Matches in SemOpenAlex for { <https://semopenalex.org/work/W1522441564> ?p ?o ?g. }
- W1522441564 endingPage "524" @default.
- W1522441564 startingPage "519" @default.
- W1522441564 abstract "Free AccessObstructive Sleep ApneaMarked Reduction in Obstructive Sleep Apnea Severity in Slow Wave Sleep Rajeev Ratnavadivel, M.B.Ch.B., Nuy Chau, B.Med.Sci., Daniel Stadler, B.Sci. (Hons), Aeneas Yeo, M.B.B.S., R. Doug McEvoy, M.D., Peter G. Catcheside, Ph.D. Rajeev Ratnavadivel, M.B.Ch.B. Address correspondence to: Rajeev Ratnavadivel, M.B.Ch.B., Adelaide Institute for Sleep Health, Repatriation General Hospital, Daw Park, South Australia 5041, Australia E-mail Address: [email protected] Adelaide Institute for Sleep Health, Repatriation General Hospital, Adelaide, Australia School of Medicine, Faculty of Health Sciences, Flinders University, Adelaide, Australia Search for more papers by this author , Nuy Chau, B.Med.Sci. School of Medicine, Faculty of Health Sciences, Flinders University, Adelaide, Australia Search for more papers by this author , Daniel Stadler, B.Sci. (Hons) Adelaide Institute for Sleep Health, Repatriation General Hospital, Adelaide, Australia Discipline of Physiology, School of Molecular and Biomedical Science, University of Adelaide, Adelaide, Australia Search for more papers by this author , Aeneas Yeo, M.B.B.S. Adelaide Institute for Sleep Health, Repatriation General Hospital, Adelaide, Australia Search for more papers by this author , R. Doug McEvoy, M.D. Adelaide Institute for Sleep Health, Repatriation General Hospital, Adelaide, Australia School of Medicine, Faculty of Health Sciences, Flinders University, Adelaide, Australia Discipline of Physiology, School of Molecular and Biomedical Science, University of Adelaide, Adelaide, Australia Search for more papers by this author , Peter G. Catcheside, Ph.D. Adelaide Institute for Sleep Health, Repatriation General Hospital, Adelaide, Australia School of Medicine, Faculty of Health Sciences, Flinders University, Adelaide, Australia Discipline of Physiology, School of Molecular and Biomedical Science, University of Adelaide, Adelaide, Australia Search for more papers by this author Published Online:December 15, 2009https://doi.org/10.5664/jcsm.27651Cited by:81SectionsAbstractPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTIntroduction:Obstructive sleep apnea (OSA) is widely accepted to improve during slow wave sleep (SWS) compared to lighter stages of NREM sleep. However, supporting data to establish the magnitude and prevalence of this effect is lacking. Consequently, we examined this phenomenon, controlling for posture, in a large group of patients investigated for OSA at an academic clinical sleep service.Methods:A detailed retrospective analysis was conducted on data obtained from each 30-sec epoch of sleep in 253 consecutive full-night diagnostic polysomnography studies performed over a 3-month period. Respiratory and arousal event rates were calculated within each stage of sleep, in the supine and lateral postures, and across the whole night, with OSA patients classified on the basis of an overall apnea-hypopnea index (AHI) ≥ 15 events/h. Central sleep apnea (CSA) patients were defined by a central apnea index ≥ 5/h. Sleep latency and time, and respiratory and arousal event rates in OSA, CSA, and non-OSA patients were compared between sleep stages and postures using linear mixed model analysis. The numbers of patients achieving reduced event rates in SWS and in the lateral posture were also examined.Results:There were 171 patients with OSA, 14 with CSA, and 68 non-OSA patients. OSA patients took significantly longer to achieve slow wave and REM sleep (p < 0.001) than non-OSA patients and had less stage 4 sleep (p = 0.037). There were striking improvements in AHI and arousal index (AI) from stage 1 to 4 NREM sleep (p < 0.001), with intermediate levels in REM sleep. AHI and AI were also markedly reduced in lateral versus supine sleep in all sleep stages (p < 0.001), with an effect size comparable to that of the slow wave sleep effect. The majority of OSA patients achieved low respiratory event rates in SWS. Eighty-two percent of patients achieved an AHI < 15 and 57% < 5 events/hour during stage 4 sleep.Conclusion:Although OSA patients demonstrate both a delayed and reduced proportion of SWS compared to non-OSA subjects, once they achieved