Portal de Periódicos da CAPES

CPAP Therapy of Obstructive Sleep Apnea in Type 2 Diabetics Improves Glycemic Control During Sleep

2008; American Academy of Sleep Medicine; Volume: 04; Issue: 06 Linguagem: Inglês

10.5664/jcsm.27347

1550-9397

Arthur Dawson, Susan L. Abel, Richard T. Loving, George Dailey, Farhad F. Shadan, John Cronin, Daniel F. Kripke, Lawrence E. Kline,

Dysphagia Assessment and Management

Free AccessCPAPCPAP Therapy of Obstructive Sleep Apnea in Type 2 Diabetics Improves Glycemic Control During Sleep Arthur Dawson, M.D., Susan L. Abel, M.P.H., Richard T. Loving, D.N.Sc., George Dailey, M.D., Farhad F. Shadan, M.D., Ph.D., John W. Cronin, M.D., Daniel F. Kripke, M.D., Lawrence E. Kline, D.O. Arthur Dawson, M.D. Scripps Clinic Sleep Center, Division of Chest and Critical Care Medicine, Scripps Clinic, La Jolla, CA Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA , Susan L. Abel, M.P.H. Scripps Clinic Sleep Center, Division of Chest and Critical Care Medicine, Scripps Clinic, La Jolla, CA , Richard T. Loving, D.N.Sc. Scripps Clinic Sleep Center, Division of Chest and Critical Care Medicine, Scripps Clinic, La Jolla, CA , George Dailey, M.D. Division of Diabetes and Endocrinology, Scripps Clinic, La Jolla, CA Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA , Farhad F. Shadan, M.D., Ph.D. Scripps Clinic Sleep Center, Division of Chest and Critical Care Medicine, Scripps Clinic, La Jolla, CA Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA , John W. Cronin, M.D. Scripps Clinic Sleep Center, Division of Chest and Critical Care Medicine, Scripps Clinic, La Jolla, CA Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA , Daniel F. Kripke, M.D. Scripps Clinic Sleep Center, Division of Chest and Critical Care Medicine, Scripps Clinic, La Jolla, CA Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA , Lawrence E. Kline, D.O. Scripps Clinic Sleep Center, Division of Chest and Critical Care Medicine, Scripps Clinic, La Jolla, CA Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA Published Online:December 15, 2008https://doi.org/10.5664/jcsm.27347Cited by:89SectionsAbstractPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTBackground:Type 2 diabetes and obstructive sleep apnea (OSA) are frequently comorbid conditions. OSA is associated with increased insulin resistance, but studies of continuous positive airway pressure (CPAP) have shown inconsistent effects on glycemic control. However, endpoints such as hemoglobin A1c and insulin sensitivity might not reflect short-term changes in glycemic control during sleep.Methods:We used a continuous glucose-monitoring system to measure interstitial glucose every 5 minutes during polysomnography in 20 patients with type 2 diabetes and newly diagnosed OSA. The measurements were repeated after an average of 41 days of CPAP (range 26–96 days). All patients were on a stable diet and medications. Each 30-second epoch of the polysomnogram was matched with a continuous glucose-monitoring system reading, and the sleeping glucose level was calculated as the average for all epochs scored as sleeping.Results:The mean sleeping glucose decreased from untreated (122.0 ± 61.7 mg/dL) to treated (102.9 ± 39.4 mg/dL; p = 0.03 by Wilcoxon paired rank test). The sleeping glucose was more stable after treatment, with the median SD decreasing from 20.0 to 13.0 mg/dL (p = 0.005) and the mean difference between maximum and minimum values decreasing from 88 to 57 mg/dL (p = 0.003). The change in the mean hemoglobin A1c from 7.1% to 7.2% was not significant.Conclusions:Our study is limited by the lack of a control group, but the results suggest that sleeping glucose levels decrease and are more stable after patients with type 2 diabetes and OSA are treated with CPAP.Citation:Dawson A; Abel SL; Loving RT; Dailey G; Shadan FF; Cronin JW; Kripke DF; Kline LE. CPAP therapy of obstructive sleep apnea in type 2 diabetics improves glycemic control during sleep. J Clin Sleep Med 2008;4(6):538–542.INTRODUCTIONObstructive sleep apnea (OSA) and type 2 diabetes mellitus are frequently comorbid conditions. Clinic-based estimates of the prevalence of OSA in type 2 diabetics range from 18% to as high as 36%,1–3 and some 50% of patients with OSA