Side Effects of Statins Might Be Overstated: CTT Collaboration

The majority of undesirable side effects linked to statin therapy are not supported by evidence from randomized, controlled trials, with just four of 66 possible side effects attributed to the drug class, according to a large analysis from the Cholesterol Treatment Trialists’ (CTT) Collaboration.

In an individual, participant-level meta-analysis of trials comparing statin therapy to placebo, as well as some studies comparing more-intensive to less-intensive statins, the researchers confirmed that the medications were associated with a small excess in changes detected by hepatic transaminase and other liver function tests—a well-documented side effect—and small increases in the risk of edema and changes in urinary composition.

However, there was no evidence of a causal relationship for the other 62 adverse outcomes, which are all listed as possible side effects in the labels of various statins, report investigators in a study published last week in the Lancet.     

“For the vast majority of medical issues listed as potential side effects in statin packaging, we found that there was no increase in those allocated statin therapy,” Christina Reith, MBChB, PhD (University of Oxford/Oxford Population Health, England), who led the analysis, told TCTMD. “This includes outcomes such as memory loss, sleep disturbance, erectile dysfunction, depression, headache, nausea, fatigue—a huge number of things that people have been worried about, and some of which have been specifically added to labels by regulators.”

As the CTT Collaboration has shown in prior meta-analyses, the LDL-lowering drugs can lead to small increases in the risks of diabetes, particularly in those who might be close to the diagnostic threshold prior to starting treatment, and muscle-related side effects. Over the years, data from nonrandomized sources, such as postmarketing surveillance data and case reports, have linked the drug class to multiple other unwanted side effects. As a result, statin labels, which are known as a Summary of Product Characteristics in Europe, now list a multitude of possible adverse outcomes.

“Statins are off patent and very affordable, but the problem has been that there’s been ongoing confusion and concern, not only in patients, but also among doctors about potential side effects of statin therapy,” said Reith. Those concerns, she added, “have really gained a lot of traction” and this has “translated into some people not being willing to start statins or stopping them even if they’re at high cardiovascular risk.” 

In an editorial, Timo Strandberg, MD, PhD (University of Helsinki, Finland), and Raul Santos, MD, PhD (Hospital Israelita Albert Einstein/University of São Paulo Medical School Hospital, São Paulo, Brazil), say that while most experts in cardiovascular prevention are aware of the safety of statins, they hope these results “get wide publicity and coverage in the lay media.” The hope is that doing so will translate into better adherence and “lead to a profound revision of drug labeling of statins to be more evidence based.”

Kausik Ray, MBChB, MD (Imperial College London, England), who wasn’t involved in the study, agreed that a label revision is warranted. The potential change, which would factor in the new meta-analysis, should also quantify the risk of these rare side effects in absolute terms. “So, even if there is an excess of a few things, the benefits [of statins] outweigh the harms by far,” he told TCTMD.  

Label Changes Needed

The meta-analysis included 19 trials, among them 4S, WOSCOPS, PROSPER, JUPITER, and the Heart Protection Study, with 123,940 participants (mean age 63 years; 72% men) randomized to statin therapy or placebo. The majority of trials (n = 16) tested a moderate-intensity statin, with a median duration of follow-up of 4.5 years. Based on a search of statin product labels—each label differed based on statin type, dose, and formulation—the investigators identified 66 listed adverse outcomes spread across 15 body systems (gastrointestinal, psychiatric, and hepatobiliary, among others).

After controlling for multiple-outcomes testing, statin therapy was associated with an excess risk of abnormal liver transaminases (absolute annual excess risk 0.09%) and other abnormal liver function tests (absolute annual excess risk 0.05%). Similarly, there was a significant, albeit small, excess risk of changes in urinary composition and edema. The rise in abnormal liver enzymes appeared related to statin intensity.  

However, “we didn’t see an increase in more serious liver sequelae, like hepatitis and liver damage,” said Reith. “You might get this increase in [liver function tests], which was previously known, but it apparently doesn’t lead to anything more serious.”

Regarding the increase in edema and change in urinary composition, researchers aren’t certain of the robustness of the finding since the risk wasn’t higher in the four trials with 30,724 participants that compared higher-intensity versus lower-intensity statins. In the editorial, Strandberg and Santos say the edema and slight changes in urinary composition “are obscure but apparently of uncertain clinical significance.”

Science Is Rigorous and Methodical

Over the years, there have been some vocal physicians arguing that statins don’t prevent heart disease or extend life, and come with considerable side effects, when used for primary prevention. In 2013, following the release of the American Heart Association/American College of Cardiology cholesterol guidelines, there was also a high-profile opinion piece in the New York Times advising doctors not to “give more patients statins,” particularly for those without evidence of heart disease.

To TCTMD, Ray noted that observational research, opinion, and randomized, controlled trial data are often lumped together “under the umbrella of science, [but] science isn’t about headlines shouting the loudest, but meticulous methodology [and] rigor.”

“Sometimes the easy answers seem logical, may support our intrinsic biases, and are appealing,” he continued. However, randomized trials are the highest bar that prove beyond reasonable doubt the safety and efficacy of medical therapies, and this threshold must be cleared before they are made available to patients. The new CTT Collaboration meta-analysis, he said, provides the most robust safety data to date and should refute prior thinking around the harms of statins.

Reith stressed the results don’t mean that statin-treated patients won’t develop some of these unwanted outcomes over time, because sleep problems, memory loss, erectile dysfunction, and headaches, as examples, commonly occur in the general population. However, “there’s not good evidence that statins are the cause,” she said.

The CTT Collaboration researchers would like to see revisions to the statin labels, as well as changes in other sources of information, such as the British Heart Foundation and National Health Service websites, to reflect the current evidence.

“We’re hoping that these findings can be conveyed there so that people—doctors and patients—are in a better place to make informed decisions on the basis of the most robust evidence,” she said.

In the UK, the Medicines and Healthcare Products Regulatory Agency recently concluded that statin labels should be updated to better reflect the risk of muscle-related events. That conclusion was based on a prior CTT Collaboration meta-analysis showing that statins led to a small relative increase in mild muscle pain or weakness, a risk that was largely confined to the first year of treatment.