Midlife Estrogen Therapy Not Linked to Dementia

From the American Academy of Neurology

SAN DIEGO—For women undergoing natural menopause, estrogen replacement therapy in midlife does not have a substantial effect on episodic memory 5 years later, according to data presented at the 58th Annual Meeting of the American Academy of Neurology.

“Research from the Women’s Health Initiative shows that hormone therapy begun later in life can increase dementia risk,” said Victor W. Henderson, MD, of Stanford University in Palo Alto, Calif. “Much less is known about the cognitive effects of estrogen in midlife, which is a time when women are more likely to consider hormone therapy for hot flashes and related menopausal symptoms.”

He continued, “Memory complaints are common around the time of menopause, and deficits in episodic memory later in life are a strong risk factor for subsequent diagnosis of Alzheimer’s disease. However, there are no longitudinal data on the potential effects of hormone therapy used specifically by middle-aged women on their subsequent memory performance. Rather, existing information pertains primarily to hormone effects in older women.”

“Much less is known about the cognitive effects of estrogen in midlife, which is a time when women are more likely to consider hormone therapy.”

—Victor W. Henderson, MD

Included were women aged 53-63 years who had menstruated within the previous 3 months and who were enrolled in the population-based Melbourne Women’s Midlife Health Project (MWMHP). The women were administered a 10-item word list memory task for immediate and delayed recall on 2 occasions, 5 years apart (ie, baseline and “Time-2”). The MWMPH cohort was established through random digit dialing of Melbourne residents.

The current analysis included 145 women who had undergone natural menopause by the time the second questionnaire was administered. At baseline, 76 were never-users of hormone therapy, 32 were past users, and 37 were current users. Results showed only a slight decline in total memory score between test sessions (a mean of 0.69 ± 4.7 points). However, mean memory scores at baseline and at the time of the second questionnaire did not differ by hormone therapy status.

P

In an analysis of covariance adjusted for baseline memory score, age, and education, there were still no changes at 5 years by hormone therapy status ( = .5). Additional potential control variables (ie, mood, apolipoprotein E genotype, and hormone therapy use at Time-2) did not alter these results.

The findings were similar when immediate and delayed recall memory performances were considered separately.

Previous cross-sectional analyses of the MWMHP cohort indicated that hormone replacement therapy status was not associated with episodic memory scores during middle age, Dr Henderson said. The present findings extend this observation to show that such treatment in midlife does not cause an appreciable change in memory performance 5 years later.

Potential shortcomings of the study include its observational design and inability to discern small between-group differences.

Dr Henderson said that future studies are needed to determine the consequences of midlife hormone therapy on memory and dementia risk in old age. This important question, he added, can be approached with laboratory models, observational studies and, most critically, properly designed, randomized clinical trials.