Previous research has suggested that estrogen may play a role in protecting women against memory loss, but now it appears that the timing of estrogen replacement therapy may be a critical factor.

A study by the Mayo Clinic in Rochester, Minnesota, looked at the relationship between estrogen and memory loss in a group of 3,000 women. About half of them had one or both ovaries removed between 1950 and 1987 for non-cancer-related reasons.

Neurologist and epidemiologist Walter Rocca led the study. He and his colleagues contacted the women more than 30 years after their surgeries (between 2001 and 2006) and gave them a short test to evaluate their memory and ability to reason. When the women could not be interviewed directly, the researchers spoke with a family member.

They found that women who had their ovaries removed before menopause had an increased risk of cognitive impairment or dementia later in life, sometimes many years after the surgery. "We also found that the younger they were at the time of the ovariectomy," says Rocca, "the more they were at risk."

Rocca and his colleagues also tracked whether the women had been given estrogen replacement treatments after the surgery. The researchers found that if the women were treated with estrogen until at least age 50, their risk of developing dementia was no greater than for women who had not had their ovaries removed.

But, Rocca notes, a Women's Health Initiative study of women over age 65 showed the opposite result. That research found that women who started taking estrogen after menopause were more likely to develop dementia than those who did not get the replacement hormone. Rocca says these findings suggest that timing is critical: estrogen treatment may be protective before age 50, but harmful later in life.

Rocca's team also looked at the same group of women to assess their risk of developing Parkinsonism, a syndrome that impairs movement. The most common form of this syndrome is Parkinson's disease.

They found a similar result as for dementia: the risk of Parkinsonism was almost twice as great for women who had had their ovaries removed, as for those who had not. And, as for dementia, the younger the women were at the time of the surgery, the higher their risk of developing Parkinsonism. "The limitation of the study on Parkinsonism is that the numbers were smaller, because Parkinsonism is less common than dementia," Rocca explains. "Dementia is ten times more common."

The two studies were published on August 29 in the online edition of Neurology, which is available at