World Heart Federation Cholesterol Roadmap 2022

Prof Kausik K. Ray, MBChB, MD, MPhil, FRCP, 1 Brian A. Ference, MD, MPhil, MSc, 2, 3 Tania Séverin, MPH, 4 Dirk Blom, PhD, 5 Stephen J. Nicholls, MBBS PhD, 6 Mariko H. Shiba, M.D., Ph.D, 7 Wael Almahmeed, MD FRCPC, 8 Rodrigo Alonso, MD, PhD, FNLA, 9 Magdalena Daccord, MSc, 10 Marat Ezhov, MD, 11 Rosa Fernández Olmo, MD, 12 Piotr Jankowski, MD, PhD, 13 Fernando Lanas, MD, PhD, 14 Roopa Mehta, PhD, FRCP, 15 Raman Puri, MD, DM, 16 Nathan D. Wong, PhD, 17 David Wood, MB ChB MSc, 18 Dong Zhao, MD, PhD, 19 Samuel S. Gidding, MD, 20 Salim S. Virani, MD. PhD, 21 Donald Lloyd-Jones, MD ScM FACC FAHA, 22 Fausto Pinto, MD, PhD, 23 Pablo Perel, MD, PhD, 24 and Raul D. Santos, MD, PhD, MSc 25

Prof Kausik K. Ray

1 Imperial Centre for Cardiovascular Disease Prevention, Department of Primary Care and Public Health, Imperial College London, Reynolds Building, St. Dunstans Road, London, GB

Brian A. Ference

2 Department of Public Health and Primary Care, British Heart Foundation, UK

3 Cardiovascular Epidemiology Unit, Centre for Naturally Randomized Trials, University of Cambrige, Cambridge, GB

Tania Séverin

4 World Heart Federation, Geneva, CH

Dirk Blom

5 Department of Medicine, University of Cape Town, Cape Town, ZA

Stephen J. Nicholls

6 Victorian Heart Institute, Monash University, Melbourne, AU

Mariko H. Shiba

7 Cardiovascular Center, Osaka Medical and Pharmaceutical University, Takatsuki, JP

Wael Almahmeed

8 Cleveland Clinic Abu Dhabi, Abu Dhabi, AE

Rodrigo Alonso

9 Center for advanced metabolic medicine and nutrition, Santiago, CL

Magdalena Daccord

10 Europe, Rochester, Kent, GB

Marat Ezhov

11 Chazov National Medical Research Center of Cardiology, Moscow, RW

Rosa Fernández Olmo

12 Cardiac Rehabilitation Unit Jaen University Hospital, Jean, ES

Piotr Jankowski

13 Department of Internal Medicine and Geriatric Cardiology and Department of Epidemiology and Health Promotion, School of Public Health, Centre of Postgraduate Medical Education, Warsaw, PL

Fernando Lanas

14 Universidad de La frontera, Temuco, CL

Roopa Mehta

15 Instituto Nacional de Ciencias Medicas y Nutricion, Salvador Zubirán, Mexico City, MX

Raman Puri

16 Department of Cardiology, Apollo Hospital, New Delhi, IN

Nathan D. Wong

17 University of California, Irvine, US

David Wood

18 Health, National University of Ireland Galway, Galway, IE

Dong Zhao

19 Beijing Institute of Heart, Lung & Blood Vessel Diseases, Capital Medical University Beijing Anzhen Hospital, Beijing, CN

Samuel S. Gidding

20 Geisinger Genomic Medicine Institute, Danville, PS, US

Salim S. Virani

21 Baylor College of Medicine/Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX, US

Donald Lloyd-Jones

22 Preventive medicine, Northwestern University, Chicago, US

Fausto Pinto

23 Lisbon School of Medicine, University of Lisbon, Lisbon, PT

Pablo Perel

24 London School of Hygiene & Tropical Medicine and World Heart Federation, London, UK and Geneva, CH

Raul D. Santos

25 Cardiopneumology Department and Lipid Clinic, Heart Institute (InCor) University of Sao Paulo Medical School Hospital and Hospital Israelita Albert Einstein, Sao Paulo, BR

1 Imperial Centre for Cardiovascular Disease Prevention, Department of Primary Care and Public Health, Imperial College London, Reynolds Building, St. Dunstans Road, London, GB

3 Cardiovascular Epidemiology Unit, Centre for Naturally Randomized Trials, University of Cambrige, Cambridge, GB

13 Department of Internal Medicine and Geriatric Cardiology and Department of Epidemiology and Health Promotion, School of Public Health, Centre of Postgraduate Medical Education, Warsaw, PL

19 Beijing Institute of Heart, Lung & Blood Vessel Diseases, Capital Medical University Beijing Anzhen Hospital, Beijing, CN

24 London School of Hygiene & Tropical Medicine and World Heart Federation, London, UK and Geneva, CH

25 Cardiopneumology Department and Lipid Clinic, Heart Institute (InCor) University of Sao Paulo Medical School Hospital and Hospital Israelita Albert Einstein, Sao Paulo, BR

CORRESPONDING AUTHOR: Prof Kausik K. Ray Imperial Centre for Cardiovascular Disease Prevention, Department of Primary Care and Public Health, Imperial College London, Reynolds Building, St. Dunstans Road, London, W6 8RP, United Kingdom ku.ca.lairepmi@yar.k

This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/.

