Abstract
Rationale: Obstructive sleep apnea (OSA) is reported to have a metabolic profile predisposing to cardiovascular disease. However, previous case-control studies have not adequately controlled for confounders.
Objectives: To determine whether OSA is associated with increased insulin resistance and related metabolic risk factors.
Methods: We performed a matched case-control study (n = 42) examining putative metabolic risks among men with OSA attending a sleep clinic (apnea–hypopnea index [AHI] > 15] compared with no OSA (AHI < 5). Participants were matched for age ± 5 yr, body mass index ± 10%, and current smoking status. They were free of diabetes, clinically demonstrable cardiovascular disease, marked hypertension, and dyslipidemia.
Measurements and Main Results: Mean ± SD AHI was higher in patients with OSA (40 ± 27) than in control subjects (3 ± 1.3, p = 0.02), and median (interquartile range) nocturnal oxygen saturation was lower (OSA, 83 [76–88]; control, 91 [90–93]%; p < 0.001). Patients with OSA had a higher median (interquartile range) homeostasis model assessment score for insulin resistance (OSA, 1.7 [0.8–4.1]; control, 1.0 [0.7–1.8] mU·mmol/L2; p = 0.02), total cholesterol (OSA, 5.6 [4.8–6.2]; control, 4.8 [4.3–5.4] mmol/L; p = 0.049), and low-density-lipoprotein cholesterol (OSA, 3.8 [2.9–4.2]; control, 3.1 [2.6–3.6] mmol/L; p = 0.04). Patients with OSA had higher 24-h and nocturnal (12-h) urinary norepinephrine excretion and plasma leptin levels, and lower insulin-like growth factor (IGF)-1 levels (all, p ⩽ 0.02). Multiple linear regression, adjusting for central obesity, age, and alcohol consumption, confirmed an independent association between OSA and metabolic risks (all, p < 0.05), with a trend for IGF-1 (p = 0.053).
Conclusions: In a sleep clinic population, men with OSA and no identifiable cardiovascular disease have increased insulin resistance and other metabolic changes that act to increase the risk of vascular disease, compared with age- and body mass index–matched attendees without OSA.
Obstructive sleep apnea has been associated with increased vascular risk. There is increasing evidence of metabolic changes occurring in this condition that predispose to cardiovascular disease.
This study shows metabolic changes in a sleep clinic population of men with obstructive sleep apnea compared with control subjects matched for body mass index and age. Changes persisted when allowance was made for central obesity.
Although a number of case-control studies have assessed the effect of OSA on metabolic risk factors, there are concerns that matching for potential confounding variables has been inadequate (7, 8). Central obesity is an important confounder because it is found commonly in OSA and is a major determinant of metabolic syndrome and insulin resistance (9, 10). We hypothesized that men attending a sleep clinic with OSA would have metabolic changes associated with increased cardiovascular risk when compared with attendees without OSA. We sought to carefully match cases and control subjects for age and body mass index (BMI) and to exclude those with clinically demonstrable cardiovascular disease or diseases known to increase vascular risk, such as diabetes and dyslipidemia. Insulin resistance may play a key role in predisposing to cardiovascular disease and so it was chosen as the a priori primary outcome. Multiple linear regression was used to assess the independent relationship between OSA and metabolic risks after controlling for confounders, including central obesity. Some of the results of this study have been reported as an abstract (11).
A matched controlled study was performed among males attending the sleep clinic of a large tertiary hospital and who had undergone full overnight polysomnography (PSG) to establish whether OSA was present. The PSG was done using the Compumedics E-series (Compumedics Ltd, Abbotsford, Australia). Sleep staging and respiratory events were scored using standard criteria (12, 13).
Cases and control subjects were selected based on the apnea–hypopnea index (AHI), i.e., the number of apneas plus hypopneas per hour slept, regardless of symptom severity. Cases had at least moderate OSA (AHI > 15) and control subjects had no OSA (AHI < 5 events per hour slept) (12).
Participants had to meet the following cardiovascular disease and vascular risk criteria: (1) no history of ischemic heart disease, cerebrovascular, or peripheral vascular disease, and normal cardiovascular examination; (2) no diabetes (no previous physician diagnosis and fasting blood glucose < 7 mmol/L; (3) no moderate–severe hypertension (no previous physician diagnosis and measured blood pressure [BP] < 160/90); and (4) no moderate–severe dyslipidemia (no previous physician diagnosis and cholesterol < 7 mmol/L, triglycerides < 5 mmol/L). No patients were taking diabetic, antihypertensive, or lipid-lowering medications.
