Higher Serum Cortisol in Midlife Tied to Impaired Memory, Lower Brain Volume

Damian McNamara

October 30, 2018

Higher serum cortisol levels in midlife are associated with impaired memory and structural brain changes, including lower brain volume, new findings from the Framingham Heart Study show.

Interestingly, for reasons not yet known, women who had higher serum cortisol levels fared significantly worse than men on the measured outcomes.

"Higher serum cortisol was associated with lower brain volumes and impaired memory in asymptomatic younger to middle-aged adults, with the association being evident particularly in women," the investigators, with first author Justin B. Echouffo-Tcheugui, MD, PhD, write.

Echouffo-Tcheugui said that on the basis of the study's findings, a higher clinical suspicion for early signs of Cushing's disease, including elevated cortisol levels, may be warranted.

The study was published online October 24 in Neurology.

Negative Impact on Heart, Brain

Previous research has shown that long-term elevation of cortisol levels is tied to negative cardiometabolic changes and decreased brain volume. Additional research links such brain volume changes with cognitive decline.

The investigators note, however, that "these studies are mainly clinical studies conducted in small and convenience samples."

They add that previous studies focused on older populations, that they did not always exclude confounders, such as dementia, and that assessment of structural changes in the brain was limited to the hippocampus and limbic areas.

In light of the widespread expression of corticosteroid receptors throughout the brain, the investigators in the current study examined a larger number of brain regions. They also looked beyond the hypothalamus-pituitary-adrenal (HPA), which is directly involved in the regulation of the stress response, said Echouffo-Tcheugui.

The investigators hypothesized that higher blood cortisol levels are associated with worse cognitive functioning, reduced brain volumes, microvascular damage, and alterations in brain microstructure.

To assess a potential link between fasting morning serum cortisol levels, cognitive performance, and brain changes on MRI, the investigators examined data on 2231 participants from the Third Generation Cohort of the Framingham Heart Study. The study population was 47% men, and the mean age was 40 years (standard deviation, 8.6 years).

Initial cognitive testing was conducted between 2008 and 2011 using the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) Word List and the Victoria Stroop Test (VST). For the CERAD Word List, total, recall, and retention scores were evaluated; for the VST, the interference score was evaluated.

Participants underwent MRI scanning at a mean follow-up of 7.7 years. At the time of scanning, the researchers measured cognition and memory using the Delayed Recall component of the Visual Reproductions Test; the Similarities Test from the Wechsler Adult Intelligence Scale; the Hooper Visual Organization Test; Trails A; Trails B; the Wechsler Memory Scale Logical Memory Test Immediate and Delayed Recall; and the Delayed Recall of the Paired Associates Learning Test.

On the basis of these results, the investigators calculated a global cognitive score.

MRI data were acquired using a 1.5T Siemens Avanto scanner (Malvern, Pennsylvania). Sequences included a 3-dimensional T1-weighted coronal spoiled gradient recalled echo acquisition, a fluid-attenuated inversion recovery sequence, and diffusion tensor imaging.

The findings were adjusted for a number of covariates and possible confounders, including age, education, smoking status, body mass index, the presence of hyperglycemia, systolic blood pressure, any antihypertensive treatment, and use of hormone replacement therapy or oral contraceptive pills.

In addition, the researchers looked for an association between serum cortisol levels and sex, APOE4 genotype, and depressive symptoms, using for the latter the 20-item Center for Epidemiologic Studies Depression Scale.

Unexplained Sex Differences

Serum cortisol levels were assessed by tertile. The median (25th - 75th percentile) serum cortisol level was 12.92 μg/dL (range, 9.82 - 17.58 μg/dL). The middle tertile "roughly corresponded to the commonly accepted range of normality in clinical practice," the researchers note.

Compared to the middle tertile, the highest cortisol tertile was associated with lower total cerebral brain volume (β [standard error (SE)] = -0.38 [0.14]; P = .008), parietal gray matter volume (β [SE] = -0.06 [0.03]; P = .046) and frontal gray matter volumes (β [SE] = -0.12 [0.04]; P = .006).

In contrast, the lowest cortisol tertile was not associated with any brain volume measures.

The researchers also assessed associations by sex. Compared to the middle tertile, the highest cortisol tertile was associated with lower total cerebral brain volume among women (β [SE]: -0.73 [0.21]; P = .001) but not among men (β [SE]: -0.07 [0.19]; P = .717).

