Pathways Forward in Cardiovascular Disease Prevention One and a Half Years After the Publication of the 2013 ACC/AHA Prevention Guidelines

Abstract

The 2013 American College of Cardiology / American Heart Association (ACC/AHA) prevention guidelines represent an important step forward in both the risk assessment and management of atherosclerotic cardiovascular disease in clinical practice. Differentiated risk prediction equations for women and African Americans were developed, and convenient 10-year and lifetime risk assessment tools were provided, facilitating their implementation. Lifestyle modification was portrayed as the foundation of preventive therapy. In addition, based on high-quality evidence from randomized controlled trials, statins were prioritized as the first lipid-lowering pharmacologic treatment, and a shared decision-making model between the physician and the patient was emphasized as a key feature of personalized care. After their publication, however, important limitations of the guidelines were also identified. This resulted in a constructive scientific debate yielding valuable insights into potential opportunities to refine recommendations, fill gaps in guidance, and better harmonize recommendations within and outside the US. The latter point deserves emphasis because when guidelines are in disagreement, this may result in non-action on the part of professional caregivers or non-adherence by patients. In this review, we discuss the key scientific literature relevant to the guidelines published in the year and a half after their release. We aim to provide cohesive, evidence-based views that may offer pathways forward in cardiovascular prevention towards greater consensus and benefit the practice of clinical medicine.

Introduction

In November 2013, the American College of Cardiology and the American Heart Association (ACC/AHA) released updated cardiovascular disease (CVD) prevention guidelines.^1–4 This collaborative set of guidelines included recommendations for atherosclerotic CVD (ASCVD) risk assessment in asymptomatic adults,¹ treatment of blood cholesterol,² lifestyle management,³ and management of overweight and obesity,⁴ to reduce ASCVD risk.

The release of the guidelines was followed by the publication of many original research articles, editorial comments, reviews, and consensus statements by a number of research groups.^5–9 Strengths of the guidelines were affirmed while some limitations were also identified. Overall, the great interest among the scientific community resulted in constructive discussion and valuable insights. These insights offer opportunities to improve the implementation of the current version of the guidelines, and to consider updates for the future.

In this review, we discuss the key publications relevant to the guidelines released in the last one and a half years, including randomized controlled trials, expert consensus documents^8,9 and guidance from other scientific societies.^10–12 Moreover, building on this evidence we explore potential pathways forward, aiming to provide viewpoints that may fill gaps in the 2013 ACC/AHA prevention guidelines and meaningfully inform future updates. We consider the possibility of better harmonized recommendations and the benefits they may offer.

Approach to global risk assessment: old paradigms, modern problems

Implementing the 2013 ACC/AHA risk assessment guideline,¹ clinicians and patients have had the benefit of using the ACC/AHA Risk Estimator application. It is accessible via smartphone and web interfaces, and it allows one to input risk factors and quickly obtain the patient’s risk estimates. The estimation equations have been integrated with electronic medical records in some health systems, thereby providing automatic risk estimation. This is a major improvement given concern with past guidelines that clinicians did not have sufficient time to calculate risk estimates.

However, the current estimator continues to rely on the traditional risk factor paradigm as the central approach to CVD risk assessment. In addition to the Framingham cohorts, three other prospective cohorts are included in the 2013 Pooled Cohort Equations used by the 10-year ASCVD risk estimator. Nevertheless, the risk factors included remain the same as those from the original Framingham Risk Score. Moreover, single-time measurements are still used, although it is a recognized limitation that this misses cumulative exposure contained in the time since diagnosis and pattern of exposure to each risk factor.¹³ Furthermore, using binomial variables like current smoking (yes/no) likely underestimates the risk in some individuals.

In addition to the limited risk discrimination that the traditional risk factor model had already shown with previous versions of the guidelines,^14,15 immediately after its release the calibration of the 2013 risk estimator came into question, with evidence of risk overestimation of up to 150% when implemented in more modern cohorts⁵ (Table 1). The issue of risk overestimation had already been observed in more contemporary years of the pooled cohorts as well as in the validation study included within the guidelines using the Reasons for Geographic and Racial Differences in Stroke (REGARDS) and the Multi-Ethnic Study of Atherosclerosis (MESA) cohorts.²²

Table 1.

