ACC, AHA Risk-Based Approach Remains Best Method for Statin Allocation in Primary Prevention

The American College of Cardiology/American Heart Association (ACC/AHA) risk-based approach for allocating statin therapy in primary prevention outperforms 2 other recently proposed methods for identifying patients who should be treated with the lipid-lowering therapy, according to a study published December 14, 2015, in the Journal of the American College of Cardiology.  

Overall, the ACC/AHA risk-based approach was better at identifying European patients who would go on to have a clinical event—and those would not—than a recently proposed screening method using RCT inclusion criteria to identify statin-eligible patients. 

The ACC/AHA risk-based method for the primary prevention of atherosclerotic cardiovascular disease (ASCVD) was also better than a hybrid approach, one that combined clinical trial inclusion criteria and the ACC/AHA risk-based approach, for identifying at-risk patients who would benefit from statins.

“The majority of clinical trials have not taken people based on the highest risk,” said senior investigator Børge Nordestgaard, MD, of Copenhagen University Hospital (Denmark), explaining a drawback of using RCT criteria to determine statin eligibility. Researchers might include specific populations to make scientific points, while trial sponsors might design trials with the goal of targeting different patients for drug therapy. With a risk-based approach, you are selecting patients with the highest risk for clinical events, “and we know from randomized trials, the higher the risk the bigger the benefit,” said Nordestgaard.

In an editorial, Vera Bittner, MD, of the University of Alabama (Birmingham, AL), writes that the cardiovascular risk calculator from the AHA/ACC does appear best for identifying subjects destined to have a future ASCVD event. She points out that the quantification of risk is just the first step, however, and that atherosclerosis remains a multifactorial disease requiring an approach focusing not only on the pharmacological control of lipids but also on smoking cessation, changes in diet, weight management, and physical activity, as well as attention to psychosocial risk factors.

“Let us remember, too, that an exclusive ‘lipocentric approach’ to ASCVD risk reduction begets large residual risk, as documented by the results of our lipid-lowering clinical trials,” writes Bittner.

Cholesterol Guidelines Now Based On Risk, Not LDL Levels

In the 2013 ACC/AHA cholesterol guidelines, men and women aged 40 to 75 years without ASCVD or diabetes who have a 10-year risk of ASCVD >7.5% are now eligible for statin therapy. To assess patient risk, a new cardiovascular risk calculator was developed to accompany the cholesterol guidelines.

Although the risk calculator was derived from diverse cohorts, it has been criticized for overestimating risk in other populations. This overestimation was a concern among physicians, because it could have the unintended effect of large numbers of patients needlessly being put on statins.

Speaking with TCTMD, Nordestgaard said these criticisms sparked some researchers to propose a “trial-based” approach for statin allocation. Using this method, statins would be offered to patients in whom RCTs support their use, ignoring individual risk assessment and absolute risk. The “hybrid” approach for identifying statin-eligible patients combines the trial-based method with the ACC/AHA risk-based approach.

In this analysis, the researchers compared the performance of the 3 approaches for statin initiation in 37,892 individuals free from diabetes, ASCVD, and statin use at baseline in the Copenhagen General Population Study (CGPS). With the trial-based approach, patients who met eligibility criteria for any one of the major primary-prevention studies (WOSCOPS, AFCAPS/TexCAPS, ASCOT-LLA, MEGA, and JUPITER) were eligible for statins. The hybrid approach required patients to have an estimated 10-year ASCVD risk > 7.5% but also meet the inclusion criteria of one of the primary-prevention studies.

Using the ACC/AHA approach, 42% of the CGPS cohort would be eligible for statin therapy. In contrast, 56% of individuals would be treated if the trial-based criteria were used. The hybrid approach yielded the fewest number of statin-eligible patients, with just 21% qualifying for therapy. Of those eligible for statins with the ACC/AHA risk-based approach, the ASCVD event rate was 9.8 per 1,000 individuals compared with 6.8 and 11.2 events per 1,000 individuals with the trial-based and hybrid approaches, respectively.

In terms of how well the approaches identify patients who develop ASCVD, the trial-based and hybrid approaches were significantly less sensitive than the ACC/AHA risk score. As for specificity, the RCT-based approach fared worse than the ACC/AHA risk score, meaning this method would result in more patients who did not require statins being treated with them. The area-under-curve for the ACC/AHA risk-based approach was moderate at 0.676, but higher than that observed with other approaches.

Regarding the net reclassification index (NRI), which was used to measure how well the models improve performance compared with ACC/AHA risk-based approach, the RCT and hybrid methods had negative NRIs (-0.21 and -0.13, respectively; P <.0001 for both), meaning they “are much worse overall in finding the people who did have an event and not treating people who did not have an event,” said Nordestgaard.

In the editorial, Bittner states that there are no prospective data on the expected benefit of statins using the risk-based approach—no trial has tested treating patients using the ACC/AHA risk threshold—but that the benefit is extrapolated from clinical trials. The trial and hybrid approach to risk stratification can be limited by the types of patients included in the studies, and “potentially, by the limited applicability of trials conducted in specific populations in other geographic regions, risk environments, or ethnicities,” she writes.

“The ACC/AHA guidelines have exactly the same approach as the European guidelines, which is to find the people at the highest risk and to give them statins,” said Nordestgaard. This approach, he added, performs “much better” than any of the other proposed methods

Validation for the ACC/AHA Model

Donald Lloyd-Jones, MD, of Northwestern University Feinberg School of Medicine (Chicago, IL), who co-chaired the writing committee that developed the 2013 ACC/AHA risk calculator, told TCTMD the latest analysis “endorses, pretty strongly, what the guidelines recommend, which is the risk-based approach.”

When devising the tool, Lloyd-Jones said they considered using an approach based on RCT inclusion/exclusion criteria but felt it would be burdensome in clinical practice. “We dismissed it for a couple of reasons, but the main reason is that it’s not very practical,” he said. “It’s very hard for a clinician to strictly apply inclusion/exclusion criteria in an office visit whereas now, with the apps that are available, it’s pretty darn easy to calculate 10-year risk.”

In addition, Lloyd-Jones said, there are at-risk patients not represented in the primary-prevention studies who could still stand to benefit from treatment. Another problem might be the patient who meets the inclusion criteria of different trials testing various statins at different doses. In such a scenario, it would be difficult to know what to prescribe, he noted. The RCT-based approach could “lead to confusion, a grey zone, and possibly undertreatment,” said Lloyd-Jones.  

Sources:
1. Mortensen MB, Afzal S, Nordestgaard BG, Falk E. 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.
2. Bittner V. Selecting patients for statin therapy in primary prevention [editorial]. J Am Coll Cardiol. 2015;66:2710-2712.

Disclosures:

• The study was supported by the Herlev and Gentofte Hospital, Copenhagen University Hospital, the Copenhagen County Foundation, and Aarhus University.
• Nordestgaard and Lloyd-Jones reports no relevant conflicts of interest.
• Bittner reports receiving research support from Amgen, AstraZeneca, Bayer, Janssen, Pfizer, and Sanofi and serving on advisory panels for Amgen and Eli Lilly.

Related Stories:

• Degree of LDL-Cholesterol Lowering on High-Intensity Rosuvastatin Determines Clinical Impact
• Knowing Genetic Risk for MI Motivates Patients, Physicians to Get Cholesterol Levels Lower
• Aggressive LDL Cholesterol-Lowering Improves Post-PCI Outcomes