This article is more than 5 years old. Information may no longer be current.
In midlife cognitive decline, roles of estrogen, testosterone remain unclear
Click Here to Manage Email Alerts
For men and women entering midlife, changes in cognitive function can be a natural part of aging. Among Americans aged at least 45 years who were surveyed in the CDC’s 2015 and 2016 Behavioral Risk Factor Surveillance System, 11% said they were experiencing a decline in memory function or mental decline, with half of those reporting limitations in performing daily tasks.
Those early symptoms may or may not be early signs of dementia, and they can differ from mild cognitive impairment defined as an intermediate stage between normal aging and dementia. Adults with mild cognitive impairment have an increased risk for progressing to Alzheimer’s disease or other dementia, with roughly 20% of this population crossing over from mild impairment to a more severe level each year.
Research increasingly suggests that changes in hormone levels during aging, particularly estrogen, are associated with risk for mild cognitive impairment and Alzheimer’s disease, whereas recent studies now suggest that hormone therapy, if given at the right time, is not associated with cognitive decline and might even be protective.
“There has been a historical interest in the effects of hormones on the ‘wiring’ of the brain,” Susan M. Resnick, PhD, a senior investigator with the Laboratory of Behavioral Neuroscience at the National Institute on Aging, told Endocrine Today. “We know from animal models that you can actually change the size of specific brain structures by doing prenatal or early neonatal interventions. We’ve known for a long time that there are organizational effects of hormones on the brain during early development, and also hormone effects later in life, especially during puberty and perhaps during changes like menopause, and we think there may also be some vulnerability to specific cognitive ability.”
For midlife men and women, research shows that the interplay between hormones and cognition is complex and variable and can be influenced by multiple factors, including other endocrine conditions, the severity of menopausal symptoms in women and disrupted sleep.
“It’s not really normal to develop dementia,” Adrian Dobs, MD, MHS, professor of medicine and director of the Johns Hopkins Clinical Research Network, told Endocrine Today. “Although in a normal population there might be some small memory changes, true dementia is not a part of normal aging. That’s where hormone questions come up in two ways: One, how do hormones contribute to this normal aging process, and how do they relate to [mild cognitive impairment] or true cognitive impairment, which would be leading to Alzheimer’s disease?”
Estrogen fears
The interest in hormones, particularly estrogen, and the role they may play in cognition stem largely from the late 1990s, when a series of investigations demonstrated that women prescribed menopausal HT had about a 29% reduced risk for developing Alzheimer’s disease, according to Pauline Maki, PhD, professor of psychiatry and psychology and associate director of the Center for Research on Women and Gender at the University of Illinois at Chicago.
“The reason that was of interest is because this association was unique,” Maki told Endocrine Today. “There is no treatment for Alzheimer’s disease, and anything that can prevent it is going to have a significant public health impact.”
Studies conducted before the 2002 landmark Women’s Health Initiative trial suggesting a reduced risk for dementia with HT in women were supported by strong biological plausibility, including neurotrophic and neuroprotective effects of estrogens and effects of estradiol on metabolic and biochemical pathways implicated in Alzheimer’s disease pathogenesis, Maki wrote in an analysis published in Climacteric in 2012.
However, in contrast to prior studies, findings from the WHI showed that combination HT —conjugated equine estrogens with medroxyprogesterone acetate — doubled a woman’s risk for developing dementia, including Alzheimer’s disease, when therapy is begun at age 65 years or older. HT also did not prevent mild cognitive impairment in trial participants.
In the WHI Memory Study, researchers found that women assigned combination HT were twice as likely to develop dementia vs. those assigned placebo, although the actual number of probable dementia cases was small. Estrogen alone, however, was neutral with respect to cognitive decline in women.
“The WHI Memory Study was done in women aged 65 years and older, so I reassure younger women that those results would not necessarily apply,” JoAnn E. Manson, MD, DrPH, FACE, professor of medicine and the Michael and Lee Bell Professor of Women’s Health at Harvard Medical School and chief of the division of preventive medicine at Brigham and Women’s Hospital and a WHI investigator, told Endocrine Today. “But we don’t have conclusive evidence that taking estrogen early in menopause will delay cognitive decline or dementia. The jury is still out. There is encouraging observational research suggesting women who take hormone therapy have a lower risk for dementia, but those are not randomized trials.”
Randomized trials, to date, have been inconclusive as to whether starting HT early in menopause benefits cognition, Manson said. Both the KEEPS trial and the ELITE trial showed neutral results — no increased risk for cognitive decline with HT, but no clear cognitive benefits for women who start HT early.
“Estrogen therapy should not be used for the treatment or prevention of cognitive impairment,” Maki said. “The data aren’t there. Our conversation is really about what might explain some of the discrepancies. Of course, the hormones are of interest because two-thirds of all Alzheimer’s patients are women.”
Menopause signal
For midlife women, mild cognitive changes often coincide with the menopause transition, a period that can also include symptoms of hot flashes, changes in mood and disrupted sleep patterns. Studies have suggested decrements in verbal memory during perimenopause, a time when hot flashes are common.
In a 2015 study, Maki and colleagues explored whether physiologically assessed hot flashes in 20 midlife women (aged 40 to 60 years) not prescribed HT were associated with greater connectivity in the default mode network, the network of brain regions active during rest. The researchers focused on networks supporting the hippocampus, a region rich in estrogen receptors and previously linked to hot flashes.
Women underwent 24 hours of physiologic and diary-recorded hot flash monitoring, functional MRI and 72 hours of sleep actigraphy monitoring. The study showed that that a greater number of physiologically monitored hot flashes, but not reported hot flashes, was associated with greater activity in the default mode network. The data support differences in default mode network connectivity that distinguish women with varying degrees of hot flash burden, Maki said.
Most of the randomized trials of estrogen in women in early menopause have not looked separately at those with moderate to severe hot flashes and disrupted sleep, according to Manson. Those women, she said, may see a cognitive benefit from estrogen therapy, although that needs to be demonstrated in a study.
“One question that has not been well-answered is, in women with significant hot flashes and night sweats and disrupted sleep longer term, would they benefit from estrogen in reducing risk for cognitive decline?” Manson said.
That is, if estrogen reduces those symptoms and improves sleep quality, could it in turn have a favorable effect on cognition, Manson said.
“Where I see this going in the era of precision medicine is identifying those women for whom persistent menopausal symptoms or other indications of some hormonal vulnerability get treated [with HT], and women who don’t have that vulnerability don’t get treated,” Maki said. “It’s not a one-size-fits-all, but if indeed we are right ... it’s a treatable risk factor for Alzheimer’s disease in women, a sex-specific one.”
Role of testosterone
In men, serum levels of total and bioavailable testosterone gradually decline with age, and these decreases are associated with reductions in muscle mass, osteoporosis, diminished sexual activity and changes in cognition, according to Ronald Tamler, MD, PhD, professor of medicine in the division of endocrinology, diabetes and bone disease at the Icahn School of Medicine at Mount Sinai and an Endocrine Today Editorial Board Member. However, cognitive decline can often overlap with psychological changes, such as dysphoria and depression, making a link between testosterone level and cognition less clear, he said.
“At this point, it looks like correlation, but not causation,” Tamler told Endocrine Today. “We get older and our memory declines. We get older and our testosterone declines. But testosterone supplementation is not a fountain of cognitive youth for memory impairment.”
Men with low testosterone will often report what some refer to as “brain fog,” Tamler said, but he called the observation anecdotal.
“Testosterone does influence the brain — it influences risk-taking and perception of reward — but it’s probably not related to cognitive decline or improvement,” Tamler said. Studies conducted in the early 2000s suggested that testosterone treatment in men may improve cognitive functions that decline with age. In a small, randomized, double-blind, placebo-controlled study published in 2001, Monique M. Cherrier, PhD, director of the University of Washington Memory Health Research Program, and colleagues analyzed the relationship between exogenous testosterone and cognitive abilities in 25 healthy men aged 50 to 80 years. Participants received weekly injections of either 100 mg testosterone enanthate or saline placebo for 6 weeks. Cognitive evaluations were conducted at baseline and 3 and 6 weeks.
At 6 weeks, the researchers observed improvements in cognition for spatial memory, spatial ability and verbal memory in men treated with testosterone vs. baseline scores and vs. men receiving placebo. In treated men, circulating total testosterone increased 116%, whereas estradiol increased an average of 73% due to aromatization of testosterone.
In a randomized, placebo-controlled, crossover study of 44 men with subjective memory complaint and low testosterone levels published in 2016, researchers assigned two groups to 24 weeks of daily testosterone cream and 24 weeks of placebo with a 4-week washout period between regimens. The researchers observed improvements in baseline Mini-Mental State Examination scores after testosterone treatment vs. baseline.
However, the Cognitive Function Trial, an evaluation of cognitive function in all participants of the seven T Trials conducted between June 2010 and June 2014, suggests a different conclusion.
In an analysis published in 2017, Resnick and colleagues evaluated cognitive function in all T Trials participants (n = 788) aged at least 65 years with serum testosterone level less than 275 ng/mL and impaired sexual or physical function or vitality. Researchers randomly assigned men to testosterone gel treatment. or placebo. A subgroup of 493 men met criteria for age-associated memory impairment based on baseline subjective memory complaints and objective memory performance (247 men in the testosterone group and 245 in the placebo group).
Among older men with symptomatic hypogonadism and low baseline testosterone, testosterone treatment for 1 year was not associated with significant improvement in delayed paragraph recall score, differences in visual memory, executive function or spatial ability.
Further sensitivity analyses, designed to assess whether there were subgroups more sensitive to testosterone and cognitive improvement, also showed no association, Resnick said.
“We were surprised when we found no effects of testosterone gel treatment on cognitive function, even though the testosterone levels clearly rose as part of that study,” Resnick said. “Again, there is this disconnect between the observational and small-scale trials and then the large trial of approximately 800 men.”
Tamler agreed that the analysis demonstrated a conflict between smaller vs. larger testosterone trials, and said more research is needed.
“That [analysis] took the wind out of the sails of a lot of people who felt that testosterone supplementation was the fountain of cognitive youth,” Tamler said.
Resnick noted that some of the smaller studies used injectable testosterone, which will have peaks and troughs, and it is possible that the timing of the injection influenced cognitive test results in those men.
“The study is an authoritative, negative answer to the question of whether testosterone improves cognition in men with low testosterone and age-associated memory impairment at baseline,” Marco Marcelli, MD, FACE, professor and acting chief of diabetes, endocrinology and metabolism at Baylor College of Medicine in Houston, told Endocrine Today. “It is clear that the data reported in this study do not support the use of testosterone therapy to improve cognitive function.”
The study had limitations, Marcelli said. Baseline alterations in age-associated memory impairment were modest; therefore, it is still possible that testosterone might improve cognition if used in a patient population with more advanced baseline cognition impairments.
“Similarly, this study does not answer the question of whether testosterone supplementation would be successful if used longer than 12 months or in patients with a testosterone level lower than 275 ng/dL,” Marcelli said. “The presence of a weak signal for improved executive function in the entire population of the T Trials introduces a new question to explore in future studies with different patient populations.”
Message for patients
According to a 2018 update of the American Academy of Neurology guideline on mild cognitive impairment, no high-quality evidence exists to support pharmacologic treatments for the condition. In patients, exercise training for 6 months is likely to improve cognitive measures, and cognitive training may improve symptoms, according to the researchers.
A patient describing symptoms of confusion or memory loss may or may not have true cognitive impairment, Tamler said, and those symptoms can best be assessed by a neurologist.
“The most important thing to do is to take a patient’s complaint seriously, and not dismiss it,” Tamler said. “The patient needs to be acknowledged and heard. Then, the patient needs to be taken on a bit of a diagnostic journey.”
Apart from male hypogonadism, there are other endocrine reasons for cognitive impairment, such as untreated hypothyroidism or vitamin B12 deficiency, according to Tamler. “Sometimes, a good endocrinologist needs to take a step back and act a little more like a generalist and frame the conversation for the patient,” Tamler said. “Testosterone supplementation has benefits, but it also has downsides, and needs to be provided to the right patient.”
Symptoms of mild cognitive impairment, Dobs said, are a common complaint and can be difficult to prove without baseline data of what a person’s memory function was like years or even decades earlier.
“I remember being in college and walking to the refrigerator and saying, ‘Now, what did I come here for?’” Dobs said. “That’s a common scenario, and there is a level of that due to the brain getting occupied with something else. It’s hard to prove very mild cognitive impairment. I try to evaluate and refer most of the patients I have, whether it’s a referral to the memory clinic or neurology, because I can’t measure it in the clinic.”
For most patients, Manson said, the best prescription for brain health is lifestyle modification.
“If you can maintain vascular health, that should lower the risk for developing dementia, and there are several studies that do suggest that lifestyle factors linked to heart health and reduced risk for stroke are also linked to reduced risk for developing dementia,” Manson said. “This includes staying physically active, having a heart-healthy diet, controlling blood pressure and cholesterol, not smoking, and being really attuned to your sleep needs. There is evidence that sleep has an important role in clearing amyloid, being restorative and possibly reducing cognitive decline.
“Sometimes people get discouraged about Alzheimer’s disease, especially when there is a family history,” Manson said. “And, certainly, there are genetic factors that can predispose a person to Alzheimer’s. But, heredity is not destiny. There are things people can do to lower their risk, and a key point to remember is what is good for the heart is good for the brain.” – by Regina Schaffer
- References:
- Cherrier MM, et al. Neurology. 2001;doi:10.1212/WNL.57.1.80.
- Maki PM, et al. Climacteric. 2012;doi:10.3109/13697137.2012.660613.
- Manson JE, et al. JAMA. 2017;doi:10.1001/jama.2017.11217.
- Petersen RC, et al. Neurology. 2018;doi:10.1212/WNL.0000000000004826.
- Resnick SM, et al. JAMA. 2017;doi:10.1001/jama.2016.21044.
- Taylor CA, et al. MMWR Morb Mortal Wkly Rep. 2018;doi:10.15585/mmwr.mm6727a1.
- Thurston RC, et al. Fertil Steril. 2015;doi:10.1016/j.fertnstert.2015.03.008.
- Wahjoepramono EJ, et al. CNS Neurol Disord Drug Targets. 2016;15:337-343.
- For more information:
- Adrian Dobs, MD, MHS, can be reached at the Johns Hopkins Clinical Research Network, Johns Hopkins Hospital, 600 N. Wolfe St., Sheikh Zayed Tower, Suite 328, Baltimore, MD 21287; email: adobs@jhmi.edu.
- Pauline Maki, PhD, can be reached at the Center for Research on Women and Gender, University of Illinois at Chicago, 1640 W. Roosevelt Road, MC980, Suite 503, Chicago, IL 60608; email: pmaki1@uic.edu.
- JoAnn E. Manson, MD, DrPH, FACE, can be reached at Harvard Medical School, 900 Commonwealth Ave., Third Floor, Boston, MA 02215; email: jmanson@rics.bwh.harvard.edu.
- Marco Marcelli, MD, FACE, can be reached at Baylor College of Medicine, Section of Diabetes, Endocrinology & Metabolism, One Baylor Plaza, BCM185, Houston, TX 77030; email: marcelli@bcm.edu.
- Susan M. Resnick, PhD, can be reached at the National Institute on Aging, Laboratory of Behavioral Neuroscience, 251 Bayview Blvd., Suite 100, Baltimore, MD 21224; email: resnicks@mail.nia.nih.gov.
- Ronald Tamler, MD, PhD, can be reached at the Icahn School of Medicine at Mount Sinai, 1 Gustave L. Levy Place, Box 1055, New York, NY 10029; email: ronald.tamler@mssm.edu.
Disclosures: Maki reports she has received a speaker’s honorarium from Mylan. Dobs, Manson, Marcelli, Resnick and Tamler report no relevant financial disclosures.
Click here to read the 
, "Can optimal glucose management reduce risk for dementia in diabetes?"