Effective action for hyperlipidaemia
Our laboratory will only measure fasting lipids:
does fasting make a lot
A random specimen is fine for screening purposes a cup of tea does nothing to your total cholesterol! Fasting is very important when measuring triglycerides but is not very important for total cholesterol or HDL. If a screening cholesterol is above 5.5mmol/l, I would repeat fasting with an HDL and triglyceride. Measuring LDL has little added value over total cholesterol and HDL.
Is it necessary to classify hyperlipidaemia as we were taught in medical school?
The important thing is whether the patient has a cholesterol level that needs treating. Further classification is only usually helpful in specialist practice for patients with very abnormal lipid profiles or a young person with xanthomata.
How effective are dietary interventions in improving lipid profile?
It is not just fat intake, but also weight loss and exercise that are important. Regular exercise increases HDL and improves the risk profile. But, the evidence is that unless major changes are made, dietary modification only lowers cholesterol by about 1 per cent.
The vast majority of cholesterol is generated endogenously and has little to do with diet. So one should not torture patients with one aspect of their diet it is better to have a balanced healthy diet.
People should increase fruit and vegetables because they are separately cardioprotective and avoid saturated fat. There is not strong evidence for adding dietary supplements such as antioxidants.
Are expensive cholesterol-lowering margarines of any benefit?
There is some evidence that such products can help lower cholesterol. Of course there are no outcome data.
How important is an HDL level when deciding whether to start drug therapy?
HDL is a strong risk predictor. It is impossible to make an accurate risk assessment without a total cholesterol: HDL ratio. Making the assumption that HDL is 1mmol/l is unacceptable. We published a study that showed the Sheffield tables, which at that time did not include HDL, led to overtreatment by 30 per cent and undertreatment by 30 per cent. In other words, if you do not measure HDL you treat the wrong people.
When assessing cardiovascular risk, how can one quantify ethnicity as a risk factor?
The data are not very robust, but I multiply risk by a factor of 1.4 for south Asian patients for heart disease. Afro-Caribbean patients are at a reduced risk of heart disease and increased risk of stroke.
How can a poor family history be quantified when calculating cardiovascular risk?
It is reasonable to multiply by 1.5 or consider a 15 per cent increased risk over 10 years if there are no other identifiable risk factors. But this can be misleading because a poor family history is often accounted for by conventional risk factors such as smoking, high blood pressure, high cholesterol or low HDL, without any additional familial problem.
So it is essential to take a detailed family history first and make an accurate risk assessment based on conventional factors. If there is a strong family history of premature cardiovascular disease, not accounted for by conventional risk factors, specialist advice should be sought.
At what age would you screen the children of a patient with very high cholesterol and at what age should treatment begin?
Cholesterol metabolism is hereditary, so offspring should be tested in their 20s and advised on a healthy lifestyle, particularly on the importance of not smoking.
If a 25-year-old has a cholesterol level of >8mmol/l and it is not a lab error, they will probably need treatment. I advise six months on exercise and diet and if blood lipids have not normalised I would start treatment. Similarly, a level of >7mmol/l and an HDL of 1, with a positive family history, requires treatment. But no one knows what is the best age to start: it may not be until the late 30s for many, and the optimal age varies.
In young patients assessing lifetime risk is more valid than 10-year risk, because at 30, the 10-year risk is low. In contrast, assessing absolute risk threshold is fine above the mid-50s, because 10-year risk equates to lifetime risk.
Is there an upper age limit for primary prevention with statins?
It is illogical to treat primary prevention differently to secondary prevention and so I have no absolute upper age limit. But a more interesting question is whether we are over-treating the elderly and under-treating the young. Elderly patients are at a high absolute risk, because age is the biggest risk factor. So treatment for a short period will give a 'high return' and apparently save a lot of lives. This is because events are delayed by a few months and the patients may die of something else.
It may be better to consider by how long events would be delayed or 'healthy life gained' rather than 'events avoided'. However, this will require a change of approach and new risk prediction tools.
What should the target be for cholesterol levels in primary and secondary prevention?
The exact answer is still unknown: certainly well below 5mmol/l and for secondary prevention as low as possible. The guidelines will get lower each year. Finding people with a high absolute risk who have not yet had an event is as important as secondary prevention.
It is important to work out the benefit to the individual from lowering the cholesterol further and there are several computer programs that do so including CardioRisk Manager1 and one on the American Heart Association website2.
These may show in an individual that there is little additional benefit, for example, of reducing blood cholesterol from 5.2 to 4.5mmol.
If increasing the dose does not make any difference should statins be swapped?
