Paracelsus' poison

Statins: Side effect paper is retracted

Statin structure. Ian Musgrave

A paper published last year in the British Medical Journal claimed that there were significant risks to taking the lipid lowering drugs called statins as primary prevention measures. The journal has just retracted these claims on the risks of satins. Why is this is so important?

All therapies have side effects, every single one of them (yes, even herbal medicines). What therapists try and do, at every step, is ensure the benefits outweigh the harms. We are able to accept anti-cancer therapies that cause severe gastrointestinal side effects and hair loss because the alternative is death. A headache tablet that caused all your hair to fall out would not be acceptable, even if it was 100% effective at curing headaches.

Of course, with most of our therapies, our choices are not so clear cut, often there is a range of risk from the disease which has to be balanced with the risk of the therapy, there may be some serious side effects, but if these are sufficiently rare you can risk trying the drug if it has substantial benefits (and stop it if the serious side effects turn up).

Coronary heart disease is the biggest killer of people in Australia; however it has been steadily falling over the past 20 years, due to a combination of preventative measures (getting people to stop smoking for example) and therapies. These therapies include targeting known risk factors for coronary heart disease including high blood pressure and high blood lipid levels.

It was clear early on that treating people with severe high blood pressure was beneficial, but it was not so clear that treating mild high blood pressure was. Now it is clear that treatment for mild high blood pressure outweighs any risk of treatment.

The same thing goes for statins; these drugs target blood lipid levels (although they may have other effects of benefit in cardiovascular disease). For people who already have cardiovascular disease and/or who have already survived a heart attack, treatment with statins is known to be very effective.

For primary prevention, that is giving statins to people who do not already have cardiovascular disease, the effect was not so clear cut. Certainly people who had a range of risk factors for developing cardiovascular disease benefitted, but did people with lower risk benefit?

Enter the 2013 Cochrane Collaboration review of statins for primary prevention of cardiovascular disease (full report here). It analysed the results from 11 clinical trials that looked at statins in people at low risk of developing cardiovascular disease. They concluded:

Reductions in all-cause mortality, major vascular events and revascularisations were found with no excess of adverse events among people without evidence of CVD treated with statins.

These reductions are much less than those found for people who have significant risk factors, or who already have cardiovascular disease. But the Cochrane paper found that small reductions cardiovascular disease out weighted the risks of side effects.

This is important, as while most of the risks are not serious (eg constipation), there is a small but real chance of developing diabetes, and a very rare chance of developing rhadomyolysis, which can cause muscle tissue to break down. So we want to be certain the incidence of serious side effects is low compared to the benefits of statins.

Enter the BMJ paper. This had two parts; in one part they re-analysed one of the key studies that the Cochrane collaboration report relied on, the [Cholesterol Treatment Trialists]’ (CCT) Collaboration, the researchers reanalysis of the CCT data suggest that low risk patients (people who had a 10% chance of having an adverse cardiovascular event over a five year period), did not actually benefit from statins. While this may be true for the CCT study, looking at all trials as a whole suggests that there is an overall benefit.

The finding that the CCT trial suggested statins were not effective in low risk populations (as opposed to the overall picture when you look at all trials) was not what set the cat amongst the pigeons, it was the claim that the incidence of serious adverse effects was much higher in statin users, stating that muscle pain was 50% higher and the incidence of diabetes as 10% higher. Overall, the paper claimed that the rate of significant side effects in statin users was between 18-20%.

This is the claim that is being retracted.

It turns out the figures came from analysis of a separate uncontrolled clinical trial and the analysis that lead to the high claims of side effects was incorrect. But the damage was done, many people focused on the “high levels of side effects” and clamoured to be taken off the drugs, regardless of their risk of heart attack or if the drug was being used from primary prevention, or secondary prevention after a heart attack.

The report exaggerated the true rate of serious side effects, but what is the true rate of side effects? Some idea of the risks (compared to placebo) can be gained at this side effect resource site. But the key issues are diabetes, liver damage and muscle damage. One study that looked at this carefully showed an elevated risk of diabetes of 0.5% (much lower than the 10% quoted in the BMJ paper) and no significant increase in liver damage of muscle damage (see also my article on the risk of diabetes in statins).

While developing diabetes is not a trivial thing, this has to be balanced against the significant drop in deaths found as well.

The bottom line is that statins are effective in treating people who already have cardiovascular disease, and have high risk of cardiovascular risk. They are less effective in treating people at low risk, and any treatment decision should be made carefully. They shouldn’t be handed out like lollies, and the headlines saying “statins are safe” are just as misleading as the orighinal article. But in low risk people the risk/benefit mix is nowhere near as serious as the BMJ paper made out.

While the claim of higher side effects has been retracted, what will happen to the paper? It is unclear if the paper itself will be retracted, or it will be amended to reflect the true state of side effects.

Either way, it may take some time to overcome the damage these incorrect claims have had on prevention of cardiovascular disease.