Selection of notable studies on ASCVD risk assessment published after the release of the 2013 ACC/AHA Prevention Guidelines.

Study	Design	Key findings	Implications for future guidelines
Ridker et al5	Compared ASCVD risk predicted by the ACC/AHA 2013 Pooled Cohort Equations, and observed event rates in the WHS, PHS, and the WHI-OS.	The 2013 Pooled Cohort Equations systematically overestimated ASCVD risk by 75–150%.	Risk estimators should be built on the most recent, multi-ethnic cohorts available and undergo extensive external validation.
Pencina et al16	Estimated the number of US adults eligible for statin therapy according to the ACC/AHA 2013 treatment of blood cholesterol guideline using NHANES data from 2005–2010, and compared with the numbers following the ATP III guidelines.	The number of US adults eligible for statin therapy increased from 43 to 56 million. Statin recommendation would increase in adults expected to have events as well as those unlikely to have them.	Sensitivity and specificity should be carefully balanced when establishing risk thresholds for therapy allocation, particularly in elderly adults. Uncertainty of risk predictions should be discussed in a physician-patient discussion.
Muntner et al17	18,498 REGARDS participants aged 45–79, a subset of 10,997 without DM, not taking statins and with LDL <190mg/dL, and a subset of 3,333 Medicare beneficiaries were followed for 5 years. Predicted risk using the Pooled Cohort Equations and observed events were compared.	There was moderate overestimation of risk by the Pooled Cohort Equations in the overall cohort. The performance improved in participants without DM, not taking statins and with LDL <190mg/dL, as well as when ASCVD events identified in Medicare claims data were added.	Clear guidance should be provided regarding the target populations in which the risk estimators are intended to be used.
Cook et al18	27,542 women aged 45–79 from the WHS followed for a median of 10 years. Predicted risk and observed event rates were compared after adjusting for effects of statins / revascularization and indication bias.	Differences in statin use and revascularization procedures did not explain the discrepancy between risk predictions and observed event rates. Event underascertainment unlikely to be the cause of risk overestimation.	Risk estimators should be built on the most recent cohorts available, approximating the secular patterns of risk factors and overall health of the populations in which they are intended to be used.
DeFilippis et al19	4,227 MESA participants aged 50–74 years without DM at baseline. Expected event rates using the 2013 Pooled Cohort Equations and 4 older FRS-based risk scores were compared with observed rates after 10 years of follow up.	Of the risk scores, 4 out of 5 including ACC/AHA 2013 Pooled Cohort Equations overestimated risk by 37–154% in men and 8–67% in women. Use of pharmacotherapies and interim revascularization did not explain the risk overestimation.	Currently available risk prediction models have significant limitations when used in modern, multiethnic cohorts. Future guidelines should combine improved risk estimators with the information provided by other risk assessment tools.
Silverman et al20	6,698 MESA participants. CHD event rates after mean follow-up of 7.1 years were compared among traditional risk factor categories (0, 1, 2, or ≥3) and FRS categories after stratification by CAC score.	Individuals with 0 traditional risk factors and CAC >300 had an event rate 3.5 times higher than individuals with ≥3 risk factors and CAC=0. Similar results were seen across categories of FRS.	CAC may have the potential to further risk stratify asymptomatic adults at the extremes of traditional risk factor burden, identifying those more likely to benefit from preventive interventions.
Martin et al21	5,534 MESA participants not on lipid-lowering therapy. CVD event rates after mean follow-up of 7.6 years were compared among categories of lipid abnormalities after stratification by CAC score.	55% of the total CVD events occurred in the 21% of participants with CAC >100. Participants with CAC=0 and 3 lipid abnormalities had lower event rates than those with CAC >100 and no abnormalities.	CAC may have the potential to match lipid-lowering therapy to ASCVD risk across the spectrum of dyslipidemia.

^aAbbreviations: WHS = Women’s Health Study; PHS = Physicians’ Health Study; WHI-OS = Women’s Health Initiative Observational Study; NHANES = National Health and Nutrition Examination Survey; ATP = Adult Treatment Panel; REGARDS = Reasons for Geographic and Racial Differences in Stroke; FRS = Framingham Risk Score; NNT = number needed to treat.