Associated Data

Abstract

Background:

Atherosclerotic cardiovascular diseases (ASCVD) including myocardial infarction, stroke and peripheral arterial disease continue to be major causes of premature death, disability and healthcare expenditure globally. Preventing the accumulation of cholesterol-containing atherogenic lipoproteins in the vessel wall is central to any healthcare strategy to prevent ASCVD. Advances in current concepts about reducing cumulative exposure to apolipoprotein B (apo B) cholesterol-containing lipoproteins and the emergence of novel therapies provide new opportunities to better prevent ASCVD. The present update of the World Heart Federation Cholesterol Roadmap provides a conceptual framework for the development of national policies and health systems approaches, so that potential roadblocks to cholesterol management and thus ASCVD prevention can be overcome.

Methods:

Through a review of published guidelines and research papers since 2017, and consultation with a committee composed of experts in clinical management of dyslipidaemias and health systems research in low-and-middle income countries (LMICs), this Roadmap identifies (1) key principles to effective ASCVD prevention (2) gaps in implementation of these interventions (knowledge-practice gaps); (3) health system roadblocks to treatment of elevated cholesterol in LMICs; and (4) potential strategies for overcoming these.

Results:

Reducing the future burden of ASCVD will require diverse approaches throughout the life-course. These include: a greater focus on primordial prevention; availability of affordable cholesterol testing; availability of universal cholesterol screening for inherited dyslipidaemias; risk stratification moving beyond 10-year risk to look at lifetime risk with adequate risk estimators; wider availability of affordable cholesterol-lowering therapies which should include statins as essential medications globally; use of adequate doses of potent statin regimens; and combination therapies with ezetimibe or other therapies in order to attain and maintain robust reductions in LDL-C in those at highest risk. Continuing efforts are needed on health literacy for both the public and healthcare providers, utilising multi-disciplinary teams in healthcare and applications that quantify both ASCVD risk and benefits of treatment as well as increased adherence to therapies.

Conclusions:

The adverse effects of LDL-cholesterol and apo B containing lipoprotein exposure are cumulative and result in ASCVD. These are preventable by implementation of different strategies, aimed at efficiently tackling atherosclerosis at different stages throughout the human life-course. Preventive strategies should therefore be updated to implement health policy, lifestyle changes and when needed pharmacotherapies earlier with investment in, and a shift in focus towards, early preventive strategies that preserve cardiovascular health rather than treat the consequences of ASCVD.

Keywords: cholesterol, low-density lipoprotein cholesterol prevention, ASCVD, lipid lowering therapy, familial hypercholesterolaemia

Introduction

Atherosclerotic cardiovascular disease (ASCVD) remains the leading cause of death globally despite major advances in our understanding of the pathophysiology of atherosclerosis, its consequences and the development of new preventive therapies [1,2]. Of concern is the observation that ASCVD is increasing both in prevalence in low- and middle-income countries (LMICs), where the vast majority of the global population reside, but also presenting there a decade or more earlier than in higher income countries ( Figure 1 ). When ASCVD impacts an individual’s life-course at such an early age, it not only has severe consequences for the person affected, but invariably impacts economic growth in affected countries, adversely affecting employability, productivity, and income potential during crucial years of the lifespan. Furthermore, adverse health of an individual impacts their caregivers and immediate family.

Age-standardized mean non-HDL cholesterol (mmol/L –1 ) across the world from the NCD Risk Factor Collaboration. Reproduced from NCD Risk Factor Collaboration (NCD-RisC). Repositioning of the global epicentre of non-optimal cholesterol. Nature 582, 73–77 (2020). https://doi.org/10.1038/s41586-020-2338-1 licensed under a Creative Commons Attribution 4.0 International License, http://creativecommons.org/licenses/by/4.0/.

a, Age-standardized mean non-HDL cholesterol in women in 1980. b, Age-standardized mean non-HDL cholesterol in women in 2018. c, Age-standardized mean non-HDL cholesterol in men in 1980. d, Age-standardized mean non-HDL cholesterol in men in 2018.

