Sleep EEGs may predict future disease, mortality

CHARLOTTE, N.C. — Evaluation of nearly 9,000 study participants showed that sleep electroencephalograms contain decodable information about the risk for future incidence of mortality, dementia and other diseases.

Haoqi Sun, PhD, of Harvard Medical School and Massachusetts General Hospital, and colleagues hypothesized in their poster at the SLEEP meeting that the risk for dementia, cerebrocardiovascular disease, psychiatric disease and mortality can be predicted by analyzing sleep microstructure.

They considered 11 outcomes in a retrospective study: dementia, mild cognitive impairment or dementia, ischemic stroke, intracranial hemorrhage, atrial fibrillation, myocardial infarction, type 2 diabetes, hypertension, bipolar disorder, depression and mortality, according to the study.

Sun told Healio that he and his colleagues also looked at 86 microstructure features from overnight sleep EEG recordings, as well as BMI and prescription medications.

They included 8,673 participants (mean age, 51 years; 51% women), who were divided into three groups: poor sleep, average sleep and good sleep. The outcome-wise mean prediction difference in the 10-year risk ratio (RR) was 2.3% for the poor sleep group, 0.5% for average and 1.3% for good, according to the study.

Researchers further reported that the top poor-to-average RRs were dementia (RR = 6.2; 95% CI, 4.5-9.3), mortality (RR = 5.7; 95% CI, 5-7.5) and mild cognitive impairment or dementia (RR = 4; 95% CI, 3.2-4.9).

Sun told Healio that the RR for mortality was 5.6 in women and 5.5 in men, “and if you compare female vs. male, depression is really high for females.”

He continued, “You can combine all features into one number with the conclusion that sleep EEGs contain decodable information about the risk of unfavorable outcomes.”

The researchers concluded in the study: “The findings strengthen the concept of sleep as a window into the brain and general health.”