adding CACS to the actual inclusion criteria for polypill studies. Participants of MESA (Multi-Ethnic Study of Atherosclerosis) who fulfilled the inclusion criteria of 4 published polypill trials were stratified according to baseline CAC score and CVD events during a 7.6-year median follow-up were compared. The results of this analysis are of great interest and can be briefly summarized as follows: 1) the rate of CVD events is low for patients with CAC = 0 (2.5 to 4.0 per 1,000 person-years), whereas patients with CAC >100 experience most of the CVD events (15.8 to 18.6 per 1,000 person-years); 2) assuming a 62% reduction of CVD events with preventive pharmacotherapy, the estimated 5-year number needed to treat (NNT) to prevent 1 CVD event ranges between 84 and 134 for the group with CAC = 0 and decreases dramatically to only 20 in patients with CAC >100; 3) the estimated NNT is still acceptable for the group with CAC >100 in case the anticipated risk reduction would be less than 62% (sensitivity analysis); and 4) CAC classification provides good discrimination in the NNT to prevent CVD events across the whole spectrum of the Framingham risk score. These findings suggest that a simple and rather inexpensive imaging method identifies patients who might benefit the most from preventive pharmacotherapy. The approach proposed by the authors is probably also cost effective because pharmacotherapy is deferred in low-risk individuals with CAC = 0, who represent 40% to 60% of subjects in the published polypill studies. Presumably, these subjects would no longer be exposed to potential side effects of the polypill, while no benefit was to be expected anyhow.
The following aspects of the proposed strategy deserve further comments. 1) What is the predictive value of CAC in asymptomatic individuals with different risk profiles drafted from the general population? 2) Will CAC ever be available and affordable at large to be applicable on a population basis? 3) Preventive pharmacotherapy with the polypill—are we even prepared to implement it?
In the assessment of cardiovascular risk, CAC is currently considered a Class IIa recommendation for asymptomatic individuals at intermediate risk (10% to 20% 10-year risk) and a Class IIb recommendation for low- to intermediate-risk persons (6% to 10% 10-year risk) (7). CAC improves Framingham CVD risk prediction through significant reclassification of subjects between risk categories. When added to a prediction model based on traditional risk factors, CAC results in reclassification of 26% of patients and 25% of events (8). Among individuals with intermediate 10-year risk, CAC permits reclassification of 16% as high risk and 39% as low risk (8). Overall, adding CAC to a classic estimation of risk places almost 80% of the population into either the highest or the lowest category of risk (8). Thus, CAC incrementally improves risk estimates in individual subjects, above and beyond the predictive power of clinical scores that are most effective at a population level.
Bittencourt et al. (6) show that adding CAC to the traditional classification of risk by Framingham score translates into higher visibility of individuals in whom pharmacological prevention would be most effective. CAC shows significant discrimination in the NNT to prevent events across the whole spectrum of traditional risk categories. Irrespective of Framingham risk, NNT to prevent coronary heart disease events is highest for CAC = 0 (between 135 and 162 in intermediate-risk individuals and between 68 and 126 in high-risk individuals). On the contrary, when CAC is >100, NNT to prevent coronary heart disease events is small even in low-risk individuals (between 38 and 56). These data suggest that a CAC-based risk strategy can guide preventive measures through better visibility of subjects having the disease substrate and who represent the actual target for risk reduction (9). Given the poor performance of classic risk scores at the individual level, the design of future trials in this field should consider using CAC in addition to risk scores to focus preventive pharmacotherapy on individuals most likely to benefit. This “de-risking” approach to risk reduction could be coined “interventional prevention.”
The availability of imaging centers in emerging countries and remote areas is a real issue that might limit the implementation of a CAC-based preventive pharmacotherapy strategy. One should also keep in mind that about one-quarter of individuals with CAC = 0 show some amount of coronary calcification 5 years later (10). This opens the discussion of serial CAC assessment and complicates the delivery of preventive pharmacotherapy beyond 5 years. On the other hand, the absence of coronary calcifications in low-risk patients is strongly correlated to absence of obstructive coronary lesions but does not exclude the presence of nonobstructive coronary heart disease in short of 10% of cases. With modern equipment, the effective radiation dose per CAC study is low, ranging between 0.7 and 1.3 mSv, which is still much higher than the effective radiation dose of a chest x-ray study (about 0.02 mSv) and might be associated with incremental risk of cancer. Lastly, CAC screening with abnormal results in otherwise healthy subjects may trigger unnecessary additional noninvasive or invasive examinations, leading to greater expenditures for healthcare systems.