Different strategies are required to break the cycle of inequalities between wealth and health, resulting in adverse health outcomes and vice versa, which has been further compounded by the global coronavirus pandemic. Those most at risk for ASCVD because of obesity, hypertension, diabetes, and dyslipidaemia were particularly vulnerable to the virus. There has also been a loss of connection for many patients with their healthcare providers, leading to reduced screening, delay in diagnosis and loss of control of these chronic risk factors [3]. This will have ripple effects in ASCVD disease incidence for years to come.

In 2012, the World Health Organization (WHO) aimed to reduce premature mortality from non-communicable diseases (NCD) by 25% by the year 2025. This was soon followed by the United Nations Sustainable Development Goals (2015), which aimed to reduce premature mortality from NCD by 30% by 2030. A decade on these goals and the agenda for health and well-being are far from being realised with only 14 countries on track [4].

WHF Roadmaps provide a framework for public health policy, which build upon compelling scientific evidence, identify current roadblocks to implementation of best practice through surveys of WHF member countries and ultimately offer potential solutions to be adapted across all regions of the world over the next decade to suit local needs. The first WHF Cholesterol Roadmap was published in 2017 [5]. Since then, there have been advances in science, treatment, and technology. The present Roadmap update builds on the original publication and reviews:

new evidence on the burden of disease, epidemiology, diagnosis, treatment, technologies and policies which can lead to improved outcomes for people living with inadequately controlled blood cholesterol and

lessons learnt from the original WHF Cholesterol Roadmap implementation in different countries and settings.

This update was developed through a review of published guidelines and research papers, in consultation with a committee, composed of experts in clinical management of cholesterol, cardiovascular risk, health systems research and patient advocacy and support.

This present document is set out into three parts. The first section addresses fundamental concepts about lipids and ASCVD which are essential to inform solutions at health policy/funding level (macro-system) as well as guiding day to day interactions at the physician-patient level (micro-systems). This includes the role of high blood cholesterol levels in the global burden of disease (population approach) as distinct from the role of cholesterol lowering to mitigate ASCVD risk in those at greatest risk (high-risk approach); which lipid parameters to measure; how to assess those at higher risk of ASCVD; special populations with common inherited cholesterol disorders; and finally, pharmacotherapy. The second section provides case studies and examples of successful implementation strategies globally. Finally, the third section builds on a survey of member countries to inform current roadblocks and provides potential solutions to be implemented.

Fundamental Concepts About Lipids and Ascvd – The 8 Pillars

There is irrefutable evidence for the causal role of LDL-cholesterol (LDL-C) in ASCVD. The writing committee recognises the impact of misinformation in social media, as well as conflicting data from dietary intervention trials, on the discussion about the role of blood cholesterol in ASCVD, which may serve as a barrier for better cardiovascular health and implementation of preventive strategies. In this regard several fundamental concepts have been established through multiple lines of scientific enquiry over many decades which are important to understand, and around which future implementation strategies can be built including improving physician and citizen health literacy and refuting misinformation. These “8 Pillars” are set out in Box 1.

Box 1 The Fundamentable Principles of the Role of Lipids and Lipoproteins and Ascvd Risk and its Prevention – 8 Pillars

Atherosclerosis results from the retention of apolipoprotein B (apo B) containing lipoproteins mostly in the form of low- density lipoproteins (LDL) in the vessel wall. LDL cholesterol (LDL-C) is not only causal but a cumulative risk factor, over the lifespan, for ASCVD.

Individuals have differential retention of apo B containing lipoproteins and hence differential vulnerability to the effects of LDL-C exposure. Therefore, LDL-C and apo B should not be considered in isolation without considering other factors.

Most cardiovascular events occur among individuals without extreme elevations in LDL-C, hence global risk should be considered.

Extreme elevations in LDL-C from birth with a monogenic basis (familial hypercholesterolaemia) are more common than previously recognised and are prevalent across all regions of the world, and their consequences are largely preventable through early screening and treatment.

ASCVD can be reduced through reductions in LDL-C through multiple different pathways, with benefit reliably quantifiable and proportional to both the absolute reduction in LDL-C and the duration of that reduction, hence treatments could be interchanged or combined as needed.

As the majority of the total atherogenic cholesterol content (non-HDL-C) consists of LDL-C and as the majority of apo B containing lipoproteins are LDL particles, LDL-C reductions will provide largely predictable parallel proportional reductions in non-HDL-C and apo B.

The increasing prevalence of cardio-metabolic diseases such as obesity and diabetes has resulted in an increase in other lipid disorders which increase ASCVD risk. These are characterised by elevated triglyceride rich apo B containing lipoproteins which are atherogenic. Therefore, pragmatically moving towards estimation of atherogenic particle number in the form of apo B measurements with a single measure, or if this is not feasible, total atherogenic cholesterol content (non-HDL-C) may improve risk assessment and measures of benefit irrespective of therapeutic modality.

The recognition that elevations in lipoprotein (a) are common (but poorly detected) and are an independent causal risk factor of ASCVD.