have type 2 diabetes or impaired carbohydrate metabolism.4 Population surveys, the Wisconsin Sleep Cohort,5 and the Sleep Heart Health Study6 estimate the prevalence of type 2 diabetes in patients with OSA (apnea-hypopnea index [AHI] > 15) to be about 15%. Obesity is common in both conditions, but OSA is associated with increased insulin resistance independent of obesity.7, 8If OSA increases insulin resistance, then one might expect treating it to increase insulin sensitivity and to improve glycemic control. However, studies of the effect of treating OSA on insulin sensitivity in diabetics and nondiabetics have yielded inconsistent and mostly negative results.9, 10 However, Babu et al11 reported an improvement in postprandial blood glucose after patients with type 2 diabetes and OSA were treated with continuous positive airway pressure (CPAP) for about 3 months.OSA causes intermittent hypoxia and sleep fragmentation, which are capable of activating the sympathetic nervous system and the hypothalamic-pituitary axis.9 By decreasing this activation, CPAP therapy might result in short-term improvements in glycemic control during sleep in patients with OSA that would be reflected less in daytime values or in slowly responding measurements such as insulin sensitivity and hemoglobin A1c (HbA1c).In this study, we used a continuous glucose monitoring system (CGMS) to measure glucose levels during polysomnography recordings of sleep in patients with type 2 diabetes and moderate to severe OSA, first before treatment and then after 4 to 13 weeks of CPAP therapy.METHODSSubjectsWe selected patients with type 2 diabetes who were on a stable diabetic regimen. All were recruited at the time of their initial consultation with a sleep physician. They were considered eligible if the history and physical examination, supplemented by overnight oximetry, indicated that the patients were likely to have at least moderate OSA. All were newly diagnosed with OSA, and none had any previous experience with CPAP. They were instructed to maintain a constant diet and not to attempt to lose weight during the study. The study received approval from the Scripps Clinic Human Subjects Committee.ProcedureThe interstitial glucose level was monitored during a full-night diagnostic polysomnogram. Patients whose AHI was greater than 15 were eligible to continue the study. They were provided with an autoadjusting CPAP unit for 1 week (REMstar Auto with Encore Smart Card, Respironics, Pittsburg, PA) and then were switched to fixed-level CPAP set at the 90th percentile of the pressure required during the autoadjusting period. CPAP was continued for at least 3 more weeks, during which patients were seen weekly to monitor their adherence and to deal with any problems in using the equipment. We required that, for patients to be eligible to continue the study, they had to use the CPAP for an average of at least 4 hours per night from the time they began fixed-level CPAP. These patients had a second polysomnogram with recording of interstitial glucose while they were on fixed-level CPAP. Subjects were seen during the second day after their polysomnogram to remove the CGMS and to download the data so that we could obtain the minimum of 36 hours of glucose recording.Interstitial GlucoseGlucose levels were monitored using the CGMS System (Medtronic-MiniMed, Northridge, CA), which records the interstitial glucose level every 5 minutes for up to 72 hours. The sensor was inserted subcutaneously in the lower abdomen in the afternoon or early evening before the study night at least 3 hours before lights out. The CGMS was calibrated against a finger-stick blood glucose, the results of which were entered 3 to 4 times daily.Subjects were sent to the hospital cafeteria with instructions to take their evening meal at the same time and in the same amount on the 2 nights of polysomnography.PolysomnographyPolysomnography was done using the Compumedics E-series system (Compumedics USA, Ltd., El Paso, TX). We followed the guidelines of the American Academy of Sleep Medicine using 2 channels of electroencephalography (C3-A2, O1-A2), 2 electrooculogram channels, submental electromyography, surface anterior tibial electromyography, electrocardiography, nasal