Hormone Therapy and Dementia: What Do We Know?

Estrogen exerts neuroprotective effects on the brain, and the reduction in endogenous estrogen after menopause increases women’s risk for Alzheimer’s disease (AD), cerebrovascular disease, or both. 

However, there is debate whether estrogen replacement — with or without progestin — is neuroprotective or whether it may in fact increase the risk for AD and other dementias.

“Research on hormone therapy has been ongoing for decades and its findings have not been straightforward,” Yuko Hara, PhD, director of aging and Alzheimer’s prevention at the Alzheimer’s Drug Discovery Foundation, told Medscape Medical News. 

Hara said that initiating hormone therapy (HT) “many years after menopause can harm cognitive function and increase dementia risk.” However, the evidence “is more limited and mixed for the effects of HT on cognitive health when initiated in the early postmenopause period, with studies finding positive, negative, and null effects on cognitive function.”

Critical Window of Time?

At the International Conference on Nutrition in Medicine, held in Washington, DC, on August 15, 2024, sponsored by the Physicians Committee for Responsible Medicine, Amani Meaidi, MD, PhD, of the Danish Cancer Institute, Copenhagen, sounded a warning bell regarding possible elevated risks for dementia associated with HT.

Acknowledging the deleterious effect that hot flashes have on quality of life in perimenopausal women, Meaidi stated that HT is effective in alleviating these debilitating symptoms. But does it come at a price, particularly where dementia or AD is concerned, she asked?

Meaidi reviewed the evolution of approaches to this question. 

Early studies suggested that estrogen replacement could reduce or at least delay the risk for AD, although it might not improve already existing AD. That perspective changed over time, as more data emerged. The Cache County Study, for example, found that previous use of HT was associated with a decreased risk of AD, but no benefit was found unless HT use exceeded 10 years which may represent women who began HT in early menopause, as opposed to more recent initiation. This gave rise to the “limited window of time” concept. 

In 2003, a substudy of the Women’s Health Initiative Memory Study (WHIMS), which included more than 4000 women aged ≥ 65 years, found an increased risk for dementia with use of HT compared with placebo. “Since then, we’ve been trying to map the safety profile of hormone therapy when it comes to protection vs risk,” Meaidi said.

Although the mean age of WHI study participants was 63 years, nearly one fifth of the cohort were aged ≥ 75 years, meaning they were well past menopause when the study started. This left open the possibility that HT, administered earlier in the menopausal transition, might confer a lower risk for subsequent dementia.

Meaidi cited a UK study that compared more than 16,000 women taking HT with over 8000 controls. It suggested that estrogen alone did not increase the risk of developing dementia, although a regimen comprising estrogen and progestin did slightly increase the risk of developing AD. Women who have undergone hysterectomy can use estrogen alone, whereas those with a uterus take progestin to mitigate the risk for endometrial cancer associated with estrogen.

On the other hand, Meaidi said, a Finnish study compared 85,000 women who received a diagnosis of AD with the same number of matched controls without such a diagnosis. The use of oral or transdermal HT was associated with an increased risk for AD, regardless of whether women used estradiol only or an estrogen-progestogen combination. In the words of the study authors, age at initiation of therapy was not a “decisive determinant” of increased AD risk and exclusive use of vaginal estradiol had no impact on AD risk.

Neither observational study was able to obtain a full exposure history of HT for the majority of the study population, Meaidi pointed out. Both included women aged ≥ 80 years, so information about past HT treatment — especially short-term use — wasn’t available in most cases. For this reason, “misclassification” of women who had been treated was likely, which would “dilute” a potential association between menopausal HT use and dementia.

Can We Rely on the Guidelines?

The Menopause Society (formerly the North American Menopause Society) issued a position statement on HT in 2017, with a 2022 update confirming the recommendation to individualize treatment to determine the most suitable type of HT, including dose, formulation, route of administration, and duration of use. HT is recommended preferably for women <60 years of age or within 10 years of menopause onset. 

According to The Menopause Society’s guideline, the benefits of treatment for vasomotor symptoms and prevention of bone loss and fracture should outweigh the risks. As for women ≥ 60 years or those who initiate HT more than 10 years from menopause onset, the guidelines note that the greater absolute risks for vascular disease and dementia make the risk-benefit ratio less favorable. 

Guidance issued by the American Academy of Family Physicians states that “because of the potential risks with long-term use of hormone therapy, clinicians should prescribe the lowest effective dosage for the shortest duration necessary to improve symptoms.” 

Meaidi suggested that, ultimately, the safety even of short-term HT initiated at a younger age lacks robust evidence. She and her colleagues therefore conducted a case-control study to specifically examine this. 

Drilling Down Into Short-Term, Early-Initiation HT

As reported by Medscape Medical News, the study compared 5589 incident dementia cases to 55,890 age-matched controls identified from data drawn from the Danish National Patient Registry (aged 50-60 years in 2000).

Median age at initiation of HT was 52 years for case patients as well as controls, and the median duration of use was just over 3.5 years. Women who had undergone hysterectomy were excluded because bilateral oophorectomy and hysterectomy are associated with an increased risk for dementia independent of HT. 

Prior to the date of dementia diagnosis, 32% of case patients and 29% of matched controls had received combined HT with estrogen and progestin. Among all the combined HT users, 66% had their last treatment more than 8 years before the date of dementia diagnosis. Only 8.7% of case patients were still using HT at the time of diagnosis.

Compared with never-users, women who had received HT had a 24% higher risk of developing all-cause dementia, with increasing durations of use associated with increased risk. Those who used HT for 1 year or less had a 21% increased risk, whereas those who used it for ≥ 12 years had a 74% higher risk. The risk was similar between continuous and cyclic HT regimens (estrogen plus progestin taken daily, and daily estrogen with progestin taken 10-14 days a month, respectively). 

Progestin-only or vaginal estrogen–only HT regimens were not associated with an increased risk for dementia.

Interpret With Caution

Srilatha Raghuram, PhD, Alzheimer’s Association director of journal operations and special projects, expressed concern about concluding that HT increases dementia risk.

Commenting for Medscape Medical News, Raghuram noted that the study population was restricted to Danish registries; this “limits the generalizability and applicability of the findings, which need replication and confirmation in larger, more diverse study populations.”

And one can’t draw conclusions about causality from observational studies, she cautioned. “Brain health is not determined by any one single factor in isolation.”

Meaidi agreed that residual confounding by indication is plausible. “Women using hormone therapy may have more severe vasomotor symptoms and may have a predisposition to both vasomotor symptoms and dementia,” she suggested.

Hara elaborated: “Having severe menopausal symptoms such as hot flashes has been associated with greater brain amyloid pathology, which is the biological marker for Alzheimer’s disease. So, we cannot rule out the possibility that women who have severe menopausal symptoms (who are more likely to take HT than those who have milder symptoms) have a higher risk for dementia, independent of the use of HT,” she said.

Hara also noted that HT carries other risks, including venous thromboembolism, stroke, coronary heart disease, and some cancers. ”Some of these effects on vascular health could contribute to risk for dementia,” she suggested.

An editorial accompanying the Danish study, coauthored by Kejal Kantarci, MD, from the Mayo Clinic, Rochester, Minnesota, and JoAnn Manson, MD, MPH, DrPH, from Harvard Medical School, raised additional concerns.

In particular, increased dementia risk with less than 1 year of HT use is “not biologically plausible.” In fact, they note, two thirds of women have subjective cognitive difficulties during the menopausal transition and may experience a “temporary decline in cognitive processing speed.” These women may be more likely to seek HT compared with women who do not experience those symptoms.

Also commenting for Medscape Medical News, Stephanie S. Faubion, MD, MBA, professor and chair of the Department of Medicine and the Penny and Bill George of the Mayo Clinic Center for Women’s Health, noted that there are a great deal of data on women who undergo premature menopause. “It’s clear that if those women don’t get estrogen, they have an increased risk of dementia,” she said. 

She suggested that this is “all about the timing hypothesis, and the age at which one enters menopause.” Most data “shows no harm associated with HT under the age of 65, and the WHIMS showed an increased risk when you initiate — but not necessarily when you continue — HT after the age of 65.”

Faubion, who is medical director of The Menopause Society, noted that the 2017 and 2022 guideline removed the 5-year duration because it was a “false duration” based on the fact that the WHI trials stopped after 5 years. “There’s no clinical reason for that, nor for putting people on the ‘lowest dose,’ which might not even alleviate their symptoms.” Rather, she said, a woman should be put on the “most appropriate dose for the appropriate amount of time, based on her symptoms and treatment goal.”

She added that the “brain fog” commonly described by women around the menopause transition does not necessarily portend a risk for dementia. “We suspect that these cognitive complaints are probably related to troughs in hormone levels, hot flashes, night sweats, mood disruptions, and sleep disturbances.” HT “might help with cognitive complaints right around the menopause transition, although there’s never been a randomized controlled trial to prove that.”

Promising Research

In their editorial, Kantarci and Manson wrote that brain imaging biomarkers might help identify the effects of HT on dementia pathophysiology at an earlier stage, “making assessment of its influence on dementia risk in trials of recently postmenopausal women feasible.”

Indeed, a 2023 cohort study of almost 2000 women (mean age, 65 years), found that use of HT was associated with higher delayed memory scores and larger entorhinal and hippocampal brain volumes, as detected by brain MRI. These areas are affected early on in the AD pathologic progression. However, benefits were found only in APOE4 carriers, and the earlier the age of HT initiation, the larger the hippocampal volume. The authors regarded their findings as confirmatory of the “critical window.”

‘Think Outside the Box’

Meaidi encouraged practitioners to “think outside the box,” when it comes to treating “the millions of [menopausal] women who suffer on a daily basis” from hot flashes.

She noted that the 2017 clinical recommendations of the Danish Society of Obstetrics and Gynecology encourage a 3-month trial of lifestyle interventions and cardiovascular risk assessment for women with moderate to severe hot flashes. If lifestyle changes prove ineffective, tailored hormonal treatment may be initiated, but only in peri- and early postmenopausal women aged ≤ 65 years. 

Hara added that there are “many lifestyle approaches that are helpful in maintaining optimal brain health.” She cited the recently published Lancet Commission report on dementia prevention, intervention, and care detailing 14 lifestyle risk factors that, when fully addressed, can prevent or delay up to 45% of dementia cases. 

She encouraged taking “concrete steps” to reduce dementia risk by addressing those factors.

Kantarci was involved in an unpaid educational activity on Alzheimer’s disease for Biogen Inc. and is the principal investigator on a study of a PET imaging ligand for Alzheimer’s disease, to which Eli Lilly and Avid Radiopharmaceuticals, donated supplies. She is funded by the National Institutes of Health and Alzheimer’s Drug Discovery Foundation. Hara, Raghuram, Manson, Faubion, and Meaidi disclose no relevant financial relationships. 

Batya Swift Yasgur, MA, LSW, is a freelance writer with a counseling practice in Teaneck, New Jersey. She is a regular contributor to numerous medical publications, including Medscape and WebMD, and is the author of several consumer-oriented health books as well as Behind the Burqa: Our Lives in Afghanistan and How We Escaped to Freedom (the memoir of two brave Afghan sisters who told her their story).