The relevance of lipids and lipoproteins to the global burden of CVD

Worldwide, CVD affected 523 million individuals in 2019 [1]. It claims over 18 million lives each year, approximately 85% of which are due to ASCVD [1]. The Global Burden of Disease (GBD) Collaboration has demonstrated that assessment of total cholesterol alone is uninformative to assess time trends as it may mask opposing trends in HDL-C and non-HDL-C which have opposite associations with the risk of CVD [6,7,8]. The rise in obesity and diabetes globally results in dyslipidaemia characterised by a fall in HDL-C and a rise in non-HDL-C and high triglyceride levels [8,9], hence health systems should pragmatically move towards using non-HDL-C or apo B as a unifying measure of atherogenic lipid risk.

LDL-C lowering has been shown to be beneficial in primary and secondary prevention with individuals at highest risk of atherothrombotic events such as myocardial infarction, stroke, revascularisation or cardiovascular death, deriving greater absolute risk reductions from LDL-C lowering ( Figure 2 ). The importance of the timing of treatment initiation and the duration of that treatment in the disease course of atherosclerosis should now inform health policies and how they are implemented (as discussed later). Additionally, the vast majority of patients globally have insufficient LDL-C lowering to minimize their individual risk of ASCVD. In part this is due to the underuse of effective doses of more potent statins as first-line therapies, low use of combination therapies and poor adherence to lipid lowering regimens resulting in insufficient reductions in cumulative cholesterol exposure.

Events avoided based on baseline absolute risk and absolute lowering of LDL-C (Duality of risk and LDL-C lowering as determinants of benefit from lipid lowering therapies). Adapted from CTT Lancet 2012 The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: Meta-analysis of individual data from 27 randomised trials [87].

LDL-C lowering is beneficial not only in those with high cholesterol levels but rather in those most at risk of the major clinical manifestations of atherosclerosis with no lower limit of benefit observed from LDL-C lowering [10]. This is supported by genetic studies, prospective cohort studies, randomised trials of cardiovascular outcomes and imaging studies assessing atherosclerosis progression/regression. Recent evidence demonstrates that it is the magnitude of LDL-C lowering rather than how it is achieved that drives benefit [11]. Thus LDL-C lowering (and by proxy non-HDL-C and apo B) should be considered in all individuals at higher risk [12] identified as such by the presence of i) a prevalent/prior manifestation of ASCVD; ii) the presence of a high risk condition such as diabetes, chronic kidney disease, or extreme elevations of BP which increase ASCVD risk even in the absence of accompanying lipid abnormalities; iii) extreme elevations of LDL-C which have a genetic basis like heterozygous familial hypercholesterolaemia (FH); iv) or those with high-CV risk due to combined effects of multiple risk factors; v) isolated elevations in atherogenic lipoproteins including triglyceride rich lipoproteins (commonly referred to as atherogenic dyslipidaemia) or elevations in lipoprotein(a); vi) those with elevated burden of subclinical atherosclerosis [13]. This approach prevents recurrent events in those with prevalent ASCVD and incident events among those with high-risk conditions or high global risk.

Measurement of blood lipids

Conventionally, lipids have been measured in blood samples obtained after eight hours of fasting, to minimise the effects of postprandial changes in TG which influence LDL-C calculations. However, using one of the newer formulas to estimate LDL-C can be achieved in non-fasting samples. Alternatively, total atherogenic cholesterol content (non-HDL-C) or total number of atherogenic particles (apo B) can be obtained from non-fasting samples offering a more comprehensive measure of atherogenic risk, especially when apo B and non-HDL-C are discordant with LDL-C such as in conditions associated with insulin resistance, such as diabetes, obesity and high TG. Where appropriate laboratory facilities exist, apo B should in time become the default measure of total atherogenic lipid burden. Pragmatically, we have multiple approaches to measuring atherogenic lipid risk (LDL-C, non-HDL-C or apo B), and any of these are better than none at all. Therefore, in resource poor environments the focus should be on making some form of cholesterol testing available as a starting point for eventually achieving best practice (eTable 2).

Finally, there is consistent data irrespective of ethnicity that higher levels of lipoprotein(a) are likely causally associated with a higher risk of ASCVD (typically > 125 nmol/L or 50 mg/dl which occurs commonly). As lipoprotein(a) cannot be estimated from other lipid measures it has to be measured directly, preferably using an isoform independent assay reporting values in terms of concentration (e.g., nmol/L) rather than mass (e.g., mg/dL). As levels do not change much over a lifetime, except in women where values further increase after menopause, or in acute inflammatory states, a single measurement once in a lifetime in adults is useful to reliably exclude significant elevations [18].

