Consano.ch

Review: Current Perspective
Principles for National and Regional Guidelines on
Cardiovascular Disease Prevention
A Scientific Statement From the World Heart and Stroke Forum*
Sidney C. Smith, Jr, MD; Rod Jackson, MBChB, PhD; Thomas A. Pearson, MD, MPH, PhD; Valentin Fuster, MD, PhD; Salim Yusuf, MBBS, DPhil; Ole Faergeman, MD, DMSc; David A. Wood, MSc; Michael Alderman, MD; John Horgan, MD; Philip Home, MA, DPhil, DM; Marilyn Hunn, BS; Scott M. Grundy, MD, PhD In the global effort to reduce suffering and death from Altered diet with increased fat and total caloric consumption CVD, the World Heart and Stroke Forum (WHSF) Guide- and increased tobacco use are prevalent lifestyle trends.
lines Task Force of the World Heart Federation (WHF) Demographic changes coupled with adverse lifestyle changes recommends that every country develop a policy on CVD will accelerate the number of deaths due to CVD worldwide, prevention. National policy should grow out of systematic many of which will be premature in the developing countries.
and ongoing dialogue among governmental, public health, Although continuation of this adverse trend is not inevitable, and professional clinical groups. National policy should set the CVD disease patterns now present in the economically priorities for public health and clinical interventions appro- developed countries are, in fact, becoming established in priate to the country. It should also be the foundation for the developing countries, as noted in the World Health Report development of national guidelines on CVD prevention, which are the focus of the present document.
Whereas the causes of CVD are common to all parts of the Cardiovascular disease (CVD) is a leading cause of global world, the approaches to its prevention at a societal or individual mortality, accounting for almost 17 million deaths annually.
level will differ between countries for cultural, social, medical, Nearly 80% of this global mortality and disease burden and economic reasons. Although national guidelines will em- occurs in developing countries. In 2001, CVD was the brace the principles of CVD prevention recommended in this leading cause of mortality in 5 of the 6 World Health report, they may differ in terms of the organization of preventive Organization (WHO) worldwide regions. Of concern in cardiology, risk factor treatment thresholds and goals, and the developing countries is the projected increase in both propor- use of medical therapies. The recommendations in this report tional and absolute CVD mortality. This can be related to an focus on clinical management of patients with established CVD increase in life expectancy due to public health advances, and those at high risk; however, it is essential that each country which reduce perinatal infections and nutritional deficiencies include a societal approach to CVD prevention. As stated in the in infancy, childhood, and adolescence, and in some countries WHO publication Integrated Management of Cardiovascular to improved economic conditions. This increasing longevity Risk,2 “Epidemiological theory indicates that, compared with provides longer periods of exposure to CVD risk factors and intensive individual treatment of high-risk patients, small im- thus a greater probability of clinically manifest CVD. The provements in the overall distribution of risk in a population will concomitant decline of infections and nutritional disorders yield larger gains in disease reduction, when the underlying (competing causes of death) also increases the proportional conditions that confer risk are widespread in the population.” burden due to CVD. Adverse lifestyle changes accompanying Each country should seek to implement national clinical guide- industrialization, urbanization, and increased discretionary lines directed toward high-risk individuals and give equal im- income increase the degree of exposure to CVD risk factors.
portance to developing low-risk population strategies.
From the Center for Cardiovascular Science and Medicine, University of North Carolina School of Medicine, Chapel Hill (S.C.S.); Section of Epidemiology and Biostatistics, School of Population Health, Faculty of Medical and Health Sciences, University of Auckland (R.J.); Department ofCommunity and Preventive Medicine, University of Rochester Medical Center, Rochester, NY (T.A.P.); Zena and Michael A. Wiener CardiovascularInstitute and Marie-Josee and Henry Kravis Center for Cardiovascular Health, Mount Sinai Medical Center, New York, NY (V.F.); Population HealthResearch Institute, McMaster University, Hamilton, Ontario, Canada (S.Y.); University Hospital, Aarhus Amtssygehus, Denmark (O.F.); National Heartand Lung Institute, Faculty of Medicine, Imperial College London, UK (D.A.W.); Albert Einstein College of Medicine, Bronx, NY (M.A.); CardiacDepartment, Beaumont Hospital, Dublin, Ireland (J.H.); Department of Medicine, University of Newcastle upon Tyne, UK (P.H.); World Heart and StrokeForum, World Heart Federation, Geneva, Switzerland (M.H.); and Center for Human Nutrition, University of Texas Southwestern Medical Center, Dallas,Tex (S.M.G.).
The Data Supplement, which contains Figures I through VI, is available with the online version of this article at http://www.circulationaha.org.
Correspondence to Sidney C. Smith, Jr, MD, Center for Cardiovascular Science and Medicine, UNC School of Medicine, CB #7075, Bioinformatics
Building, 130 Mason Farm Rd, Chapel Hill, NC 27599-7075. E-mail scs@med.unc.edu *World Heart and Stroke Forum, World Heart Federation, Geneva, Switzerland.
(Circulation. 2004;109:3112-3121.)
2004 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org
DOI: 10.1161/01.CIR.0000133427.35111.67
3112
Smith et al
Principles for Guidelines on CVD Prevention
3113
Nomenclature and Profile of the Various
Emerging risk factors are factors that are correlated with Cardiovascular Risk Factors
CVD risk in prospective or case-control studies, but the A great advance in the prevention of CVD has resulted from strength of their correlation and/or their prevalence in the the identification of measurable factors that predict the population is less than that for the major risk factors. For this development of CVD. These factors are termed risk factors.
reason, the emerging risk factors generally are not included in Several risk factors are direct causes of CVD; these are risk-prediction equations. Among the emerging risk factors termed major risk factors and include tobacco smoking, high are various lipid factors [triglycerides, apolipoproteins, li- blood pressure, high serum LDL cholesterol, and elevated poprotein(a), and lipoprotein subfractions] and nonlipid fac- glucose. A low level of HDL cholesterol is also considered a tors (insulin resistance, prothrombotic markers, and proin- major risk factor because it independently predicts the inci- flammatory markers). Similarly, subclinical atherosclerosis dence of CVD. A final major risk factor is advancing age; may also be useful in predicting the risk of CVD events.
chronological age is considered a risk factor because it also Because the emerging risk factors are not incorporated into independently predicts CVD. Age per se does not cause CVD risk predictions, their use in clinical practice must be indi- but may reflect the accumulation of atherosclerosis, the vidualized and based on clinical judgment. Most importantly, severity of which predicts the likelihood of suffering a major they should not be given more priority in risk assessment than CVD event. Persons who have multiple major risk factors that given to the major risk factors.
