The trusted source for
healthcare information and
Abstract & Commentary
Synopsis: In a cross-sectional analysis of HRT always-users and never-users in their mid-60s, always-users displayed slightly better performance in delayed recall and attention.
Source: Smith YR, et al. Fertil Steril. 2001;76: 1101-1107.
This study compared 16 women with a mean age of 65 who had used HRT in the form of conjugated equine estrogens since menopause to 13 women with a mean age of 67 who had never used HRT using standardized psychometric inventories. No statistically significant differences between the groups were found for general demographic, intellectual, and psychological measures. A total of 10 tests were administered. On 2 tests, there were differences in scores between always-users and never-users. The domains that differed were attention and delayed recall (nonverbal memory).
The main strength of this study is the comparison of HRT always-users with never-users. There are also some significant limitations. First, the women were relatively young. The hypothesis being tested was that HRT would retard the age-related decline in cognitive prowess. However, age-related losses of cognitive and psychomotor functioning typically do not appear until after age 70 and do not become pronounced until after age 85. Thus, the groups were too young for major differences in performance to have emerged. Second, psychometric inventories are notoriously insensitive instruments. There is typically large variation in performance due primarily to longstanding interindividual differences in aptitudes and experiences. Even when the same individual takes the same battery of tests twice, however, there is still a lot of performance variation. This lack of sensitivity means that large groups of women would need to be studied for group differences to be detected. In this study, only a total of 29 postmenopausal women were studied. Third, the study was cross-sectional, so, again, to get beyond the expected differences due primarily to pre-existing endowments, one would need to study 2 large cohorts. In a prospective study, one could measure performance in the same individuals before and after a given treatment, but that type of study would take a long time to complete. For a study like this, one would prefer a treatment interval in the range of 20 or 30 years. Obviously, we cannot expect to see data like that being reported anytime soon. If we began such a study today, by the time it was finished, most of us would be too old to remember the question that we were trying to answer.
What advice can one give to patients about this important topic? The best that one can do is offer an educated guess based on available biological data. I emphasize that there is a wealth of biological data demonstrating that the brain is a target tissue for sex steroids. Sex steroids have a panoply of important effects on just about every known neurotransmitter system in the brain, including the cholinergic neurons that mediate memory and learning. Sex steroids increase synaptic density, cause glial to secrete nerve growth factors, promote blood flow, and in general enhance neuronal viability and connectivity. Available data suggest that to achieve full neuroprotection, one needs to take an estrogen that binds to both estrogen receptor subtypes, alpha and beta.
Given the above limitations that describe presently available data, I would be reluctant to discard the hypothesis. At this point, negative results are more likely to be due to limitations in study design rather than a valid negative result. While HRT use may pose some risks for some tissues or bodily systems, no one has yet suggested that HRT use damages the brain. All that is being debated is the type and magnitude of the benefits expected and the age at which the expected benefits will manifest. As always, the decision about HRT rests with a careful delineation of priorities, medical risks, and fears. Like osteoporosis and heart disease, no one is immune from age-related cognitive declines. Since estrogen started after dementia is clinically evident is of no apparent use, it would appear that one should recommend starting HRT at the time of menopause and continuing it indefinitely.
Dr. Berga is Professor and Director, Division of Reproductive Endocrinology and Infertility, University of Pittsburgh.