SWS, AHI, and AI markedly improved in most patients.Citation:Ratnavadivel R; Chau N; Stadler D; Yeo A; McEvoy RD; Catcheside PG. Marked reduction in obstructive sleep apnea severity in slow wave sleep. J Clin Sleep Med 2009;5(6):519-524.INTRODUCTIONObstructive sleep apnea (OSA) is a common disorder. Twenty-four percent of adult males and 9% of adult females experience at least 5 events of disordered breathing per hour of sleep.1 The diagnosis and severity of obstructive sleep apnea is typically determined from the apnea hypopnea index (AHI),2 a measure of the total number of complete or partial upper airway obstruction (apnea or hypopnea) events lasting 10 sec or more, divided by total sleep time. AHI is known to be strongly influenced by factors such as sleep posture,3–5 head position,6,7 and REM sleep.4,5 In clinical studies, OSA severity frequently changes substantially over the course of a night, even in severely affected patients. It has also been widely accepted that obstructive sleep apnea improves during slow wave sleep (SWS). However, with the exception of a few preliminary reports based on small patient samples,8–10 there appears to be remarkably little published evidence available to assess the magnitude and prevalence of this effect.If improvements in sleep disordered breathing in SWS amongst OSA patients are substantiated as being both common and marked it may be a phenomenon worthy of further investigation, for while much has been learned about the risk factors for OSA and OSA pathogenesis, only a relatively small degree of the variance in OSA severity is currently explained by these various mechanisms.11 It seems likely that non-anatomical factors mediating upper airway and respiratory control stability, such as ventilatory chemosensitivity, plant gain control factors, arousal threshold, and the magnitude of post-arousal ventilatory overshoot-undershoot responses,11 would be involved in SWS-mediated improvements in sleep apnea severity. Ultimately, better knowledge of the extent and mechanisms of SWS-mediated improvements in OSA could lead to new therapeutic approaches for this disorder.Therefore, the purpose of this study was to investigate in detail, the effect of NREM sleep stages on the frequency of respiratory and arousal events in a large cohort of patients referred for the investigation of possible OSA.METHODSPatientsThe study was approved by the Repatriation General Hospital research and ethics committee. De-identified full-night diagnostic polysomnography studies of all patients investigated in the Adelaide Institute for Sleep Health laboratory during a 3-month period between January and March 2005 were retrospectively analyzed in detail. All studies were recorded and primarily analyzed using a Compumedics E-series system and software (ProFusion, Compumedics Inc, Melbourne, Australia) in an attended laboratory setting. EEG was recorded using electrodes at C3/A2 and C4/A1 scalp locations. Eye movements were measured via 2-channel electrooculography (left and right EOG) and ECG, submental chin EMG, leg movements, snoring, and posture were also continuously recorded. Nasal pressure cannulae were used to record airflow and thoracoabdominal bands and an uncalibrated sum signal used to assess ventilatory effort. Finger pulse oximetry was used to determine arterial blood oxygen saturation. Sleep recordings were analyzed by an accredited sleep technologist using 30-sec epochs and Rechtschaffen and Kales criteria for staging sleep.12 Arousals and respiratory events were scored according to ASDA13 and AASM criteria2 respectively.AnalysisFor each 30-sec epoch within each sleep study, sleep posture was categorized according to prone, lateral (left and right lateral combined) or supine, and the total number of respiratory and arousal events determined. Given that few subjects slept prone, this posture was not analyzed. Total respiratory (apneas plus hypopneas) and arousal event rates irrespective of posture and in supine and lateral postures were separately determined within each sleep stage (NREM 1-4 and REM) as the total number of events divided by total sleep time for each posture and stage combination for which there was at least 5 min (10 epochs) of total sleep. Respiratory event rates were further divided into types (obstructive, hypopnea, central, and mixed events) for each stage and posture. A minimum of 5 minutes within each posture and/or stage was chosen to reduce the impact of more extreme event rates estimated from short periods (i.e., poor event rate resolution). Based on previous normative data published by our group using American Academy of Sleep Medicine 1999 criteria,2 total sleep AHI ≥ 15 /h was considered the cut off for the diagnosis of OSA.14 Patients with ≥ 5 central events per hour were considered to have central sleep apnea.15 The sample was divided on this basis into OSA, non-OSA, and CSA groups.Statistical AnalysisSleep stage (1-4 and REM) and posture (supine vs lateral) effects in all 3 groups were examined using linear mixed model analysis using an autoregressive covariance structure (SPSS version 16, SPSS Inc, Chicago, IL). In the event of significant mixed model effects, relevant post hoc comparisons were performed using Student t-tests with Dunn-Sidak correction for multiple comparisons.16 The strength of relationships between respiratory and arousal event rates in NREM sleep in the supine and lateral postures in the OSA and non-OSA groups were examined on a within-subject basis for each individual in which 3 or all 4 NREM stages were available for analysis. Effect sizes for posture and stage effects were estimated from within subject differences in lateral versus supine values in stage 2 sleep and differences between stage 2, SWS and REM sleep in the supine posture respectively. All data are presented as mean ± SEM. P-values < 0.05 were considered significant.RESULTSA total of 253 overnight diagnostic studies were performed during the time period of interest. Fourteen patients had a central apnea index ≥ 5/h; 171 of the remaining patients had an AHI ≥ 15/h, and 68 had an AHI < 15/h. Baseline demographic and sleep study data for the groups are summarized in Table 1. The patient population was typical of diagnostic referrals primarily for evaluation of OSA and was comprised predominantly of males who were middle-aged and overweight to obese. Approximately 70% of patients received a positive diagnosis for OSA. This group was significantly older, comprised relatively more males, and had a higher BMI than the non-OSA group (Table 1).Table 1 Patient CharacteristicsNNAgeBMIAHIAICAI(% Total)(% Males)(y)(kg/m2)(/h)(/h)(/h)Non OSA (AHI < 15/h)68 (27%)36 (53%)47.5 ± 1.928.9 ± 0.89.3 ± 0.414.3 ± 1.00.2 ± 0.1OSA (AHI ≥ 15 /h)171 (68%)120 (70%)*54.1 ± 1.1*31.8 ± 0.5*37.7 ± 1.6*23.6 ± 0.9*0.8 ± 0.1*CSA (CAI ≥ 5 /h)14 (6%)13 (93%)*53.9 ± 5.232.2 ± 1.454.3 ± 6.3*27.2 ± 4.8*16.8 ± 3.9*BMI, body mass index; AHI, Apnea-hypopnea index; AI, arousal index; CAI, central apnea index. Values are numbers and (percentage) of patients and mean ± SEM.*indicates p < 0.05 versus the non-OSA group.There was no difference in total sleep time between OSA and non-OSA patients (Table 2). OSA and non-OSA patients spent similar proportions of total sleep time in lateral (60.8 ± 2.1 vs 66.2 ± 3.2%, p = 0.159); and supine sleep (42.3 ± 2.3 vs 35.6 ± 3.6%, p = 0.119) respectively. As expected, OSA patients had significantly more stage 1 and less stage 4 sleep compared to the non-OSA group, both in absolute terms (group × stage effect p = 0.037) and as a proportion of total sleep time (Table 2). OSA patients took significantly longer to attain the first epoch of slow wave (39.3 ± 3.5 vs 25.6 ± 2.6 min, p = 0.005) and REM sleep (147.4 ± 6.1 vs 123.3 ± 6.8 min, p = 0.015) following sleep onset compared to the non-OSA group.Table 2 Time Spent in Each Sleep Stagenon-OSA (AHI < 15)OSA (AHI ≥ 15)CSA (CAI ≥ 5/h)Stagemin%min%min%133.9 ± 4.210.4 ± 1.245.4 ± 2.3*15.3 ± 0.8*38.7 ± 5.214.4 ± 2.52153.6 ± 7.247.4 ± 1.5148.1 ± 3.747.8 ± 0.9139.9 ± 14.147.5 ± 3.1350.3 ± 3.316.5 ± 1.150.2 ± 2.416.0 ± 0.648.1 ± 8.115.8 ± 2.1437.0 ± 3.411.7 ± 1.122.3 ± 2.0*7.2 ± 0.7*21.8 ± 5.1*6.5 ± 1.3*REM54.3 ± 3.616.2 ± 0.850.6 ± 2.015.7 ± 0.551.5 ± 8.417.7 ± 2.6Total Sleep316.2 ± 10.1100310.1 ± 5.7100293.7 ± 22.0100Values are time (minutes) spent in each sleep stage and percentage of total sleep time, mean ± SEM.*p < 0.05 vs non-OSA group.Changes in AHI and AI as a function of sleep stage and OSA diagnosis are shown in Figure 1A and further separated as a function of supine versus lateral postures in Figure 1B. There was a striking NREM sleep stage dependence in the frequency of respiratory (stage effect p < 0.001) and arousal events (p < 0.001), with a progressive reduction in event rates from stage 1-4, with REM sleep at an intermediate level. Compared to stage 2, the REM AHI was ∼40% higher (effect size 0.55), while SWS AHI was ∼50% less (effect size 0.94). The reduction in respiratory and arousal event rates was noted in all patient groups. Total sleep event rates were weighted towards stage 2 values, consistent with the predominance of stage 2 sleep. The ratio of respiratory events to arousal events also showed strong stage (p < 0.001) and group (p < 0.001) effects. In REM sleep, respiratory events occurred considerably more frequently relative to arousal events (AHI/AI ratio 3.9 ± 0.2, compared to NREM sleep 1.2 ± 0.2, p < 0.001) but remained essentially unchanged within NREM sleep. The frequency of arousals exceeded the frequency of respiratory events in all NREM sleep stages in non-OSA patients, whereas the opposite was true for OSA patients AHI/AI 0.7 ± 0.3 vs 1.7 ± 0.2, p < 0.001. The respiratory event rate was strongly correlated with the arousal index, overall r2= 0.82 ± 0.02.Figure 1 Respiratory and arousal event frequencies as a function of sleep stage in all postures combined (A) and in the lateral and supine postures alone (B) in OSA (AHI ≥ 15 /h, n = 171), and non-OSA (AHI < 15/h, n = 68) and CSA (CAI ≥ 5/h, n = 14) patients. Values are mean ± SEM.Download FigureAs shown in Figure 1B posture had substantial effects on respiratory and to a lesser degree arousal event rates in both OSA and non-OSA patient groups. Respiratory event rates in the lateral postures were in the order of 50% to 60% of supine values in both groups (p < 0.001), while arousal event rates were in the order of 60% to 80% of supine values in lateral postures, (p < 0.001). Thus respiratory event frequency in the supine posture was higher, relative to arousal frequency, p = 0.018. However, a similar pattern of decline in respiratory and arousal frequencies were noted from stage 1 to 4 NREM sleep. The CSA group also had a reduced rate of respiratory events in the lateral posture (p = 0.009) and a trend towards reduced arousal frequency in the lateral posture (p = 0.057).Respiratory event types in the OSA and CSA groups were further differentiated into obstructive, central, mixed, and hypopneas as shown in Figure 2. All types of respiratory events showed a strong sleep stage effect in the OSA group, (p < 0.001). In the CSA group, obstructive and central events showed a stage effect, p = 0.048 and p < 0.001, respectively, whereas hypopneas and mixed apneas did not have a significant stage effect.Figure 2 Obstructive apnea, hypopnea, central apnea, and mixed apnea event frequencies as a function of sleep stage in OSA, (AHI ≥ 15/h) and CSA (CAI ≥ 5/h) groups. Values are mean ± SEM.Download FigureSleep stage effects were large and of similar or greater magnitude when compared to posture effects. In the OSA patient group, the effect size of the difference between stage 2 and SWS (stages 3 and 4 combined) was 1.1 for AHI and 1.4 for AI, while that of the difference between supine and lateral postures in stage 2 sleep was 0.8 for AHI and 0.7 for AI. The corresponding effect sizes in the non-OSA group were weaker: 0.6, 1, 0.6, and 0.3, respectively.Most patients attained relatively low event rates in SWS. For example, 82% of OSA patients exhibited < 15 respiratory events/h and 57% < 5 events/h in stage 4 sleep.DISCUSSIONThis is the first large systematic observational study of the effects of NREM sleep stage on OSA severity. OSA patients demonstrated significantly reduced stage 4 and increased stage 1 sleep, and a significantly longer latency to achieve SWS compared to patients without OSA. There was striking NREM sleep stage dependence in OSA severity in both OSA patients and non-OSA patients, with progressive reductions in respiratory and arousal event rates from stage 1 to stage 4 sleep, that were systematically improved during lateral sleep. Respiratory and arousal event rates were strongly associated throughout NREM sleep. Patients with CSA showed similar patterns.Previous studies of sleep-stage dependence of OSA severity have generally focussed on REM versus NREM sleep comparisons and shown elevated AHI in REM compared with NREM,4,5 while ignoring the changes in AHI within NREM sleep. The present study, conducted in a large clinic patient population, showed a marked reduction in AHI during