Sleep on it: fitful slumber tied to diabetes risk.

Many people have brief bouts of interrupted breathing during the night that cause fluctuations in blood pressure and heart rate, decreased concentrations of oxygen in the blood, and other effects. Small studies have implicated this disturbed slumber, known as sleep apnea, in the development of diabetes and other chronic diseases (SN: 7/14/01, p. 31).

Now, the results of a large study led by Naresh M. Punjabi of the Johns Hopkins Medical Institutions link apnea with two conditions--glucose intolerance and impaired insulin function--that are associated with the onset of type 2 diabetes. "We think the sleep problems are more likely contributing to glucose intolerance and diabetes instead of those conditions explaining the apnea," says coauthor Rachel Givelber of the University of Pittsburgh.

The findings, which appear in the Sept. 15 American Journal of Epidemiology, could have important implications for identifying and managing diabetes, says Givelber. "Based on this study," she says, "if you have glucose intolerance or diabetes, it might be wise to treat the sleep [apnea] because that may be contributing to the condition."

Sleep apnea, often accompanied by loud snoring (SN: 3/11/00. p. 172), occurs because the airways narrow as the surrounding muscles relax during sleep.

In their investigation, the researchers analyzed data on 2,656 of the people who participated in the Sleep Heart Health Study sponsored by the National Heart, Lung, and Blond Institute in Bethesda, Md. That study was conducted from 1994 to 1999 in several U.S. cities to investigate possible cardiovascular consequences of sleep apnea. The study also acquired data on blond-glucose concentrations because diabetes is a known risk factor for cardiovascular disease.

Data on breathing patterns, respiratory effort, number of arousals from sleep, body position, and blood-oxygen concentrations were recorded in each participant's home, then transferred to a central location for evaluation. Blood samples were obtained after overnight tasting to check glucose concentrations. Punjabi's team focused on the participants who were at least 40 years old and hadn't been previously diagnosed with diabetes.

The researchers used statistical methods to sidestep any effects of age, gender, and weight. The team found that the people whose sleep was interrupted most frequently at night--15 times or more each hour--were most likely to show glucose intolerance and impaired insulin function. Those with no sleep apnea seldom showed those conditions. "We were able to show that, even accounting for overweight and body shape, people with high sleep apnea had more diabetes and glucose intolerance than we had expected," says Givelber.

The study "moves the field incrementally but importantly forward," says Paul E. Peppard, a sleep researcher at the University of Wisconsin-Madison. "Other studies in this area haven't been so rigorously designed."