Just as estrogen's reputed effects on physical health have been hotly debated, so has its influence on cognitive functioning. While a number of studies have found beneficial effects of estrogen replacement therapy in postmenopausal women--especially for verbal and visual memory--others have found no such evidence. A major, randomized study now under way will help settle the question--but its results aren't due for five years.
"It looks like hormone replacement therapy may protect against age-related cognitive decline and against Alzheimer's disease, but it doesn't look particularly hopeful as a treatment for Alzheimer's disease," concludes neuropsychologist Pauline Maki, PhD, of the National Institute on Aging (NIA). Maki adds, "Results from the large, randomized trials will provide the necessary data to confirm those observational findings."
The large body of work investigating the connection between estrogen and cognition was in large part sparked by a seminal 1988 study by McGill University psychologist Barbara Sherwin, PhD. Sherwin followed women who had undergone surgery to remove the ovaries and uterus, a procedure that causes a precipitous drop in blood estrogen levels. Women who received estrogen replacement therapy after surgery performed better on tests of verbal memory and other cognitive functions than did women who did not receive estrogen, the study found. That finding has now been repeated many times.
Other research has indicated that estrogen lowers the risk of developing Alzheimer's disease, although it appears not to slow or reverse the disease's course once symptoms have set in. Two longitudinal studies, each following hundreds of women over time to see who developed Alzheimer's disease, found that those who received estrogen during or shortly after menopause--even for as little as six months--were less likely to later develop Alzheimer's disease than were women who had not used estrogen replacement therapy.
There are several likely explanations for such effects, Sherwin says. For example, estrogen increases the concentration of an enzyme needed to synthesize acetylcholine, a brain chemical that's critical for memory. Estrogen also enhances communication between neurons in the hippocampus, an area of the brain that is important for verbal memory.
These chemical connections between estrogen and memory were unknown when Sherwin started working in the area 20 years ago, and that made it difficult to find funding, she says.
"Even though I was seeing this in women, it sounded like hocus pocus because there wasn't a good scientific rationale for why estrogen should do anything to memory," Sherwin remembers. "Now that we're aware of those mechanisms, it makes sense for us to go ahead and launch all these clinical studies."
Jury still out
University of California, San Diego, epidemiologist Elizabeth Barrett-Connor, MD, is more cautious about estrogen's effects on cognition, however. For one thing, Barrett-Connor warns, much of the evidence that estrogen deters memory loss in healthy postmenopausal women has stemmed from observational studies. Such studies can be deceiving, she points out, because women who have more education and who tend to be healthier overall are also more likely to use estrogen, making it difficult to interpret the association between estrogen and cognition.
Second, Barrett-Connor argues, although some small, short-term studies that have randomly assigned study participants to receive either estrogen or a placebo have indeed been promising, others have shown no overall benefit to cognition.
"I think the jury is out," Barrett-Connor says. "We need more studies that have more than 20 people and that last for more than eight weeks."
Sherwin agrees that larger, randomized studies will be crucial. But she disagrees that, overall, research on estrogen has shown mixed results. She argues that many studies that have found no effect of estrogen have used tests of cognitive function that are too blunt to differentiate different aspects of cognitive function.
"If you just read the literature superficially, you'd come out saying that it's really inconsistent," she says. "But if you bring some neuropsychology to bear on trying to understand it, it's not inconsistent. You wouldn't expect [estrogen] to affect all tests of cognition, because that's not how the brain works. There is no drug that I know of that affects all brain functions, except maybe an anesthetic."
But not all observational studies that have shown no effect of estrogen have used blunt measurement techniques, notes Susan Resnick, PhD, a neuropsychologist who studies estrogen and cognition at NIA. She suggests an additional explanation for the mixed findings.
"Some of the difference between the results of studies that showed an effect and those that did not may be explained by differences in the ages of women studied," Resnick speculates. "Most women who use hormone therapy take estrogens around the time of the menopause, and even short-term exposure seems to offer some decrease in the risk for Alzheimer's disease. This may suggest that there is some critical period during which estrogen is most effective in reducing cognitive decline. We don't know if that's the case, but it's one possibility."
The best hope for resolving estrogen's effects on cognition lies with an ongoing clinical trial that will track the risk and progression of dementia in more than 8,000 women who have been randomly assigned to receive either hormone replacement therapy or a placebo.
The investigation, begun in 1991, is headed by Wake Forest University psychologist Sally Shumaker, PhD, and is a piece of the Women's Health Initiative, a 15-year, multicenter study examining several aspects of older women's health. A companion study initiated by Resnick and developed and performed in collaboration with Shumaker and colleagues will examine the effects of estrogen on long-term cognitive change among a smaller sample of women without dementia. The results of both studies are expected in 2005.
In the meantime, some researchers have sought a new kind of evidence, using brain imaging methods to understand estrogen's influence on cognition.
In a 1998 study, Resnick, Maki and colleagues at NIA and Johns Hopkins University compared brain activation patterns in 15 estrogen users and 17 nonusers. Using positron emission tomography imaging to trace blood flow in the brain while study participants completed tests of verbal and visual memory, the group found that the tasks prompted different patterns of brain activation for estrogen users compared with nonusers.
More recently, Maki and Resnick reported that, compared with non-estrogen users, estrogen users showed increased blood flow over two years in the hippocampus and other brain regions known to be part of a memory circuit and to be impaired early in Alzheimer's disease. Those results were published in the journal Neurobiology of Aging (Vol. 21).
"The neuroimaging evidence suggesting a biological effect of estrogen on the brain provides a different kind of support than does behavioral evidence of memory performance," says Maki. "The fact that we found effects in the hippocampus makes it especially compelling."
In addition, Resnick suggests, the neuroimaging findings may provide a clue to how estrogen influences cognition during normal aging: "If you argue that having more blood flow may help protect against tissue loss and development of disease, then this offers one possible route--but not the only route--through which estrogen may protect against memory loss."
Shumaker suggests that Resnick and Maki's findings may carry an additional implication.
Their work, and that of similar investigations, "has much broader implications than being able to track the real or imagined effects of estrogen," she says. "The imaging tests are more sensitive than traditional neuropsychological tests [of cognitive performance], and they may give us a mechanism for testing people early on and detecting structural changes that might be diagnostic of Alzheimer's disease."