SWS compared with stages 1 and 2 NREM sleep. These findings are similar to those in previous preliminary reports8–10 but extend these earlier observations by showing that the SWS effects are of a similar or greater magnitude than posture effects and that most patients, even those with moderate to severe OSA, can achieve low respiratory event rates in SWS. The magnitude of the improvement we observed amongst OSA patients in AHI during SWS versus stage 2 is similar to the deterioration in AHI observed in REM sleep compared with stage 2 sleep. The improvement in AHI frequency from stage 1 to 4 NREM sleep occurred regardless of the scored event type.There are several possible reasons for the improvement in OSA during SWS. One possibility is that the upper airway becomes more neuromechanically stable with the onset of SWS. Alternatively, upper airway function could improve for some other reason, allowing patients to progress to deeper sleep. There is some evidence that the upper airway is better able to resist collapse during induced airway occlusion in SWS,18 but this finding was not replicated when passive airway function was measured by the PCRIT technique.19 In healthy volunteers, upper airway resistance is thought to be significantly higher during SWS than light sleep,20 despite genioglossal EMG activity being higher in this state,21 suggesting that, the upper airway is neuromechanically disadvantaged in SWS.An alternate reason for OSA improvement with increasing NREM sleep depth may relate to sleep stage modulation of arousal propensity, and potentially to sleep stage modulation of post-arousal ventilatory responses. Poor airway function frequently triggers arousal, and there is emerging evidence that arousal responses themselves promote conditions that predispose the airway to re-collapse.22 The arousal index in this study was strongly NREM sleep stage dependent with a consistent relationship between arousal and respiratory indexes. Respiratory arousals from NREM sleep are thought to occur when progressively increasing respiratory effort or drive reaches a particular effort threshold.23 This threshold is significantly higher during SWS than light sleep.24,25 Arousal frequency following respiratory loading is reduced during SWS in healthy subjects.26,27 Studies have also shown much higher thresholds to non-respiratory stimuli during SWS than stage 2 sleep.28,29 Thus, reduced propensity to arouse from sleep during SWS may be an important cause for improvement in OSA.Although protective upper airway dilatory responses are thought to be impaired during sleep,30,31 the upper airway does have the ability to respond to increasingly negative upper airway pressure and chemostimuli such as hypoxia/ hypercapnia,32,33 possibly in combination. Reduced arousal propensity during SWS may permit upper airway neuromuscular drive (chemo and/or mechanoreceptor mediated) to achieve sufficient activity to allow steady-state ventilation compatible with sustained sleep. Arousals from sleep may in fact promote ongoing cyclical breathing via post-arousal hyperventilation and a subsequent state of reduced drive,34 which may favor airway collapse. Theoretical models of OSA predict that a low arousal threshold predisposes to cyclical breathing,35 although this remains to be shown in OSA patients.CSA patients were analyzed separately and showed a similar pattern of improvement in overall AHI from stage 1 to 4 NREM sleep. When the event type was analyzed, obstructive and central events improved significantly, whereas mixed events and hypopneas did not show a statistically significant change, unlike the OSA group. This likely reflects the small sample of CSA subjects studied (n = 14). Given the known strong connection between central apneas and arousal,36 the reduction in AHI and CAI with increasing NREM stage is not surprising. However, this had not been described previously.There are a number of limitations of this study. Given its observational nature, the causes of improved AHI during slow wave sleep remain speculative. Respiratory and arousal events were scored according to AASM and ASDA criteria respectively. Consequently, subcriterion events, either respiratory or arousal were not considered. However, given that the same criteria were applied across all sleep stages by a single technician within each patient, altered classification