About 67 percent of patients with Alzheimer's disease are women. The decrease in estrogen levels during menopause may make women more vulnerable to Alzheimer's.

Early studies done using animal models suggest estrogen may prevent the formation of beta-amyloid plaques found in the brains of patients with Alzheimer's. Estrogen may also help reduce inflammation and protect the blood-brain barrier.

Recent studies on the effects of estrogen have produced conflicting results, however. Some have found hormone replacement therapy (HRT) had no effect on dementia risk. Some reported it may lower the risk, and still other studies found it had a harmful effect.

Receiving HRT was associated with a 39 percent lower chance of being diagnosed with dementia and less risk of memory problems or functional decline.

To get a clearer picture of whether estrogen helps or hinders the aging brain, a team of researchers from Stanford University and Johns Hopkins analyzed data from two large studies that included more than 21,460 women.

They concluded that estrogen-only HRT may have a protective effect against Alzheimer's.

Women in one data set had been followed for about three to five years starting at an average age of 71. Older women in this group, over 1,950, took estrogen-only HRT because combination therapy with estrogen and progestin may increase the risk of dementia. More than 19,500 said they did not take it.

Autopsies were done to check for three signs of Alzheimer's disease in the brains of almost 3,000 participants in one of the data sets, the National Alzheimer's Coordinating Center. Women in this group who took HRT were less likely to have these signs of Alzheimer's disease in their brain — tau protein tangles, beta-amyloid plaques or neuritic plaques — beta-amyloid plaques surrounded by damaged nerve cells.

Eighteen percent of HRT users had no signs of Alzheimer's disease in their brain compared to ten percent of those who did not take HRT.

Forty percent of those who took HRT had all three signs of Alzheimer's disease versus 51 percent of those who did not take HRT.

After the researchers accounted for age, race, education level, genetics and hypertension, or high blood pressure, they calculated that women who took HRT in this group had 35 percent lower odds of showing signs of Alzheimer's at autopsy.

The other data set involved more than 720 participants in the Alzheimer's Disease Neuroimaging Initiative. These women underwent brain scans or had their blood and cerebral spinal fluid tested for biomarkers of Alzheimer's disease while they were living. Hormone replacement therapy was associated with higher levels of beta-amyloid protein in the blood and spinal fluid, so less of the protein was deposited as plaques in the brain.

When the researchers analyzed the data from both sets, they found that receiving HRT was associated with a 39 percent lower chance of being diagnosed with dementia and less risk of memory problems or functional decline.

It's important to note that many of the women in the earlier data sets had started taking HRT when they were over age 70. The current recommendation for relief of menopausal symptoms is for women to start taking combined estrogen-progestin HRT in their late 40s or early 50s, then stop taking it before they turn 60.

Hormone replacement therapy was associated with higher levels of beta-amyloid protein in the blood and spinal fluid, so less of the protein was deposited as plaques in the brain.

Women should talk to their doctor to see if they are a candidate for HRT and what kind of hormone replacement may be best. Estrogen-only HRT is prescribed only for a woman who has had her uterus removed. The progestin in combined estrogen-progestin HRT protects those with a uterus against uterine cancer.

The current study shows there is an association between estrogen-only HRT and lower dementia risk, but this type of research cannot prove that HRT has a role in lowering that risk. Randomized clinical trials and observational studies will be needed to uncover what causes may underlie this association and to help women make more informed decisions about care.

The study and a related editorial are published in Neurology.