Reference: Front Aging Neurosci. 2023 Oct 23:15:1260427
Practice Point: Starting HRT in midlife - within 10 years of the LMP - may reduce the risk of dementia, tipping the scales in favor of HRT, at least when it comes to memory.
EBM Pearl: In meta-analyses, a “common (or fixed) effect” model assumes that there is one true effect size underlying all included studies. A “random effects” model assumes the true effect could vary from one study to another due to heterogeneity among studies.
Hormone replacement therapy (HRT): the age-old (pun intended) example of weighing the risks and benefits. That scale is hard to interpret though, because much of the data on HRT involves differing studied populations, ages, treatment durations, dosing, formulations, etc. One of the risks or benefits is dementia, which is of particular concern for women who represent 60% of those affected by Alzheimer disease (AD). Moreover, the number of people living with AD is expected to triple by 2050.
A recent meta-analysis published in Frontiers in Aging Neuroscience analyzed data from trials, cohorts, and case series investigating the effects of HRT on memory, using outcomes of AD and/or all-cause dementia. The authors reasoned that mechanistically, the effects of HRT likely depend on duration of use and timing of initiation (when patients start HRT relative to when menopausal symptoms begin), something that has been called the “healthy-cell bias hypothesis”. In response, investigators stratified data by these factors as well as study type and HRT formulation (estrogen alone vs estrogen + progesterone). Data were analyzed using both common effect and random effects models, although due to heterogeneity between studies, the random effects model is most likely to approximate the truth in this meta-analysis (and is what we report below).
The well-executed meta-analysis included six total arms of the two RCTs from The Women's Health Initiative Memory Study (WHIMS) and 45 observational studies. The only RCTs related to HRT and memory enrolled only postmenopausal women aged 65 or older, which is nearly 15 years past the average age of menopause onset, 51. The pooled effect from RCTs demonstrated a significant increase in dementia risk compared to placebo (RR 1.43). Upon subgroup analysis, these results were driven primarily by those taking estrogen + progesterone (RR 1.64, 95% CI 1.20-2.25), with nonsignificant findings in the estrogen-only arm. Because the timing of initiation was so late (>65 years old), data from RCTs were analyzed separately from observational data, which showed an overall significant reduction in dementia with HRT. Hmmm. Let’s break the observational data down further.
- HRT formulation: HRT with estrogen-only was associated with a decreased dementia risk (RR 0.85 [95% CI 0.77-0.95]). Estrogen + progesterone showed a nonsignificant effect.
- Timing of initiation: HRT initiated in midlife showed a risk reduction (RR 0.87, 95% CI 0.79-0.95), whereas HRT not started until 10 years after menopause showed an increased dementia risk (RR 1.07, 95% CI 1.03-1.12).
- Duration of treatment: Patients taking HRT for more than 3 years had a reduction in dementia risk (RR 0.82, 95% CI 0.73-0.92). For those taking HRT for less than 3 years, the reduction was nonsignificant (RR 0.89, 95% CI 0.78-1.02).
- Timing of initiation relative to HRT formulation: In what is portrayed as the pièce de résistance, investigators put it all together with a beautiful, if not slightly misleading figure. Figure 8 depicts a 31% risk reduction for those who initiated estrogen-only HRT at midlife. This effect was neutral when it was taken in late-life. However, for those taking estrogen + progesterone, the results were all nonsignificant. We hesitate to speak of “trends”, but for estrogen + progesterone, there was a nonsignificant risk reduction if started at midlife, and a nonsignificant increase in risk when taken in late life.
When it comes to HRT and dementia risk, the conclusions are primarily driven by observational data because those evaluated in existing RCTs did not initiate treatment during what is claimed to be a critical window. The bottom line for now is that starting HRT as close to the onset of perimenopause as possible seems to reduce the risk of developing dementia. However, this reduction is diminished if progesterone must be taken with estrogen, and the majority of patients will fall into this category. Starting estrogen + progesterone in late-life pretty clearly increases dementia risk, so that should be weighed heavily against any benefits.
For more information, see the topic Alzheimer Dementia in DynaMed.
DynaMed EBM Focus Editorial Team
This EBM Focus was written by Katharine DeGeorge, MD, MS, Senior Deputy Editor at DynaMed and Associate Professor of Family Medicine at the University of Virginia. Edited by Alan Ehrlich, MD, FAAFP, Executive Editor at DynaMed and Associate Professor in Family Medicine at the University of Massachusetts Medical School; Dan Randall, MD, MPH, FACP, Deputy Editor at DynaMed; Nicole Jensen, MD, Family Physician at WholeHealth Medical; Vincent Lemaitre, PhD, Medical Editor at DynaMed; Elham Razmpoosh, PhD, Postdoctoral Fellow at McMaster University; Hannah Ekeh, MA, Senior Associate Editor at DynaMed; and Jennifer Wallace, BA, Associate Editor at DynaMed.