generally are more likely to experience a CVD event than Concept of Total CVD Risk
those with a single risk factor. Many prospective epidemio- In general, the benefits of interventions on particular risk logical studies provide estimates of the relative contributions factors are related more to the magnitude of the preinterven- of each major risk factor to CVD risk. Prediction equations tion total CVD risk than to relative risk associated with a have been developed from these estimates and can be used to single, specific risk factor. Therefore, determination of total estimate risk for individuals. Risk estimate based on risk CVD risk is critical to recommendations on the effective and equations is termed total CVD risk.
efficient management and control of CVD risk at both In clinical practice, it is convenient to categorize total risk population and individual levels. Total CVD risk is a measure estimates into high, intermediate, and lower risk. Patients of the number of events in a defined population per unit of with established CVD are said to be at high risk because they time (eg, CVD events per 1000 in 55- to 64-year-old men per are highly likely to experience new CVD events in the next 10 year). In effect, a total risk compares a person’s or popula- years. However, some asymptomatic patients with multiple tion’s risk with a zero risk. The combined effects of all risk risk factors, particularly those with type 2 diabetes, may carry factors determine total CVD risk, and often, modest increases as high a risk for future CVD events as patients with in multiple risk factors have a greater impact on CVD risk established CVD. These persons with multiple risk factors than a significant increase in 1 risk factor.
likewise are said to be at high risk. Multiple risk factors also For example, a 46-year-old woman with high blood pres- typically are required to elevate persons to the intermediate- sure (170/100 mm Hg), but who is a nonsmoker, is nondia- risk category, whereas most persons with only a single risk betic, and has a total cholesterol level of 5.5 mmol/L and factor are at lower short-term risk. Nonetheless, even single HDL cholesterol level of 1.5 mmol/L, has an absolute CVD risk factors, if severe and sustained, can lead to premature risk of Ͻ4/100 in 5 years. In contrast, a 62-year-old smoking CVD and should not be ignored in clinical practice.
man without diabetes and with a lower blood pressure Other risk factors, in addition to the aforementioned major (150/90 mm Hg) but with a slightly higher total cholesterol risk factors, may further contribute to total risk. They are level (6.0 mmol/L) and a slightly lower HDL level underlying risk factors and emerging risk factors. The under- (1.2 mmol/L) has an absolute CVD risk of Ͼ20/100 in the lying risk factors are overweight/obesity, physical inactivity, next 5 years. Moreover, although blood pressure–lowering atherogenic diet, socioeconomic and psychosocial stress, drugs would reduce the relative CVD risk by at least one family history of premature CVD, and various genetic and quarter in both patients, the woman’s risk would fall from 4% racial factors. To some extent, the underlying risk factors to 3% (ie, a 1% absolute risk reduction in the next 5 years), affect risk by acting through the major risk factors, and they whereas the man’s risk would fall from 20% to 15% (ie, a 5% also appear to influence risk in ways unrelated to the major absolute risk reduction in the next 5 years). Appropriately, risk factors. Although these underlying risk factors likely add most practitioners would treat the first patient in concordance an independent component to total risk, their contribution has with national guidelines; unfortunately, many clinicians been difficult to distinguish in prospective studies from their might not start a blood pressure–lowering drug in the second effects on major risk factors; for this reason, they generally are not included in clinical predictive equations. Nonetheless,the underlying risk factors apparently affect population base- Total CVD Risk and Policy Development: Efficacy
line risk. Thus, the available predictive equations may not be and Cost Issues
applicable equally to all populations. The major risk factors Several risk factors that are only moderate often incur a are similar in relative predictive power in different popula- greater total risk in the short term than does a single, severe tions, but absolute estimates of risk are variable. Differences risk factor. Risk assessment in both individuals and popula- in the underlying risk factors probably account for much of tions must take this fact into account. The greatest efficacy of treatment occurs in patients who are at highest risk. Thus, 3114
Circulation
June 29, 2004
persons who are at higher total risk will attain greater Education Program (Adult Treatment Panel II),5 published in reductions in absolute risk with any given lowering of risk the same year, also recommended, for the first time, assessing factors. Giving priority in risk-reduction therapies to patients and managing lipids in the context of other cardiovascular at higher total risk will produce a substantial reduction in total risk factors. The principle of global risk was also supported CVD events. Furthermore, more high-risk individuals will by conclusions of the 27th Bethesda Conference6 (Matching benefit; in other words, the number needed to treat over a the Intensity of Risk Factor Management with the Hazard of given period of time to achieve prevention of 1 CVD event Coronary Disease Events), followed by the Sixth report of the will be fewer in higher-risk persons than in lower-risk Joint National Committee on High Blood Pressure,7 the AHA Prevention V Conference,8 the International Task Force on One important issue to consider in CVD prevention is cost Coronary Heart Disease,9 the WHO/International Society of of medical management. In traditional medical practice, Hypertension Guidelines for Management of Hypertension,10 priority in spending has gone to treatment of persons who the National Cholesterol Education Program (Adult Treat- already manifest disease. However, there is increasingly a ment Panel III),11 and, most recently, the Third Joint Euro- demand on the part of society to prevent the chronic diseases pean Societies’ Task Force on CVD Prevention in Clinical that rob individuals of health in their later years. Among the Practice12 and the Seventh report of the Joint National latter, heart disease, stroke, and chronic renal failure are high Committee on High Blood Pressure.13 All of these guidelines on the list. Moreover, advances in medical practice now make since 1993 have embraced, to different extents, the principle it possible to prevent or to delay the onset of these diseases.
of multifactorial or global risk assessment as a basis for Consequently, prevention is assuming increasing importance.
deciding whom to treat with drugs, although patients with On the other hand, adding prevention to conventional medical hypertension and end-organ damage such as renal failure or practice increases the cost and cuts into the overall healthcare younger patients with hypertension and dyslipidemias whose budget of individuals and nations. Consequently, the health short-term risk may be low also can benefit from medical policy of each nation must determine what portion of the total therapy directed toward a single risk factor. Because physi- healthcare budget can go into prevention and what portion cians deal with the whole patient and therefore every aspect must go into treatment of existing disease. Medical econo- of their risk of CVD, the principle of total risk assessment and mists have attempted to compare benefits of prevention and management is consonant with the practice of medicine.
treatment through estimations of “cost-effectiveness.” Esti- A number of tools for estimating risk of CHD or other mates of benefits have been made for both and have been atherosclerotic diseases have been developed over the past 10 expressed in terms of quality life extension. These estimates years, including risk score charts, risk assessment algorithms, suggest that the greatest cost-effectiveness for prevention and computer software programs. They are all based on the occurs for individuals at high short-term risk, whether they same principle, and many have used Framingham data.14 have established CVD or not. Most of the more economically Ideally, coronary or CVD risk prediction should be based on privileged nations can readily afford to institute preventive a prospective population cohort study undertaken in the measures in high-risk individuals. However, considerable population to which the risk score is to be applied. This is controversy exists about where to draw the line for primary because total risk of CVD may differ from one country to prevention in lower-risk persons with the use of public funds.
another, and the contribution of individual risk factors may In societies with higher socioeconomic levels, primary pre- also differ to some extent from one part of the world to the vention in the clinical setting can be employed in other. The published examples of coronary or CVD predic- intermediate-risk persons. In less economically privileged tions include the Systematic Coronary Risk Evaluation societies, even high-risk prevention may strain available (SCORE)15 Project in liaison with the Third Joint European resources. Regardless of healthcare policy for clinical inter- Societies’ risk charts (Data Supplement Figures II and III); vention, in all societies, public health measures can be the New Zealand cardiovascular risk assessment and manage- instituted for primary prevention, and these are highly cost- ment chart (Data Supplement Figure IV), which provides effective. These include programs to discourage cigarette estimates of both CVD risk and the likely benefit of therapy smoking, to promote appropriate nutrition, and to encourage to lower blood pressure or lipids; the Sheffield Tables16; the physical activity. The cardiology and medical communities Joint British Societies’ coronary risk prediction chart and can play a major role in public health efforts for primary associated software program17; and the ATP III 10-year Risk Estimates for men and women11 using Framingham PointScores14 (Data Supplement Figures V and VI), which are also Total CVD Risk for Specific Individuals
available as a computer program. Most of these risk tools are The principle of assessing the total or global risk associated based on the Framingham function. In addition, computer with multiple risk factors was first introduced in New software programs are available based on the PROCAM Zealand in 1993, in relation to the management of blood study of men in Germany,18 PRECARD from a prospective pressure.3 The following year, the European Society of cohort study of Danish men and women,19 and the European Cardiology, European Atherosclerosis Society, and European Society of Hypertension proposed total multifactorial risk as First, when we use the European Society of Cardiology’s the primary determinant of drug treatment for both blood SCORE charts (Data Supplement Figure II or III) as an pressure and blood lipids in preventing the development of example, an individual’s short-term risk of developing a CVD coronary heart disease (CHD).4 The US National Cholesterol event (myocardial infarction or stroke) over the next 10 years Smith et al
Principles for Guidelines on CVD Prevention
3115
is found by locating the appropriate box in the chart based on CVD at a population and individual level should be consid- the knowledge of age, gender, smoking status, systolic blood ered as a continuum from low to high risk: those at highest pressure, and total cholesterol level. The New Zealand chart risk are patients with clinically manifest CVD, followed by (Data Supplement Figure IV) estimates CVD rather than individuals without known CVD at different levels of risk CHD, but the risk is over a shorter period, 5 years rather than from high to low. The risk for an individual within a 10. Systolic and diastolic pressure are both used, as well as population is not just a function of their absolute ranking in the ratio of total to HDL cholesterol. The total cholesterol/ relation to others but on the overall risk of the population in HDL cholesterol ratio improves coronary risk prediction, which they live. A “low-risk” individual in a high-risk particularly for women and for those in the middle range of population may actually be at higher total CVD risk than a cholesterol. The Framingham tables (Data Supplement Fig- “high-risk” individual in a low-risk population. The risk of an ures V and VI) produce a numerical score that also corre- individual should always be judged in the context of the CVD sponds to a short-term 10-year CHD risk (myocardial infarc- tion and CHD death). Although all of these charts, tables, and An assessment of the determinants of total CVD risk computer programs estimate CHD or CVD risk for an should be a major determinant of priority setting for CVD individual, it must to be emphasized that some individuals prevention and management policy at both the clinical and will be at higher risk than is evident from these calculations.
population level, and guideline recommendations should Patients with clinically established CHD, other atheroscle- emphasize interventions on all CVD risk factors rather than rotic disease, and diabetes; patients with hypertension asso- ciated with end-organ damage or familial dyslipidemias;patients with a family history of premature CVD; and those Established Atherosclerotic Vascular Disease
with low HDL cholesterol or raised triglyceride levels also People who present with symptoms or history of atheroscle- may be at higher risk than indicated by the charts. Use of the rotic vascular disease (AVD), ie, CHD, stroke, or peripheral Framingham risk function has certain limitations. Although it arterial disease, are at high risk of recurrent nonfatal and fatal depends on the population, it or any other algorithm that is cardiovascular events. Although the initial prognosis of these derived from a different region may not accurately predict patients is determined by the extent of tissue damage to heart total risk in another population. Nevertheless, coronary risk or brain, the longer-term prospects are strongly influenced by charts or computer programs can have several useful func- the extent of their atherosclerotic process, lifestyle, and other tions: An individual’s total risk of developing a CHD or CVD risk factors responsible for expression of atherosclerotic event over a defined time period can be read from a chart disease. CHD dominates the clinical presentation of athero- sclerosis and accounts for a large majority of CVD patients.
Second, relative risk can readily be estimated by compar- Of those with other manifestations of atherosclerosis in the ing the risk in one cell with any other in the same age group form of stroke or peripheral arterial disease, many will also or with a table of average or low risk.
have CHD, which is a frequent cause of death. Population- Finally, the chart can be used to illustrate the effect of based autopsy studies have shown a strong correlation be- changing from one risk category to another.
tween the severity of atherosclerosis in one arterial territory Although young people are generally at lower risk, this will and involvement of other arterial beds. Therefore, the pre- rise steadily as age increases. In the European recommenda- vention of atherosclerosis and its complications is the same tions, short-term risk estimates for clinical decisions in young regardless of which arterial territory becomes symptomatic.
adults and subjects in early middle age are made to project For practical purposes, no distinction needs to be made risk to age 60 years. For example, if the projected risk to age between those presenting with CHD and other forms of 60 years places a person in the high-risk category, this person atherosclerosis in terms of lifestyle intervention and risk can be treated accordingly with more intensive monitoring factor management for blood pressure, lipids, and hypergly- and earlier intervention. In this way, individuals with low cemia. However, specific drug therapies may differ according CVD risk today, but who will become high risk in the long to the clinical expression of atherosclerotic disease and its term unless there is lifestyle and, where appropriate, thera- complications (eg, preference for ␤-blockers or angiotensin- peutic intervention, can also be identified and treated earlier.
converting enzyme [ACE] inhibitors for blood pressure con-trol in CHD patients).
Concept of a Continuum From Low- to
CHD is the most common clinical manifestation of athero- High-Risk CVD Prevention
sclerosis. Sudden cardiac death in the community is often the The concept of total CVD risk also challenges the traditional first manifestation of CHD and is the terminal event in more classification of prevention into tertiary, secondary, and than half of CHD patients.20 Acute myocardial infarction and primary. Most patients with established CVD have developed unstable angina account for approximately one third of all symptomatic disease because they are at high risk, and the cases, whereas exertional angina is the most common clinical management of these multiple risk factors will over the manifestation of this disease, accounting for more than one longer term determine their risk of recurrent disease. Healthy half of all cases presenting in the community. Because the individuals at high risk are usually no different (and many majority of individuals with CHD survive their first symp- will already have asymptomatic atherosclerosis) from those tomatic presentation, the potential to reduce the risk of who have declared their disease; all are at high risk of recurrent events and death is considerable. Surveys of con- developing and dying from a CVD event. Thus, prevention of temporary clinical practice around the world, however, show 3116
Circulation
June 29, 2004
that lifestyle and risk factor management, including the use of uals is to intervene in order to slow disease progression, if prophylactic medical therapies, falls far short of evidence- possible to induce regression, and to reduce the risk of based national guidelines on CVD prevention. To further thrombotic complications, thereby reducing the risk of a first reduce the risk of recurrent CHD events and death, the nonfatal or fatal coronary or other atherosclerotic disease standards of preventive care must be raised.
Patients with atherosclerosis of the carotid, vertebral, and Before screening technology is used in routine clinical cerebral arterial circulations can present with transient epi- practice, the following screening criteria should be met: (1) sodes of cerebral ischemia (transient ischemic attack) or a full The noninvasive technique for detecting CHD or other stroke (either thrombotic or embolic), which can leave them atherosclerotic disease is valid, precise, easy, and acceptable.
temporarily or permanently disabled. Prevention of hemor- (2) The risk of symptomatic disease, eg, angina, CHD death, rhagic stroke is not included here because the pathology is not or stroke, has been quantified. (3) The screening strategy, usually atherosclerosis; however, because of its association intervention, and follow-up policy are defined. (4) Trained with hypertension as a modifiable risk factor, it must be staff and facilities for screening and intervention are avail- considered to benefit from medical treatment. The risk of able. (5) Screening and intervention results in a reduction in recurrent cerebrovascular disease is determined by multiple clinical events: CHD and other atherosclerotic morbidity and risk factors, particularly hypertension. In addition, patients mortality. (6) Screening has no adverse effects. (7) Cost of with cerebrovascular disease due to thrombosis usually have screening and intervention is affordable, appropriate for the CHD as well. Therefore, their risk factors should be managed healthcare system, and justified by the outcome.
on the assumption that they have CHD in order to broadly For CHD, CT can identify coronary calcification as a reduce their risk of CVD events. Although it is beyond the surrogate for coronary atheroma. In addition, CT and MRI are scope of this discussion, embolic stroke associated with atrial evolving technologies for the detection of epicardial disease.
fibrillation deserves attention as a major preventable entity.
The impact of coronary atheroma on perfusion of the myo- Atherosclerosis of the peripheral arteries usually presents cardium can also be objectively assessed noninvasively with clinically with aneurysmal dilatation of the aorta, aortic a variety of techniques including radionuclide scintigraphy, dissection, and, most commonly, progressive ischemia (“in- stress echocardiography, and exercise ECG testing. However, termittent claudication”) of the lower limbs. Although an most of these techniques detect obstructive coronary athero- aortic aneurysm or dissection can be life threatening, athero- ma, and each has limitations as a sensitive and specific test sclerosis of the lower limb arteries is usually not, although for the diagnosis of CHD in an asymptomatic individual.
patients can develop critical ischemia of the foot requiring Asymptomatic atherosclerotic disease of the aorta, carotid, amputation. However, almost all patients with atherosclerosis brachial, and lower limb arteries can also be detected by of the peripheral arteries also have CHD and therefore are at noninvasive techniques, including MRI, carotid ultrasound, increased risk of a nonfatal coronary event or coronary death.
brachial reactivity, ankle-brachial pressure index, and tibial The risk factors for atherosclerosis of the peripheral arteries artery blood flow velocity by Doppler ultrasound.
are the same as those for the coronary circulation, although More research is needed to evaluate the incremental value smoking is a particularly powerful risk factor for atheroscle- and cost-effectiveness of these techniques compared with rosis of the aorta and lower limbs. Therefore, patients with conventional risk factors in assessing the absolute risk of peripheral atherosclerotic disease should also have their risk developing symptomatic disease. Randomized controlled tri- factors managed in the same way as those with CHD to als are also required to evaluate the impact of noninvasive reduce their risk of CVD complications. In addition, periph- screening and intervention programs for CHD or other eral arterial disease is a powerful predictor of major coronary atherosclerotic disease on subsequent CVD morbidity and events. Therefore, the presence of peripheral atherosclerotic mortality. Until such evidence is available, screening for disease places a person in a high-risk category.
asymptomatic disease with advanced technologies should be Patients with CHD or other atherosclerotic disease are considered investigational, with the exception of ankle- considered to be at high risk. There is no practical utility in brachial pressure index, and studies should be performed to further quantifying their total risk of a future CVD event confirm their cost-effectiveness before adoption for use in a because risk stratification will not alter recommendations for target goals of risk factor therapy.
High-Risk Populations
Asymptomatic Atherosclerotic Vascular Disease
The rising prevalence of CVD worldwide is in part a The medical technology to detect asymptomatic atheroscle- reflection of a rising prevalence of CVD risk factors in many rotic disease is already available for coronary atherosclerosis, nations. Among these are increasing prevalence rates of carotid/vertebral atherosclerosis, and peripheral arterial dis- cigarette smoking, hypertension, lipid disorders, diabetes, and ease. This technology has revealed the ubiquity of AVD, as older people. Changing life habits across broad populations is understood by pathologists many years ago. Emerging meth- responsible for the emergence of most of these risk factors.
odologies can aid in the detection of AVD before clinical Cultural changes are such that multiple risk factors in symptoms. The cost of this technology emphasizes the individuals are common. To stem the rising tide of CVD benefits and importance of primordial preventive strategies, worldwide, it will be necessary to attack the causes of CVD as discussed elsewhere in this article. The objective of risk factors. These underlying causes include increasing detecting asymptomatic AVD in apparently healthy individ- obesity, decreasing physical activity, and changes in the Smith et al
Principles for Guidelines on CVD Prevention
3117
composition of the diet. To modify these underlying risk Lifestyle
factors, CVD specialists must team with primary healthcare Intervention in relation to tobacco cessation, healthy food providers, epidemiologists, and public health officials to choices, weight control, and physical activity is the founda- modify behavioral characteristics of individuals. CVD spe- tion of preventive cardiology. Diets associated with a low cialists can assist in the identification of problem areas and CVD risk will differ in terms of food composition around the serve as a catalyst for change. It is logical that preventive world. Although pharmacological, interventional, and device- efforts in whole populations should be broad based and oriented interventions may depend on national economic directed toward reducing all the risk factors simultaneously.