air flow by thermistor and/or nasal pressure transducer, respiratory movements (thoracic and abdominal belts), snoring, body position, and pulse oximetry. The American Academy of Sleep Medicine 1999 criteria were used for identifying apneas and hypopneas.12Data AnalysisThe sleep stage was scored for each 30-second epoch of the polysomnogram. To account for the time lag between the blood glucose and the interstitial glucose levels, 5 minutes were added to the time of each polysomnography epoch. Since the epochs were 30 seconds each, whereas the CGMS reported the glucose level every 5 minutes, the same glucose level was attributed to a block of 10 polysomnography epochs. The actual difference between the polysomnography epoch time and the CGMS time could range from 4.5 to 9.5 minutes. This is consistent with published estimates of the time delay between the interstitial glucose level recorded by the CGMS and the plasma glucose level.13, 14The primary endpoint was the mean interstitial glucose for all epochs scored as sleep from lights out to lights on (Gsleepmean).Secondary endpoints were the average glucose from lights out to lights on (Gnocmean), the average glucose during epochs scored as awake after sleep onset (Gwasomean), the standard deviation (SD) of the sleeping glucose (Gsleepsd), the maximum sleeping glucose (Gsleepmax), and the difference between the maximum and minimum sleeping glucose (Gsleepdiff).We also calculated the mean and SD of the 24-hour interstitial glucose and the postprandial level after breakfast. We determined the postprandial glucose by finding the peak interstitial value during a 2-hour window after the time the subject recorded for beginning breakfast. The values were then averaged over the period 30 minutes before and 30 minutes after the peak time. This should be similar to the postprandial value that Babu and associates described as decreasing after CPAP therapy.11 They recorded the mean values over a 1-hour period "on the basis of correlation with the food diary."As an additional assessment of glucose variability, we calculated MAGE, the mean amplitude of glycemic excursions, as described by Service et al.15 MAGE is designed to include major and to exclude minor fluctuations in the glucose level by averaging the peak-to-trough differences of changes exceeding 1 SD of the values observed over 24 hours. MAGE has shown a close correlation with urinary 8-iso prostaglandin F2α, an indicator of oxidative stress, and predicts it better than does the 24-hour coefficient of variation of glucose levels.16, 17 We also calculated MAGE over the 8-hour interval from 2300 to 0700 during which our subjects were fasting and were sleeping most of the time, though we recognize that all of the published data on MAGE have calculated it over 24 to 48 hours.Statistical AnalysisMost of the data were not normally distributed, and so we used 2-tailed nonparametric tests—Wilcoxon signed rank test for comparisons and Spearman rs for correlation. We used GraphPad Prism version 5 for Macintosh (GraphPad Software, San Diego, CA) with an α of 0.05 to reject the null hypothesis.RESULTSSubjectsA total of 28 subjects completed the first night of the study. One was dropped because her AHI proved to be less than 15. Two were dropped because the CGMS recording was unsatisfactory. Three additional subjects were not able to sleep with CPAP for the required 4 hours per night. This left 22 subjects who then completed the second study night while using CPAP. One of these was dropped from the analysis because he showed many central apneas during his night on CPAP with no improvement in the AHI. One additional subject recruited early in the study was excluded from the final analysis because she was the only type 1 diabetic and the only subject on an insulin pump. That left 20 subjects who completed the 2 study nights, and their data were included in the analysis. Their characteristics are described in Table 1. Eighteen of the 20 had a body mass index greater than 30, and 10 had a body mass index greater than 40. Sixteen of the 20 had severe sleep apnea (AHI > 30). In 8, the diabetes was not well controlled (HbA1c > 7.0). Eighteen were on only oral medications, and 2 were on insulin plus oral medications.Table 1 Chacteristics of the 12 Men and 8 Women Who Took Part in the StudyMeanSDRangeAge, y59.810.234–75BMI, kg/m239.68.023–59Duration of diabetes, y9.87.70.3–25AHI, no./h6330.419–113CPAP use, h/day5.81.04.6–7.3Days on CPAP before second PSG41.017.026–96HbA1c, %7.11.34.8–10.3BMI refers to body mass index; AHI, apnea-hypopnea index; CPAP, continuous positive airway pressure; PSG, polysomnogram; HbA1c, glycated hemoglobin.Clinical and Polysomnography Findings with CPAPThe clinical data are summarized in Table 2. Twelve subjects gained weight from the first to the second study night, 6 of them more than 2 kg. Four subjects had an AHI above 10 with CPAP, though the AHI was greatly improved from before treatment (74 to 113 before treatment decreasing to 23 to 29 after). In 1 of the 4, all of the events on CPAP were central apneas, whereas, in the other 3, they were a mixture central events and hypopneas that were difficult to characterize as obstructive or central.Table 2 Changes with CPAP TherapyUntreatedCPAPp Value MeanSDMeanSDWeight, kg117.928.8118.929.00.02ESS, score11.24.96.13.9< 0.001AHI, no./h6330.47.99.5< 0.001HgA1c, %7.11.37.21.3NSCPAP refers to continuous positive airway pressure; ESS, Epworth Sleepiness Scale (24-point scale, normal 0–10); AHI, apnea-hypopnea index; HbA1c, glycated hemoglobin.The polysomnography results are summarized in Table 3. After treatment, the subjects showed an increased total sleep time with less time awake after sleep onset and less time in stage 1 and more time in stage 3–4 and in stage REM. The number of arousals was significantly reduced with treatment.Table 3 Polysomnography ResultsUntreatedCPAPp ValueTotal sleep time, min367 (332–419)372 (348–415)NSWASO, min71 (49–105)61 (34–88)0.02Sleep stage, min 143 (33–66)27 (9–44)NS 2241 (175–278)233 (199–255)NS 334 (11–53)37 (13–82)NS 446 (26–68)69 (52–81)NSSleep efficiency, %79 (73–89)85 (78–91)NSArousal index, no./h Respiratory-related51 (30–74)2 (1–7)< 0.001 Total64 (38–79)17 (12–24)< 0.001Data are presented as median (interquartile range). CPAP refers to continuous positive airway pressure; WASO, wake time after sleep onset.Interstitial Glucose LevelsThe sleeping and nocturnal glucose levels are summarized in Table 4. The mean sleeping glucose level decreased with CPAP treatment, and it was less variable, as indicated by the decreased SD and the reduced difference between the maximum and minimum sleeping values. There was not much difference between Gnocmean and Gsleepmean, and the latter also decreased significantly on the CPAP night. The mean sleeping glucose level decreased in 10 of the 11 subjects whose Gsleepmean was greater than 100 mg/dL (5.5 mmol/L), and, in the other, it increased by only 2 mg/ dL (0.11 mmol/L). There was no decrease in the 9 subjects whose Gsleepmean was less than 100 mg/dL (untreated 84.3 mg/dL; treated 85.1 mg/dL), but, even in these subjects, there was a tendency for Gsleepmax to decrease (untreated 140.7 mg/dL (7.74 mmol/L and with CPAP, 118.2 mg/dL [6.50 mmol/L]; p = 0.07). There was no significant correlation between the change in Gsleepmean and the AHI (rs = −0.02, p = 0.95).Table 4 Interstitial Glucose LevelsUntreatedCPAPpMeanSDMeanSDMean sleeping (Gsleepmean)120.761.1101.939.00.03SD of sleeping values (GsleepSD)a20.013.1–31.513.09.3–18.20.004Maximum sleeping value (Gsleepmax)173.771.1135.542.6< 0.001Maximum - minimum sleeping value (Gsleepdiff)86.946.556.023.80.002Mean lights out to lights on (Gnocmean)123.962.0103.939.10.03Mean awake after sleep onset (Gwasomean)128.162.799.738.40.02Postprandial breakfast198.063.4196.659.6NS24-hour mean valueb153.752.0142.243.90.0324-hour SDa,b33.125.3–54.332.730.0–47.7NSMAGE 8a44.838.9–77.543.031.5–53.10.06MAGE 24a,b77.644.1–122.682.266.3–95.2NSResults are presented as mg/dL; to convert to mmol/L, divide by 18. MAGE refers to the mean amplitude of glycemic excursions; MAGE 8, the MAGE calculated over 8 hours from 2300 to 0700; MAGE 24, the MAGE calculated over 24 hours beginning at midnight.aMedian and interquartile rangebData from 2 subjects excluded because of missing dataGsleepwaso was also significantly decreased on the treatment night. On the untreated night, there was a nonsignificant tendency for Gsleepwaso