Re: Statins and The BMJ: Let's start again
I am pleased to see that the review panel requested that further contributions to the statin debate be channelled through medical media and not the lay press. In my view, Professor Collins’ initial unscientific resort to the newspapers disqualifies him from undertaking any further data analysis, something I have already criticised in a rapid response to the BMJ. It would have been ridiculous to withdraw the original article on the basis of one small (in context) error but, just as Roy Meadow was unjustly pilloried for a similar type of error, mud sticks. The Cochrane Collaboration remains the only completely independent potential reviewer that cannot be criticised by one of the existing parties to the debate. Or does it? Nigel Hawkes’ article raises some doubts.
There is no doubt that some muscle “side-effects” are coincidental. I concluded that from an observational study of statin users who contacted me following an article of mine in the "Daily Mail". Neither do I doubt that statins are the only cause of muscle side-effects – other cholesterol-lowering drugs may do it. The principle of re-challenge may be important. In my own case I re-challenged with a whole range of different statins once the apparent side-effect of the previous one wore off. In each case, though with variable intervals, the side-effects recurred and were accompanied by biochemical disturbance (a rise in muscle enzymes). Since stopping all anti-cholesterol therapy they have never recurred. This is a pattern I have had patients report to me also. I confess that when I also developed severe myalgia (and a significantly raised creatine kinase) on ezetimibe I was weary of science and did not re-challenge.
I still maintain that we may be looking through the wrong end of the telescope. Cholesterol levels may be an irrelevant epiphenomenon, or at best a marker for treatment compliance. We should re-define cardiac risk omitting cholesterol levels from the algorithm. The effect of statins may be (indeed in my view is much more likely to be) a function of their anti-inflammatory properties. If it is true, then the C-reactive protein level would be a better biochemical marker. The hypothesis would also explain the apparent paradox of a rise in cholesterol when active rheumatoid arthritis is treated, despite such treatment reducing the already substantially elevated cardiac risk. Likewise the relation between side-effects and coenzyme Q10 levels may be critical in determining whether there is a particular group of people who should not take statins because they are at risk of side-effects, and this requires investigation. Given the current controversy, investigation of these hypotheses may distract folk from hysterical polemic.
Of course, for as in Adam all die, reducing cardiac causes of death increases the deaths from other things…
Competing interests: I have suffered from significant muscle side-effects from cholesterol-lowering drugs