In 2014 the Centers for Disease Control and Prevention reported that over the past 32 years, the number of adults in the United States in whom diabetes is diagnosed had nearly quadrupled.1 Now more than 30 million people—or 9.3% of the United States population—are estimated to have diagnosed or undiagnosed diabetes. Each year, more than 200,000 deaths occur among people with diabetes in the United States, making it the country's seventh leading cause of death. In addition to those who already have type 2 diabetes mellitus (T2DM), the Centers for Disease Control and Prevention estimates that 86 million adults in the United States have prediabetes, which markedly increases the risk of the development of T2DM and cardiovascular disease.1 Undoubtedly the alarming increase in overweight and obesity has played a pivotal role in the rise of prediabetes and T2DM.
In recent years, evidence has accumulated to indicate that obstructive sleep apnea (OSA) is both a risk factor for the development of T2DM2–4 and an exceptionally frequent comorbidity with an adverse effect on glycemic control.5–8 Multiple studies, including approximately 1,400 patients with T2DM, have shown that the prevalence of OSA (assessed by polysomnography) ranges between 58% to 86%.5,7,9–14 There is a significant association between increasing OSA severity and reduced glycemic control and this association persists after controlling for multiple potential confounders.5,7,8 Therefore, based on current estimations, of the nearly 30 million people in the United States with T2DM, approximately two-thirds may have OSA. Given these important associations, it seems appropriate to consider OSA as a modifiable risk factor in patients with pre-diabetes and T2DM. However, substantial variability has been observed with respect to the effect of continuous positive airway pressure (CPAP) treatment on glycemic control in patients with OSA. This has in turn led to a great deal of uncertainty among clinicians as to whether treatment of OSA has a meaningful effect on glycemic control.
It is important to consider that the association between OSA and glycemic control may be dependent on the sleep stage in which apneas and hypopneas occur. Obstructive apneas and hypopneas during rapid eye movement (REM) sleep are longer in duration, are associated with greater oxygen desaturation, and lead to greater surges in sympathetic activity compared to events in non-REM sleep.15,16 Indeed, in a cross-sectional analysis of 115 patients with T2DM, glycemic control assessed by hemoglobin A1c (HbA1c) was predicted by the frequency of obstructive events during REM sleep, but not during non-REM sleep.7 Because REM sleep predominates in the early morning hours before typical awakening, the benefits of CPAP therapy in patients with T2DM may not be achieved with the typical CPAP use of 3 to 4 h/night.7 In this context, two recent laboratory-based proof-of-concept studies implemented a protocol that consisted of nightly in-laboratory CPAP treatment over an entire 8-h night across 7 nights17 or 14 nights,18 thereby ensuring optimal adherence to CPAP therapy (i.e., eliminating OSA across the entire sleep period, including all REM sleep). These two studies demonstrated that effective resolution of OSA improves glycemic control in participants with prediabetes and T2DM.17,18 However, field studies utilizing home CPAP therapy over longer periods of time have yielded conflicting results. Two randomized clinical trials found no effect of 3 or 6 mo of home CPAP therapy on HbA1c in patients with T2DM when compared to either sham CPAP19 or usual care.20 One potential reason for the negative findings may have been insufficient CPAP use. Notably, the mean nightly CPAP use in these two randomized controlled trials was 3.6 and 4.9 h/night, respectively.19,20 In contrast, a recent randomized controlled trial reported a significant reduction in HbA1c of 0.4% after 6 mo of CPAP therapy in patients with T2DM. Of note, in this study the mean adherence to CPAP was longer than in the two negative clinical trials, averaging 5.2 h/night.21 Higher levels of CPAP adherence in individuals with prediabetes with severe OSA has also been reported to improve insulin sensitivity.22
From a mechanistic standpoint, a rigorous laboratory-based study using frequent blood sampling technique across the entire 24-h period revealed that one week of effective CPAP therapy in patients with T2DM leads to a significant reduction in plasma glucose levels without a significant change in serum insulin levels.23 This suggests that effective treatment of OSA improves insulin resistance. The only counterregulatory hormone that improved after one week of effective CPAP therapy was norepinephrine.
In this issue of the Journal of Clinical Sleep Medicine, Ioachimescu and colleagues provide new evidence that optimal adherence is critical in order to derive metabolic benefits from CPAP therapy.24 In a nonrandomized, point-of-care, comparative effectiveness study, 925 patients (86% men, 61% black, 65% obese, 28% with T2DM) who had been seen at the Atlanta VA Sleep Clinics for initial evaluation of OSA and underwent in-laboratory polysomnography were retrospectively identified. The reported median apnea-hypopnea index (AHI) for the entire cohort was 11 events/h (interquartile range [IQR] 5.5 to 20.7 events/h). OSA was diagnosed in 738 patients (52% mild, 30% moderate, and 19% severe OSA) and 720 patients were started on CPAP therapy. The cohort was then followed prospectively. The median duration of CPAP follow-up was 5 mo (IQR 3 to 14 mo). Overall CPAP adherence was low with a median usage on only 34.5% of the nights and a median of 4.8 h/night (IQR 2.8 to 6.5 h/night). The investigators further categorized their patients into three different CPAP adherence groups: good adherence (≥ 4 h/night on ≥ 70% of the nights, n = 174), excellent adherence (≥ 6 h/night on ≥ 80% of the nights, n = 118), and outstanding adherence (≥ 8 h/night and ≥ 90% of the nights, n = 36). The authors note no significant changes in HbA1c and fasting blood glucose in patients with OSA who showed good adherence to CPAP therapy. However, in the subgroup with outstanding adherence, there was a significant improvement in delta fasting blood glucose and a trend toward improved delta HbA1c. The authors also report interesting findings with regard to patients with normal glycemic control at baseline. In this subgroup, 19% of patients who used CPAP less than 4 h/night developed impaired fasting glucose. In contrast, incident impaired fasting glucose was 8% and 5% in the excellent and outstanding CPAP adherence categories, respectively.
There are several important limitations to this large study; therefore, these interesting findings must be interpreted with caution. The observational nature of the study does not address causality and only provides associative, not pathophysiologic, links between OSA therapy and glucose metabolism. Another notable limitation is the variable and relatively short follow-up period. Moreover, the fact that most patients enrolled were men limits the generalizability of the results. The modest effect of CPAP on various glycemic measures could have also been affected by the overall mild OSA in this cohort (median AHI 11, 52% with mild OSA). As the authors note, poor CPAP adherence despite a closed VA system with frequent follow-ups is evident and disappointing. Several questions remain unanswered by this study. In a real world cohort of patients with OSA and varying glycemic indices, what is the optimal level of CPAP adherence that would lead to a clinically meaningful improvement in glycemic control? Further research is needed to address important questions regarding the ways in which various OSA treatment modalities can be utilized to significantly affect cardiometabolic health in patients with prediabetes and T2DM.
In line with prior laboratory-based proof-of-concept studies,17,18 Ioachimescu and colleagues have provided incremental evidence that effective treatment of OSA may lead to improvement in glucose metabolism dysregulation. Although such a high level of CPAP adherence may be difficult to attain for most of our patients, novel and better tolerated therapeutic approaches may eventually allow us to effectively treat OSA during the entire sleep period.
Dr. Mokhlesi is supported by National Institutes of Health grant R01HL119161, has served as a consultant to Philips/Respironics, and has received research support from Philips/Respironics. He has also received an honorarium from Zephyr Medical Technologies and has served on the advisory board of Itamar Medical. Dr. Kaur has indicated no financial conflicts of interest.