Participants were also excluded for the following: (1) age younger than 18 or older than 70 yr; (2) severe medical illness; (3) excessive alcohol intake (average alcohol consumption > 30 g/d); (4) inability to understand the study due to language difficulties or mental illness; or (5) previous treatment for OSA with continuous positive airway pressure, mandibular repositioning splint, or upper airway surgery.
Participants were matched by age ± 5 yr, BMI [BMI = weight/(height)2] ± 10%, and current smoking status.
An initial assessment occurred in the morning after participants were asked to fast and abstain from caffeinated drinks and cigarette consumption for 12 h. A medical examination was performed and weight, height, and waist and hip circumference were measured, and BP was recorded using a validated automated sphygmomanometer (Dinamap; Critikon, Inc., Tampa, FL) according to current guidelines (14). Blood tests were taken for glucose, insulin, lipids, insulin-like growth factor (IGF)–1, TNF-α, leptin, and adiponectin. Insulin resistance was estimated using the homeostasis model assessment (HOMA) score (15) [fasting serum insulin (mU/L) × fasting plasma glucose (mmol/L)/22.5].
At a subsequent visit, two consecutive 12-h (“day” and “night”) urine samples were collected for catecholamine excretion using our standard methods.
Additional detail on the methods of making the above measurements is provided in the online supplement.
All analyses were performed blind to case-control status.
Power calculations indicated that 21 pairs of participants were needed for a power of 90% to detect a real difference in insulin resistance between cases and control subjects of 0.5 the standard deviation of the differences between groups (i.e., a moderate effect size) (16) at p < 0.05.
Statistical analyses were performed using SPSS (release 11.0 for Windows; SPSS, Inc., Chicago, IL). Unmatched comparisons were made using the unpaired t test (parametric) and the Mann-Whitney U test (nonparametric data). Matched comparisons were made using the paired t test (parametric) and the Wilcoxon's rank test (nonparametric data).
To assess the independent association between OSA and metabolic risks, a multiple linear regression model was used with OSA as a “dummy” independent variable (i.e., OSA: yes [AHI > 15] or no [AHI < 5]), after adjusting for central obesity, age, and alcohol consumption. To further explore the relationship between insulin resistance and OSA, indices of sleep-disordered breathing were entered as independent variables. Residuals were examined and skewed variables were log transformed. Statistical significance was considered to occur at p < 0.05.
The study was approved by both hospital ethics committees, and written consent was obtained from all participants.
Twenty-one pairs of male participants were recruited and completed the study. Cases and control subjects were similar in age (both 46 yr), weight (OSA, 86.7; control, 86.4 kg), and BMI (both 28 kg/m2), as expected (Table 1). Measures of central obesity were also similar between the two groups (Table 1). There was no significant change in weight between the overnight PSG and the study assessments (mean [95% confidence interval (CI)] difference, PSG assessments: OSA, 0.4 [−0.2, 0.9]; control, 1.4 [−0.1, 2.8] kg). The groups had similar blood pressures, weekly alcohol consumption, and current smoking status (Table 1). There were no differences in medications used between groups (number of any medications: OSA, 9; control, 16; asthma medications: OSA, 1; control, 6; antidepressants: OSA, 1; control, 3; nocturnal sedatives: OSA, 0; control, 3; nonsteroidal antiinflammatories: OSA, 1; control, 2; and miscellaneous: OSA, 5; control, 8; two subjects receiving dopaminergic agents did not have catecholamines measured; all p > 0.05).