A subsequent, more comprehensive study of the 10-year calibration and discrimination performance of the risk estimator in MESA revealed persistent concerns about overestimation without evidence of substantially improved discrimination compared to prior risk scores.¹⁹ Differences in socioeconomic status and secular trends of CVD risk factors not captured in risk scores, along with changing incidence patterns of CVD and overall health between the cohorts used for developing the Pooled Cohort Equations and modern populations, may explain this phenomenon.⁶ Further research in this area will enhance our understanding regarding the reasons for such overestimation.

Following recommendations of the 2013 ACC/AHA cholesterol guideline,² clinicians and patients decide upon lipid lowering therapy according to absolute risk estimates. In this context, risk overestimation together with the reduction of the risk threshold for considering statin therapy means expanded consideration of statin therapy in the general population.^16,23 Even though recent studies have found that this may result in the accurate assignment of a greater proportion of asymptomatic patients with subclinical atherosclerotic disease or significant coronary stenoses to statin therapy,^24,25 there has been concern about it leading to the unnecessary treatment of a larger number of individuals unlikely to have events.¹⁶ Importantly, the risk estimator does not lead to automatic statin prescription, but rather the results must be considered in the context of a clinician-patient risk discussion.

Such discussion needs to be informed by the best possible data. Based on what we have learned over the past 18 months, it is a priority for future iterations of the guidelines to develop risk estimators incorporating the most recent cohorts available in order to mimic as much as possible the population in which they are intended to be used. Moreover, risk estimators could be updated and recalibrated on a regular basis, incorporating the insights from the latest, highest-quality demographic and epidemiologic evidence available. Nevertheless, the fundamental limitations of the traditional risk factor-based model²⁶ underscore the need to explore alternative risk assessment approaches. Thus, the performance of non-linear mathematical models departing from standard linear modeling should be assessed. Specifically, in the electronic medical record era, models including information on cumulative exposure to each risk factor may soon be feasible, and will likely enhance risk assessment.

Risk prediction: one size does not fit all

Over the past year and a half, clinicians and patients have been able to use separate risk prediction equations for women and African Americans. This is an important improvement as it accounts for effect modification by sex and race in the relationship between cardiovascular risk factors and the development of CVD. On the other hand, in clinical practice we are still applying, with caution, non-hispanic-whites’ equations in other non-white, non-African American racial/ethnic groups. There are currently more than 53 million Latinos living in the US, and Asians are the second fastest growing group.^27,28 Studies have shown coronary heart disease (CHD) event and mortality rates among Latinos and Chinese-Americans living in the US being lower than or similar to those of US non-hispanic whites,²⁹ whereas individuals of South Asian ancestry are known for their relatively high rates of ASCVD at early ages.³⁰ Thus, equations from non-hispanic-whites may over- or underestimate risk in other groups.

It has been unclear how to best handle such issues in clinical practice. Future guidelines can make it a priority to further account for the multi-ethnic nature of the US population and provide more specific guidance. In particular, once sufficient evidence from prospective studies is available, risk estimators can incorporate specific equations for each group.

Addition of stroke as part of the ASCVD outcome

The 2013 ACC/AHA guidelines are about ASCVD risk reduction and therefore clinical practice this past year and a half has been focused on preventing strokes in addition to CHD. Clinicians are now more easily able to acknowledge the presence of shared risk factors and common upstream pathways between CHD and some types of stroke, as well as for the need for joint preventive efforts. Moreover, using a broader clinical endpoint is an area of agreement among guidelines, though there is still room to harmonize the specific components of the endpoint.

Nevertheless, there has been some uneasiness about expanding the endpoint to include because not all strokes may be atherosclerotic in nature. Accordingly, the use of an ASCVD risk estimate that includes risk of any stroke may result in the treatment of individuals at high risk of non-atherosclerotic strokes, such as patients with atrial fibrillation, which may not benefit from the same cholesterol lowering interventions. In addition, incorporating stroke to the risk estimation has yielded an even greater sensitivity to chronological age compared to previous CHD risk scores. Consequently, many middle-aged men and almost every elderly individual with seemingly normal lipid profiles are exceeding the 7.5% risk threshold for considering statin therapy primarily on the basis of their chronological age.

Based on these observations, future iterations of the guidelines could consider separate CHD and ASCVD risk estimates, as valuable, complementary sources of information to be integrated by the clinician and patient in the context of a thoughtful discussion. In addition, guidance on the discussion and accurate management of other concurrent risk factors that may be particularly relevant in an individual’s ASCVD risk, such as hypertension or tobacco use, could be expanded.³¹

Key role of advanced risk assessment tools

Since November 2013, clinicians have had the opportunity to further adapt their individual practices for the use of advanced risk assessment tools. We are learning within the scope of each of our practices, as we have risk discussions with our patients, when treatment decisions are uncertain. When uncertainty arises, coronary artery calcium (CAC) is generally the most accurate tool for CVD risk assessment.^1,32–37 In the NHLBI-funded, multi-ethnic, population-based MESA study, CAC showed evidence of providing additional risk information over and above chronological age,³⁸ traditional risk factor burden²⁰ and lipid abnormalities.²¹ Of note, a CAC score >100 has been associated with high event rates in multi-ethnic cohorts,²¹ whereas a CAC score of zero (CAC=0) is associated with very low event rates.³⁹ Moreover, now down to a cost of <$100 at many imaging centers, CAC may be a valuable, helpful tool for the cost-effective allocation of preventive therapies, particularly in intermediate risk patients.^40–42

On the other hand, prior studies on CAC have been limited by short to intermediate follow-up periods. However, recent retrospective cohort analyses suggest that the prognostic value of CAC extends to 10-years of follow-up, at least for all-cause mortality.⁴³ Finally, CAC is associated with radiation exposure (1 mSv), and carotid ultrasound for plaque and IMT is an alternative to avoid radiation, particularly in young adults.

In view of these findings, future guidelines can consider an approach to risk assessment that combines the traditional risk factor-based paradigm with a more personalized atherosclerosis-imaging model, actively incorporating the knowledge of the patient’s actual burden of atherosclerosis to the risk assessment process. It is clear that there are serious limitations to restricting “uncertainty” to those patients with 5% to <7.5% ASCVD risk,⁴⁴ and the target group in which advanced risk assessment tools such as CAC would be considered could be expanded, including individuals with a 10-year ASCVD predicted risk of 5 – 10% or 5 – 15%.^13,45 This may include intermediate and low-intermediate risk individuals with uncommon risk factors.¹¹ In these scenarios, early cardiovascular risk assessment should be considered,¹² and clear guidance regarding the use of these tools in the context of a clinician-patient discussion should be provided.

Thus, the finding of a CAC score >100 may trigger the discussion for starting statin therapy. CAC may also be used for motivating statin-reluctant patients, aid decision-making in patients at risk of drug-drug interactions, and for the allocation of non-statin therapies.^41,42 Finally, CAC testing may also be particularly valuable in middle-aged and elderly adults with optimal levels of risk factors in whom the high predicted risk is directly linked to their chronological age. In such patients, a CAC=0 could be used as a rationale to emphasize lifestyle interventions and defer statin therapy. In MESA (mean age 62 years) 50% of participants had CAC=0 at baseline.³²

High sensitivity C-reactive protein and the ankle-brachial index were also recommended as tools for further risk assessment by the 2013 ACC/AHA risk assessment guideline.¹ However, recent studies have shown that the performance of high sensitivity C-reactive protein as a tool for further risk assessment is modest, with multi-factorial variability being one of its key limitations.^46,47 Regarding the ankle-brachial index, currently available evidence suggests that its addition to traditional risk factors has limited value for predicting CVD, particularly in non-elderly adults.⁴⁸ Finally, even though the 2013 guidelines recommended against carotid intima-media thickness assessment,¹ the combined assessment of carotid intima-media thickness and carotid plaque⁴⁹ was not addressed by the guideline.