factors, lifestyle interventions can be implemented At the same time, even within a single nation there can be subpopulations that are at higher risk than others because of Physicians and other health professionals should set an either genetic or racial factors or unique exposures to envi- example for patients with AVD, high-risk individuals, and the ronmental factors. Indeed, many epidemiological studies general population by not smoking themselves. A physician’s reveal that socioeconomic status can be an independent firm advice that a patient should stop smoking is the most predictor of risk. Some of the excess risk dependent on important first step. The goal is complete cessation and advantaged socioeconomic status can be explained by the avoidance of passive smoking. The only important difference major, independent risk factors and their root causes. How- between current recommendations is in the use of nicotine ever, other factors such as psychosocial stress, behavioral replacement therapy, especially for patients with AVD. Nic- factors, and access to the medical care system may raise the otine chewing gum and transdermal nicotine patches can risk in persons who are economically disadvantaged.
double the cessation rates compared with a placebo. The use Physicians should play an increasing role in public health of nicotine patches has been tested successfully in patients medicine. Not only can their national health societies take an who have coronary disease without any adverse effects, but active part in public health issues, but they have an opportu- caution in the use of nicotine replacement therapies is still nity to convey the public health (preventive) messages in required. Patients should not smoke while they are using their daily interactions with patients. When patients are being these nicotine delivery preparations because doing so may treated for various medical complaints, the physician should exacerbate symptoms. The antidepressive drug bupropion is not overlook the chance to deliver a broader message. In fact, an additional treatment to help individuals quit.
societal changes that lead to an increased prevalence of An atherogenic diet contributes to CVD in many popula- cardiovascular risk factors also provide greater access to tions. A healthy diet is low in saturated and trans-fatty acids personalized healthcare. The prevention message should be and low in dietary cholesterol. The amount of saturated and trans-fatty acids in the diet should be Ͻ10% of total calories, The public health approach to prevention of CVD has and the dietary cholesterol intake should be Ͻ300 mg/d. A several components, including government policy, educa- useful recommendation for individuals at high risk is to tional efforts, industrial policy, and testing for risk factors.
reduce the quantity of food they consume by 20% to 25%, Physicians are appropriately involved at every level. Partic- reduce animal fats, and decrease the amount of salt added to ularly important are screening programs for risk factors.
foods in cooking and at the table. A good example of a diet Screening is best done in the medical setting where appro- low in saturated fat and cholesterol is the traditional Medi- priate follow-up and advice are available. However, when terranean diet; in this diet, unsaturated fats replace most of the prevalence of certain risk factors is identified in subpopula- saturated fat. The traditional Japanese diet is also low in tions, mass screenings may be more efficient and saturated fat but high in complex carbohydrates. Both of these diets are associated with the best life expectancy in the world.
For prevention of CHD and other AVD, the best advice is to Management of Specific and Total Risk
use a diet low in saturated fatty acids by replacing them in Factors in Patients and Populations
part with monounsaturated and polyunsaturated fatty acids as The overall objective of CVD prevention in patients with well as with complex carbohydrates. These principles are clinically established AVD or asymptomatic individuals at reflected in all the recommendations. Physicians should high risk is the same: to reduce the risk of subsequent major emphasize the importance of diet in relation to reducing CHD or other AVD events. Secondary prevention for patients weight, lowering blood pressure and blood cholesterol, con- with established AVD has traditionally been distinguished trolling blood glucose in diabetic patients, and reducing the from primary prevention for asymptomatic high-risk individ- propensity to thrombosis. Alcohol should be considered in the uals, but this distinction is artificial because the majority of context of dietary advice. Although moderation in the use of individuals at high risk are also likely to have advanced alcohol should always be advised, further restriction may be subclinical atherosclerosis. Thus, prevention of CVD at a necessary in those who are overweight (to reduce calories), population and individual level should be considered a particularly in patients with elevated blood pressure and continuum. Those at highest risk are patients with clinically elevated serum triglycerides. The intake of salt (sodium manifest CVD, followed by asymptomatic CVD individuals chloride) should also be reduced to Ͻ5 g/d in patients with and by individuals with a high risk factor profile. Because the high blood pressure. The goals of dietary counseling have to biology of AVD and the distribution of risk overlap in these be defined on a national basis, together with the practical 3 groups, a high intensity of lifestyle intervention and risk recommendations for translating such goals into the selection, preparation, and consumption of foods.
3118
Circulation
June 29, 2004
The goals and recommendations for weight management and mortality. Microalbuminuria in diabetic and in nondia- vary on the basis of geographic region. In Western Europe betic patients is also associated with increased risk. Systolic and the United States, body mass index Ͻ25 but Ͼ20 kg/m2 blood pressure is as strongly, or even more strongly, associ- is associated with the lowest risk of CVD and CHD. As ated with CVD risk as diastolic blood pressure. In some people become overweight (BMI Ͼ25 and Ͻ30 kg/m2), CVD clinical trials of hypertension, cardiovascular events correlate risk increases, and with obesity (BMI Ͼ30 kg/m2) all-cause more closely with achieved systolic pressure than diastolic mortality increases, largely because of an increase in CVD pressure. Recent trials on isolated systolic hypertension have mortality. Overweight is also associated with an increased added to evidence regarding the importance of systolic blood risk of stroke. Central adiposity, defined as an increased pressure in risk assessment and management.
intra-abdominal fat mass, is associated with an adverse risk International, continental, and national guidelines now factor profile, including insulin resistance, and, as assessed by recommend that treatment of hypertension in healthy individ- waist-to-hip circumference ratio, is more strongly associated uals be based on both the systolic and diastolic values and the with risk of CHD and other CVD than general adiposity coexistence of other atherosclerotic disease risk factors and assessed by body mass index. Reducing weight will reduce aforementioned comorbidities. For some individuals the level blood pressure and plasma LDL cholesterol, raise HDL of blood pressure is deemed sufficiently high to merit cholesterol, and lower triglycerides and will decrease glucose antihypertensive treatment in its own right, regardless of its intolerance. It should be emphasized that the aforementioned clinical context. For others a decision to treat is taken in the numbers for BMI have been determined for Western Europe context of absolute CHD or CVD risk. The definition of high and the United States and may be lower for other regions and risk differs between guidelines in terms of both the levels of systolic and diastolic blood pressure and the levels of abso- The relevance of physical activity in helping weight lute CHD or CVD risk. The optimal blood pressure to be control and favorably modifying other risk factors should be achieved by treatment has not been established in randomized explained. A balance in caloric intake and energy expenditure controlled trials, but the blood pressure goal is the same for is fundamental to any program that seeks to alter and international, continental, and some national guidelines, and maintain ideal body weight. Regular physical activity is the risk of events has been shown to increase continuously associated with a lower risk of death from CVD and CHD.