Vascular Risks | OSA (n = 21) | Control (n = 21) | Mean Difference (95% CI) |
|---|---|---|---|
| Age, yr | 46 ± 10.2 | 46 ± 9.7 | 0.29 (−1.2, 1.8) |
| Weight, kg | 86.7 ± 9.7 | 86.4 ± 11.4 | 0.28 (−4.6, 5.2) |
| Body mass index, kg/m2 | 28.4 ± 3.4 | 27.9 ± 3.6 | 0.57 (−0.35, 1.5) |
| Waist:hip ratio | 0.96 (0.92–0.97) | 0.96 (0.91–1.00) | −0.002 (−0.04, 0.04) |
| Waist circumference, cm | 98 ± 8 | 97 ± 10 | 0.86 (2.0, −3.3) |
| Current smokers, n | 2 | 2 | —* |
| Alcohol consumption, g/wk | 70 (15–130) | 20 (10–80) | 2.0 (−2.5, 6.4) |
| Systolic BP, mm Hg | 122.5 ± 13 | 119.6 ± 13 | 2.9 (−5.9, 11.7) |
| Diastolic BP, mm Hg | 71.3 ± 8 | 73.1 ± 7 | −1.8 (−7.3, 3.6) |
The cases had severe OSA (mean ± SD AHI = 40 ± 27 events/h) and the control subjects had no OSA (AHI = 2.8 ± 1.4, p < 0.001). The groups had similar total sleep time and sleep efficiency (Table 2). Subjects with OSA had less slow-wave sleep, a higher arousal index, and were more hypoxic than control subjects (Table 2). The Epworth Sleepiness Scale score showed a trend to increased sleepiness in the OSA group compared with controls (Table 2). Seven control subjects had periodic leg movement syndrome (PLMS) (17), with a mean ± SD periodic leg movement index of 18 ± 13 per hour slept, and the remainder had simple snoring.
OSA (n = 21) | Control (n = 21) | Mean Difference (95% CI) | p Value | |
|---|---|---|---|---|
| Apnea–hypopnea index, events per hour slept | 40 ± 27 | 2.8 ± 1.5 | 37.2 (25.2, 49.3) | < 0.001 |
| Arousal index, Events per hour slept | 44 ± 19 | 19 ± 12 | 25.2 (11.8, 38.6) | 0.001 |
| Oxygen saturation < 90% time, s | 147 (9–739) | 0 (0–0) | 2309 (−345, 4,962) | 0.001 |
| Lowest oxygen saturation, % | 83 (76–88) | 91 (90–93) | −10.5 (−15.8, −5.3) | < 0.001 |
| Total sleep time, min | 392 ± 68 | 363 ± 70 | 29.3 (−16.8, 75.4) | 0.2 |
| Sleep efficiency index, % | 77 ± 15 | 78 ± 12 | 0.8 (−6.4, 8.0) | 0.8 |
| Total slow-wave time, min | 28 ± 21 | 44 ± 26 | −15.9 (−29.6, −2.2) | 0.025 |
| Periodic leg movement index, number per hour slept | 0 (0–0) | 15 (9–16) | −6.4 (−11.4, −1.3) | < 0.001 |
| Epworth Sleepiness Scale score | 12 ± 6 | 9 ± 5 | 3.0 (−0.8, 6.9) | 0.1 |
Participants with OSA had higher levels of serum insulin and showed a trend to higher serum glucose (Table 3). Median (interquartile range [IQR]) insulin resistance (HOMA) was higher in those with OSA (OSA, 1.7 [IQR, 0.8–4.1]; control, 1.0 [IQR, 0.7–1.8]; p = 0.02) (Table 3). Patients with OSA had higher total and low-density-lipoprotein (LDL) cholesterol, but no difference in high-density-lipoprotein (HDL) cholesterol or triglycerides (Table 3). There was a weak trend to increased metabolic syndrome (18) among the patients with OSA (n = 5) compared with control subjects (n = 1; p = 0.19). Twenty-four hour and nocturnal urinary norepinephrine levels were higher in the patients with OSA (Table 3). Leptin levels were higher in the OSA group (OSA, 7.8; control, 4.1; p = 0.01). IGF-1 levels were reduced in participants with OSA (Table 3). TNF-α and adiponectin were similar in both groups (p ⩾ 0.8) (Table 3).
