According to principal investigator Sally E. Shaywitz, MD, of the Yale University School of Medicine, the researchers used a technology known as functional magnetic resonance imaging (fMRI), which produces computer-generated images of the brain during thinking, reading or remembering. Briefly, the technology measures the increase in blood flow that occurs in active areas of the brain. The researchers compared brain activation patterns in postmenopausal women when they were taking estrogen to when they were taking a placebo. They found that estrogen changed the brain activation patterns of the postmenopausal women and that the changes resembled brain activation patterns typically seen in younger compared to older people. The study appears in the April 7 issue of the Journal of the American Medical Association.
"Taking estrogen after menopause has both benefits and risks," said Duane Alexander, MD, Director of the NICHD. "This finding provides important new information for postmenopausal women and their physicians to consider when making the complex, highly individual decision as to whether to take estrogen supplements."
For the study, Dr. Shaywitz and her coworkers analyzed the brain activation patterns of 46 women. In the first phase of the study, half of these women were chosen at random to receive 1.25 mg of a standard estrogen preparation for postmenopausal women, and the other half received a placebo. The brain functioning of both groups was then observed with fMRI technology. At the end of 21 days, all the women went through a 14 day "washout period" where they did not receive either estrogen or placebo. In the second phase of the study, the two groups were switched, with the estrogen group receiving a placebo, and the placebo group receiving estrogen therapy. The brain functioning of both groups was again examined with fMRI. Neither the researchers nor the women knew which women were taking estrogen and which were taking placebo.
The researchers tested the types of memory function called upon frequently during any given day. To assess their memory, the women underwent fMRI while they performed two different sets of memory tasks: one verbal and the other non-verbal. In each sequence, the different components of memory (registration of the stimulus to be remembered; storage of the stimulus; and retrieval of the stimulus) were assessed. In the first task, designed to measure the ability to remember words, the women were shown a sequence of nonsense words such as "gex," "hib," and "mek" on a computer screen. The screen went blank for 20 seconds, and then a new sequence was shown: two of the previous words, along with two new nonsense words. The women were then asked to identify the words they had seen previously.
To test their non-verbal memory, the women were shown abstract symbols, using similar testing. The women were shown 3 characters from the Tamil alphabet, then a blank screen for 20 seconds, then asked to distinguish 2 previous characters from 2 they had never seen before. (Tamil is a language spoken in southern India and Sri Lanka. People who do not read Tamil perceive its characters as abstract symbols.)
During storage of nonsense words, women on estrogen demonstrated significantly more activation in the inferior parietal lobule, a brain region known to be associated with storing verbal information in a speech-based or phonetic code. Such phonetic codes are believed to be critical for reading and for holding information in memory for brief periods of time (e.g., remembering a telephone number).
In addition, the researchers noted that while on estrogen, the women showed a pattern of increased activation in left frontal brain regions during registration of the stimuli, and increased activation in right frontal regions during retrieval of the stimuli. This pattern has been observed in younger compared to older participants in previous imaging studies.
Shaywitz noted that whether on estrogen or the placebo, the women received roughly the same numerical scores for the tests (within the 90 percent range). The tasks were designed to be relatively easy and not effortful. As a result, Dr. Shaywitz speculated that the test may not have been difficult enough to identify differences between the two groups of women. She added that fMRI is highly sensitive and may be able to detect changes in brain activation patterns before they are apparent in test scores of learning or behavior.
"These findings show that it's possible to alter brain organization in older individuals," Dr. Shaywitz said. "These alterations suggest a plasticity in the memory systems of mature women and that these neural systems are neither fixed nor immutable."
The study conducted by Dr. Shaywitz and her colleagues is an outgrowth of NICHD's research program in reading disabilities, explained Reid Lyon, Ph.D., Chief of NICHD's Child Development and Behavior Branch and the project officer for the Yale study. Specifically, the study drew heavily from methods developed by Dr. Shaywitz and her colleagues to study reading and dyslexia.
In their paper, Dr. Shaywitz and her coworkers state that fMRI is a sensitive new tool to detect estrogen's effect on the brain. "These results are encouraging and suggest that the use of functional imaging together with protocols examining, for example, different doses, treatment lengths and washout periods, may herald a new era in efforts to understand the effects of estrogen on cognitive function in postmenopausal women," the researchers wrote.
A release describing a previous imaging study of reading by the Yale team appears on the Mass Media page of the NICHD website at http://www.nih.gov/nichd/, "NICHD-Funded Researchers Map Physical Basis of Dyslexia."