Physical activity helps to prevent obesity, is associated with In clinical practice, it is important to set a blood pressure lower levels of plasma LDL cholesterol and triglycerides and target for an individual and to try to achieve it with a higher levels of plasma HDL cholesterol, and lowers blood minimum of side effects. Several classes of drugs have been pressure. Exercise-based cardiac rehabilitation in patients shown in randomized controlled trials to reduce the risk of with established coronary disease has been shown to reduce CHD and CVD: diuretics, ␤-blockers, calcium antagonists, total cardiovascular and coronary mortality.
ACE inhibitors, and angiotensin receptor blockers. Localcosts and patient characteristics should be taken into account Blood Pressure
in the selection of antihypertensive drugs. The Antihyperten- Hypertension is a major cause of stroke and contributes to an sive and Lipid Lowering Treatment to Prevent Heart Attack increased risk of recurrent myocardial infarction in patients (ALLHAT)21 results indicate improved or comparable car- with CHD. Treatment of hypertension is therefore important diovascular outcomes among patients given thiazide diuretics as a primary and secondary prevention strategy. Several trials for treatment of hypertension and lend support to the potential in coronary patients with blood pressure–lowering drugs, ␤-blockers and ACE inhibitors, particularly after myocardialinfarction, have demonstrated a reduction in both recurrent Blood Lipids and Lipoproteins
myocardial infarction and all-cause mortality.13 Similarly, A strong, independent relationship exists between serum LDL treatment of high blood pressure has been shown to lower the cholesterol levels and risk for CHD and to a lesser extent for occurrence of fatal and nonfatal stroke. International, conti- other CVD end points. The relationship between other serum nental, and national recommendations advise treating hyper- lipids (HDL cholesterol and triglycerides) and the risk of tension in patients with established atherosclerotic disease, atherosclerotic disease is more complex. Like blood pressure, and a blood pressure target of Ͻ140/90 mm Hg is common to the relationship between serum LDL cholesterol and risk of all. In healthy individuals there is agreement across all CVD (principally CHD) increases continuously as LDL recommendations that the decision to start treatment depends cholesterol levels rises, starting from levels that are consid- both on the blood pressure level and the overall CHD or CVD ered to be within the so-called normal range. Therefore, like risk as well as the presence of subclinical CVD or end-organ blood pressure, the dividing line between individuals requir- damage. Markers such as left ventricular hypertrophy, a ing clinical intervention is determined operationally by epi- marked reduction in glomerular filtration rate, proteinuria, demiological data, randomized controlled trials, and eco- and retinal hemorrhages and/or exudates with or without nomic considerations. Standard risk equations have a papilledema are all associated with an increased risk at any diminished reliability in familial dyslipidemias, particularly given blood pressure level. Echocardiography is a more familial hypercholesterolemia. Affected patients are at very sensitive marker of left ventricular hypertrophy than electro- high risk of aggressive premature atherosclerosis and suffer cardiography, and echocardiographic left ventricular hyper- early coronary morbidity and mortality. For these patients, trophy is associated with an increased risk of CVD morbidity lipid-lowering therapies and other forms of treatment are Smith et al
Principles for Guidelines on CVD Prevention
3119
essential regardless of the presence of other cardiovascular control has been shown in randomized controlled trials to risk factors. Although these other risk factors also need to be reduce the risk of microvascular complications in both type 1 effectively addressed in patients with these familial dyslipid- and type 2 diabetes. In addition, in the UK Prospective emias, lowering LDL cholesterol should be the primary Diabetes Study of type 2 diabetes, there was a favorable trend for glycemic control reducing the risk of myocardial infarc- For patients with established CHD or other atherosclerotic tion.22 Blood pressure reduction in the same trial significantly disease and even for those with diabetes or hypertension, reduced the risk of myocardial infarction, and this result is there is randomized controlled trial evidence that modifying consistent with the subgroup analyses of patients with diabe- lipids, principally reducing LDL cholesterol, irrespective of tes in other primary prevention trials of hypertension that the initial values, reduces the risk of recurrent coronary showed a reduction in cardiovascular morbidity and mortality disease, stroke, and all-cause mortality. Thus, for asymptom- at least as good as that seen in nondiabetic individuals.
atic individuals, international, continental, and nationalguidelines now recommend that treatment of blood lipids, in Other Risk Factors
the absence of familial dyslipidemia, should be based on Although risk assessment is principally focused on aspects of lifestyle, blood pressure, lipids, and diabetes, there are other Hyperlipidemias secondary to other diseases are common, risk factors for CHD and other AVD. These include psycho- including abuse of alcohol, hypothyroidism, diseases of the social factors, markers of inflammation, thrombogenic fac- kidney and liver, and diabetes, particularly in the presence of tors, insulin resistance, and genetics. However, the benefit of a nephropathy. Therefore, it is always important to exclude clinical interventions directed to each of these factor profiles these diseases with an appropriate clinical assessment and remains to be determined through controlled clinical trials.
tests before introducing drug therapy.
Although goals for total cholesterol and LDL cholesterol Prophylactic Medical Therapies
have been set, there is insufficient evidence to justify goals In individuals at high multifactorial risk of developing CHD for triglycerides and HDL cholesterol. Instead, these mea- or other AVD, there is evidence from randomized controlled surements should be used to identify individuals at high trials that prophylactic aspirin reduces risk.23 There is grow- multifactorial risk of CHD or other atherosclerotic disease ing agreement across international, continental, and national and possibly used as secondary considerations in the selection guidelines that persons at intermediate or high risk (Ͼ10% of lifestyle and drug interventions.
per 10 years) for hard CHD events (myocardial infarction or Several classes of lipid-lowering drugs have been shown in CHD death) may benefit from 75 to 160 mg/d of aspirin. For randomized controlled trials to reduce clinical events: patients with established CHD or other atherosclerotic dis- 3-hydroxy-3-methylglutaryl coenzyme A reductase (statins), ease, aspirin (Ն75 mg) or other platelet-modifying drug is fibrates, bile acid sequestrants (resins), and nicotinic acid universally recommended. The meta-analysis of antiplatelet derivatives. All 4 classes of drugs, but not all drugs within trials after myocardial infarction demonstrates a significant each class, have been shown in clinical trials to reduce reduction in all-cause mortality, vascular mortality, nonfatal myocardial infarction and sudden death.11 A new class, reinfarction of the myocardium, and nonfatal stroke for those cholesterol absorption blockers, reduces LDL cholesterol but receiving antiplatelet therapy. In several studies of anticoag- has not yet been tested in clinical trials to determine the effect ulation after myocardial infarction, systemic anticoagulants on cardiovascular morbidity and mortality. The statin drugs reduced the risk of all-cause mortality and coronary death.23 are the most widely used of the lipid-lowering drugs because This drug class is used selectively in patients at high risk of they are highly effective in lowering LDL levels and because systemic embolization or in patients unable to take aspirin.
