Chicago — Too little sleep — or too much — may raise the risk of developing heart disease, according to a study of nearly 72,000 nurses.

Women who averaged five hours or less of sleep a night were 39 per cent more likely to develop heart disease than women who got eight hours. Those sleeping six hours a night had an 18 per cent higher risk of developing blocked arteries than the eight-hour sleepers.

And nine or more hours of shuteye was associated with a 37 per cent higher risk of heart disease. Researchers could not explain that finding, but suggested those women might have slept more because of underlying illnesses.

"People should start thinking of adequate sleep not as a luxury but more as a component of a healthy lifestyle," said Dr. Najib Ayas, a sleep disorders specialist who was at Harvard-affiliated Brigham and Women's Hospital in Boston when he led the study.

The researchers suggested that getting enough sleep may be nearly as important to heart health as eating right and exercising. And they pointed out a recent poll that found that about one in three Americans has long-term sleep deprivation.

The study was published in Monday's Archives of Internal Medicine.

The researchers could not say for certain whether the findings apply to men, too. But other research strongly suggests so.

Previous studies of men and women found short-term sleep deprivation can raise blood pressure and levels of the stress hormone cortisol, lower glucose tolerance and lead to variations in heart rate — all precursors of heart disease.

Phyllis Zee, director of the sleep disorders centre at Northwestern Memorial Hospital, said the findings show that doctors should be asking their patients about sleep habits. And if those patients are losing sleep by choice, "they may want to rethink their priorities," Dr. Zee said.

Researchers examined 10 years of data on 71,617 participants in Harvard's Nurses' Health Study, which tracked female nurses for a variety of studies. The women were ages 45 to 65 and had no sign of heart disease at the outset in 1986. Over 10 years, 934 of the women had non-fatal heart attacks or died of heart disease.

The study relied on the nurses' recollection of their sleep patterns rather than directly measuring their sleep.

The researchers were also from the Harvard School of Public Health and Harvard Medical School, all in Boston.