Other tools for risk assessment and motivation for lifestyle change in young adults

In addition to a 10-year ASCVD risk estimate, the 2013 ACC/AHA risk estimator provides clinicians with a lifetime risk estimation in adults 20–59 years of age. This initiative provides an appealing tool for further informing ASCVD risk and motivating health behaviors. Cumulative exposure is a critical factor for the eventual development of atherosclerosis,⁵⁰ and preventive interventions may be more effective when implemented at early ages. Specifically, lifestyle changes in young adults with low predicted 10-year risk (due to their young age) but prevalent cardiovascular risk factors may result in large long-term health benefits.⁵¹ In such patients, lifetime risk estimation may be particularly helpful forestalling false reassurance when the predicted 10-year risk is low or the CAC score is zero.

Nevertheless, the 2013 ACC/AHA lifetime risk tool has limitations that must be noted. First, lifetime risk predictions are built on a very limited set of traditional risk factors categorized as binary predictors. Second, despite being intended to be used mainly in young adults, the cohort used for developing the estimator was middle-aged. Third, conclusive evidence is lacking regarding its performance in clinical practice, particularly motivating patients for lifestyle change or improving adherence to recommended pharmacotherapies. Finally, communicating lifetime risk can be as least as difficult as communicating 10-year risk.³¹ Further research is warranted in order to better understand the limitations of currently available tools, as well as to aid the development of comprehensive, accurate, externally validated lifetime risk estimators that could be incorporated in future iterates of the guidelines.

Other risk assessment/motivational tools such as relative risk and the “risk age” are currently recommended by other scientific societies.^11,12 Once more data is available regarding the performance of these tools, they could also be considered for their incorporation in future ACC/AHA guidelines. We encourage patient-oriented research to understand how patients understand and respond to different forms of risk communication.

Statin therapy and other lipid-lowering treatments

A healthy lifestyle was portrayed in the 2013 ACC/AHA prevention guidelines as the foundation of cardiovascular prevention.^2–4 Indeed, the Work Group of the Treatment of Blood Cholesterol Guideline noted that a healthy lifestyle was the background intervention in most of the studies considered.² The central role of lifestyle in CVD prevention is one of the key agreements between the ACC/AHA and other scientific societies.^10–12 Accordingly, healthy lifestyles and the best tools for their implementation in each individual patient should be thoroughly addressed in a clinician-patient discussion.

Additionally, future iterations of the guidelines can consider providing detailed guidance regarding the potential role of lifestyle as a first step therapy before pharmacologic interventions, as well as on the timing and methods for risk reassessment after lifestyle change. This may be particularly relevant when patients are motivated and their risk can be reasonably decreased to a level where no pharmacologic intervention would be needed.

Regarding pharmacologic treatments, accounting for the extensive, high-quality evidence supporting their cardiovascular benefits, statins were prioritized by the 2013 ACC/AHA cholesterol guidelines as the first lipid-lowering pharmacologic treatment option. The Work Group of the guideline identified four “statin benefit groups”, and absolute risk was emphasized as the key criterion for therapy intensity and allocation.² This has likely led to significant changes in some practices where statins were not being prioritized among drug therapy options.

Meanwhile, there has been considerable confusion about the continued role of non-statin lipid-lowering agents in clinical practice. The 2013 ACC/AHA guidelines are often interpreted as mainly suggesting against use of non-statins. They can be considered in statin intolerant individuals, as well as in patients on maximally-tolerated statin therapy with a less than anticipated response, although assessment of adherence to statin therapy and lifestyle changes is also to be considered in such a case.

Nevertheless, there are other clinical scenarios in which clinicians and patients may reasonably use non-statin therapy.⁵² Regarding adding non-statin therapy to background monotherapy, a landmark trial provided results after the release of the guidelines. In the Improved Reduction of Outcomes: Vytorin Efficacy International Trial (IMPROVE-IT), 18,144 secondary prevention patients were randomly assigned to simvastatin plus ezetimibe or simvastatin plus placebo. After a mean follow-up of six years, the mean low-density lipoprotein cholesterol (LDL-C) levels in the ezetimibe group were 54 mg/dL, compared to 70 mg/dL in the placebo group, and there was a statistically significant 6% relative risk reduction in major adverse cardiovascular and cerebrovascular events in the ezetimibe group. Furthermore, the number needed to treat (NNT) was 50 over 6 years, overall on top of comprehensive secondary prevention therapy.

