Into the Fray: AHA's First Scientific Statement on Statin Safety

Patrice Wendling

December 12, 2018

In its first scientific statement specifically aimed at reviewing statin harms, the American Heart Association (AHA) concludes that the benefits of these commonly used drugs greatly outweigh the risks.

"With the exception of hemorrhagic stroke and the possible exception of newly diagnosed diabetes mellitus and some cases of autoimmune necrotizing myositis, statin adverse effects can almost always be reversed by stopping treatment. In contrast, an MI or ischemic stroke permanently damages an individual's heart or brain and can be fatal," the committee said.

"Thus, in the patient population in whom statins are recommended by current guidelines, the benefit of reducing cardiovascular risk with statin therapy far outweighs any safety concerns."

One in four Americans older than 40 years of age are taking statins, and about 10% will discontinue statin therapy because of adverse-effect symptoms or concerns, according to the report, published online December 10 in Arteriosclerosis, Thrombosis, and Vascular Biology.

"This is one of those things that is very high on people's minds, probably because so many people are on these medications," said Larry B. Goldstein, MD, a neurologist at the University of Kentucky, Lexington, and a coauthor of the scientific statement. "It's a big deal and despite all of the studies that have been done, these issues are still out there."

On the highly controversial issue of muscle symptoms, the report states that "there is little if any difference (at most 1%) in the incidence of muscle symptoms between the statin and placebo" in double-blind randomized controlled trials (RCTs) in a broad array of patients.

It puts the risk for statin-induced myopathy, including rhabdomyolysis, at less than 0.1% of patients at maximal recommended doses. Myopathy is defined as unexplained muscle pain or weakness accompanied by increases in creatine kinase (CK) above 10 times the upper limit of normal (ULN).

"There is increasing appreciation of the role of patient expectations of harm as the cause of muscle and other symptoms in statin-treated patients," say the authors, who suggest these expectations are generated from clinician warnings about the risk for rhabdomyolysis and negative media coverage.

Commenting on the report, Professor Jane Armitage, FRCP, professor of clinical trials and epidemiology at Oxford University, United Kingdom, said, "It would be wonderful if it caused a positive upswing in people taking up statins, but I think that's unlikely. But at least it means when there is bad press coming out about statins, there's somewhere where people can go to read an authoritative, balanced account of what the data show."

Armitage, who served as a reviewer of the 44-page document, said the committee was very fair in its assessment of the evidence and included randomized and high-quality observational data.

She added, "I think the real strength of the document is that it absolutely focuses on the randomized evidence, but I think that will be cause for comment."

Armitage coauthored a 2016 meta-analysis that found a similar risk for muscle symptoms without significant CK elevations over 5 years using published data from RCTs supplemented with unpublished data from industry-sponsored statin trials. The analysis drew sharp criticisms of bias and ignited calls for disclosure of the unpublished patient-level data, held by the Cholesterol Treatment Trialists' (CTT) Collaboration.

"A document on statin safety is necessary but I'm not sure this is moving the ball forward," said Rita Redberg, MD, professor of medicine, University of California, San Francisco, who is among those seeking release of the raw data.

"If you look at the trials, they do very little collection of adverse-event data; what they concentrate on is CPK rises, but most people who have muscle problems from statins don't have CPK rises, so that's not a very useful number to give," she said. "What we really need is the number of people who have actual muscle problems, which in observational data has been estimated to be 20% to 30%."

Redberg, whose many reports on statins are not among the AHA report's 369 references, described the 0.1% risk for statin-induced myopathy as a "big red flag" for the document. She highlighted a recent modeling study that suggests that risk thresholds in clinical practice guidelines are too low, largely because they fail to take into consideration the full range of adverse events associated with statins.

Armitage said the CTT is in a good position to pick up any adverse events that might have been missed, but "as I say, they are likely to be rare or only a very small signal."

Provisional analyses will be out "as soon as possible" from the CTT, which recently published a protocol for analyses of pooled patient-level adverse events from statin RCTs, but she would not elaborate on the timing. Release of the raw data has been promised for years, observed Redberg.

Paul Thompson, MD, chief of cardiology, Hartford Hospital, and professor of medicine, University of Connecticut, said the limitation of any scientific statement is that it relies on the available evidence and that long-term follow-up is lacking.

"For example, the diabetes issue is a real side effect of statins, but it's a very low frequency, primary shown from the JUPITER trial, which was a study that only went for about 2 years," he said. "So what happens with those people who don't get treated for 2 years, but get treated for 10 and 20 years? There are a lot of things that we don't know and those limitations are present in a statement like this."

