ABSTRACT & COMMENTARY
Obstructive Sleep Apnea and Incident Diabetes
By Barbara A. Phillips, MD, MSPH
Professor of Medicine, University of Kentucky; Director, Sleep Disorders Center, Samaritan Hospital, Lexington
Dr. Phillips serves on the speakers bureau for PotomaCME.
Severe obstructive sleep apnea is associated with a 30% increase in risk of incident diabetes, controlling for multiple relevant confounders.
Kendzerska T, et al. Obstructive sleep apnea and incident diabetes: A historical cohort study. Am J Respir Crit Care Med 2014; Jun 4. [Epub ahead of print.]
These Canadian authors set out to undertake a rigorous study of the relationship between obstructive sleep apnea (OSA) and incident diabetes. They collected data from adults who did not have diabetes at baseline and who underwent a sleep study, then followed these people for a median time of about 5 years. The primary outcome was time from the diagnostic sleep study to the diagnosis of incident diabetes. During that time, 10,149 people had sleep studies. Of these, 8678 did not have diabetes at baseline. Over the follow-up of about 6 years, 11.7% (1017) of these people developed diabetes. After controlling for age, sex, body mass index (BMI), smoking, multiple medical comorbidities, and income, those people with a severe sleep apnea (defined as an apnea plus hypopnea index [AHI] > 30 events per hour of sleep) had a 30% higher risk of incident diabetes than those with AHIs < 5. In addition to AHI, time spent with oxygen saturations < 90%, heart rate, and neck circumference were associated with incident diabetes. The older and heavier the patients were, the less effect AHI had on risk of incident diabetes.
The authors also looked at the effect of continuous positive airway pressure (CPAP) (assessed by submission of CPAP insurance claims) on the risk of diabetes development. A claim for CPAP treatment did not have a significant effect on the risk of diabetes development.
COMMENTARY
The prevalence of both OSA and diabetes is increasing, at least partly because the prevalence of obesity is increasing. However, OSA is also associated with intermittent hypoxemia, arousals from sleep, and neurohumoral changes that may affect insulin sensitivity.1 A few small studies have suggested an independent link between OSA and risk of diabetes, but those preliminary reports were not as large or as rigorously done as this one. Because these investigators were able to include health data for all residents of Ontario, they had a huge and rich source of data. (Indeed, legislation prohibits private delivery of services covered under the Ontario Health Insurance Plan, including laboratory testing!) The findings that degree and duration of hypoxemia and higher heart rate were associated with increased risk of diabetes in this study support the current hypothesis that oxidative stress (resulting from sympathetic activation or hypoxemia) promotes glucose intolerance.
This study adds strong evidence to the notion that the relationship between sleep apnea and diabetes is not mediated by obesity alone. Stronger proof, of course, would be data demonstrating that CPAP treatment improves diabetic outcomes (which was not demonstrated in this study that was not designed to address this issue). A recent review2 of 22 articles reported that CPAP use results in significant changes in glucose metabolism, including glycosylated hemoglobin (HbA1c), postprandial or nocturnal glucose, and insulin sensitivity. In addition, a recent report from the United Kingdom based on National Health Service (NHS) data, noted "Initiating treatment with CPAP in OSA patients with type 2 diabetes leads to significantly lower blood pressure and better controlled diabetes and affords a cost-effective use of NHS resources."
An under-the-radar finding of the current study was that the time spent with oxygen saturation below 90% (that is, degree and duration of oxygen desaturation) predicted incident diabetes more strongly than did AHI, and that AHIs below 30% were not associated with statistically significant increases in incident diabetes after controlling for confounders. These and other findings (including the huge numbers of people at risk for OSA) are likely to further the migration of the diagnostic approach toward sleep apnea out of the sleep lab and into the home.
What does all of this mean for our patients? It is probably cost-effective to test for sleep apnea in patients who fit the profile (obese, sleepy, > 55 years of age, witnessed apneas, hypertension, big necks), particularly if the diagnosis can be done with an oximetry-based home study.
REFERENCES
- Punjabi NM, Polotsky VY. J Appl Physiol 2005;99:1998-2007.
- Gallegos L, et al. Hosp Pract 2014;42:31-37.
- Guest JF, et al. Diabetes Care 2014;37:1263-1271.