In the on-treatment analysis, the NNT was 38.⁵³ Considering these results, the use of ezetimibe would broadly fit into the treatment of blood cholesterol guideline framework for the use of proven and safe drugs shown in randomized controlled trials to reduce ASCVD. Nevertheless, therapeutic decisions must be individualized, and the pros, cons, cost-effectiveness, and benefit/harm balance of each alternative should be thoroughly addressed in a clinician-patient discussion.

The IMPROVE-IT trial yielded other relevant insights. First, the relative risk reduction in the simvastatin plus ezetimibe group per mmol/L lowering of LDL-C was nearly the same as expected from statin monotherapy trials, suggesting that the “pleiotropic” effects of statins may be mediated by their lipid lowering action. Second, the safety profile of the combination simvastatin-ezetimibe was excellent, providing a valuable example of a safe treatment strategy beyond statins alone offering net clinical benefit without being outweighed by adverse effects.

Use of fixed cholesterol treatment goals

Based on randomized controlled trial data, the 2013 ACC/AHA cholesterol guideline panel found “insufficient evidence for or against” the use of LDL-C treatment goals as means for guiding therapy, as well as for up-titrating statin treatment to further reduce cholesterol levels.² Thus, unlike Adult Treatment Panel III and the European Society of Cardiology¹¹ and Canadian Cardiovascular Society¹² guidelines, the new ACC/AHA recommendations do not target fixed LDL-C or non–high-density lipoprotein cholesterol (non-HDL-C) goals. Rather, they recommend lipid measurement at baseline, 1 to 3 months after statin initiation, and yearly thereafter to check for the expected percentage decrease of LDL-C levels.

However, there are several potential issues with the approach. First, it is in discord with other guidelines around the globe, which has led to confusion. Second, clinicians commonly are not able to determine the percent reduction in LDL-C that a patient has achieved. Third, patients who begin treatment with a high LDL-C may not attain a clinically optimal LDL-C level with statin monotherapy even if the expected percent reduction is achieved. Fourth, the emphasis in the guideline on removing LDL-C goals led to the misperception among some clinicians over the past year that lipid levels no longer need to be followed up.

In contrast, other organizations such as the National Lipid Association⁵⁴ have published approaches that continue to emphasize fixed LDL-C treatment goals to aid in assessing the adequacy of treatment. Nevertheless, risk- and lipid-based paradigms are not mutually exclusive and can be complementary. At baseline, obtaining the most accurate assessment of risk is crucial in deciding who is most likely to have an event in the future, whereas in follow-up, lipid measurements can serve as a marker of therapeutic response, promote adherence, motivate lifestyle improvements, and guide discussions about add-on pharmacologic therapy for patients who are clearly established as high-risk.

The IMPROVE-IT trial, combined with a re-evaluation of all current best evidence, may enable the next ACC/AHA guideline panel to re-assess the potential role of lipid goals using proven therapies. IMPROVE-IT further reinforces the central role of LDL-C and non-HDL-C lowering in ASCVD risk management and shows that the effect is not specific to only statins. Moreover, on the basis of the totality of current best evidence, most expert groups in the US and throughout the world recommend fixed on-treatment lipid goals.^7,10–12,54

The 2013 ACC/AHA guidelines recommend a focus on the percentage reduction and allow assessment for LDL-C <100 mg/dL when the percentage reduction cannot be determined. So, it would not be a major change to bring back guidance on on-treatment lipids, at least for specific patients in whom fixed goals can guide a more aggressive management of high absolute risk. This may be particularly valuable in those patients with LDL-C levels remain high after a 30–50% relative reduction, as well as in those with a high burden of subclinical disease.²¹

A key issue often overlooked when guidelines are in disagreement, as is currently present, is the resultant confusion which often contributes to either non-action on the part of the professional caregiver or non-adherence by the patients. Ultimately, guidelines clearly agree on the central role of LDL-C and atherogenic lipid lowering in therapy, the need to assess treatment adherence, and the need to assess the adequacy of the existing treatment plan.^2,11,12 Current recommendations may give the impression of less fundamental agreement than truly exits and harmonizing recommendations will likely benefit clinical care.