Atorvastatin was initially said to be much more effective than simvastatin, but this was probably simply because atorvastatin was licensed to a higher effective dose.
Now the licence for simvastatin has changed, and at higher doses of simvastatin the effects are approximately equivalent. So if a patient is on maximal dose there is very little point in changing statin. If you have not reached target cholesterol on a statin, it may be the stage to refer to a specialist, because adding a fibrate gives a greater chance of myopathy and patients need to be monitored more carefully.
How often should patients with raised
lipids be monitored?
Initially I check bloods six-monthly and then I check two-yearly. HDL must be included in these reassessments.
Is there any evidence that a very low cholesterol level is harmful?
The associations reported between low cholesterol and disease are probably because terminal illnesses can lower cholesterol and not because low cholesterol causes disease.
Most evidence suggests a linear relationship between lowering cholesterol and reducing cardiovascular risk, rather than a J-shaped curve, so the lower the better.
I have had a couple of patients with secondary hypercholesterolaemia in association with hypothyroidism, but the cholesterol levels do not settle when thyroid function is normalised. How should one manage this?
The evidence is that normalising thyroid function does very little to cholesterol levels. So cholesterol should be rechecked and treated on its own merit.
How should patients with isolated hypertriglyceridaemia be managed?
There is not enough data on this subject yet. There is a statistical relationship between triglycerides and HDL that could be responsible for the risk associated with high triglycerides.
A very high triglyceride, more than around 10mmol/l, is associated with pancreatitis, but there is no indication for treatment otherwise. There is no evidence that treating a triglyceride of 5mmol/l alters cardiovascular risk.
It is very important to have a fasting sample when assessing triglycerides. Hypertriglyceridaemia associated with alcoholism is very difficult to control unless the patient stops drinking and I would not treat unless it was over 10.
Many poorly controlled diabetics have a raised triglyceride in association with raised cholesterol and the first-line treatment is a statin. If this does not work consider adding or substituting a fibrate.
What is the evidence for other reported benefits of statins such as possible risk reduction for Alzheimer's disease?
A large number of diseases, including types of dementia, have atherosclerosis as their basis, and there are many beneficial effects of statins. We do not yet know whether statins prevent dementia, but there is a good biological rationale for assuming that they should delay onset of vascular dementia in high-risk individuals.
Managing patients who fall outside guidelines
Sometimes targets seem impossible to attain, or are limited by side-effects. Is it acceptable to discuss this with patients and accept suboptimal control?
It is important to apply common-sense. Guidelines show you the evidence base and what to aim for. However, you often have to deviate because of side-effects or patient preference. This is fine as long as you know when and why you deviate from evidence-based practice and that it is a rational choice.
Some patients have been started on statins by specialists or previous GPs and have a relatively low cardiovascular risk. Should they continue?
We are sometimes referred patients to our clinic for cardiovascular risk reduction who were started on a statin because of apparent angina, but the exercise stress test proved negative. If a risk prediction shows low risk of cardiovascular events we would then recommend stopping the statin and re-assessing risk.
How do you suggest managing the worried well with a moderately raised cholesterol, who fall outside guidelines but wish to take a statin?
I do a risk assessment and discuss it with the patient, comparing their risk with that of the general population. People often only want to be as healthy as their age- and sex-matched peers. Many then elect to not have a statin. In the rare anxious low-risk patient who still wants drug therapy, you have to decide whether to refuse or go against the guidelines and prescribe. Remember, many American cardiologists are taking statins themselves!
Side-effects of treatment
Why does myopathy occur as a side-effect of statins?
The mechanism is not really known, but it happens with the statins and the fibrates, which suggests it is something intrinsic to cholesterol metabolism.
If a patient has an adverse effect with one statin, can one change to another safely?
Some are more likely to cause myopathy than others depending on how they are metabolised, for example cerivastatin, which was withdrawn. But there is a high chance that a patient who has had myopathy with one agent will again. Muscle aches and pains are not a contraindication, it is only if myalgia is associated with a raised creatinine kinase. If so, stop the drug and it is usually reversible.
When is disordered liver function an indication for stopping a statin?
Liver function must be significantly disordered, at least a doubling of transaminases, before considering stopping the statin. This is probably a class effect. If the potential therapeutic gain from a statin is very large, then a specialist may, in consultation with the patient, continue the statin, but if there is
a low absolute benefit, then few risks in terms of side-effects should be taken.
1. Deanfield J. CardioRisk Manager. London:
BMJ Books, 1999
2. American Heart Association website: www.americanheart.org