The preoccupation the medical and scientific communities have with cholesterol have convinced many of us, doctors included, that ‘cholesterol is bad’ so ‘cholesterol reduction is good’. But, as I’ve pointed out before, even if cholesterol does jeopardise health, the fact that something reduces cholesterol does not make it inherently healthy. Even if arsenic and cyanide were found to be effective cholesterol-reducing agents, it wouldn’t make sense to swig back these poisons each day, would it?
If this notion seems far-fetched, consider the fact that the cholesterol-improver torcetrapib had to be pulled from development when it was discovered to be killing people. The effects on health are the key thing, not its effects on cholesterol. Failure of regulators to focus on this critical distinction is how we have governments spending money on a drug called ezetimibe which reduces cholesterol but has never been shown to have benefits for health and has been linked with an increased risk of death from cancer.
Another class of drug that requires some scrutiny, I think, is the ‘fibrates’ such as clofibrate, gemfibrozil and fenofibrate. These work mainly to lower levels of ‘unhealthy’ blood fats known as triglycerides, while raising levels of ‘healthy’ high density lipoprotein-cholesterol (HDL). But what evidence do we have that fibrates benefit health (rather than blood fat levels)? The most recent decent review of the evidence I could find was published in 2010 in the Lancet medical journal [1]. This review included data from 18 studies. Here are the highlights of this review:
- Fibrates had no impact on stroke risk
- Fibrates did not reduce the risk of dying from cardiovascular disease
- Fibrates did not reduce the overall risk of dying
- Fibrates doubled the risk of having raised blood levels of creatinine (usually a sign of kidney damage)
One positive finding from this review, though, was that Fibrates reduce the risk of ‘coronary events’ (such as heart attacks) by 13 per cent. Some may use this justify the use of fibrates, but what is the impact of this benefit in the real world? The real effect of a 13 per cent reduction in risk depends on the underlying risk to begin with. Basically, if risk is relatively low, than a 13 per cent reduction in risk may not mean much in real terms. Probably a better measure of benefit is the impact of fibrates on ‘absolute risk reduction’. Say the risk of heart disease is reduced from 3 per cent to 2 per cent. Relative risk has fallen by 33 per cent. The real (absolute) reduction in risk is 1 per cent, however.
The fibrate review tells us that 5 years of treatment with fibrates reduces the absolute risk a cardiovascular outcome (e.g. heart attack) by 0.6 – 1.65 per cent. That equates to an absolute risk reduction of about 0.12 – 0.33 per cent per year.
Another way we can express these statistics is as what is termed the ‘number needed to treat’ or NNT. This is basically a calculation of the number of people needed to be treated to prevent one cardiovascular event over a specified period of time. This figure is calculated by dividing 100 by the absolute reduction in risk. So, between 303 and 833 people would need to be treated with fibrates for a year to prevent one cardiovascular event.
Can the prescribing of fibrates really be described as ‘evidence-based’? Only if by ‘evidence’ we mean that hardly anyone who takes them will benefit and no-one will have their life saved. The fact is fibrates are shockingly ineffective overall. The only real winners are the companies that make and sell these things, I reckon.
References:
1. Jun M, et al. Effects of fibrates on cardiovascular outcomes: a systematic review and meta-analysis. Lancet 2010;375(9729):1875-84