Metabolic Risk Factor | OSA (n = 21) | Control (n = 21) | Mean Difference (95% CI) | p Value |
|---|---|---|---|---|
| Insulin resistance (HOMA), mU·mmol/L2 | 1.7 (0.8–4.1) | 1.0 (0.7–1.8) | 1.2 (0.1, 2.2) | 0.02 |
| Insulin, mU/L | 8.0 (4–18.5) | 5.0 (4–9) | 4.6 (0.5, 8.7) | 0.02 |
| Glucose, mmol/L | 4.7 ± 0.5 | 4.5 ± 0.3 | 0.3 (−0.02, 0.6) | 0.07 |
| Cholesterol, mmol/L | 5.6 (4.8–6.2) | 4.8 (4.3–5.4) | 0.5 (0.01, 1.0) | 0.049 |
| LDL, mmol/L | 3.8 (2.9–4.2) | 3.1 (2.6–3.6) | 0.5 (0.06, 1.0) | 0.044 |
| HDL, mmol/L | 1.2 (1.0–1.4) | 1.2 (1.1–1.6) | −0.05 (−0.2, 0.1) | 0.6 |
| Triglycerides, mmol/L | 1.1 (0.8–1.8) | 1.0 (0.7–1.4) | 0.1 (−0.15, 0.41) | 0.5 |
| IGF-1, μg/L | 120 (100–150) | 140 (120–170) | −24.9 (−44.6, −5.2) | 0.02 |
| Leptin, ng/ml* | 7.8 (4.2–14.2) | 4.1 (1.8–6.7) | 4.2 (1.2, 7.2) | 0.01 |
| TNF-α,* pg/ml | 1.2 (1.0–2.2) | 1.4 (1.0–1.9) | 0.7 (−0.6, 2.0) | 0.9 |
| Adiponectin,* μg/ml | 4.35 (3.09–5.43) | 4.30 (3.14–6.25) | −0.01 (−1.34, 1.32) | 0.8 |
| Urinary norepinephrine, nmol/24 h† | 348 (270–442) | 224 (174–310) | 149 (27, 270) | 0.02 |
| Urinary norepinephrine: night, nmol/12 h† | 140 (99–195) | 88 (49–145) | 102 (13, 191) | 0.005 |
| Urinary norepinephrine: day, nmol/12 h† | 183 (117–281) | 124 (91–160) | 47 (−17, 110) | 0.12 |
| Urinary epinephrine, nmol/24 h† | 38 (22–67) | 35 (24–60) | 14.1 (−4.6, 32.9) | 0.09 |
| Metabolic syndrome, n | 5 | 1 | —‡ | 0.19 |
Multiple linear regression confirmed an independent association between metabolic risks and OSA after controlling for central obesity, age, and alcohol consumption, with a trend for IGF-1 (p = 0.053) (Table 4).
Metabolic Risk Factor | Adjusted R2 | β Coefficient (95% CI) | p Value |
|---|---|---|---|
| Insulin resistance (HOMA), mU·mmol/L2 | 0.18 | 1.13 (0.10, 2.16) | 0.03 |
| Insulin, mU/L | 0.22 | 4.48 (0.46, 8.51) | 0.03 |
| Glucose, mmol/L | 0.04 | 0.24 (−0.04, 0.52) | 0.09 |
| Cholesterol, mmol/L | 0.22 | 0.49 (0.06, 0.92) | 0.028 |
| LDL, mmol/L | 0.21 | 0.49 (0.10, 0.89) | 0.016 |
| HDL, mmol/L | 0.09 | −0.05 (−0.21, 0.11) | 0.6 |
| Triglycerides, mmol/L | < 0.05 | 0.12 (−0.23, 0.48) | 0.5 |
| IGF-1, μg/L | 0.23 | −20.9 (−42.2, 0.3) | 0.053 |
| Leptin, ng/ml* | 0.64 | 0.22 (0.03, 0.4) | 0.02 |
| TNF-α,* pg/ml | 0.08 | 0.09 (−0.08, 0.26) | 0.3 |
| Adiponectin,* μg/ml | 0.16 | −0.03 (−1.4, 1.3) | 0.9 |
| Urinary norepinephrine, nmol/24 h† | 0.23 | 0.18 (0.05, 0.31) | 0.007 |
| Urinary norepinephrine, night, nmol/12 h† | 0.32 | 0.22 (0.05, 0.39) | 0.014 |
| Urinary norepinephrine: day, nmol/12 h† | 0.10 | 0.15 (0.01, 0.29) | 0.04 |
| Urinary epinephrine, nmol/24 h† | < 0.05 | 0.07 (−0.17, 0.31) | 0.57 |
AHI (log transformed; β [95% CI]: 0.42 [0.15, 1.9]; p = 0.02) and lowest oxygen saturation (β [95% CI]: −0.08 [−0.03, −0.14]; p = 0.005), but not arousal index, were predictive of insulin resistance.
