How to use the Coronary Risk Prediction Charts for Primary Prevention

Risk assessment tools estimating absolute coronary or cardiovascular risk should be used to identify high risk people for primary prevention. They are an aid to making clinical decisions about how intensively to intervene on lifestyle and whether to use antihypertensive and lipid lowering medication. They should not replace clinical judgment. There is a calculator programme in the separate article on Primary Prevention of Cardiovascular Disease (CVD).

• All charts are based on groups of people with untreated levels of blood pressure, total cholesterol and HDL cholesterol. The use of these charts is not appropriate for patients who have existing diseases which already put them at high risk, e.g. pre-existing cardiovascular disease, familial lipid disorders, renal dysfunction or established hypertension or diabetes with associated target organ damage.
• The charts should not be used to decide whether to introduce antihypertensive medication when blood pressure is persistently at or above 160/100 or when target organ damage due to hypertension is present. In both cases antihypertensive medication is recommended regardless of CVD risk.
• Risk assessment tools are likely to underestimate risk in people with a family history of premature CVD or in people from certain high-risk ethnic groups such as South Asians.
• Inaccuracies in estimating risk may also occur in people already taking antihypertensive treatment.

• The Framingham Heart Study began in 1948 and more than 10,000 residents of Framingham, Massachusetts have been enrolled.
• The study has documented population characteristics such as blood pressure, diabetic status and smoking, together with causes of death for over 50 years. From these data, various risk factors for CHD, stroke and total CVD (including heart failure and peripheral vascular disease) have been defined and quantified.
• The Framingham study population was mainly Caucasian and people with pre-existing cardiovascular disease were excluded. The performance of the Framingham risk scores varies considerably between populations and evidence supporting the use of cardiovascular risk scores for primary prevention is scarce.¹
• However some studies have shown that the Framingham equations generally do give a reliable prediction of risk in northern European and UK populations.² Work is continuing to develop further cardiovascular risk prevention tools, especially for specific groups, e.g. people with diabetes and people of Afro-Caribbean or Asian origin.
• Although information taken from the Framingham Heart Study is based on observational data, this has mostly been supported by interventional studies that show the modification of defined risk factors is beneficial.

Considered to be the most accurate, the Joint British Societies Coronary Risk Prediction Chart is the preferred method for estimating CVD risk.² The charts are printed at the back of the BNF but using computer calculators may be easier (for example our CVD calculator). Others are available on the internet sites below.

The use of these charts is not appropriate for patients with established cardiovascular disease, familial hypercholesterolaemia or other inherited dyslipidaemias, chronic renal dysfunction or type 1 or 2 diabetes mellitus.

Method

• To estimate an individual's absolute 10 year risk of developing CVD choose the table for his or her gender, diabetes (yes/no), smoking status (smoker/non-smoker) and age.
• Within this square define the level of risk according to the point where the coordinates for systolic blood pressure and the ratio of total cholesterol to high density lipoprotein (HDL) cholesterol meet.
• Higher risk individuals are defined as those whose 10 year CHD risk exceeds 15%, which is equivalent to a combined risk of CHD and stroke (cardiovascular risk) of >20% over the same period.

Patient data required is as follows

The initial blood pressure and the first random (non-fasting) total cholesterol and HDL cholesterol are used to estimate an individual's risk. However, the decision on using drug therapy should generally be based on repeat risk factor measurements over a period of time.

• Gender
• Age (years)
• Systolic BP (mm Hg)
• Smoking status (yes/no): Smoking status should reflect lifetime exposure to tobacco and not simply tobacco use at the time of assessment (e.g. those who have given up smoking within 5 years should be regarded as current smokers for the purposes of the charts).
• Total cholesterol
• HDL cholesterol (if no HDL cholesterol result is available, then assume this is 1 mmol/l and the lipid scale can be used for total serum cholesterol alone).

Inaccuracies of the charts

CHD risk is also higher than indicated in the charts for:³

• Those with a family history of premature CHD (male first degree relatives aged less than 55 years and female first degree relatives aged less than 65 years) which increases the risk by a factor of approximately 1.5.
• Those with raised triglyceride levels.
• Women with premature menopause.
• Those who are not yet diabetic, but have impaired fasting glucose (6.1- 6.9 mmol/l).
• As the person approaches the next age category. Risk increases exponentially with age so the risk will be closer to the higher decennium for the last four years of each decade.
• In some ethnic minorities the risk charts underestimate CVD risk because they have not been validated in these populations. For example, in people originating from the Indian subcontinent it is safest to assume that the CVD risk is higher than predicted from the charts (1.4 - 1.5 times).
• In patients already receiving antihypertensive therapy in whom the decision is to be made about whether to introduce lipid-lowering medication or vice versa the charts can act as a guide, but unless recent pretreatment risk factor values are available it is generally safest to assume that CVD risk is higher than that predicted by current levels of blood pressure or lipids on treatment.

There are an increasing number of alternative risk prediction scores, particularly focused on specific groups, e.g. people with diabetes, ethnic populations.

• ASSIGN: developed in Scotland and includes an index of deprivation and also family history.^4,5
• Reynolds Risk Score: provides a greater accuracy for assessment of cardiovascular risk in women.⁶
• UK Prospective Diabetes Study Risk Engine for people with type 2 diabetes:⁷
  • Provides risk estimates and 95% confidence intervals, in individuals with type 2 diabetes not known to have heart disease, for:
    • Non-fatal and fatal coronary heart disease
    • Non-fatal and fatal stroke
  • These can be calculated for any given duration of type 2 diabetes based on current age, sex, ethnicity, smoking status, presence or absence of atrial fibrillation and levels of HbA1c, systolic blood pressure, total cholesterol and HDL cholesterol.
• INDANA risk calculator:⁸ focuses on patients with raised blood pressure.⁹
• Other risk calculators include the Sheffield Table for Primary Prevention of Cardiovascular Disease,¹⁰ the New Zealand tables and tables proposed by the European Society of Cardiology.¹¹