Overall approaches to prevention

Primordial prevention and population health

The first step in the public health approach to ASCVD prevention is to change the milieu that promotes risk factor development as early as possible [19]. While primary prevention is about treating risk factors to prevent ASCVD, primordial prevention is avoiding the development of risk factors in the first place. It begins with changes in lifestyle, environmental and social conditions to prevent risk factor development from birth. Many risk factors have their origins early in life since this is the time when lifestyles are formed. Whilst genetics may be considered the ‘loaded gun’, environment and lifestyle effectively ‘pull the trigger.’ Many of these changes require policy changes, affordable healthy life choices and improvements in health literacy. Population level changes at a time of maximal developmental plasticity offer the best hope of setting individuals on a more favourable trajectory long-term towards better cardiovascular health. Recommendations regarding this remain the same as in the 2017 Roadmap; public health interventions should incentivise a healthy lifestyle. These can simply be put as keep physically active, an important necessity as many jobs have a sedentary nature, avoid tobacco use, maintain ideal body weight for ethnicity, reduce salt consumption and where feasible adopt the traditional Mediterranean-type diet with substitution of monounsaturated and polyunsaturated for saturated fat [20,21]. In particular, plant-based foods should be encouraged and incentivised [22,23,24]. Specifically, these approaches should focus on children or young adults, that is, start as early as possible when it comes to primordial prevention.

Primary Prevention

Current approach: 10-year ASCVD risk assessment

In the absence of high-risk conditions such as genetic dyslipidaemias (which include familial hypercholesterolaemia), diabetes mellitus, chronic kidney disease, a global approach to assess/predict 10-year risk of fatal and non-fatal ASCVD events should be used as those at highest risk derive greater absolute risk reductions from lipid lowering therapies (with smaller numbers needed to treat). These approaches should be region specific where available data allow, re-calibrated and updated as appropriate incorporating additional relevant risk factors that may emerge and allowing for secular trends to provide more reliable estimations of absolute risk. Risk calculators which use fatal events only disenfranchise younger individuals and women and their use is discouraged in favour of calculators assessing both fatal and non-fatal events. The decision of who to screen for ASCVD risk, and which risk threshold to use to initiate pharmacological treatments if healthy diet and lifestyle changes do not sufficiently attenuate risk is a matter for national (or local) policy which considers available resources. In the presence of conditions such as previous manifestations of ASCVD, genetic dyslipidaemias which include familial hypercholesterolaemia, diabetes mellitus and chronic kidney disease, risk calculators are not necessary to initiate therapy [25,26]. Categorising individuals into different risk categories provides the rationale and helps tailor the intensification of any intervention.

Global ASCVD risk calculators could be in the form of charts or preferably available as a web-based tool or Apps on a smartphone to be easily accessible. These should not only be used to provide estimates of risk (whom to potentially treat) but could be used to help shared decision-making, and where feasible, provide visualisation not only of potential risk but also of the magnitude of any potential benefit obtained from reductions in atherogenic lipids (and other factors). Ideally these could be automatically integrated into electronic medical record systems where the required data elements are available. This could aid/improve patient understanding and acceptance plus adherence to a given treatment regimen. Examples of contemporary calculators are shown in eTable 3. All risk prediction tools have strengths and limitations and should be used in conjunction with clinician judgement. As shown in the recent SCORE2 risk calculator [25], baseline risk in part related to societal, environmental and potentially genetic differences varies widely across geographical regions. Hence risk calculators which have been closely calibrated to the population of interest should be used where feasible, to avoid over or underestimation of risk. This has been done for different regions with the new WHO risk calculator [27]. This inevitably means that for the same level of risk factors, predicted event rates will vary, hence the threshold for treatment may vary to avoid treating too many or too few people. That said, in many countries the rate of CVD in indigenous or first world communities is very high and risk scores don’t work so well – so the implications for lipid lowering are more important.

In general, formal 10-year risk assessment should not be used among individuals with high-risk conditions, such as FH (as discussed later), type 2 diabetes or chronic kidney disease. However, risk of ASCVD is a continuum even among individuals with high-risk conditions such as diabetes. It is now recognised that diabetes is not a ‘coronary heart disease risk equivalent’, but rather absolute risk varies considerably by presence or absence of microvascular disease/end organ-damage, duration of diabetes and presence or absence of additional risk factors, duration or high CAC score [28]. Diabetes is included as a variable in some risk equations but can also be pragmatically categorised in those with moderate, high or very-high risk, that is, the presence of diabetes alone places an individual as a minimum into the moderate risk category [26,29]. Currently, the severity of renal impairment may be used to categorise patients into high (eGFR 30-60) or very high-risk categories (eGFR < 30).