Echouffo-Tcheugui did not have a definitive explanation for the sex disparity. However, he said, there a couple of possibilities. One is that women are under more stress. He also noted that "there may be biologic reasons," such as a difference in response to hormone exposures.

"Measuring stress is not easy...and stress is not the only reason for a rise in cortisol," he said.

When researchers compared participants with regard to APOE4 status, they found no significant associations with cortisol levels and structural changes to the brain.

Adjusted multivariable analyses showed that the highest tertile of cortisol was associated with poor global cognition (β [SE] = -0.18 [0.05]; P = .001) and with poorer performance on the Hooper Visual Organization Test (β [SE] = -0.08 [0.05], P = .002) and the Trails B/Trails A score (β [SE] = -0.02 [0.01], P = .001) compared to participants in the middle tertile.

Again, the lowest tertile of cortisol was not associated with reduced cognitive function.

Strengths, Weaknesses

The researchers found that, compared to the middle cortisol tertile, the highest cortisol tertile was independently associated with lower fractional anisotropy within tracts covering 13.3 mm3 of the white matter.

The splenium and body of corpus callosum region and the superior and posterior part of the corona radiata were the white matter tracts most strongly associated with cortisol levels.

The investigators found no association between serum cortisol and gray matter density in both overall and subgroup analyses.

In contrast to the lack of brain structure changes, the investigators found that the presence of the APOE4 allele in the highest tertile of cortisol was associated with poorer cognitive performance on Trails A (P = .021) compared to those in the middle tertile. They reported no such association among those without the APOE4 allele (P = .430)

A diagnosis of major depressive disorder was not associated with brain structural outcomes or cognitive performance on the basis of cortisol levels.

The decline in cognitive performance among those with higher cortisol levels may be partially explained as disruption of information transfer by white matter damage, the researchers note.

The study's strengths include its community-based design, its large sample size, the exclusion of participants with dementia, and the adjustment for potential confounders, including accounting for the presence of the APOE4 allele.

Potential limitations of the research include the cross-sectional nature of the study, which prevented the investigators from determining whether changes in HPA axis dysfunction preceded or came after changes in brain volume and cognition. In addition, the single measure of serum cortisol may not reflect long-term exposure.

Generalizability of the findings remains uncertain because the Framingham Heart Study cohort "includes relatively young participants of European ancestry and consequently is not representative of all age groups in the general population of the United States," the investigators note.

Echouffo-Tcheugui wants to conduct further research to assess other measures of cortisol, such as urinary or salivary levels, and to track cortisol levels over time to determine its trajectory.

Novel Findings

"The findings are novel and ought to be pursued both to understand antecedent effects and impacts on symptomatic cognitive impairment in later life," said David S. Knopman, MD, a clinical neurologist at the Mayo Clinic in Rochester, Minnesota, and a Fellow of the American Academy of Neurology.

However, Knopman had several reservations about the research.

"To be honest, I have no idea what these results mean for cognitive aging, and certainly not for specific diseases, such as Alzheimer's, with abnormal accumulation of beta amyloid or neurofibrillary tangles, nor vascular conditions, such as arteriolosclerosis and/or microvascular changes in the brain," he said.

He pointed out that the changes in cognitive functioning and on MRI findings between the middle-aged participants and the patients who were in their 20s "are still both very modest."

He also noted that the study did not address the basis for higher vs lower cortisol levels.

"The 'Pop Science view' would be that it has something to do with stress, but this article did not identify any particular underlying condition that was associated with the variations in cortisol levels," he noted.

Knopman said the findings do not support patients' asking their physicians to check their cortisol levels.

"Thus, I can't relate these observations to some antecedent condition, nor to any conceptual model of brain aging, and I certainly don't think the observations have any relevance whatsoever for clinical practice," he said.

The National Heart, Lung, and Blood Institute supports the Framingham Heart Study. Grants from the National Institutes of Health, the National Institute of Neurologic Disorders and Stroke, and the National Institute on Aging provided additional support. Dr Echouffo-Tcheugui and Dr Knopman have disclosed no relevant financial relationships.

Neurology. Published online October 24, 2018. Abstract

Follow Damian McNamara on Twitter: @MedReporterFor more Medscape Neurology 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.
Post as: