I’ve had a few emails today alerting me to reports of a study concerning the use of statins in healthy individuals. The study in question is a meta-analysis (groping together of similar studies) of statin trials . Part of this meta-analysis involved assessing the impact of statin therapy in individuals deemed to be at relatively low risk of cardiovascular events such as heart attacks and strokes. One of the stand-out findings of this study is that statins led to a statistically significant reduction in risk of ‘major vascular events’. This was even true for individuals at less than 10 per cent risk of vascular events over a 5-year period. This has led to the suggestion that statins used might be widened to even people at low risk of cardiovascular problems.
Before we swallow this idea, though, it is perhaps a good idea to see just how effective statins were found to be in this meta-analsysis. First of all, what is meant by ‘major vascular events’? Actually, this is a term that includes many different potential outcomes including fatal and non-fatal heart attacks and strokes and ‘revascularisation’ procedures (such as placing tubes called stents in the coronary arteries). When a lot of different outcomes are grouped together, it makes it much more likely that a ‘statistically significant’ results will emerge.
When the outcomes are narrowed a little, the results are less impressive. For example, when we look at risk of death from any vascular event (a heart attack or stroke), we find that statins did not reduce risk in individuals deemed to be at low risk (<10 per cent over 5 years). This, by the way, was even true for those who had known vascular disease.
The ‘positive’ findings from this study have, as is often the case, been expressed as reductions in relative risk. The risk of vascular events overall was 21 per cent lower for each 1 mmol/l (39 mg/ml) reduction in levels of low density lipoprotein cholesterol (LDL-C). However, when overall risk is low, then a relative risk reduction might not amount to much in real terms.
We’re told by the authors this meta-analysis that treating with statins prevented 11 major vascular events for every 1000 people treated for a period of 5 years. Put another way, 91 people would need to be treated for 5 years to prevent one major vascular event. Or in other words, only about 1 per cent of people treated with statins for 5 years will benefit (and about 99 per cent won’t).
Overall, lowering LDL-C by 1 mmol/l was found to reduce the risk of death by 9 per cent over a 5-year period. Again, this might sound like a positive finding to some, but the actual reduction in risk of death was 0.2 per cent per year. What this means is that at this level of cholesterol reduction, 500 individuals would need to be treated with statins for a year for one person to have his/her life saved.
The authors of this meta-analysis give us some soothing reassurances about the fact that the benefits of statins vastly outweighing the risks of adverse events such as myopapthy (muscle pain and weakness). They quote of the excess incidence of myopathy as 0.5 cases per 1000 people over 5 years. However, the source they quote is based on diagnosing myopathy once the marker for muscle damage (creatine kinase) is at least TEN TIMES the upper limit of normal. Many individuals will have significant pain and weakness with much lower levels of creatine kinase. Statins are also linked with adverse effects on the liver and kidneys, and increase risk of diabetes too.
Despite the very positive interpretation of the data by the study authors and the media, this meta-analysis shows us again what previous evidence has revealed: statins are highly ineffective in terms of improving health and saving lives. And their risks are generally downplayed.
Collectively, the authors of the meta-analysis are referred to as the Cholesterol Treatment Trialists’ (CTT) Collaborators, including researchers from Clinical Trial Service Unit and Epidemiological Studies Unit at Oxford University. The conflicts of interest statement which accompanies this paper informs us that: “Some members of the writing committee have received reimbursement of costs to participate in scientific meetings from the pharmaceutical industry.” I suppose this may account, at least in part, for a data interpretation that appears so heavily biased towards statins.
Cholesterol Treatment Trialists’ (CTT) Collaborators. 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. The Lancet epub 17th May 2012