This study shows that male sleep clinic patients with moderate or severe OSA (AHI > 15) had numerous metabolic changes predisposing them to vascular disease compared with attendees without OSA (AHI < 5) carefully matched for age and BMI. These metabolic changes were present in the absence of identifiable vascular diseases or other disorders known to predispose to vascular disease, such as diabetes. They included higher levels of insulin resistance, total and LDL cholesterol, serum leptin and urinary norepinephrine excretion, and a reduction in serum IGF-I levels. Multiple linear regression with OSA as a predictor, controlling for waist circumference, age, and alcohol consumption, confirms an independent association between OSA and these metabolic risks, with a trend for an association with decreased IGF-1.
Previous case-control studies of the association between OSA and metabolic risks have not made adequate allowance for confounders (7, 8). Obesity is strongly associated with metabolic changes predisposing to cardiovascular disease. A central distribution of obesity, in particular, has a major role in explaining insulin resistance, leptin levels, and the metabolic syndrome (9, 10) and is strongly associated with OSA. Adjusting for central obesity confirms an independent association between OSA and metabolic risks (insulin resistance, leptin, IGF-1, catecholamines) and may explain the lack of association (adiponectin, TNF-α) and novel associations (lipids) we found with other risks.
The primary outcome in this study is insulin resistance. We assessed this using the HOMA score, which is a well-validated measure of insulin resistance (15). Previous case-control studies yielded conflicting findings on the relationship between insulin resistance and OSA (19–23). The association between OSA and insulin resistance independent of central obesity remains uncertain. For example, the study by Vgontzas and colleagues (22) found higher levels of insulin resistance among men with OSA (n = 14) compared with BMI-matched control subjects (n = 11) but patients with OSA also had increased “visceral” (central) fat. The present case-control study is the first to show increased insulin resistance in patients with OSA compared with control subjects after allowance for central obesity and other important confounders. Multiple linear regression also demonstrates a dose–response relationship between insulin resistance and common indices of OSA severity (AHI, oxygen desaturation).
Limited case-control data (20, 24) suggest OSA is associated with a pattern of dyslipidemia typical of the metabolic syndrome (decreased HDL cholesterol, increased total:HDL cholesterol ratio and/or triglycerides). By contrast, the current study is the first to report an increase in LDL and total cholesterol among patients with OSA. The current study has allowed for central obesity, which may explain the discrepancy with previous work. Insulin resistance could increase total and LDL cholesterol by decreasing the catabolism of LDL cholesterol, probably by down-regulation of the LDL receptors (25).
We found an increase in norepinephrine levels in OSA, confirming the majority of investigations (8). Previous studies have been criticized for poor allowance for confounders (8); by contrast, our groups were well matched for common confounders affecting catecholamines (age, BMI, BP) and participants were not taking potentially interacting drugs.
IGF-1 levels (reflecting growth hormone secretion) were reduced in the OSA group, in keeping with a small case-control study (26) and correlational data (27) Reduced IGF-1 levels may be an additional explanation for the increase in LDL cholesterol (28).
In contrast to a previous study (29), the present study confirms most case-control studies (5, 6, 24) showing increased serum leptin in OSA, despite similar levels of obesity, suggesting OSA is a leptin-resistant condition.
We found no association between OSA and adiponectin, possibly because our groups had similar central obesity. Our findings are in keeping with Makino and colleagues who found adiponectin was correlated with visceral fat rather than OSA (30).
It is postulated that hypoxia may induce insulin resistance by producing reactive oxygen species and inflammatory cytokines. We did not find OSA was associated with increased levels of TNF-α, in contrast to Vgontzas and colleagues (6). Hypoxia, which is believed to induce inflammatory cytokines, was less severe among our patients with sleep apnea (mean ± SE minimum oxygen saturation = 80.4 ± 2.5%) compared with those of Vgontzas and colleagues (mean ± SE minimum oxygen saturation = 74.6 ± 3.3%) (6).
Unlike the current study, previous case-control studies (20, 19, 6, 21, 22) compared patients recruited from a hospital-based sleep clinic with control subjects, many, or all, of whom were recruited from the community. Such designs may produce a selection bias (31) because seeking medical attention that culminates in referral to an outpatient clinic of any type may select a group with increased risk of vascular disease relative to that found in the general community.
In keeping with epidemiologic investigations of vascular risk in OSA (2–4), we defined cases and control subjects by their AHI, regardless of daytime symptoms. By contrast, others have incorporated daytime symptoms of OSA into their case definitions (6, 20). Our findings indicate increased metabolic risks among patients with OSA even though some report minimal or no daytime OSA symptoms.
The current study excluded participants with known cardiovascular disease or diseases predisposing to vascular disease and so may have excluded those patients with OSA most predisposed to vascular disease. Furthermore, our matching constraints may have excluded some patients with OSA with the metabolic syndrome and those with more severe desaturation (e.g., control subjects with high BMIs). These aspects of our study design are likely to have made it more difficult to find differences between the groups. Despite this, we found multiple metabolic risks in those with OSA compared with matched control subjects.
Although we report a large number of outcome measures, the primary outcome measure (insulin resistance) was selected a priori. Furthermore, 8 of our 15 outcome measures were positive, a finding that cannot be explained merely by chance.
We did not study females as to avoid potential interactions with cyclic hormonal changes, and so can only generalize our findings to males. However, in the regression analysis by Ip and colleagues (21), the association between AHI and insulin resistance did not show a gender effect. Hence, we believe it is likely that females with OSA would have increases in vascular risk factors similar to the substantial changes we observed in males.
The authors thank research nurses Vicki Stenning and Apolonia Maria Arnolda for their assistance during the study. They also thank Dick Chan for analyses of adipocytokines.
| 1. | Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230–1235. |
| 2. | Shahar E, Whitney CW, Redline S, Lee ET, Newman AB, Javier NF, O'Connor GT, Boland LL, Schwartz JE, Samet JM. Sleep-disordered breathing and cardiovascular disease: cross-sectional results of the Sleep Heart Health Study. Am J Respir Crit Care Med 2001;163:19–25. |
| 3. | Peker Y, Hedner J, Norum J, Kraiczi H, Carlson J. Increased incidence of cardiovascular disease in middle-aged men with obstructive sleep apnea: a 7-year follow-up. Am J Respir Crit Care Med 2002;166:159–165. |
| 4. | Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:1378–1384. |
| 5. | Phillips BG, Kato M, Narkiewicz K, Choe I, Somers VK. Increases in leptin levels, sympathetic drive, and weight gain in obstructive sleep apnea. Am J Physiol Heart Circ Physiol 2000;279:H234–H237. |
| 6. | Vgontzas AN, Papanicolaou DA, Bixler EO, Hopper K, Lotsikas A, Lin HM, Kales A, Chrousos GP. Sleep apnea and daytime sleepiness and fatigue: relation to visceral obesity, insulin resistance, and hypercytokinemia. J Clin Endocrinol Metab 2000;85:1151–1158. |
| 7. | Punjabi NM, Ahmed MM, Polotsky VY, Beamer BA, O'Donnell CP. Sleep-disordered breathing, glucose intolerance, and insulin resistance. Respir Physiol Neurobiol 2003;136:167–178. |
| 8. | Coy TV, Dimsdale JE, Ancoli-Israel S, Clausen J. Sleep apnoea and sympathetic nervous system activity: a review. J Sleep Res 1996;5:42–50. |
| 9. | Carr DB, Utzschneider KM, Hull RL, Kodama K, Retzlaff BM, Brunzell JD, Shofer JB, Fish BE, Knopp RH, Kahn SE. Intra-abdominal fat is a major determinant of the National Cholesterol Education Program Adult Treatment Panel III criteria for the metabolic syndrome. Diabetes 2004;53:2087–2094. |