to be greater than Gsleepmean (p = 0.055), whereas Gsleepwaso was no greater than Gsleepmean on the treatment night.The mean 24-hour glucose, beginning at midnight of the polysomnography night, also decreased significantly during CPAP treatment. However, the change with CPAP in the mean daytime glucose from 0700 hours to 2300 hours (167.4 ± 52.9 to 159.3 ± 51.6) was not significant (p = 0.36). We found no significant change in the variability of the 24-hour glucose levels when assessed by either the SD or MAGE. There was a tendency for MAGE, calculated from 2300 to 0700 hours, to decrease with CPAP therapy, but the change was not statistically significant. There was no significant change in the HbA1c or in the breakfast postprandial glucose level.DISCUSSIONSleeping and nocturnal hyperglycemia were reduced and the sleeping interstitial glucose level was less variable during CPAP treatment in this group of mostly obese type 2 diabetics with moderate to severe, mostly severe, OSA.We also found a decrease in the mean 24-hour glucose but not in the HbA1c. The lack of change in the HbA1c may be explained in part by the fact that only 3 of our subjects had 60 days or more of CPAP before the second PSG. Babu and associates showed a significant reduction in HbA1c in those subjects with a baseline level greater than 7%.11 However, among their subjects who adhered well to CPAP therapy, the HbA1c decreased consistently only in those who were treated for 60 days or more. The diabetes was less well controlled in the subjects of the Babu study than in ours, with a mean baseline HbA1c of 8.3%, but we found no decrease with CPAP therapy even in our 8 subjects whose baseline was above 7%.West and associates10 also reported that there was no significant change in the HbA1c and no change in insulin sensitivity after 3 months of CPAP therapy. The diabetes of their subjects was also less well controlled than the diabetes of the subjects in our study, with a mean HbA1c of 8.5% before treatment.Our study was not designed to evaluate the effect of treatment on postprandial glucose levels but, because of the report by Babu et al that postbreakfast levels were lower after treatment,11 we analyzed our data to see if we could show a similar improvement. We found no significant change with treatment. Though our subjects fasted until 0700 when they had blood drawn, we did not require them to take breakfast at a specific time or to eat a standard meal. Therefore, any effect of CPAP on the postprandial glucose could have been obscured by variations in the amount and timing of the meal.Harsch and associates showed an improvement in insulin sensitivity after 3 months of CPAP therapy in a group of obese type 2 diabetics with OSA.18 However, they had only 8 subjects in their series after excluding data from 1 who had a significant loss of weight during the study period.Hassaballa and associates reported a significant decrease in the HbA1c after at least 3 months of CPAP therapy in a group of obese type 2 diabetics, but their findings were based on a retrospective chart review.19 They also reported a weight gain in the small number of their subjects for whom the information was available.The studies cited above evaluated the effect of 3 months of CPAP therapy on glycemic control, but Pallayova et al recently reported that the mean and SD of the overnight interstitial glucose decreased rapidly with the initiation of CPAP therapy. They found a significant improvement even during a CPAP titration night.20The intermittent hypoxemia associated with OSA has been shown to produce surges of sympathetic activation,21, 22 whereas arousals during sleep can induce bursts of cortisol release in normal humans.23 Punjabi and associates suggested that intermittent hypoxemia and sleep fragmentation alter the autonomic, hypothalamic-pituitary, and somatotropic axes and also increase the levels of inflammatory cytokines and adipokines, thereby affecting glucose metabolism.9 These short-term effects of sleep apnea could worsen hyperglycemia and cause fluctuations of the blood glucose during the night that would diminish with CPAP therapy. The improvement in these nocturnal disturbances of glucose metabolism might have relatively little effect on daytime glucose metabolism and might not be reflected by measurements of insulin sensitivity or by a change in the HbA1c.Tasali et al have shown that, in normal subjects, selective suppression of slow-wave sleep produces increased insulin resistance and decreased glucose tolerance.24 Though our subjects showed an increase in slow-wave sleep on the CPAP night, the increase was small and not statistically significant.We found that glucose levels tended to be greater during wake after sleep onset than during sleep on the untreated night but not on the CPAP night. This suggests that CPAP therapy may have reduced short-term rises in glucose levels provoked by arousals and respiratory events.Our data suggest that glucose levels decreased with CPAP therapy only during the night. However, Babu et al found that, with CPAP, postprandial glucose levels were lower after all 3 meals,11 and so it is likely that there are persistent effects of CPAP therapy on daytime glycemic control that our study lacked the power to demonstrate.Even if CPAP did not decrease the 24-hour mean blood glucose, it could have a beneficial effect on the course of type 2 diabetes by reducing the nocturnal fluctuations in the glucose level. It has been shown that variability of the fasting blood glucose level is an independent risk factor for mortality and for retinopathy in type 2 diabetics.25, 26The major limitation of our study was the lack of a control group. West et al10 included a group treated with sham CPAP, and they pointed out that glycemic control might be improved during a study if people modified their behavior, knowing that they were being monitored. It is not likely that our subjects were more adherent to their diets during the study because there was actually a statistically significant gain in weight from the untreated night to the night on CPAP. We instructed our subjects not to attempt to lose weight, and these obese diabetics probably interpreted this as a license to ease up on their usual efforts to diet. However, we cannot be certain that they did not take their medications more consistently. Another limitation of our study was our inability to ensure that subjects had the same caloric intake at the same time on the 2 polysomnography nights.Only 3 of 28 subjects recruited to the study (11%) were dropped because of nonadherence to the CPAP therapy, an adherence rate that is much better than is typically reported in the literature.27 Adherence was encouraged by the weekly visits with our research nurse,28 but probably more important was that this was a highly motivated group of subjects who participated in the study without any financial incentive. The study was further biased toward compliant subjects by our exclusion of the data from the 3 subjects whose use of CPAP was less than 4 hours per night. The new diagnosis of sleep apnea might also have encouraged closer adherence to their diabetic regimen. Because our study included only the most compliant subjects, our conclusion that CPAP therapy of diabetics with OSA improves their glycemic control may not be generalizable to other patients who have the 2 conditions.Our findings need confirmation by a larger study that randomly assigns subjects to sham-CPAP and effective-CPAP arms, but our findings suggest that screening type 2 diabetics for OSA and treating those with moderate to severe sleep-disordered breathing could improve the management of their hyperglycemia and might favorably influence their long-term prognosis.DISCLOSURE STATEMENTThis was not an industry supported study. The authors have indicated no financial conflicts of interest.REFERENCES1 West SD, Nicoll DJ, Stradling JRPrevalence of obstructive sleep apnoea in men with type 2 diabetesThorax20066194550, 16928713 CrossrefGoogle Scholar2 Elmasry A, Lindberg E, Berne C, et al.Sleep-disordered breathing and glucose metabolism in hypertensive men: a population-based studyJ Intern Med200124915361, 11240844 CrossrefGoogle Scholar3 Einhorn D, Stewart DA, Erman MK, Gordon N, Philis-Tsimikas A, Casal EPrevalence of sleep apnea in a population of adults with type 2 diabetes mellitusEndocr Pract20071335562, 17669711 CrossrefGoogle Scholar4 Meslier N, Gagnadoux F, Giraud P, et al.Impaired