Add another study to the growing and often contradictory welter of studies postmenopausal women need to weigh before deciding whether to begin a program of estrogen replacement.
The new report by researchers at the University of Southern California Health Sciences Center, published in the Archives of Internal Medicine, followed 8,800 women in a California retirement community for eight years.
The study found that those who took estrogen had a 20 percent lower death rate from all causes, but especially from heart disease and stroke, than those who did not.
Women who took estrogen longer had lower death rates; among those who took the hormone for 15 years, mortality was 40 percent lower.
Previous studies of estrogen replacement use have shown increased rates of some kinds of uterine and breast cancer.
Estrogen replacement has been shown to ease the symptoms of menopause and to prevent osteoporosis -- the condition in which bones become thin and brittle.
More recently, it has been shown to protect against heart disease and stroke, the major killer of postmenopausal women.
Some studies have indicated that adding the hormone progesterone to the replacement therapy can prevent cancer of the endometrium, the lining of the uterus, but that also seems to block the cardiovascular benefit.
The women in the California study took only estrogen.
Researchers concluded that the risk-benefit ratio for estrogen replacement was so complex that "despite the substantial overall mortality benefit," women on the therapy and their physicians should "place high priority" on watching for such symptoms of endometrial cancer as uterine bleeding or for breast lumps.