From a pragmatic perspective for day-to-day practice, primary prevention includes those without overt major clinical manifestations of ASCVD. Imaging, especially in the coronary arteries like coronary artery calcium scores or computed tomography coronary angiography, is being used increasingly during the course of routine clinical practice, although cost has limited their use in LMICs. Where evidence of atherosclerosis is found, most guidelines recommend using this as a risk enhancer in the setting of primary prevention, thus evidence of risk enhancing factors should result in more aggressive control of risk factors due to reclassifying these people into a higher risk category.

Future approach: Lifetime risk estimation

The main implication of the cumulative exposure hypothesis is that maintaining low levels of LDL-C throughout life reduces cumulative exposure to the number of lipoprotein particles that become trapped within the artery wall over time, thus slowing the rate at which atherosclerosis develops and delaying the age at which the underlying plaque burden is large enough to increase the risk of having an acute ASCVD event. However, to maximize the benefit of reducing plaque burden by reducing cumulative exposure to LDL-C, it is also necessary to reduce exposure to other causes of arterial wall injury that increase the risk of having an acute event at all levels of plaque burden. Therefore, the cumulative exposure hypothesis suggests that the most effective way to reduce the lifetime risk of ASCVD events is by reducing cumulative exposure to LDL-C to slow atherosclerotic plaque progression, while also reducing cumulative exposure to other causes of arterial wall injury including elevated blood pressure, diabetes, and tobacco smoking.

Very long-term prospective follow-up of the Framingham Heart Study demonstrates that persons who maintain low levels of LDL-C and blood pressure throughout their lives have a very low lifetime risk of atherosclerotic cardiovascular events [30]. The PDAY risk score based on the relationship of risk factors to measured atherosclerosis in 15–34 year-olds strongly predicts premature ASCVD [31]. In addition, Mendelian randomization studies demonstrate that persons who are naturally randomized to genetic variants associated with lower lifetime exposure to LDL-C have a substantially lower lifetime risk of ASCVD [32,33]. The same observation is true for BP. The latter studies agree closely with the 80% reduction in cardiovascular morbidity and mortality observed in Finland over the past 40 years [34]. Beginning in the 1970s, Finland initiated a coordinated public health programme to reduce cardiovascular morbidity and mortality by reducing exposure to saturated fats to lower plasma cholesterol levels, and by discouraging the practice of preserving meat with cured salts to lower blood pressure levels [34]. Over the following 40 years, this programme resulted in a reduction of 1.5 mmol/L (58 mg/dL) in the population mean level of total cholesterol, an 8.7 mmHg reduction in the population mean level of systolic blood pressure (SBP), and a corresponding 80% reduction in cardiovascular morbidity and mortality. Assuming that most of the total cholesterol reduction was due to reduction in LDL-C in response to lower consumption of saturated fats, the magnitude of this reduction in cardiovascular morbidity and mortality is nearly identical to what would have been predicted by long-term exposure to the same magnitude of lower LDL-C and SBP in Mendelian randomization studies, thus providing powerful real-world evidence to support the cumulative exposure hypothesis.

Although some guidelines have referenced the value of lifetime risk [25,35,36], current clinical practice does not focus on lifetime risk [37], or on reducing cumulative exposure to the causes of ASCVD. Instead, current clinical practice guidelines recommend both informing the decision to initiate LDL-C lowering therapies, and titrating the intensity of those therapies, based on a person’s estimated short-term risk of developing an acute cardiovascular event over the next 10 years (29). However, short-term risk estimating algorithms are mathematically dominated by age because acute atherosclerotic cardiovascular events tend to occur later in life after a person develops a substantial underlying atherosclerotic plaque burden. As a result, informing the decision to initiate LDL-C lowering therapy based on short-term 10-year risk has the practical effect of inviting us to wait until a person develops a large atherosclerotic burden before initiating therapy to lower LDL-C to slow the progression of atherosclerosis.

Indeed, current clinical practice guidelines explicitly focus on using pharmacologic therapy to lower LDL-C to prevent atherosclerotic cardiovascular events among persons with a high estimated 10-year risk of experiencing an acute ASCVD event; and on intensification of pharmacologic LDL-C lowering therapies to achieve even lower LDL goals among persons who have already experienced an ASCVD event. Achieving these more intense plasma LDL goals almost always require the use of two or more pharmacologic LDL-C lowering therapies. Recent advancements in embedding causal inference and causal effects into artificial intelligence and machine learning algorithms (‘causal AI’) permit an estimation of the effect of lower LDL-C or systolic blood pressure (SBP) in discrete time-units of exposure, conditional on previous cumulative exposure to account for the amount of plaque burden and arterial wall injury that has accumulated prior to the initiation of interventions to lower LDL-C or SBP. This method allows an estimate of the benefit of lowering LDL-C, SBP or both beginning at any age (and extending for any duration of time) on the risk of developing an acute atherosclerotic cardiovascular event. Studies using these causal AI methods suggest that there is a stepwise increase in the reduction in the risk of atherosclerotic cardiovascular events for each decade earlier that LDL-C lowering is initiated [32] ( Figure 3 ).

Benefit of reducing cumulative exposure to LDL on the lifetime risk of atherosclerotic cardiovascular disease. Panel A shows the effect of reducing LDL by 50% from a population median of 3.5 mmol/L (135 mg/dl) resulting in an absolute difference of 1.75 mmol/L (67.7 mg/dL) on the lifetime risk of experiencing a major atherosclerotic cardiovascular event (defined as fatal or non-fatal myocardial infarction, fatal or non-fatal ischemic stroke, or coronary revascularization) if LDL lowering is started at ages 30, 40, 50 or 60 years and continued up to age 80 years as compared to either no LDL reduction of lifelong exposure to the same magnitude of lower LDL. Panel B shows the effect on the lifetime risk of experiencing a major atherosclerotic cardiovascular event up to age 80 years from reducing LDL by 33% beginning at age 40 years, or by 50% beginning at age 55 years. Greater benefits are observed if LDL-C lowering is begun at an earlier age.

These studies suggest that modest sustained reductions in LDL-C beginning earlier in life are associated with a lower risk of ASCVD events at all ages as compared to more intense LDL-C lowering started later in life. Indeed, these studies suggest that ‘residual risk’ of experiencing an acute ASCVD event despite intense LDL-C lowering initiated later in life may be explained by the extent of the plaque burden that accumulates prior to the initiation of LDL-C lowering therapy. Although aggressive LDL-C lowering initiated later in life can slow (or perhaps even arrest) plaque progression, the plaque burden that accumulated prior to the initiation of LDL-C lowering still persists and may disrupt the subsequent formation of a thrombus overlying the disrupted plaque, obstructing blood flow and leading to an acute cardiovascular event. As a result, reducing the cumulative exposure to LDL-C as a strategy to slow the progression of atherosclerotic plaque may both substantially reduce the lifetime risk of ASCVD and substantially reduce the residual risk of acute atherosclerotic events if implemented globally.

Familial hypercholesterolemia

Familial hypercholesterolaemia is an autosomal co-dominant condition which results in life-long elevations in LDL-C from birth leading to premature cardiovascular disease (from age 30 in men and age 40 in women) and premature deaths. Heterozygous FH (HeFH), results when an individual inherits one affected gene from a parent. Several recent studies have emerged that should inform future public health strategies directed towards early case finding and initiating treatments early in the life course.

The global prevalence of HeFH is 1:311 and is present in all WHO regions at similar levels [38]. However, the frequency in communities with founder effects is higher, being as high as 1:70 [39]. Even though it affects every WHO region, there is little data from as many as 130 countries, meaning that HeFH may not even be recognised and is thus a missed opportunity. The FHSC (Familial Hypercholesterolemia Studies Collaboration) global registry demonstrated that HeFH diagnosis globally occurs in the fourth decade of life with only 2.1% diagnosed before the age of 18 years, with age of diagnosis in the Asia-Pacific region worst, underscoring the need for system-wide changes enabling early diagnosis [40]. At diagnosis, 17.4% had coronary artery disease (CAD), 2.1% stroke and 5.2% peripheral arterial disease with 11.3% having premature CAD [40] consistent with data suggesting that among patients with premature CAD the prevalence of FH could be as high as 1:17–25 [38].

The lack of widespread availability of genetic testing means that the commonest method of diagnosis globally is using clinical criteria with ~75% using the Dutch Lipid Clinic Network criteria (DLCN) [40], consisting of biochemical measurements, physical examination and family history. Women are diagnosed several years later than men and are less likely to have CVD at diagnosis. This means that contributions (from premature ASCVD) to the DLCN score will be absent, making LDL-C, physical examination and family history even more important, to even consider a diagnosis of FH in women. Furthermore, with increasing age, metabolic causes of elevated LDL-C become more common. As the prevalence of risk factors such as hypertension, diabetes and increased BMI increase with age (40), if patients with FH are identified earlier there is a greater chance that unhealthy lifestyles and other behaviours might be avoided reducing the likelihood of lifestyle associated risk factors later in life. Treatment beginning in adolescence dramatically reduces premature myocardial infarction and death in this condition [41]. Case finding through screening in childhood (as discussed later) is attractive for multiple reasons, not least because LDL-C is less likely to be influenced by metabolic disorders in this age group, increasing the likelihood that elevated cholesterol has a genetic basis.

Homozygous FH (HoFH) occurs in approximately 1:300 000. Affected individuals have extreme elevations in LDL-C – typically > 13 mmol/L (500 mg/dL) – and manifestations of ASCVD or aortic or supra-aortic stenosis before the age of 20 which is invariably fatal by aged 40 if untreated [43]. Those affected carry two affected genes in pathways related to cholesterol metabolism. Recently available data on HoFH from 38 countries showed that the average age of diagnosis was 12 years, with average LDL-C levels of 14.7 mmol/L (570 mg/dL) and with 9% already having evidence of ASCVD or aortic valve disease at diagnosis [44]. There are significant differences in how patients are treated between high- and low-to-middle-income countries which impacts health outcomes. Patients in high-income countries (HICs) had LDL-C levels on treatment of ~4 mmol/L (155 mg/dL) and were more likely to receive newer therapies with four or more drug combinations than in non-high-income countries. The latter by contrast mostly use two drug combinations and had on treatment LDL-C levels of about 9 mmol/L (350 mg/dL). The average age of first cardiovascular event in low-middle income countries was at 24.5 years and at 37 years in high income countries. Cardiovascular risk related to high-LDL-C in part could be attenuated with an earlier age of diagnosis as well as using four or more lipid lowering treatments, including apheresis. Notably, as many of these patients have severely diminished LDL-R function, hence traditional therapies that work through the LDL-R pathway like statins, ezetimibe and PCSK9 inhibitors may not achieve the magnitude of LDL-C lowering necessary to reduce atherosclerosis progression. This necessitates the need for newer therapies that are independent of the LDL-R like lomitapide and evinacumab. These therapies are currently extremely expensive but potentially lifesaving.

Secondary Prevention

Lipid modification therapy to prevent ASCVD

As the risk of incident ASCVD or recurrent ASCVD events is multi-factorial, a global approach is required to prevent adverse clinical outcomes. As such, modification of all risk factors starting with lifestyle are needed, and for those higher risk primary prevention individuals (to prevent incident disease) or those with prevalent clinical ASCVD, using pharmacological interventions to target different causal pathways offers independent and additive benefits [51,52]. The greater the baseline absolute risk the greater the absolute benefit from any intervention, hence the aim of lipid modification is first to lower all atherogenic lipids (LDL-C, non-HDL-C and apo B), and second to match the degree of lowering and achieved on treatment LDL-C levels to the level of risk, that is, those at greatest risk should achieve greater reductions in and have lower on treatment levels (eTable 5).

Pharmacological lipid lowering therapies

LDL-C

Healthy diet and lifestyle have been covered in prior roadmaps. For some these are insufficient. Details of currently available and future therapies, their mechanisms of action and where they should be used are available in detail in the supplementary material although they may not all be available in all regions of the world. Therapies fall into two categories; oral agents which require daily dosing and injectable therapies which require less frequent dosing. Statins remain first line therapy. They are generic and generally safe and available throughout the world. They should be initiated at the outset using effective doses that offer the greatest reductions in LDL-C to avoid delays in treatment escalation (for instance 40–80 mg atorvastatin, 20–40 mg rosuvastatin, i.e., regimens that could provide ~50% lowering). Routine monitoring for adverse effects, such as liver function, are not required unless clinically indicated. Measuring LDL-C is needed to assess response to therapy and adherence. Multiple studies have shown a lack of association between achieved LDL-C levels and adverse events. The commonest adverse effect (myalgia) may be mitigated by cessation of therapy and re-challenge at a similar dose or by a reduction in dose of the statin or reduction in the frequency of dosing (once weekly, twice weekly, alternate doses) or switching to an alternative statin [53,54,55]. Ezetimibe is also now available as a generic therapy in many parts of the world and in many countries as a combination pill with statins. Bempedoic acid is the third oral agent which has now become available in north America, Europe and elsewhere and the combination of this with ezetimibe offers similar reductions in LDL-C to a high-intensity statin alone [56], which is a possible option in those with intolerance to statins. Injectable therapies against PCSK9 include monoclonals (mAbs), which require twice monthly or monthly dosing, and inclisiran, a small interfering RNA (siRNA) therapeutic which requires twice yearly dosing. These are now available in many countries. Uptake of injectable therapies to date has largely been limited by cost, as these are more expensive than oral therapies.

Combination therapies

An inevitable consequence of the lowering of recommended cholesterol goals around the world is the need for the use of multi-drug combinations targeting different pathways in cholesterol regulation. We have increasingly recognised that many people with hypertension require multiple agents to achieve effective lowering. We need to acknowledge this may be the same for cholesterol and ask the question of how we can develop smart ways to do this? The current stepwise approach often results in clinical inertia, which in turn results in inadequate control of cholesterol, demonstrated almost serially by recurrent surveys and registries [57]. Oral combinations of two therapies are already available in the form of a single pill; statins (at various doses) and ezetimibe and bempedoic acid plus ezetimibe which may reduce pill burden and help long-term adherence. Potential combinations of oral and injectable therapies and the potential LDL-C reductions achievable are shown in Table 1 .

Table 1

Potential LDL-C reductions achievable through different combinations of lipid lowering treatments.