The clinician-patient risk discussion and the shared decision-making model

A shared decision-making model between the clinician and the patient was emphasized in the 2013 ACC/AHA prevention guidelines, particularly in the treatment of blood cholesterol document.² Bringing more attention to shared decision making is one of the key achievements of the guidelines, and it is in agreement with the recommendations from other scientific societies^11,12 and expert groups.^8–10 Guideline recommendations provide a general framework for decision making, but they should not replace clinical judgment nor disregard patients’ preferences. Accordingly, any algorithm should be critically evaluated in the context of the individual being assessed. Indeed, as both the target of the intervention and the main healthcare provider on a daily basis, the patient is the center of the prevention process. Encouraging greater patient–provider dialogue may result in improved adherence to consensual lifestyle and/or pharmacologic interventions¹¹ and accordingly, yield better health outcomes.⁵⁵

A clinician-patient discussion should not only guide treatment choices, but be fully incorporated into the ongoing ASCVD risk assessment-management process. Accordingly, given the relevance of such discussion, practical guidance regarding the key elements may be useful to clinical practice. Considerations include patient preferences, precision of the risk estimate and opportunities for a more personalized assessment, willingness of the patient to participate in their ongoing care and change lifestyle habits, treatment and dose being proposed, potential benefits and harms of each treatment option, and price.⁴⁵ Limited data are available regarding the effectiveness of different risk communication strategies, how to handle discussion of controversies in primary prevention, and more broadly on the effect of different risk discussion approaches on downstream clinical outcomes. Tools such as the Mayo Clinic Statin Decision aid are already available online, and in the next years high-quality patient-oriented studies are critically needed to help determine the optimal methods for structuring clinician-patient risk discussions.

Conclusion

It has been an exciting and productive time in the year and a half since the landmark 2013 ACC/AHA prevention guidelines were released. Continuous scientific discussion in the following years together with standardized evaluation of their clinical effectiveness will further improve future iterations of the guidelines and likely result in more unified recommendations across various guideline committees throughout the world.

CVD risk assessment tools should be built on the latest, highest-quality evidence in order to attain the most accurate absolute risk assessment. In addition, atherosclerosis imaging could be more strongly emphasized in future guidelines as a means to refine risk assessment built on the knowledge of the patient’s burden of atherosclerosis. Among such tools, CAC appears as the best test for refinement of risk assessment. Regarding risk management, with data continuing to support the notion that lower LDL-C is better, cholesterol treatment goals and non-statin therapies shown to prevent ASCVD events should be reincorporated in future versions of the guidelines, with the critical notion that lowering should occur through safe, proven strategies. Finally, a patient-physician discussion should address all these features in order to fully incorporate patient’s expectations and preferences to the risk assessment-management process.

Highlights.

1. The 2013 ACC/AHA prevention guidelines represent an important step forward in the risk assessment and management of ASCVD.

2. The fundamental limitations of the traditional risk factor-based model underscore the need to explore alternative risk assessment approaches.

3. Atherosclerosis-imaging tools such as coronary artery calcium could be more strongly emphasized in future guidelines as powerful tools for refining risk assessment.

4. Cholesterol treatment goals and non-statin therapies shown to prevent ASCVD events should be reincorporated in future versions of the guidelines, with the critical notion that lowering should occur through safe, proven strategies.

5. As highlighted by the 2013 ACC/AHA guidelines, a clinician-patient risk discussion is a critical element in the ongoing ASCVD risk assessment-management process and further research in this area is warranted.

6. Continuous scientific discussion together with standardized evaluation of clinical effectiveness will improve the development and implementation of future iterates of the guidelines.

Abbreviations and Acronyms

ACC/AHA

American College of Cardiology / American Heart Association

ASCVD

Atherosclerotic Cardiovascular Disease

CAC

Coronary Artery Calcium

CHD

Coronary Heart Disease

CVD

Cardiovascular Disease

IMPROVE-IT

Improved Reduction of Outcomes: Vytorin Efficacy International Trial

MESA

Multi-Ethnic Study of Atherosclerosis