Estrogen therapy in postmenopausal women: effects on cognitive function and dementia.

CONTEXT Several studies have suggested that estrogen replacement therapy in postmenopausal women improves cognition, prevents development of dementia, and improves the severity of dementia, while other studies have not found a benefit of estrogen use. OBJECTIVE To determine whether postmenopausal estrogen therapy improves cognition, prevents development of dementia, or improves dementia severity. DATA SOURCES We performed a literature search of studies published from January 1966 through June 1997, using MEDLINE, manually searched bibliographies of articles identified, and consulted experts. STUDY SELECTION Studies that evaluated biological mechanisms of estrogen's effect on the central nervous system and studies that addressed the effect of estrogen on cognitive function or on dementia. DATA EXTRACTION We reviewed studies for methods, sources of bias, and outcomes and performed a meta-analysis of the 10 studies of postmenopausal estrogen use and risk of dementia using standard meta-analytic methods. DATA SYNTHESIS Biochemical and neurophysiologic studies suggest several mechanisms by which estrogen may affect cognition: promotion of cholinergic and serotonergic activity in specific brain regions, maintenance of neural circuitry, favorable lipoprotein alterations, and prevention of cerebral ischemia. Five observational studies and 8 trials have addressed the effect of estrogen on cognitive function in nondemented postmenopausal women. Cognition seems to improve in perimenopausal women, possibly because menopausal symptoms improve, but there is no clear benefit in asymptomatic women. Ten observational studies have measured the effect of postmenopausal estrogen use on risk of developing dementia. Meta-analysis of these studies suggests a 29% decreased risk of developing dementia among estrogen users, but the findings of the studies are heterogeneous. Four trials of estrogen therapy in women with Alzheimer disease have been conducted and have had primarily positive results, but most have been small, of short duration, non-randomized, and uncontrolled. CONCLUSIONS There are plausible biological mechanisms by which estrogen might lead to improved cognition, reduced risk for dementia, or improvement in the severity of dementia. Studies conducted in women, however, have substantial methodologic problems and have produced conflicting results. Large placebo-controlled trials are required to address estrogen's role in prevention and treatment of Alzheimer disease and other dementias. Given the known risks of estrogen therapy, we do not recommend estrogen for the prevention or treatment of Alzheimer disease or other dementias until adequate trials have been completed.