they are well tolerated. Increasingly, other lipid-lowering In a meta-analysis of ␤-blockers after myocardial infarc- drugs are used in combination with statins in patients with tion, there was also evidence of a significant reduction by severe hyperlipidemias or complex dyslipidemias.
therapy in all-cause mortality and in particular sudden cardiac Blood Glucose
death, as well as nonfatal reinfarction.24 The benefit was Mounting evidence suggests that aggressive blood glucose greatest in those with left ventricular dysfunction or su- lowering with insulin in patients with myocardial infarction, praventricular or ventricular tachyarrhythmias. Therefore, a both during the hospital admission and 1 year after it, reduces ␤-blocker is recommended in patients with no contraindica- mortality. Although there is no specific randomized con- trolled trial evidence for blood pressure lowering in patients A meta-analysis of ACE inhibitors has confirmed a similar with atherosclerotic disease and diabetes, the subgroup anal- benefit in regard to all-cause mortality for this drug class in yses of patients with diabetes and myocardial infarction in patients with myocardial infarction with symptoms or signs trials of ␤-blockers and ACE inhibitors have shown a similar of heart failure at the time of acute myocardial infarction, in treatment benefit for patients with and without diabetes.
those with impaired systolic ventricular function (ejection Similarly, there is no direct trial evidence of cholesterol fraction Ͻ40%), and in patients at high risk with preserved lowering in patients with diabetes, but subgroup analyses in systolic function.25 Because most trials of ␤-blockers and large statin trials showed reductions in CHD events at least as ACE inhibitors were single-drug trials, the use of both drugs large in patients with diabetes as in nondiabetic patients.11 In versus one or the other has not been studied. Patients with individuals with diabetes but no symptomatic AVD, glucose clinical CHF after myocardial infarction have also been 3120
Circulation
June 29, 2004
shown to have a benefit from angiotensin receptor blocker level of risk at which to intervene, and the risk factor therapy comparable to that from ACE inhibitors.
thresholds themselves. All of these differences should be As mentioned, for asymptomatic individuals, international, resolved at a national level by taking account of the scientific continental, and national guidance now recommends that evidence and the resources available to deliver effective irrespective of the initial LDL cholesterol values, treatment with LDL-lowering drugs, in the absence of familial dyslip-idemia, should be based on absolute risk.
International Call to Action to Address
the Challenge
Screening Relatives
Because CVD is a global problem, societies of cardiology can A detailed family history of CHD or other atherosclerotic and will benefit through international professional collabora- disease should be part of the assessment of all patients. The tion. The International Heart Health Conferences issued risk of CHD increases when a first-degree family relative has declarations on prevention of CVD in 1992 (Victoria Decla- a history of premature CHD. Risk factor screening should be ration),26 in 1996 (Catalonia Declaration),27 in 1998 (Singa- considered in first-degree relatives of any patient developing pore Declaration),28 and most recently in 2001 (Osaka Dec- CHD at an early age: before 55 years in men and before 65 laration).29 The Singapore Declaration particularly is a years in women. In this context, the multifactorial risk will be valuable description of the intellectual and organizational higher than that estimated from the coronary risk chart. When principles that should underlie programs to prevent CVD. The familial dyslipidemia is suspected, particularly familial hy- principles are broadly divided into those pertaining to the percholesterolemia (family history of premature CHD, blood structure of preventive programs and those pertaining to the cholesterol Ͼ8.0 mmol/L, with or without stigmata or hyper- political will to proceed to action. Preventive efforts have lipidemia), screening all first-degree relatives with a full been mounted by international organizations with more spe- cific agendas for longer periods of time, and thus the presentdocument is a logical extension of international collaborative National and International Guidelines on
efforts that have been in place since 1992.
CVD Prevention
The idea of political will, consistent with an activist agenda, is that prevention will get nowhere if clinicians, Similarities and Differences
researchers, and others who want to advance the cause of An international consensus has emerged among guidelines CVD prevention do not accept personal responsibility to regarding priorities for CVD prevention, risk factor assess- assume a leadership role. In the section on physical and ment, and management, including the use of drug therapies.
organizational infrastructure of prevention, the Singapore However, this consensus mainly comes from guidelines Declaration specifies the importance of nongovernmental developed in the United States, Europe, Australia, and New organizations and professional health organizations such as Zealand. For much of the world, especially the developing the WHF. The WHSF quite specifically requests in this countries in the Asia-Pacific region, Africa, and South Amer- document that continental and national societies of cardiolo- ica, there are few data on risk factors and CVD and few gy and related professional organizations assume leadership published guidelines. The strongest agreement across inter- of continental and national programs to prevent further national, continental, and national guidelines is for patients increases in the occurrence of CVD. Societies of cardiology with established CHD or other AVD. These patients are have the professional authority to not only ask government to recognized by cardiologists and other physicians as the top allocate resources for care of patients with CVD but also to priority for prevention, and there is general agreement on the ask government, be it continental, national, or local, to need for lifestyle intervention, blood pressure reduction, incorporate prevention of CVD into legislation whenever cholesterol reduction, and the use of prophylactic drug therapies: aspirin, ␤-blockers, ACE inhibitors, and LDL-lowering drugs. Although the same or similar blood pressure Strategic Principles for the Development of
goal has been specified in all guidelines, this is not so for National Clinical Guidelines
cholesterol. There are some differences between guidelines On the basis of and following the sequential approach of this on cholesterol goals for patients with CHD and CVD, but this document, the WHSF of the WHF recommends 10 strategic is of practical importance only to a small minority of patients; principles to serve as a template for the development of most have cholesterol levels that are untreated and remain above the standards of the most conservative of cholesteroltargets. Otherwise, it is important to set a treatment target for 1. Governments, national societies, and foundations should LDL cholesterol in patients with AVD at a national level. The collaborate to develop clinical and public health guide- same principle applies to patients with diabetes mellitus.
lines for CVD prevention that target risk factors.
