No Link Between Statins and Cognitive Decline

Batya Swift Yasgur, MA, LSW

November 22, 2019

A new longitudinal study suggests that statin therapy may not be associated with a decline in cognition and, in fact, may confer some protection for individuals with heart disease or diabetes who are at risk for dementia.

Australian investigators followed more than 1000 older adults over a 6-year period, assessing memory and global cognition at 2-year intervals. They used MRI to evaluate total brain volume, as well as the volumes of the hippocampus and parahippocampus.

They found no difference in the rate of decline in memory or global cognition between statin users and nonusers, nor did they find differences in brain volume changes.

When statin treatment was initiated during the observation period, it was associated with a reduction in the rate of memory decline.

"We found that the slow decline in memory observed with the aging in older people was the same in people who took statins, compared to those who did not," lead author Katherine Samaras, MD, PhD, head of the Clinical Obesity, Nutrition and Adipose Biology Laboratory at the Garvan Institute, Sydney, Australia, told | Medscape Cardiology.

"In those people with dementia risk factors, such as heart disease, the decline in aspects of memory was slower over the 6-year period," said Samaras, who is also an endocrinologist at St. Vincent's Hospital, Sydney.

The study was published online November 18 in the Journal of the American College of Cardiology.

Case Reports

"Consumers and doctors have been concerned for some years that statins may cause memory loss [and] we are all concerned about case reports of people experiencing memory or cognitive changes," Samaras said.

These reports prompted the US Food and Drug Administration to require a black box warning on statin drugs, she noted.

"There are good randomized trial data showing no detrimental effects of statins on limited measures of cognition; however, this is the first study to comprehensively evaluate cognition over four cognitive domains and for as long as 6 years," she added.

Participants were drawn from the Sydney Memory and Ageing Study (MAS), a longitudinal observational study of cognition in community-dwelling Australians without dementia who were aged 70 to 90 years (n = 1037; 98% white; mean age, 79 ± 4.8 years).

Every 2 years during the study period, participants completed an array of standardized questionnaires regarding medical conditions and sociodemographic factors and underwent an examination that yielded information regarding weight, height, blood pressure, and diagnoses such as type 2 diabetes mellitus.

Participants were divided into four categories:

  • Statin ever-users vs never-users

  • Continuous statin users during observation vs never-users

  • Users of specific statins (simvastatin, pravastatin, and atorvastatin) vs never-users

  • Those who initiated statin use during the observation period

Of the total number of participants, there were 397 never-users and 642 ever-users, including 99 individuals who initiated statin treatment during the study period.

The primary outcome measures were cognition (measured using a comprehensive assessment of global condition that evaluated processing speed, language, visuospatial ability, and executive function) and memory.

Apolipoprotein E (ApoE) genotype was assessed, together with laboratory measures such as lipid and plasma glucose levels.

Participants underwent structural brain MRI and volumetry at baseline and at 2 years. These assessments focused on total brain volume and hippocampal and parahippocampal volumes. Complete data were available for 573 participants.

Covariates used in linear mixed modeling analyses included age, sex, education, conditions such as heart disease and hypertension, smoking status, and ApoE-epsilon 4 (ApoEε4) status.

Protective Interaction?

At baseline, the average duration of statin use was 9.1 years; 68% of participants were continuous users.

Statin ever-users differed from never-users in that they were slightly younger and tended to have a higher body mass index. In addition, in these patients, heart disease, diabetes, stroke, and hypertension were more prevalent.

The linear mixed model analyses, adjusted for covariates, showed that at baseline, statin ever-users were similar to never-users in memory and global condition and also in results of specific memory tests.

No significant differences in these measures between ever-users and never-users were detected over the 6-year observation period.

Interestingly, for those who used statins continuously during the 6 years, baseline performance in memory and global cognition was statistically significantly higher, although the rate of decline in these parameters was similar to that of the never-users.

These findings were consistent when each statin was analyzed separately.

The 99 participants who initiated statin treatment during the observation period exhibited "attenuation" in the rate of memory decline over subsequent measures, as determined on the basis of test-wise significance level (B = .066; P = .038).

In particular, among those who initiated statin therapy, the rate of decline in the domain of long-delayed recall (RAVLT-7) was attenuated (B = .092 P = .002).

These findings remained unchanged, regardless of the presence or absence of dementia risk factors.

A "protective interaction" was observed between statin ever-use, heart disease, and 6-year slower decline in RAVLT-7 score, although among participants who did not have heart disease, declines in RAVLT-7 total scores were comparable for ever-users and never-users.

A similar protective interaction was found between ApoEε4 carriage, ever-use, and the rate of decline in RAVLT-7 score, whereas among noncarriers, the rate was similar for ever-users and never-users.

Total brain volume, hippocampal volume, and parahippocampal volume at baseline showed no differences either at baseline or at 2 years when adjusted for a variety of covariates.

Moreover, three-way analyses that included dementia risk factors as well as diabetes found that diabetes moderated the relationship between statin use and change in parahippocampal volume over 2 years (P = .016). Volume was comparable for those with dementia who did not have diabetes, but the rate of decline was slower among those with diabetes plus dementia risk factors, compared to never-users (B = −.584, P = .039), the authors write.

"People with diabetes have a three- to fivefold higher risk of dementia, but those with diabetes taking a statin had less decline in volume and in some of these specific brain regions, compared to people with diabetes who had never taken statins," Samaras commented.

Message of Hope?

Commenting on the study for | Medscape Cardiology, Constantino Iadecola, MD, Anne Parrish Titzell Professor of Neurology and director and chair, Feil Family Brain and Mind Research Institute, Weill Cornell Medicine, New York City, noted that there are "some suggestions that statins may lead to increased cognitive impairment, and this paper suggests that in a well-controlled study, this may not be the case."

In fact, the data "suggest that, although not proven, statins may be beneficial to preventing cognitive impairment, which feeds into another line of research that has shown that use of statins may prevent or delay Alzheimer's disease from setting in," said Iadecola, who is the coauthor of an accompanying editorial.

The study may therefore be a "message of hope to the Alzheimer's disease community, because there is currently basic scientific evidence suggesting that statins may help the brain get rid of amyloid," he stated.

"Consumers who take these medications and doctors who prescribe them need to be able to discuss if there are any untoward adverse effects," Samaras added. "This study helps expand our evidence base and offer reassurance for those patients who stand to benefit from statin therapy but express concern about media reports."

The study was supported by the Australian Government's National Health and Medical Research Council. Samaras and Iadecola have disclosed no relevant financial relationships. Disclosures for other authors are listed on the original article.

J Am Coll Cardiol. Published online November 18, 2019. Abstract, Editorial

For more Medscape Cardiology news, join us on Facebook and 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.