To date, all studies investigating preventive pharmacological strategies at the population level have used surrogate endpoints to evaluate CVD risk reduction. The recently published UMPIRE (Use of a Multi-Drug Pill in Reducing Cardiovascular Events) trial randomized 2004 subjects with established CVD (history of coronary heart disease, ischemic cerebrovascular disease, or peripheral artery disease) or with an estimated 5-year CVD risk of 15% or greater to a single fixed-dose polypill or with usual care (11). The trial showed that a fixed-dose combination strategy resulted in better control of blood pressure and low-density lipoprotein cholesterol and improved adherence at 15 months as compared with usual care; granted, these are patients with known disease (11). The ongoing large international HOPE (Heart Outcomes Prevention Evaluation)-3 trial is investigating the effect of a combined therapy (statin, angiotensin receptor antagonist, and thiazide diuretic) on CVD events in 12,500 individuals without known heart disease at average risk (12). This trial is aimed at providing robust evidence supporting more widespread application of preventive pharmacotherapy. Until then, many healthcare providers are uncertain about its value and seem reluctant to prescribe the polypill for primary prevention in individuals at low or intermediate risk, for a number of reasons (13). The ideal composition of the polypill and the best dosage to avoid adverse effects and drug interactions remain subjects of debate. This is of importance because adverse effects negatively impact compliance and adherence to therapy. One may fear that healthy adults at moderate risk endure side effects that limit physical performance and interfere with personal and professional activities for the sake of prevention. Data from the PILL Collaborative Group (14) show that the polypill arm had significantly more adverse effects than the placebo group and that roughly 1 out of 20 participants stopped medication because of such effects. Others are concerned that the development of preventive pharmacotherapy might undermine the achievements of preventive medicine with respect to the implementation of indispensable lifestyle-modification measures, especially smoking cessation, weight reduction, and physical exercise. Technological and medical progress must not lead to an inert society where individuals assess periodically their risk, pursue unhealthy habits, and take medication to reassure themselves or to please their physicians.
Despite, or perhaps because of, these potential shortcomings, refining risk prediction by adding CAC to classic scores as proposed by Bittencourt et al. (6) is an innovative approach to preventive strategies. Unlike implementation of healthy diet and lifestyle modification, preventive pharmacotherapy in asymptomatic healthy subjects remains controversial. Because long-term drug prescription unavoidably comes with side effects, it is conceivable that the potential benefit of polypill prescription will be confounded. As a result, polypill trials may run the risk of showing neutral or even negative net effects. Using CAC to identify subjects at intermediate to high risk who carry the disease substrate at a pre-clinical state for inclusion in such trials seems like an attractive “de-risking” strategy for preventive pharmacotherapy.
↵∗ 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. Wijns has received research grants, honoraria, and consultant fees to the Cardiovascular Research Center Aalst from various device and pharmaceutical companies not directly relevant to this editorial; and is a co-founder, stockholder, and board member of Argonauts Partners, Genae Americas, and Cardio3 BioSciences, none directly relevant to this editorial. Dr. Rusinaru has reported that he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- ↵World Health Statistics 2013. Cardiovascular diseases. Available at: http://www.who.int/mediacentre/factsheets/fs317/en/index.html. Accessed September 7, 2013.
- Wald N.J.,
- Law M.R.
- Lonn E.,
- Bosch J.,
- Teo K.K.,
- Pais P.,
- Xavier D.,
- Yusuf S.
- Bittencourt M.S.,
- Blaha M.J.,
- Blankstein R.,
- et al.
- Greenland P.,
- Alpert J.S.,
- Beller G.A.,
- et al.
- Nasir K.,
- Rubin J.,
- Blaha M.J.,
- et al.
- Min J.K.,
- Dunning A.,
- Lin F.Y.,
- et al.
- ↵Heart Outcomes Prevention Evaluation-3 (HOPE-3). Available at: http://clinicaltrials.gov/show/NCT00468923. Accessed September 7, 2013.
- Virdee S.K.,
- Greenfield S.M.,
- Fletcher K.,
- McManus R.J.,
- Hobbs F.D.,
- Mant J.