Thursday, April 17, 2014

Yet More on Statins and the Recent Guidelines

Not to make this blog the “Statins Blog” or anything, but since I have been on a tear regarding the unfortunate ACC/AHA guidelines on cholesterol:
--and since my pals at Primary Care Medical Abstracts keep feeding me more ammunition, I wanted briefly to mention two more commentaries that take aim at the guidelines (subscriptions probably required to access).

First, John Abramson, who’s been mentioned here numerous times, and his colleagues wrote in BMJ about the CTT meta-analysis, on which the new guidelines heavily relied. They challenged the rosy picture painted by the CTT by drilling down and recalculating the CTT’s own numbers. The CTT, recall, claimed that if one carefully summed the data from numerous previous clinical trials (none of which showed any reduction in all-cause mortality from taking statins for primary prevention), you could see that statins in low-risk patients save lives and prevent strokes and heart attacks.

So when Abramson’s team went back and crunched the numbers themselves they found, for the populations included in these studies:

  • No significant difference in all-cause mortality
  • You’d have to treat 140 low risk patients for 5 years to prevent one heart attack or stroke
  • The studies overall either fail to report any adverse reactions to statins, or else report adverse reactions at a much lower rate than has been shown in independent (non-manufacturer-sponsored) studies. If the rates of adverse reactions in the independent studies are valid, it is very likely that the chance of suffering an adverse reaction from a statin is notably greater than the chance of preventing a bad outcome.
Skip now to another old acquaintance, John Ioannidis, commenting more recently in JAMA. Ioannidis addresses the new guidelines head-on and is more interested in two issues. One is the total global impact—he calculates as his title suggests that if applied across the world, the “statinization” of humankind would result in at least 1 billion folks being told they needed statins, which would be a huge impact on the health systems of those nations and would conceivably shift resources away from much more desperately needed stuff. The main issue Ioannidis focuses on is the risk calculator employed by the guideline—a calculator that’s new, and that was shown right from the get-go to have significant weaknesses, that have been since confirmed. He gives us the math to show why it’s not very reliable.

Ioannidis takes a step back then and asks the question—we’ve been in the business of trying to calculate risk of coronary artery disease for more than 30 years; so how come we still can’t get it right? He also notes that the decision on the part of the guideline writers that a 10 year risk of developing cardiovascular disease of 7.5% ought to be the cutoff for recommending statins is a completely arbitrary number, unsupported by any empirical evidence—the “right” number could be 2% or 20% for all we know.

What these commentaries have in common is that both point out how messed up the basic data are due to all the major studies being supported by the drug industry, and how serious conflicts of interest contaminate this entire enterprise.

Abramson JD, Rosenberg HG, Jewell N, Wright JM. “Should People at Low Risk for Cardiovascular Disease Take a Statin?” BMJ 347:f6123, 2013.

Ioannidis JPA. “More Than a Billion People Taking Statins? Potential Implications of the New Cardiovascular Guidelines.” JAMA 311:463-464, Feb. 5, 2014.

No comments: