Statins and Your Memory: "Forget" About It?

Christie M. Ballantyne, MD; Vijay Nambi, MD, PHD


J Am Coll Cardiol. 2021;77(25):3157-3159. 

Therapy with statins has been established as standard of care in the prevention of cardiovascular diseases (CVDs). As with any other therapy, the benefit of the therapy must be weighed against its potential risk to help guide clinical utility and patient acceptance. For statins, fortunately, the risk for serious adverse consequences (e.g., rhabdomyolysis) is low. However, statins have been associated with several other side effects, including liver function abnormalities, diabetes, muscular pain and weakness, and changes in memory.[1] As evidence accumulated and one statin (cerivastatin) was taken off the market, the concern related to hepatotoxicity of statins has abated to a large part (except perhaps in those with existing liver dysfunction), and treatment guidelines no longer recommend routine monitoring of liver function. However, the association of statins with risk for diabetes emerged over time and currently seems well established, although for the greater part, this risk is not factored into clinical decision making given the overall benefits of statins in patients at risk for diabetes who have indications for statin therapy. From a patient perspective, statin-associated muscle symptoms (SAMS) and memory changes have consistently been of greater importance, as they affect quality of life. A higher incidence of SAMS has consistently been documented in both observational and randomized controlled statin studies and is a key reason for statin discontinuation in clinical practice. Since the initial description, subsequent efforts have identified ethnicities, statins, genes, and underlying medical conditions associated with a higher risk for SAMS, whereas contemporary studies, including n-of-1 randomized studies,[2] have shown that the prevalence of SAMS may not be as high as described in observational cohorts. The emergence of alternative lipid modification approaches has also helped in the care of patients with SAMS.

Statin-associated changes in memory, however, have been inconsistent, as observational reports of both improvement and deterioration of memory emerged. Clearly, prevention of stroke can help with preservation of memory, and potential benefit in patients with Alzheimer's disease (AD) was noted in observational studies. However, reports and observations of deterioration of memory and slowness in thinking also emerged. Potential mechanisms of interest and concern included extremely low cholesterol level and its impact on the brain; alternatively, statin-specific effects were also considered, especially regarding water- versus lipid-soluble statins and their ability to cross the blood-brain barrier, resulting in subsequent toxicity. Hence, independent randomized controlled trials and other studies embedded in randomized controlled trials were conducted to study the effect of lipid-lowering therapies, including statins, on cognition[3] and for the most part have produced neutral findings, with one study suggesting minor decrements in cognition.[4] However, these studies were limited by short duration of follow-up, younger populations, or use of tests that may not be sufficient to identify smaller changes in cognition. It is on this background that in this issue of the Journal, Zhou et al.[5] report on the effect of statin use on memory and cognition using a battery of tests in older patients on statins and enrolled in the ASPREE (Aspirin in Reducing Events in the Elderly) trial.

Over a median follow-up period of 4.5 years, statins were not associated with incident dementia, mild cognitive impairment, or cognitive change, although there was a trend toward an increase in AD (hazard ratio: 1.33; 95% confidence interval: 1.00 to 1.77; p = 0.05) in 18,846 participants (median age 74 years, 56.4% women, 31.3% on statins). In addition, a significant interaction between baseline cognitive ability and statins was noted: patients in the lowest quartile of baseline cognitive ability on statins had higher hazards for dementia and change in episodic memory, one component of cognitive testing most associated with AD. The lipo- or hydrophilicity of the statin and low-density lipoprotein cholesterol (LDL-C) levels at baseline did not affect the results.

Overall, the analysis was well done, and its main strengths are a large cohort with a battery of standardized tests that allowed the investigators to track both cognition and incidence of dementia and its subtypes over time. Additionally, the investigators were able to look at the question of whether the lipophilicity of the statin affected the changes in cognition. However, there were several limitations, most of which the investigators acknowledge. First, although this was a post hoc analysis (which has its own limitations) of a randomized trial, the randomization was in regard to therapy with aspirin, not statins. Additionally, the statin doses and achieved LDL-C levels over the study period were not known. Furthermore, patients who were treated with statins were more likely to have diabetes, hypertension, chronic kidney disease, and obesity, all of which are known to increase risk for cognitive decline, and, as might have been predicted, statin users therefore had significantly lower scores for global cognition and episodic memory. Finally, the apolipoprotein E genotype was missing in about 35% of the study population, and hence the investigators did not factor this into their analysis; however, a subanalysis would be useful, especially in light of their findings regarding AD.

So where does this study leave us with respect to lipid lowering, statins, and dementia? On the basis of accumulated evidence, lipid lowering in the short term does not appear to result in improvement or deterioration of cognition irrespective of baseline LDL-C levels and medication used. The present study further provides additional information that the lipo- or hydrophilicity of the statin does not affect changes in cognition. However, the potential increased risk for AD, especially among patients with baseline cognitive impairment, requires further investigation. As the investigators point out, possible explanations for their findings pertaining to AD include chance, increased statin use in higher risk patients, and belated initiation of statins. However, another important consideration is the broader question of risk stratification and primary prevention of CVD in elderly patients given that heart failure supplants atherosclerotic CVD as the major CVD at an older age.[6] Hence, efforts to clarify underlying CVD risk in elderly patients are critical, and tools such as calcium scoring[7] and biomarkers may help identify not only high-risk patients but also what cardiovascular outcomes they are at risk for and hence guide appropriate preventive strategies, especially if the concern for statin-related changes in cognition is further substantiated.

Lingering questions such as the one raised by this analysis regarding potential adverse effects of statins in patients with mildly impaired cognition can be answered only in randomized controlled trials in the appropriate age group and population and with appropriate testing and adequate follow-up. However, although 2 such studies are already in progress (NCT02099123 and NCT04262206), each of these trials has design issues related to inclusion criteria, such as diabetes, chronic kidney disease, baseline cognitive function, and duration of prior treatment with statin therapy, that will still leave us with some unanswered questions. In the meantime, practicing clinicians can have confidence and share with their patients that short-term lipid-lowering therapy in older patients, including with statins, is unlikely to have a major impact on cognition.