The risk for statin-induced newly diagnosed diabetes is about 0.2% per year, the AHA statement concludes, but it notes that the estimate depends on the underlying risk for diabetes in the population studied and will be greater in patients with prediabetes or clinical characteristics, such as metabolic syndrome.

The risk for statin-induced severe liver toxicity is about 0.001%, whereas dose-dependent asymptomatic elevations in transaminases above three times the ULN are more common, occurring in about one in 100 people in clinical trials. The report notes these are usually transient and typically not associated with signs or symptoms of liver disease.

Thompson said a scientific statement on statin safety is needed because many people are concerned that patients who could benefit from statins are not taking or staying on these lifesaving drugs.

"I was a reviewer on all three drafts and it was originally written almost to say statins don't do anything," he said. "I think the final document is well balanced, but it took some pushing to get to that approach."

Beyond Muscle Symptoms

Although the reviewers found no increased risk for hemorrhagic stroke with statin use in primary prevention patients, interpretation of data from secondary stroke prevention studies suggests an increased risk. However, "the absolute risk is very small, and the benefit in reducing overall stroke and other vascular events outweighs that risk."

There was no convincing evidence for a causal relation between statins and peripheral neuropathy, cognitive dysfunction, sleep disturbances, Alzheimer's disease, Parkinson's disease, cancer, cataracts, tendonitis, or tendon rupture.

Although statin-related muscle symptoms might garner most of the attention, Goldstein said he expects pushback on the document's stance on statins and cognitive function. The US Food and Drug Administration issued a warning in 2012 that statins might be associated with memory loss or confusion, although a subsequent review found no link between statins and cognitive decline.

Interactions, Patient Populations

The new scientific statement includes a chart on principal drug interactions that increase the risk for myopathy, usually through higher plasma concentrations of statin or active metabolites.

"The only established case in which statins affect the action of another drug is warfarin (and potentially other vitamin K antagonists), via an unknown mechanism," it says.

On the basis of randomized cardiovascular outcomes trials in patients 65 years and older treated with statins for about 3 to 5 years, the reviewers conclude there is no evidence that statins are unsafe in older patients, but caution that the risk for myopathy/rhabdomyolysis, although rare, could be about twice that in younger patients.

Because of multiple comorbidities and concomitant medications in older patients, "clinicians must carefully evaluate benefit versus risk of statin therapy, including the potential for drug interactions, priorities of care, and patient preferences," they state.

Asked about the quality of the evidence, Goldstein pointed out that "this is a scientific statement, it is not a guideline. We do not make recommendations per se with formal ratings on the quality of evidence or on the level of strength of the recommendations."

Nevertheless, he said, they performed an exhaustive review of all the available data, "such as it is," and discuss the relative issues with each particular data source.

Goldstein said that during the creation of the report, he was unaware that the AHA/American College of Cardiology (ACC) was simultaneously preparing its new lipid guidelines, but that the two documents are complementary.

"This needs to be taken into context with formal guideline recommendations, such as the recently published AHA/ACC guidelines for lipid management, which also consider risk and benefit," he said. "This is almost like an adjunct."

Goldstein reports no relevant conflicts of interest; information on coauthor disclosures is listed in the paper.

 Armitage reports being a coapplicant on a grant to the University of Oxford from The Medicines Company for investigation of a lipid-lowering agent, and previously was a coapplicant on grants to the University of Oxford from Merck for lipid-lowering studies, including statins, and from Bayer, Mylan/Abbott/Solvay for the ASCEND study. She also reports previously serving as a principal or coprincipal investigator of the Heart Protection Study, SEARCH, and HPS2-THRIVE studies, which all included statins and were funded by grants from Merck. Redberg reports no relevant financial conflicts of interest.

Thompson reports serving as a speaker for Regeneron, Sanofi, Amgen and Amarin; serving as a consultant for Amgen, Regeneron, Esperion, and Sanolfi; and receiving research support from Sanofi, Regeneron, Esperion, Amgen, Dalcor, and Amarin. He also reports holding stock in AbbVie, Abbott, CVS, General Electric, Johnson & Johnson, Medtronic, Myokardia, and Sarepta; and having provided legal consultation on exercise-related cardiac events and statin myopathy.

Arterioscler Thromb Vasc Biol. Published online December 10, 2018. Article

Follow Patrice Wendling on Twitter: @pwendl. For more from | Medscape Cardiology, follow us on Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.