http://www.the-scientist.com/yr2001/oct/research3_011015.html
The Scientist 15[20]:21, Oct. 15, 2001

Estrogen Replacement and Cognition: Ready for Prime Time?
Definitive answers about ERT effects are down the road.
By Harvey Black

While estrogen replacement therapy shows promise in helping post-menopausal women preserve important cognitive abilities such as memory, its effectiveness is still being questioned.  In studies at the National Institutes of Health and at the University of California, Los Angeles, researchers have demonstrated that in some women, this hormone alters brain blood flow and improves performance on certain mental tests.  But other studies are not as definitive, suggesting that improved cognitive abilities could be associated with a decrease in menopausal symptoms.  "The epidemiologic data we have is not that mature," says Stanley Birge, clinical director of the Older Adult Health Center at Washington University.  "But I think if you add up the negative studies and the positive studies, it does fall to the side of recommending.  It probably is effective in preserving the brain."

But don't advise treatment right now, some researchers say.  "Not yet," says Pauline Maki, an investigator with the National Institute on Aging.  "There haven't been any [looks] at large numbers of women on cognitive outcomes."  Natalie Rasgon, assistant professor of psychiatry at UCLA and director of the menopause-related mood disorders research program, shares Maki's opinion.  "It will take us some, probably a few more years to tease it out.  But as a researcher I believe there are niches for estrogen.  We just haven't hit on them," she says. 

A  Medscape (free registration required) article  from The Brown University Geriatric Psychopharmacology Update considers the evidence for the effects of not only estrogen but testosterone on brain function

Extracts from http://womenshealth.medscape.com/Manisses/GPU/2001/v05.n09/gpu0509.01/gpu0509.01.html#1
Can Sex Hormones Protect Against or Treat Alzheimer's?
[Brown Univ Geriatric Psychopharm Update 5(9):1-3, 2001. © 2001 Manisses Communications Group, Inc.]

During the last ten years, a significant number of studies have been conducted that explore the possibility of a connection between sex hormones and dementia.

In spite of numerous clinical trials [some of which are described below}, definitive findings that support or question the role of estrogen in preventing, delaying onset or diminishing symptoms of Alzheimer's disease have not been reached. 

Estrogen: Hope or Hype?
Since estrogen promotes neuronal sprouting and enhances cholinergic activity in the brain, researchers rationalize that its use may prevent or adequately treat Alzheimer's. In addition to having anti-inflammatory and anti-oxidative properties, estrogen has been found to lower apolipoprotein E levels. Also, human studies have shown that estrogen appears to increase cerebral blood flow and glucose metabolism in a several areas of the brain, all of which serves to stave off Alzheimer's disease.

Speaking recently at the American Association for Geriatric Psychiatrists annual meeting, Kristine Yaffe, M.D., assistant professor in the Department of Psychiatry, Neurology and Epidemiology at the University of California, San Francisco, noted that seven small, published trials of estrogen treatment for women diagnosed with Alzheimer's disease have produced conflicting evidence regarding the potential benefits of the hormone. Four of the studies were randomized and placebo-controlled, two  reported improvement on some, but not all cognitive tests and two studies determined that no benefit existed when estrogen was compared with placebo on any of the tests.

What about Testosterone?

Researchers have noted that males between the ages of 60 and 70 are not as apt to develop Alzheimer's disease, as are women in this same age range. But by the age of 84, the disease begins to affect both men and women equally. Since men typically secrete more of three androgens -- testosterone, methyltestosterone and epitestosterone -- than women, researchers have pursued the theory that these male hormones might offer neuroprotection against the onset of Alzheimer's disease. 

To date, there is little information available regarding the effects of androgens in the
brain.....

Further commentary at http://www.the-scientist.com/yr2001/oct/research3_011015.html

Postmenopausal Estrogen Replacement Therapy and the Risk of Alzheimer Disease at http://archneur.ama-assn.org/issues/v58n3/rfull/noc00173.html#r21
Minor extracts:
Background 

Previous studies have examined the relation between  postmenopausal estrogen replacement therapy (ERT) and the risk of  Alzheimer disease (AD). The findings have been inconsistent, since some studies have been interpreted as showing a protective effect while others have reported no effect.  ......

In this cohort-based study with an average follow-up of more than 5 years, we found no material evidence that current ERT use in postmenopausal women reduced the risk of developing AD. The risk estimate comparing all current ERT users with nonusers was 1.18  (95% CI, 0.59-2.37). For ERT users who received the drug for 5 yearsor more compared with nonusers, it was 1.05 (95% CI, 0.32-3.44). Odds ratio estimates were similar in women who used unopposed estrogens and for those who also used progestins.

In summary, our findings indicate that ERT use in postmenopausal women is not associated with a substantially reduced risk of AD, and highlight the need for restraint in advocating postmenopausal ERT for this purpose.

Therese van Amelsvoort, Jacqueline Compton and Declan Murphy 
[email protected]
Trends in Endocrinology and Metabolism 2001, 12:273-276

Abstract        (extract) 
There is increasing evidence from animal and in vitro studies to suggest that estrogen might have neuroprotective effects, and several plausible physiological mechanisms have been proposed. However, it is not yet fully understood how estrogen affects the human brain. <snip> The current data from humans suggest that the use of estrogen hormone-replacement therapy (HRT) in healthy, postmenopausal women might reduce the risk of developing Alzheimer's disease (AD) and preserve certain aspects of cognitive function. <snip> However, there is very little evidence at present that HRT is an effective treatment for established AD.

Estrogen failed to improve cognitive or functional outcomes in this  1-year study of women with mild to moderate AD and hysterectomies.  Similar to previous reports, we found a benefit of low-dosage estrogen  on the MMSE after brief exposure (2 months; P = .05), but the benefit  did not persist with continued treatment. In fact, patients receiving  estrogen appeared to decline more than those receiving placebo on 1  global clinical measure, the CDR, despite the greater use of donepezil  in the estrogen-treated patients. Overall, the results of this study do  not support the role of estrogen in the treatment of AD. 

 To date, this study is the largest and the longest study to examine  estrogen as a treatment for women with AD. Given that patients  receiving estrogen did no better or worse than patients receiving  placebo, the use of a larger sample size would not have changed this result.

Estrogen deficiency has been proposed as a cause of memory loss in postmenopausal women. If true, men should have less memory loss with age than women. The present study is designed to examine the postulated effect of estrogen on memory by studying the effect of gender on the age-related decline in cognitive function. 

These weak or absent gender differences in decline in cognitive function with age do not support the thesis that estrogen deficiency is associated with a decline in cognitive function in postmenopausal women.

 Abstract: [Emphasis added]

The potential role of estrogen in the prevention of Alzheimer's disease (AD) and other forms of dementia is an exciting area of research. There is evidence that loss of estrogen after menopause is associated with subclinical impairment in some aspects of neuropsychological function. Case-control studies of estrogen and AD provide mixed results, but some studies have concluded that estrogen replacement therapy (ERT) is associated with reduced risk of developing AD. There is preliminary evidence from clinical trials that estrogen treatment provides modest short-term improvement in women with AD. The research to date, however, has been too limited to provide sufficient evidence to warrant widespread use of estrogen to prevent or treat AD. Large-scale clinical trials such as the Women's Health Initiative Memory Study may offer the information necessary to adequately assess the role of ERT in preventing dementia in postmenopausal women. [Medscape Mental Health 2(7), 1997. © 1997 Medscape, Inc. 

Evidence to date provides a rationale for considering the role of estrogen in preventing or treating AD. There is evidence that estrogen improves performance of selected memory tests in otherwise healthy women who are estrogen deficient. Retrospective case-control studies have yielded mixed results. Some studies showed a protective effect of ERT on the development of AD, while others have not demonstrated this benefit. Several small case-control studies were inconclusive. Finally, a small number of trials of estrogen in women with AD have suggested benefits in cognition. These included open trials and placebo-controlled trials. However, the number of patients involved in these trials was small. 

 It is possible that estrogen replacement may have a role in primary prevention--delaying the age at which susceptible women show diagnosable symptoms of AD--as well as in secondary prevention--slowing the progression of AD once it is diagnosed. Only placebo-controlled studies of large sample size can determine whether estrogen replacement has clinically significant protective effects against AD and whether the potential benefits outweigh the risks. 

 At present there is not sufficient scientific evidence to recommend ERT on a routine basis to prevent or delay AD. The available evidence, while exciting, demands additional research that will address the benefits and risks of ERT in the amelioration of dementia and Alzheimer's disease.