criteria and inter-patient scoring differences were unlikely to substantially influence the main within patient findings of this study. Our non-OSA group comprised of clinic referred patients and may not have been a “normal” group. Nevertheless, progressively lower respiratory and arousal event rates across NREM sleep stages, regardless of sleep posture in non-OSA and OSA patients remains consistent with important NREM sleep stage effects on OSA severity. It seems that regardless of the level of underlying anatomical compromise and propensity for upper airway collapse, most people can overcome this in SWS and achieve improved or near normal respiratory and arousal event frequencies.We conclude that OSA, CSA, and non-OSA subjects demonstrate marked reductions in respiratory and arousal event rates during slow wave compared to light NREM and REM sleep. Further investigation into potential differences in upper airway behaviour and arousal responses to ventilatory challenges in light compared to SWS are warranted, and may lead to an improved understanding of the pathophysiology of OSA and ultimately, to improvements in therapy for this disorder.DISCLOSURE STATEMENTThis was not an industry supported study. Drs. Ratnavadivel and Yeo have received the use of equipment from Philips Respironics. Dr. McEvoy has received research support and use of equipment from Respironics Research Foundation, use of equipment from ResMed, and research support from Fisher & Paykel.REFERENCES1 Young TPalta MDempsey JSkatrud JWeber SBadr SThe occurrence of sleep-disordered breathing among middle-aged adultsN Engl J Med199332812305, 8464434CrossrefGoogle Scholar2 AASM Task ForceSleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical researchSleep19992266789, 10450601CrossrefGoogle Scholar3 Cartwright RDEffect of sleep position on sleep apnea severitySleep198471104, 6740055CrossrefGoogle Scholar4 Pevernagie DAShepard JWRelations between sleep stage, posture and effective nasal CPAP levels in OSASleep1992151627, 1579791CrossrefGoogle Scholar5 Cartwright RDDiaz FLloyd SThe effects of sleep posture and sleep stage on apnea frequencySleep1991143513, 1947599CrossrefGoogle Scholar6 Isono STanaka ATagaito YIshikawa TNishino TInfluences of head positions and bite opening on collapsibility of the passive pharynxJ Appl Physiol20049733946, 15020573CrossrefGoogle Scholar7 Isono STanaka AIshikawa TTagaito YNishino TSniffing position improves pharyngeal airway patency in anesthetized patients with obstructive sleep apneaAnesthesiology200510348994, 16129972CrossrefGoogle Scholar8 Dinner DSSantin JGodoy JLudera M.Relationship of sleep apnea and sleep stages in obstructive sleep apneaAssociation of Professional Sleep Societies1989p.224Google Scholar9 Guchu RFindlay LWoodson HFabrizio MSuratt PUpper airway stability is increased during slow-wave sleep in obstructive sleep apneaAm Rev Respir Dis1991143A796Google Scholar10 Dingli KFietze IAssimakopoulos TQuispe-Bravo SWitt CDouglas NJArousability in sleep apnoea/hypopnoea syndrome patientsEur Respir J20022073340, 12358354CrossrefGoogle Scholar11 White DPPathogenesis of obstructive and central sleep apneaAm J Respir Crit Care Med2005172136370, 16100008CrossrefGoogle Scholar12 Rechtschaffen AKales AA manual of standardized terminology, techniques and scoring system for sleep stages of human subjects1968Bethesda, MDNational Institutes of HealthGoogle Scholar13 EEG arousals: scoring rules and examples: a preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders AssociationSleep19921517384, 11032543CrossrefGoogle Scholar14 Banks SBarnes MTarquinio NPierce RJLack LCMcEvoy RDFactors associated with maintenance of wakefulness test mean sleep latency in patients with mild to moderate obstructive sleep apnoea and normal subjectsJ Sleep Res200413718, 14996038CrossrefGoogle Scholar15 Morgenthaler TIKagramanov VHanak VDecker PAComplex sleep apnea syndrome: is it a unique clinical syndrome?Sleep20062912039, 17040008CrossrefGoogle Scholar16 Ludbrook JRepeated measurements and multiple comparisons in cardiovascular researchCardiovasc Res19942830311, 8174149CrossrefGoogle Scholar17 Redline SKirchner HLQuan SFGottlieb DJKapur VNewman AThe effects of age, sex, ethnicity, and sleep-disordered breathing on sleep architectureArch Intern Med200416440618, 14980992CrossrefGoogle Scholar18 Issa FGSullivan CEUpper airway closing pressures in obstructive sleep apneaJ Appl Physiol1984575207, 6381440CrossrefGoogle Scholar19 Penzel TMoller MBecker HFKnaack LPeter JHEffect of sleep position and sleep stage on the collapsibility of the upper airways in patients with sleep apneaSleep200124905, 11204057CrossrefGoogle Scholar20 Henke KGDempsey JAKowitz JMSkatrud JBEffects of sleep-induced increases in upper airway resistance on ventilationJ Appl Physiol19906961724, 2121702CrossrefGoogle Scholar21 Basner RCRingler JSchwartzstein RMWeinberger SEWeiss JWPhasic electromyographic activity of the genioglossus increases in normals during slow-wave sleepRespir Physiol199183189200, 2068416CrossrefGoogle Scholar22 Younes MRole of arousals in the pathogenesis of obstructive sleep apneaAm J Respir Crit Care Med200416962333, 14684560CrossrefGoogle Scholar23 Gleeson KZwillich CWWhite DPThe influence of increasing ventilatory effort on arousal from sleepAm Rev Respir Dis1990142295300, 2382892CrossrefGoogle Scholar24 Berry RBBonnet MHLight RWEffect of ethanol on the arousal response to airway occlusion during sleep in normal subjectsAm Rev Respir Dis199214544552, 1736755CrossrefGoogle Scholar25 Berry RBAsyali MAMcNellis MIKhoo MCWithin-night variation in respiratory effort preceding apnea termination and EEG delta power in sleep apneaJ Appl Physiol199885143441, 9760338CrossrefGoogle Scholar26 Zavodny JRoth CBassetti CLMathis JDouglas NJGugger MEffects of sleep fragmentation on the arousability to resistive loading in NREM and REM sleep in normal menSleep20062952532, 16676786CrossrefGoogle Scholar27 Gugger MMolloy JGould GA, et al.Ventilatory and arousal responses to added inspiratory resistance during sleepAm Rev Respir Dis198914013017, 2817592CrossrefGoogle Scholar28 Catcheside PGOrr RSChiong SCMercer JSaunders NAMcEvoy RDMild hypoxia does not suppress auditory arousal from NREM sleepSleep20062961923, 16774151CrossrefGoogle Scholar29 Williams HLHammack JTDaly RLDement WCLubin LResponses to auditory stimulation, sleep loss, and the EEG stages of sleepElectroencephalogr Clin Neurophysiol19641626979, 14128872CrossrefGoogle Scholar30 Wheatley JRMezzanotte WSTangel DJWhite DPInfluence of sleep on genioglossus muscle activation by negative pressure in normal menAm Rev Respir Dis1993148597605, 8368629CrossrefGoogle Scholar31 Malhotra APillar GFogel RB, et al.Genioglossal but not palatal muscle activity relates closely to pharyngeal pressureAm J Respir Crit Care Med2000162105862, 10988130CrossrefGoogle Scholar32 Stanchina MLMalhotra AFogel RB, et al.Genioglossus muscle responsiveness to chemical and mechanical stimuli during non-rapid eye movement sleepAm J Respir Crit Care Med20021659459, 11934719CrossrefGoogle Scholar33 Lo YLJordan ASMalhotra A, et al.Genioglossal muscle response to CO2 stimulation during NREM sleepSleep2006294707, 16676780CrossrefGoogle Scholar34 Jordan ASEckert DJCatcheside PGMcEvoy RDVentilatory response to brief arousal from non-rapid eye movement sleep is greater in men than in womenAm J Respir Crit Care Med200316815129, 14525799CrossrefGoogle Scholar35 Khoo MCGottschalk APack AISleep-induced periodic breathing and apnea: a theoretical studyJ Appl Physiol199170201424, 1907602CrossrefGoogle Scholar36 Xie AWong BPhillipson EASlutsky ASBradley TDInteraction of hyperventilation and arousal in the pathogenesis of idiopathic central sleep apneaAm J Respir Crit Care Med199415048995, 8049835CrossrefGoogle Scholar Previous article Next article FiguresReferencesRelatedDetailsCited by Messineo L and Eckert D Obstructive Sleep Apnea Phenotyping to Understand Pathophysiology an" @default.
- W1522441564 created "2016-06-24" @default.
- W1522441564 creator A5001755883 @default.
- W1522441564 creator A5040394806 @default.
- W1522441564 creator A5042509048 @default.
- W1522441564 creator A5053789318 @default.
- W1522441564 creator A5058046345 @default.
- W1522441564 creator A5071548469 @default.
- W1522441564 date "2009-12-15" @default.
- W1522441564 modified "2023-10-18" @default.
- W1522441564 title "Marked Reduction in Obstructive Sleep Apnea Severity in Slow Wave Sleep" @default.
- W1522441564 cites W153968381 @default.
- W1522441564 cites W156142978 @default.
- W1522441564 cites W1744121058 @default.
- W1522441564 cites W1968771781 @default.
- W1522441564 cites W1988343534 @default.
- W1522441564 cites W1994514417 @default.
- W1522441564 cites W1997557472 @default.
- W1522441564 cites W2036292392 @default.
- W1522441564 cites W2040308656 @default.
- W1522441564 cites W2043364256 @default.
- W1522441564 cites W2047789059 @default.
- W1522441564 cites W2066544084 @default.
- W1522441564 cites W2067665240 @default.
- W1522441564 cites W2085918185 @default.
- W1522441564 cites W2097207628 @default.
- W1522441564 cites W2101348180 @default.
- W1522441564 cites W2104274405 @default.
- W1522441564 cites W2123841491 @default.
- W1522441564 cites W2123926550 @default.
- W1522441564 cites W2139496586 @default.
- W1522441564 cites W2149350222 @default.
- W1522441564 cites W2159767433 @default.
- W1522441564 cites W2160967022 @default.
- W1522441564 cites W2177382690 @default.
- W1522441564 cites W2185686289 @default.
- W1522441564 cites W2239259475 @default.
- W1522441564 cites W2263515572 @default.
- W1522441564 cites W2264784497 @default.
- W1522441564 cites W2279041817 @default.
- W1522441564 cites W2286275570 @default.
- W1522441564 cites W2333343699 @default.
- W1522441564 cites W3022302886 @default.
- W1522441564 cites W3166543270 @default.
- W1522441564 cites W44341116 @default.
- W1522441564 cites W75661469 @default.
- W1522441564 cites W2616546062 @default.
- W1522441564 doi "https://doi.org/10.5664/jcsm.27651" @default.
- W1522441564 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/2792966" @default.
- W1522441564 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/20465017" @default.
- W1522441564 hasPublicationYear "2009" @default.
- W1522441564 type Work @default.
- W1522441564 sameAs 1522441564 @default.
- W1522441564 citedByCount "137" @default.
- W1522441564 countsByYear W15224415642012 @default.
- W1522441564 countsByYear W15224415642013 @default.
- W1522441564 countsByYear W15224415642014 @default.
- W1522441564 countsByYear W15224415642015 @default.
- W1522441564 countsByYear W15224415642016 @default.
- W1522441564 countsByYear W15224415642017 @default.
- W1522441564 countsByYear W15224415642018 @default.
- W1522441564 countsByYear W15224415642019 @default.
- W1522441564 countsByYear W15224415642020 @default.
- W1522441564 countsByYear W15224415642021 @default.
- W1522441564 countsByYear W15224415642022 @default.
- W1522441564 countsByYear W15224415642023 @default.
- W1522441564 crossrefType "journal-article" @default.
- W1522441564 hasAuthorship W1522441564A5001755883 @default.
- W1522441564 hasAuthorship W1522441564A5040394806 @default.
- W1522441564 hasAuthorship W1522441564A5042509048 @default.
- W1522441564 hasAuthorship W1522441564A5053789318 @default.
- W1522441564 hasAuthorship W1522441564A5058046345 @default.
- W1522441564 hasAuthorship W1522441564A5071548469 @default.
- W1522441564 hasBestOaLocation W15224415641 @default.
- W1522441564 hasConcept C110539466 @default.
- W1522441564 hasConcept C111335779 @default.
- W1522441564 hasConcept C111919701 @default.
- W1522441564 hasConcept C118552586 @default.
- W1522441564 hasConcept C126322002 @default.
- W1522441564 hasConcept C164705383 @default.
- W1522441564 hasConcept C2524010 @default.
- W1522441564 hasConcept C2775841894 @default.
- W1522441564 hasConcept C2776006263 @default.
- W1522441564 hasConcept C2777935920 @default.
- W1522441564 hasConcept C2778205975 @default.
- W1522441564 hasConcept C2781326671 @default.
- W1522441564 hasConcept C33923547 @default.
- W1522441564 hasConcept C41008148 @default.
- W1522441564 hasConcept C42219234 @default.
- W1522441564 hasConcept C522805319 @default.
- W1522441564 hasConcept C71924100 @default.
- W1522441564 hasConceptScore W1522441564C110539466 @default.
- W1522441564 hasConceptScore W1522441564C111335779 @default.
- W1522441564 hasConceptScore W1522441564C111919701 @default.
- W1522441564 hasConceptScore W1522441564C118552586 @default.
- W1522441564 hasConceptScore W1522441564C126322002 @default.
- W1522441564 hasConceptScore W1522441564C164705383 @default.
- W1522441564 hasConceptScore W1522441564C2524010 @default.