Transcript: Review Article Ignores Most Statin Side Effects
The Lancet’s November 2016 edition contains a 30 page review article calling for doctors to stop worrying and start loving their statins. The article’s abstract is also massive, which sometimes means that the author is using it to editorialize not to summarize, as is the case here. The gist of the abstract is that some doctors are irrationally fearful of side effects and this irrational fear is putting their patients in danger. The conclusion of the abstract lays it on thick:
“the heart attacks or strokes that may occur if statin therapy is stopped unnecessarily can be devastating.”
This review attracted criticism from of a number of doctors and writers who I respect and I’m going to summarize their concerns before I add a few of my own.
The first is from healthInsight UK, a site who’s mission is to create a discussion forum for professionals who are critical of current approaches to preventing and treating chronic disease. The article points out something I am embarrassed to say I didn’t realize, that when pharmaceutical companies foot the bill, all the research data can be kept secret. In other words, we’re forced to take the word of the authors.
Wow. I wasn’t sure I could believe that so I had to google it. I found that yes indeed study results are considered confidential. Although the NIH just recently said that when they’re paying for it, the data can’t be kept secret, or the NIH won’t pay for the next study.
The next critique is from a man I’ve known for over a decade, Dr. Malcolm Kendrick a GP working in the UK.
Dr Kendrick points out that Lancet article’s author has made a career out of working for the drug companies. Whenever an article critical of statins is published, Sir Collins rushes in to defend statin use. He seems to come as close as he can to accusing those who don’t use statins of being uncaring and perhaps even medically irresponsible.
My take on this article is that it’s also flawed by the failure to consider all of the side effects of statins in it’s analysis of how many will be helped versus how many will be harmed.
Thanks to HealthInsightUK, who does some math the authors should have done, we can better understand that the numbers proposed in the article are 2 percent harm and 15 percent help.
But those numbers are a result of including only three of the possible side effects.
Other established side effects should have been included in his analysis.
But they weren’t because the so called randomized controlled trials cited in the articles did not find evidence of several important side effects. However I’m not convinced all the articles cited are as randomized as they claim to be.
Here’s why. This is one of the trials cited in the Lancet article. This table shows that a full one third dropped out of the trial. We don’t get any information as to why. But its very possible that after starting the drug a side effect developed.
In other words, a run in period is an opportunity to effectively screen out those people who experience side effects. In my view, any study that has a run in period is of no value in determining the potential for a drug to cause side effects. But most randomized controlled drug trials do have run in periods and doctors don’t realize that it means people with side effects were screened out.
This is just one trick statisticians can use to take control of the conversation around benefits and harms of drugs and drown out concerns doctors may have based on basic understanding of science, like biochemistry. For example, the fact that the brain is composed of 15 percent cholesterol and some statins impair the brain’s ability to manufacture this essential chemical. And that’s why
Perhaps the most important [kind of side effect excluded] are the cognitive effects,
which I see in my friends taking statins as well as my patients who have gotten prescriptions from other specialists. Because the symptoms come on so gradually, most people don’t make the connection to the drug and as long as they stay mild will assume they’re simply a result of aging.
Other important side effects that were not built into the mathematical analysis include liver damage, which is why you need blood testing done before you can start statins,
About one in ten thousand suffer kidney damage serious enough to land them the hospital, but we’re not looking for less serious damage and this article concludes that we should.
Muscle damage is discussed in the article, and we’re starting to understand why. Statins actively disrupt the process of normal muscle maintenance and promote atrophy.
So that means there is another side effect not considered by the Lancet, the weakness that results from muscle atrophy.
The heart is a muscle, and its not surprising given the known risk of muscle damage on statins that heart failure is a potential side effect. What did surprise me when I read this is the authors suggest its actually a good thing that statins weaken heart muscle.
Low cholesterol levels have long been associated with increased risk of cancer, and for certain groups of people, like those over age 70, or those with prior cancer, starting statins is more likely to trigger cancer development.
And the last common complication of statin therapy I’ll mention is infection. This study evaluated risk of infection while in the hospital after a stroke. Pneumonia and bladder infections are the most common reasons for extended ICU time and death after a stroke, and those on statins are seven times more likely to experience one of these.
Now let’s return to the Lancet article. I would like to bring to attention to this statement the authors make. Here they say, regarding the side effect of muscle pain, that placebo-controlled trials have shown “definitively” that almost all of the symptomatic adverse events aren’t actually caused by statins because people on placebo get them just as often.
That would be a worthy point if it had in fact been shown definitively. But in the body of the article they only cite a grand total two such trials—hardly definitive. What is far more outrageous is that the group of people supposedly getting placebo, hence placebo controlled, were in reality also getting injections of an experimental cholesterol-lowering drug called PCSK-9. So the claim that this was a placebo controlled trial would seem outright false, although, admittedly, the study description is remarkably unclear.
You can imaging that a busy doctor who reads just the abstract, which sings the praises of statins and overtly recommends that doctors should be prescribing them more often, would be left with the impression that the case for more widespread statin use has been settled. Those doctors who actually read the entire article and been able to compare its contents against the other articles they’ve also actually read in their entirety would come to a very different conclusion.
In my view, statins are only useful for people who are unwilling to change the habits that damage their lipoproteins. Please see links for more information on the website.