2. Evidence-based guidelines should incorporate profes- For healthy individuals, there is also agreement across sional judgment on the translation of such evidence into international, continental, and national guidelines on the effective and efficient care addressing all areas of CVD principle of basing the decision to treat blood pressure or lipids on absolute multifactorial risk of CVD. However, the 3. The assessment of total CVD risk should be based on practical application of this principle differs between guide- epidemiological risk factor data appropriate to the popu- lines in terms of the method of risk calculation, the absolute Smith et al
Principles for Guidelines on CVD Prevention
3121
4. Policy recommendations and guidelines should empha- 9. Assmann G, Carmena R, Cullen P, et al. Coronary heart disease: reducing size a total risk approach for CVD prevention.
the risk: a worldwide view: International Task Force for Prevention of 5. The intensity of interventions should be a function of the Coronary Heart Disease. Circulation. 1999;100:1930 –1938.
10. World Health Organization, International Society of Hypertension, total risk of CVD, with lower treatment thresholds for Guidelines Subcommittee. Guidelines for the management of hyper- tension. J Hypertens. 1999;17:151–183.
6. National cardiovascular societies/foundations should 11. Expert Panel on Detection, Evaluation and Treatment of High Blood promote routine prospective collection of validated na- Cholesterol in Adults. Executive summary of the third report of theNational Cholesterol Education Program (NCEP) (Adult Treatment Panel tional vital statistics on the causes and outcomes of CVD III). JAMA. 2001;285:2486 –2497.
for use in the development of national policies.
12. DeBacker G, Ambrosini E, Borch-Johnsen K, et al. The Third Joint 7. National professional societies should inform policymak- European Societies’ Task Force on CVD Prevention in Clinical Practice: ers of risk factor targets and drug therapies for prevention executive summary. Eur Heart J. 2003;24:1601–1610.
of CVD that are culturally and financially appropriate to 13. Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. The Seventh Report of the Joint their nation and ask the government to incorporate National Committee on Prevention, Detection, Evaluation and Treatment prevention of CVD into legislation whenever relevant.
of High Blood Pressure. JAMA. 2003;289:2561–2572.
8. National professional societies/foundations should facil- 14. Wilson PWF, D’Agostino RB, Levy D, et al. Prediction of coronary heart itate CVD prevention through education and training disease using risk factor categories. Circulation. 1998;97:1837–1847.
15. Conroy RM, Pyorala K, Fitzgerald AP, et al. Estimation of ten year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur 9. National professional societies should assess the achieve- ment of lifestyle, risk factor, and therapeutic targets 16. Ramsay LE, Haq IU, Jackson PR, et al. Targeting lipid-lowering drug therapy for primary prevention of coronary disease: an updated Sheffield 10. Health professionals should include prevention of CVD table. Lancet. 1996;348:387–388.
as an integral part of their daily clinical practice.
17. British Cardiac Society, British Hyperlipidaemia Association, British Hypertension Society, endorsed by the British Diabetic Association. JointBritish recommendations on prevention of coronary heart disease in Although the focus of these recommendations is clinical, it clinical practice. Heart. 1998;80(suppl 2):1–29.
is recognized that a population approach to CVD prevention 18. Assmann G, Cullen P, Schulte H. The Munster Heart Study (PROCAM): is the foundation of all clinical strategies in preventive results of follow-up at 8 years. Eur Heart J. 1998;19(suppl A):A2–A11.
19. Thomsen TF, Davidsen M, Ibsen H, et al. A new method for CHD cardiology. The WHSF strongly endorses the World Health prediction and prevention based on regional risk scores and randomized Report 2002 recommendations that urge countries to adopt clinical trials: PRECARD and the Copenhagen Risk Score. J Cardiovasc policies and programs to promote population-wide interven- 20. American Heart Association. Heart Disease and Stroke Statistics: 2004 tions such as reducing use of tobacco, reducing saturated fat Update. Dallas, Tex: American Heart Association; 2003.
in the national diet and salt in processed foods, encouraging 21. ALLHAT Officers and Coordinators for the ALLHAT Collaborative higher consumption of fruits and vegetables, and encouraging Research Group. Major outcomes in high-risk hypertensive patients ran-domized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic. JAMA. 2002;288:2981–2997.
22. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood glucose References
control with sulfonylureas or insulin compared with conventionaltreatment and risk of complications in patients with type 2 diabetes.
1. World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. Geneva, Switzerland: World Health Orga- 23. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomized trials of antiplatelet therapy for prevention of death, myo- 2. Integrated Management of Cardiovascular Risk: Report of a WHO cardial infarction and stroke in high risk patients. BMJ. 2002;324:71– 86.
Meeting, July 2002. Geneva, Switzerland: WHO Library; 2002.
24. Freemantle N, Cleland J, Young P, et al. Beta blockade after myocardial 3. Jackson R, Barham P, Maling T, et al. The management of raised blood infarction: systemic review and meta regression analysis. BMJ. 1999;318: pressure in New Zealand. BMJ. 1993;307:107–110.
4. Pyorala K, deBacker G, Graham I, et al. Prevention of coronary heart 25. ACE Inhibitor Myocardial Infarction Collaborative Group. Indications for disease in clinical practice: recommendations of the Task Force of the ACE inhibitors in the early treatment of acute myocardial infarction: European Society of Cardiology, European Atherosclerosis Society and systematic overview of individual data from 100,000 patients in ran- European Society of Hypertension. Eur Heart J. 1994;15:1300 –1331.
domized trials. Circulation. 1998;97:2002–2212.
5. Expert Panel on Detection, Evaluation and Treatment of High Blood 26. The Victoria Declaration: On Heart Health. In: Declaration of the Cholesterol in Adults. Executive summary of the second report of the Advisory Board of the International Heart Health Conference; 1992; National Cholesterol Education Program (NCEP) (Adult Treatment Panel II). JAMA. 1993;269:3015–3023.
27. The Catalonia Declaration: Investing in Heart Health. In: Declaration of 6. 27th Bethesda Conference. Matching the intensity of risk factor man- the Advisory Board of the Second International Heart Health Conference; agement with the hazard for coronary disease events. J Am Coll Cardiol.
28. Pearson T, Bales VS, Blair L, et al. The Singapore Declaration: forging 7. Joint National Committee on Prevention, Detection, Evaluation and the will for heart health in the next millennium. CVD Prevention. 1998; Treatment of High Blood Pressure. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High 29. The Osaka Declaration: Health, Economics and Political Action: Blood Pressure. Arch Intern Med. 1997;157:2413–2445.
Stemming the Tide of Cardiovascular Disease. In: Declaration of the 8. Smith SC Jr, Greenland P, Grundy SM. AHA Conference Proceedings, Fourth International Heart Health Conference; 2001; Osaka, Japan.
Prevention Conference V: Beyond secondary prevention: identifying thehigh-risk patient for primary prevention: executive summary. Circulation.
Ⅲ cardiovascular diseases Ⅲ prevention

Source: http://www.consano.ch/WorldHeartFederationGuidelines062004.pdf