For that reason, many medical scientists began recommending against the combined therapy in the absence of such symptoms, saying the risks of estrogen plus progestin outweighed the benefits.
The latest findings from Women's Health Initiative studies provide new evidence that the combined hormone therapy significantly boosts the risks of dementia and strokes in postmenopausal women while not improving what scientists call "global cognitive function" -- how the brain works.
"The strength of the findings last year, as well as the new ones, is not only the large number of study participants, but also that they were randomly assigned to receive daily estrogen plus progestin or placebo," said Dr. Gerardo Heiss, professor of epidemiology at the University of North Carolina at Chapel Hill School of Public Health.
The Women's Health Initiative Memory Study (WHIMS), which showed roughly a doubling of dementia among those in the treatment group, involved 4,532 women, while the new stroke investigation involved 16,608.
As a member of its steering committee, Heiss helped design the overall Women's Health Initiative study, which is producing useful data for numerous ancillary studies. He is co-author of the stroke report, whose lead author is Dr. Sylvia Wasserthiel-Smoller of the Albert Einstein College of Medicine. Drs. Sally A. Shumaker and Stephen R. Rapp of Wake Forest University are first authors of the WHIMS dementia and global cognitive function studies, respectively.
"What is new in the WHI paper on stroke is that we now include all the strokes with 5.6 years follow-up, adjudicated by neurologists with 258 patients instead of the 212 published last year," Heiss said. "The refined classification by neurologists allowed us to analyze the findings by types of stroke."
Heiss and colleagues found estrogen-plus-progestin therapy boosted the risk of strokes caused by clots but not by bleeding, he said. They found a 44 percent increase in ischemic stroke but no evidence of increased risk of hemorrhagic stroke.
"There was no subgroup of women who were not at increased risk of stroke if they were taking the combination therapy," he said. "The increased risk occurred in both in women with hypertension and those without it, although it appeared later in those with normal blood pressure and in those at low risk of stroke at baseline."
Stroke risk was also elevated in the combination therapy group among women with vasomotor symptoms, Heiss said.
"Overall, this paper indicates that the increased risk of ischemic stroke among women who were taking the hormone therapy is consistent with the reported increased risk in coronary heart disease and also with dementia attributable to an increased propensity to small strokes and cerebral infarcts (blood flow blockages)."
UNC participated in the Women's Health Initiative Memory Study, a substudy of the WHI Hormone Program, as one of the clinical examination sites, he said. Under the direction of Carol Murphy as principal investigator, the UNC site enrolled and examined 254 women aged 65 and older, as one of 39 sites involved in that investigation.
"Because of the potential harm and lack of benefit found, we recommend that older postmenopausal women not take the combination hormone therapy to prevent dementia, and we hope doctors will incorporate what we've learned in their recommendations to women," said lead author Shumaker. She and colleagues found that after four years of follow-up, 61 women were diagnosed with dementia, including 40 among the 2,229 in the hormone group compared with 21 among the 2,303 who received an inactive substance known as a placebo.
In an accompanying commentary, Dr. Christine Yaffe of the University of California at San Francisco said more than a third of women and 20 percent of men aged 65 and older eventually will develop dementia. As the U.S. population ages, the number of people with Alzheimer's disease and other forms of dementia is expected to quadruple over the next 50 years at great cost to families and society.
Until discrepancies between earlier observational studies suggesting a benefit of estrogen on dementia and the new, more rigorous randomized trials suggesting harm are resolved, "the findings in this issue of JAMA underscore that hormone therapy should be prescribed only for temporary use to treat menopausal symptoms."
The National Heart, Lung and Blood Institute sponsors the continuing Women's Health Initiative investigations. Wyeth-Ayerst Research Laboratories supplied the drugs.
Note: Heiss can be reached at (919) 962-3253, Shumaker and Rapp through Barbara Hahn or Karen Richardson at Wake Forest University Baptist Medical Center at (336) 716-4587.
School of Public Health Contact: Lisa Katz, (919) 966-7467
News Services Contact: David Williamson, (919) 962-8596
By DAVID WILLIAMSON
UNC News Services