glucose-insulin metabolism in males with obstructive sleep apnoea syndromeEur Respir J20032215660, 12882466 CrossrefGoogle Scholar5 Reichmuth KJ, Austin D, Skatrud JB, Young TAssociation of sleep apnea and type II diabetes: a population-based studyAm J Respir Crit Care Med200517215905, 16192452 CrossrefGoogle Scholar6 Punjabi NM, Shahar E, Redline S, Gottlieb DJ, Givelber R, Resnick HESleep-disordered breathing, glucose intolerance, and insulin resistance: the Sleep Heart Health StudyAm J Epidemiol200416052130, 15353412 CrossrefGoogle Scholar7 Ip MS, Lam B, Ng MM, Lam WK, Tsang KW, Lam KSObstructive sleep apnea is independently associated with insulin resistanceAm J Respir Crit Care Med20021656706, 11874812 CrossrefGoogle Scholar8 Punjabi NM, Ahmed MM, Polotsky VY, Beamer BA, O'Donnell CPSleep-disordered breathing, glucose intolerance, and insulin resistanceRespir Physiol Neurobiol200313616778, 12853008 CrossrefGoogle Scholar9 Punjabi NM, Polotsky VYDisorders of glucose metabolism in sleep apneaJ Appl Physiol20059919982007, 16227461 CrossrefGoogle Scholar10 West SD, Nicoll DJ, Wallace TM, Matthews DR, Stradling JREffect of CPAP on insulin resistance and HbA1c in men with obstructive sleep apnoea and type 2 diabetesThorax20076296974, 17557769 CrossrefGoogle Scholar11 Babu AR, Herdegen J, Fogelfeld L, Shott S, Mazzone TType 2 diabetes, glycemic control, and continuous positive airway pressure in obstructive sleep apneaArch Intern Med200516544752, 15738376 CrossrefGoogle Scholar12 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 ForceSleep19992266789, 10450601 CrossrefGoogle Scholar13 Boyne MS, Silver DM, Kaplan J, Saudek CDTiming of changes in interstitial and venous blood glucose measured with a continuous subcutaneous glucose sensorDiabetes20035227904, 14578298 CrossrefGoogle Scholar14 Steil GM, Rebrin K, Mastrototaro J, Bernaba B, Saad MFDetermination of plasma glucose during rapid glucose excursions with a subcutaneous glucose sensorDiabetes Technol Ther200352731, 12725704 CrossrefGoogle Scholar15 Service FJ, Molnar GD, Rosevear JW, Ackerman E, Gatewood LC, Taylor WFMean amplitude of glycemic excursions, a measure of diabetic instabilityDiabetes19701964455, 5469118 CrossrefGoogle Scholar16 Monnier L, Mas E, Ginet C, et al.Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetesJAMA200629516817, 16609090 CrossrefGoogle Scholar17 Wentholt IM, Kulik W, Michels RP, Hoekstra JB, Devries JHGlucose fluctuations and activation of oxidative stress in patients with type 1 diabetesDiabetologia20085118390, 17994218 CrossrefGoogle Scholar18 Harsch IA, Schahin SP, Bruckner K, et al.The effect of continuous positive airway pressure treatment on insulin sensitivity in patients with obstructive sleep apnoea syndrome and type 2 diabetesRespiration2004712529, 15133345 CrossrefGoogle Scholar19 Hassaballa HA, Tulaimat A, Herdegen JJ, Mokhlesi BThe effect of continuous positive airway pressure on glucose control in diabetic patients with severe obstructive sleep apneaSleep Breath2005917680, 16283228 CrossrefGoogle Scholar20 Pallayova M, Donic V, Tomori ZBeneficial effects of severe sleep apnea therapy on nocturnal glucose control in persons with type 2 diabetes mellitusDiabetes Res Clin Pract200881e811, 18440086 CrossrefGoogle Scholar21 Leuenberger U, Jacob E, Sweer L, Waravdekar N, Zwillich C, Sinoway LSurges of muscle sympathetic nerve activity during obstructive apnea are linked to hypoxemiaJ Appl Physiol1995795818, 7592221 CrossrefGoogle Scholar22 Smith ML, Niedermaier ON, Hardy SM, Decker MJ, Strohl KPRole of hypoxemia in sleep apnea-induced sympathoexcitationJ Auton Nerv Syst19965618490, 8847442 CrossrefGoogle Scholar23 Spath-Schwalbe E, Gofferje M, Kern W, Born J, Fehm HLSleep disruption alters nocturnal ACTH and cortisol secretory patternsBiol Psychiatry19912957584, 1647222 CrossrefGoogle Scholar24 Tasali E, Leproult R, Ehrmann DA, Van Cauter ESlow-wave sleep and the risk of type 2 diabetes in humansProc Natl Acad Sci U S A200810510449, 18172212 CrossrefGoogle Scholar25 Muggeo M, Zoppini G, Bonora E, et al.Fasting plasma glucose variability predicts 10-year survival of type 2 diabetic patients: the Verona Diabetes StudyDiabetes Care2000234550, 10857967 CrossrefGoogle Scholar26 Gimeno-Or