Background
Obstructive sleep apnea (OSA) is characterized by sleep-related airway obstructions that produce apnea. These events provoke arousals and cause oxygen desaturations and heightened sympathetic activity during sleep and waking hours [
1] that may play a role in the development of insulin resistance [
2]. Obesity is a strong risk factor for OSA [
3] and both obesity and OSA are associated with increased insulin resistance and diabetes [
4].
Hormones involved in the regulation of body weight and glucose metabolism include ghrelin, leptin, adiponectin and resistin. Ghrelin is an orexigenic hormone and it has been proposed as a cause of increased appetite and obesity [
5]. Administration of ghrelin increases adiposity, food intake and body weight [
6]. It also regulates glucose homeostasis increasing glucose levels and decreasing insulin secretion [
7]. Leptin is a hormone secreted by adipocytes in proportion to fat mass. It is elevated in obesity and its administration suppresses appetite and induces weight loss [
8]. Resistin and adiponectin are also adipocyte-derived hormones linked to obesity, insulin resistance, and diabetes. Adiponectin levels inversely correlate with BMI and are lower in individuals with diabetes whereas resistin directly correlates with obesity and insulin resistance.
Whether treatment of OSA can reverse insulin resistance and prevent body weight gain is controversial. Because hypoxemia-induced sympathetic activation is thought to be the source of the endocrine abnormalities often seen in patients with OSA, and continuous positive airway pressure (CPAP) effectively reverses hypoxemia in patients with OSA, we hypothesized that CPAP will decrease insulin resistance, ghrelin and resistin levels and increase adiponectin levels in a group of obese individuals with OSA.
Discussion
Our study suggests that glucose metabolism is disturbed in obese patients with OSA and that weight change rather than hypoxia is the major long-term modulating factor in insulin resistance after CPAP treatment in this population. These findings also suggest that CPAP alone may not reduce body weight, and that in the face of weight gain CPAP treatment may not reduce insulin resistance and leptin or increase adiponectin in obese subjects. The results of our regression analyses where the predictive value of BMI and nadir SpO2 was explored support this hypothesis given that changes in BMI but not changes in nadir SpO2 predicted changes in insulin, insulin resistance and leptin.
We did not observe any changes in blood pressure, or heart rate after CPAP treatment in contrast to what most [
15‐
17] but not all studies [
18‐
20] have reported. Possible explanations for this discrepancy include: 1) A higher body weight in our cohort compared to others or the fact that body weight remained stable or increased in our cohort. This could have negated the beneficial effects of CPAP on these outcomes as suggested by a previous report that showed that the course of hypertension in OSA is more closely linked to weight loss than to elimination of sleep apnea by CPAP [
16]; 2) Different duration of CPAP treatment (6 months in our study v. 1-2 months in other reports); 3) Time of the day at which BP was assessed given that CPAP effects on BP are reportedly more pronounced during sleep and we monitored our patients in the morning; 4) Methods of BP measurement since this factor has been shown to influence results [
17]; and 5) We did not power the study to detect differences in these outcomes so a negative result should be interpreted with caution.
Several reports have demonstrated an association between OSA and insulin resistance [
2,
21‐
24]. However, the effect of CPAP therapy on insulin resistance remains controversial (recently reviewed in [
25]). Some reports failed to detect an improvement in insulin sensitivity [
26], others showed an improvement in glucose levels only during sleep [
27,
28] and others showed an almost immediate improvement, especially in non-obese patients [
13]. In our study, we found increased insulin resistance after 6 months of CPAP use. This insulin resistance was associated with weight gain indicating that body weight plays a major role in determining insulin resistance in obese CPAP-treated patients with OSA. These results are in agreement with those reported by Ip and others [
21]. The apparently divergent findings between our results and those previously reporting an improvement in insulin sensitivity also may relate to differences in sample timing. Our assessment was done 6 months after starting treatment whereas most reports have been done between 48 hours and 3 months after starting CPAP. It is possible that CPAP use has only a transient effect on insulin sensitivity and that changes in body weight are a much more important factor in the long-term regulation of insulin sensitivity.