| 10. | Cnop M, Landchild MJ, Vidal J, Havel PJ, Knowles NG, Carr DR, Wang F, Hull RL, Boyko EJ, Retzlaff BM, et al. The concurrent accumulation of intra-abdominal and subcutaneous fat explains the association between insulin resistance and plasma leptin concentrations: distinct metabolic effects of two fat compartments. Diabetes 2002;51:1005–1015. |
| 11. | McArdle N, Philpott J, Hillman D, Beilin LJ, Watts JF. Case control study: sleep apnoea clinic patients have insulin resistance and dyslipidaemia. Eur Respir J 2004;48:2098. |
| 12. | American Academy of Sleep Medicine Task Force. Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research: the report of an American Academy of Sleep Medicine Task Force. Sleep 1999;22:667–689. |
| 13. | Rechtschaffen A, Kales A. A manual of standardized terminology and scoring system for sleep stages of human subjects. Washington, DC: U.S. Government Printing Office; 1968. NIH Publication No.204. |
| 14. | The sixth report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med 1997;157:2413–2446. |
| 15. | Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–419. |
| 16. | Altman DG. Statistics and ethics in medical research: III. How large a sample? BMJ 1980;281:1336–1338. |
| 17. | Lugaresi E, Cirignotta F, Coccagna G, Montagna P. Nocturnal myoclonus and restless legs syndrome. Adv Neurol 1986;43:295–307. |
| 18. | World Health Organization. Definition, diagnosis and classification of diabetes mellitus and its complications: report of a WHO consultation. Geneva, Switzerland: World Health Organization; 1999. |
| 19. | Davies RJ, Turner R, Crosby J, Stradling JR. Plasma insulin and lipid levels in untreated obstructive sleep apnoea and snoring; their comparison with matched controls and response to treatment. J Sleep Res 1994;3:180–185. |
| 20. | Coughlin SR, Mawdsley L, Mugarza JA, Calverley PM, Wilding JP. Obstructive sleep apnoea is independently associated with an increased prevalence of metabolic syndrome. Eur Heart J 2004;25:735–741. |
| 21. | Ip MS, Lam B, Ng MM, Lam WK, Tsang KW, Lam KS. Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med 2002;165:670–676. |
| 22. | Vgontzas AN, Bixler EO, Chrousos GP. Metabolic disturbances in obesity versus sleep apnoea: the importance of visceral obesity and insulin resistance. J Intern Med 2003;254:32–44. |
| 23. | Shin C, Kim J, Kim J, Lee S, Shim J, In K, Kang K, Yoo S, Cho N, Kimm K, et al. Association of habitual snoring with glucose and insulin metabolism in nonobese Korean adult men. Am J Respir Crit Care Med 2005;171:287–291. |
| 24. | Ip MS, Lam KS, Ho C, Tsang KW, Lam W. Serum leptin and vascular risk factors in obstructive sleep apnea. Chest 2000;118:580–586. |
| 25. | Mazzone T, Foster D, Chait A. In vivo stimulation of low-density lipoprotein degradation by insulin. Diabetes 1984;33:333–338. |
| 26. | Gianotti L, Pivetti S, Lanfranco F, Tassone F, Navone F, Vittori E, Rossetto R, Gauna C, Destefanis S, Grottoli S, et al. Concomitant impairment of growth hormone secretion and peripheral sensitivity in obese patients with obstructive sleep apnea syndrome. J Clin Endocrinol Metab 2002;87:5052–5057. |
| 27. | Grunstein RR, Handelsman DJ, Lawrence SJ, Blackwell C, Caterson ID, Sullivan CE. Neuroendocrine dysfunction in sleep apnea: reversal by continuous positive airways pressure therapy. J Clin Endocrinol Metab 1989;68:352–358. |
| 28. | McCallum RW, Petrie JR, Dominiczak AF, Connell JM. Growth hormone deficiency and vascular risk. Clin Endocrinol (Oxf) 2002;57:11–24. |
| 29. | Barcelo A, Barbe F, Llompart E, Mayoralas LR, Ladaria A, Bosch M, Agusti AG. Effects of obesity on C-reactive protein level and metabolic disturbances in male patients with obstructive sleep apnea. Am J Med 2004;117:118–121. |
| 30. | Makino S, Handa H, Suzukawa K, Fujiwara M, Nakamura M, Muraoka S, Takasago I, Tanaka Y, Hashimoto K, Sugimoto T. Obstructive sleep apnoea syndrome, plasma adiponectin levels, and insulin resistance. Clin Endocrinol (Oxf) 2006;64:12–19. |
| 31. | Cole P. Introduction. In: Breslow NE, Day NE, editors. Statistical methods in cancer research. Vol. 1. Lyon, France: International Agency for Research on Cancer; 1980. pp. 14–40. |
