Disordered sleep in mid- to late life is associated with an increased risk for future cognitive impairment and may alter the dynamics of the Alzheimer’s disease–associated protein amyloid-beta, according to several studies.

It’s important that physicians recognize disordered sleep as a modifiable risk factor, Dr. Kristine Yaffe said in an interview.

“Please attend to sleep hygiene in your elderly patients,” said Dr. Yaffe, director of the memory disorders clinic at the San Francisco VA Medical Center. “Some people think that 50% of elderly people have some kind of sleep complaint. They are common and they are treatable.”

Dr. Yaffe and her colleagues studied 1,309 elderly women who completed several days of sleep observation as part of a 15-year longitudinal study. They measured the motor activity of the women (mean age, 82 years) during sleep via wrist actigraphy. All of the participants had several neuropsychological evaluations and cognitive measurements during the study. A subset of 298 patients also underwent polysomnography.

After 5 years, women with sleep-disordered breathing were twice as likely to develop mild cognitive impairment or dementia as were those who slept normally. There was a similar risk level for women who had delayed sleep acrophase – difficulty falling asleep before the early morning hours and trouble waking up before late morning or early afternoon.

Women with greater nighttime wakefulness were more than twice as likely to show impaired global cognitive functioning, and twice as likely to have delayed verbal recall.

“We already know that sleep deprivation and abnormal sleep patterns are associated with falls and increased morbidity and mortality,” said Dr. Yaffe, who also is a professor in the departments of psychiatry, neurology, epidemiology, and biostatistics at the University of California, San Francisco. “This is the first study showing that disordered sleep is a risk factor for later cognitive problems.

“It’s important for clinicians to check for sleep problems and excessive daytime sleepiness as a possibly treatable cause of later cognitive problems,” she said at the Alzheimer’s Disease International Conference 2012. “Sleep habits are as important in prevention of dementia as diabetes and obesity.”

Dr. Elizabeth Devore

Elizabeth Devore, Sc.D., of Brigham and Women’s Hospital, Boston, found similar results in her analysis of sleep data extracted from the longitudinal Nurses’ Health Study. The sleep study included information on 15,263 women, who were aged 70 or older at the time of their first cognitive evaluation. These subjects were followed for 6 years with cognitive testing every other year.

The women reported their sleep duration and quality at ages 40-65 and at ages 54-79. A normal night’s sleep was considered to be 7 hours.

When comparing sleep duration to later cognitive status, Dr. Devore found that:

• Those who slept 5 hours per day or less had lower average cognition than those who slept 7 hours per day.

• Those who slept 9 hours per day or more also had lower average cognition than those who slept 7 hours per day.

• Too little or too much sleep was cognitively equivalent to 2 years of aging.

When the researchers evaluated the effects of change in sleep duration from mid- to later life, they observed that women whose sleep changed by 2 hours per day or more had worse cognitive function than those with no change in sleep duration, independent of their initial sleep duration.

“Either too little or too much sleep, and sleep duration changes over time, might contribute to cognitive decline in older adults,” she said. “This is up-and-coming research and has great implications for public health. We need to do more research in this area, especially looking at sleep duration over the course of life. It’s simple, and down the road it could lead to the development of sleep and circadian rhythm strategies to target therapy.”

Dr. Devore also found that shortened or extended sleep patterns were also entwined with abnormalities in the processing of amyloid-beta (Abeta) peptides, which are believed to cause neuronal dysfunction in Alzheimer’s disease. Subjects with these extremes of sleep duration showed a skewed proportion of Abeta-40 and Abeta-42 peptides in cerebrospinal fluid, suggesting that more amyloid could be accruing in brain plaques.

Dr. Yafei Huang and her associates at Washington University, St. Louis, also found that the circadian pattern of Abeta secretion is altered in Alzheimer’s disease. Her study involved three groups: 12 Alzheimer’s patients with a mean age of 72 years, 8 age-matched controls, and 10 young, healthy controls with a mean age of 36 years.

Each subject underwent hourly sampling of cerebrospinal fluid (CSF) and plasma for 36 hours during waking and sleep times. Dr. Huang plotted the circadian secretion patterns of Abeta-40 and Abeta-42, and the amyloid precursor protein (APP).

While there were no significant time-linked associations of Abeta in plasma, there were differences in the CSF.

Overall, both Abeta-40 and -42 in CSF showed a linear increase during waking time, and dips during sleep. The age-matched controls and healthy young subjects – who were presumably amyloid negative – had a steady increase in APP, Abeta-40, and Abeta-42 during waking times. But this finding was absent in subjects with Alzheimer’s.

“In this group, the Abeta-40 increased much less, and the change in Abeta-42 was almost absent,” Dr. Huang said.

This suggests that amyloid processing is impaired in patients with the disease. Different enzymes split the APP molecule into the benign Abeta-40, which is secreted into the CSF, and the toxic Abeta-42, which forms brain plaques. A low CSF Abeta-42 level indicates abnormal APP cleavage, leading to retained Abeta-42.

Some studies have suggested that hypoxia alters the enzymatic cleavage of APP, allowing more Abeta-42 production. This finding could have implications in people who experience years of sleep apnea, said Dr. Constantine Lykestos, who moderated the press briefing where the studies were presented.

“There is emerging evidence that hypoxia is associated with amyloidosis, but it’s too new a theory to really know,” said Dr. Lykestos, director of the memory and Alzheimer’s treatment center at Johns Hopkins University, Baltimore. “The jury is still out on this relationship.”

Dr. Yaffe, Dr. Huang, and Dr. Devore reported no relevant financial disclosures.