Ghrelin is an appetite-increasing hormone postulated as a contributor to OSA-associated obesity as ghrelin levels were elevated in one report [
14]. In the same study, fasting total (the sum of active and inactive) ghrelin levels decreased after 2 days of CPAP. Another study reported equivalent fasting total ghrelin levels in obese subjects with OSA and BMI matched controls without OSA [
29]. In our study, we measured active ghrelin instead of total ghrelin because 75% of the circulating peptide is biologically inactive and the ratio between inactive and active ghrelin changes in different clinical scenarios [
9]. Since ghrelin is suppressed by food intake, ghrelin levels were measured while fasting and during the OGTT. Our results show that 6 months of CPAP treatment significantly decreased fasting active ghrelin levels but that postprandial levels of this hormone remained unchanged. This is in agreement with a recent report of fasting active ghrelin levels being decreased by CPAP after one month of treatment [
30]. Although ghrelin inversely correlates with body weight in the setting of obesity, we did not found any association between changes in ghrelin levels and changes in BMI, CPAP use or changes in the ESS in this setting. Ghrelin correlated with changes in insulin resistance, suggesting that other factors besides body weight may play a role in its regulation including changes in insulin sensitivity. Insulin administration has been shown to suppress circulating ghrelin levels in some [
31] but not all studies [
32]. Plasma insulin levels and insulin resistance correlate inversely with ghrelin. This association was BMI-independent in some studies [
33]. However in a study using euglycemic hyperinsulinemic clamp method, insulin sensitivity did not correlate with ghrelin concentrations [
34]. Independent of metabolic factors, ghrelin may also act as a sleep-inducing hormone. Ghrelin levels increase after sleep deprivation [
35] and slow wave sleep is enhanced after ghrelin administration [
36]. Based on these data, we postulate that the fasting ghrelin level increase seen in patients with OSA is a compensatory response to poor-quality sleep and could explain why fasting ghrelin levels decreased after CPAP use.
Leptin is secreted by adipocytes in proportion to body fat, being elevated in obese individuals and decreasing with weight loss. Leptin-deficient animals exhibit respiratory depression and CO
2 retention. Leptin administration to these animals increases minute ventilation and improves lung mechanics [
37]. These animal experiments suggest that an increase in leptin levels in patients with OSA may represent a compensatory response to hypoxia. Consistent with this hypothesis, elevated leptin has been described in OSA patients compared to BMI-matched controls. This elevation in leptin was reversed by CPAP treatment [
14,
38], although this was associated with a decrease in fat accumulation in some studies [
39] that may have accounted at least partially for the changes in leptin. Others have reported that leptin levels are similar in obese OSA patients when compared to non-OSA controls and that these levels do not change significantly after 1 month or 1 year of CPAP [
30,
40]. In agreement with the latter study, our data showed that leptin levels remained stable after CPAP use. Taken together, these data suggest that if CPAP has an effect on leptin levels, it is short-lasting.
The role of resistin in diabetes remains a matter of debate. Circulating resistin levels directly correlate with BMI and have been shown to decrease with weight loss [
41]. Resistin also directly correlates with insulin resistance in some studies, but not in others [
42,
43]. In our study, resistin levels did not change after 6 months of CPAP and its levels did not correlate with changes in body weight, insulin and other adipokines or sleep parameters. In agreement with our data, resistin levels were stable after 2 days and 2 months of CPAP use in a group of subjects with OSA, suggesting that resistin is unlikely to play an important role in the insulin resistance or obesity seen in OSA [
13].
Adiponectin is decreased in obese individuals and in those with type 2 diabetes. It is thought to play a role in many of the metabolic complications suffered by these patients including metabolic syndrome and cardiovascular disease. However, its role in patients with OSA remains controversial. Elevated adiponectin was found in subjects with OSA when compared with non-OSA controls in one report and diminished in another [
44,
45]. In agreement with prior reports of adiponectin levels after CPAP use [
46], we report here that adiponectin levels remained unchanged after 6 months of CPAP treatment. Harsch et al. had previously reported a decrease in adiponectin levels after 48 hrs of CPAP use but levels returned to baseline at 3 months. The data suggest that chronic CPAP treatment does not play a role in the regulation of adiponectin levels.
Although the study was powered a priori using published data [
13,
14], the small sample size is a limitation of this study. Other limitations include the lack of data on changes in dietary habits; physical activity and body composition that could help us better understand the effects of CPAP on hormonal regulation. Also, it would have been useful to compare changes in body weight and other parameters with a non-interventional group of controls. However, such a group was not included in our design because these subjects have a clinical indication for CPAP use and delaying its use would have been unethical. Our study was powered to detect significant differences in insulin resistance and ghrelin levels. Consequently, we cannot conclude that the lack of changes in leptin, adiponectin and resistin levels in this relatively small sample would not be seen in a larger sample. Significant associations detected during simple correlation analysis should be interpreted with caution given the number of variables compared which increase the chance for a type I error. Future studies should include a larger number of patients along with an assessment of dietary habits; physical activity, energy expenditure, anthropometrics (i.e. waist-to-hip ratio) and body composition in order to better understand the effects of CPAP in this setting.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
JMG and AS participated in the design of the study and in writing the manuscript. HS recruited patients and collected the data. MH and RN performed the PSG studies. JG performed the statistical analysis and hormonal assays. JG, HS, RN, MH, AS reviewed and approved the final version of the manuscript.