By Michael Crawford, MD
Professor of Medicine, Chief of Clinical Cardiology, University of California, San Francisco
Dr. Crawford reports no financial relationships relevant to this field of study.
SYNOPSIS: The new vascular disease risk calculator discriminates who will experience a vascular event in the near future better than using a trial entry criteria approach or a hybrid approach.
SOURCES: Mortensen MB, et al. Primary prevention with statins: ACC/AHA risk-based approach versus trial-based approaches to guide statin therapy. J Am Coll Cardiol 2015;66:2699-2709.
Bittner V. Selecting patients for statin therapy in primary prevention: If we could only predict the future. J Am Coll Cardiol 2015;66:2710-2712.
The new American College of Cardiology/American Heart Association (ACC/AHA) risk assessment equation was based on decades-old cohort studies and may overestimate risk in modern cohorts. Accordingly, a trial-based approach, which uses statin trial population characteristics rather than a risk assessment, has been recommended. Also, a hybrid approach combining both methods has advocates. Investigators compared these three methods for statin allocation in the Copenhagen General Population Study (CGPS). CGPS enrolled randomly selected Danish adults between 40-75 years of age. The study excluded diabetics, those with vascular disease, statin users, and patients with an LDL-cholesterol > 189 mg/dL or an estimated 10-year risk of atherosclerotic cardiovascular disease (ASCVD) of < 7.5%. Trial-based statin allocation was based on the entry criteria of five primary prevention trials. The hybrid approach required a risk assessment of > 7.5% and meeting criteria for inclusion in one of the five trials demonstrating statin benefit. Of the total population of 57,892 eligible patients (57% women), more were statin eligible using the trial-based approach vs the risk equation (59% vs 42%). The hybrid approach reduced the number of eligible patients vs the risk equation (21% vs 42%). Of those eligible for statin therapy, the ASCVD rate per 1000 person years was 9.8 with the risk equation, 6.8 with trial-based approach, and 11.2 with the hybrid approach. The ACC/AHA risk equation discriminated better than the other two approaches between who would and would not have an ASCVD event (area under receiver operating characteristic curve 0.68 for the risk equation, 0.57 for the trial-based approach, and 0.61 for the hybrid approach). The authors concluded that use of the ACC/AHA risk equation approach will prevent more ASCVD deaths than the other two approaches while treating fewer patients than the trial-based approach.
COMMENTARY
Ever since the introductions of the new risk calculator and the concept that we should treat risk instead of lipid levels, there has been controversy, with concerns at both ends of the spectrum. Age heavily influences the new risk calculator. It is possible that younger patients with mildly elevated lipids and a bad family history (which isn’t part of the formula) will not be treated, yet should be. More likely is the concern that older individuals with reasonable lipid levels will be treated unnecessarily. This would translate into higher healthcare costs and potential adverse effects in this group, which is why the trial-based and hybrid approaches were advanced.
Dr. Bittner asks whether this study will resolve the primary prevention dilemma of who to treat with statins. Unfortunately, the answer is no. The risk calculator approach was best at identifying those destined to experience an event. However, we don’t know if these are the same individuals who will benefit from statin therapy. We would need a randomized trial to answer that question, but it is unlikely we will ever see such a trial. Also, the 8% of the population who had events, but were not identified by any of the three approaches, are a concern. Data only presented in the online data tables show that this 8% had higher Lp(a) levels than the rest of the population, so perhaps we should add Lp(a) as a treatment indicator. Also, none of these methods consider family history, yet other studies have shown it is important in risk prediction.
There are limitations to this study. First, they studied a largely white, European population; however, the lipid-lowering trials had the same weakness. Also, clinical trials tend to exclude more people. In fact, some enrolled < 10% of those screened. Thus, the trial approach may not represent a real patient population, whereas this large population study would. One advantage of the risk equation is that it can be adjusted over time as new data are generated, but trial results are frozen in time. The follow-up period in this study was only 5 years, so 10-year risk had to be extrapolated. In addition, treatment was not evaluated in this study and some higher-risk patients, who were not on statins at baseline (an exclusion criteria), may have been put on statins during the study period.
What should we do? Using the risk calculator as a first step makes sense and if statin therapy is indicated, deploy them. For those with a 10-year ASCVD event rate < 7.5%, individual judgment should be used. For example, if there is a family history of coronary artery disease at young ages, I would start statins regardless of the risk calculator. In the elderly with a risk > 7.5%, a discussion of the pros and cons of statin therapy would be beneficial, especially in those > 75 years of age in whom we have little data, but don’t expect them to be different than younger patients. Finally, the whole patient should be considered and attention given to other risk factors besides lipid values, smoking, hypertension, and age (all in the risk formula), such as diet and exercise.
