Sleep Medicine 8 (2007) 773–778 www.elsevier.com/locate/sleep
Comparison of sleep parameters at titration and subsequent compliance between CPAP-pretreated and non-CPAP-pretreated patients with obstructive sleep apnea Masaaki Suzuki a
, Hanako Saigusa a, Taiji Furukawa
Department of Otolaryngology, Teikyo University School of Medicine, 2-11-1 Kaga, Itabashi-ku, 173-8605 Tokyo, Japan b Department of Cardiology, Teikyo University School of Medicine, 2-11-1 Kaga, Itabashi-ku, 173-8605 Tokyo, Japan Received 19 May 2006; received in revised form 5 January 2007; accepted 13 January 2007 Available online 23 May 2007
Abstract Background and purpose: Obstructive sleep apnea–hypopnea syndrome (OSAHS) patients undergo continuous positive airway pressure (CPAP) treatment for the ﬁrst time on titration night, and then the eﬀect of overnight CPAP treatment is estimated immediately. The purpose of this study is to compare the eﬀects of CPAP-pretreated and non-pretreated on patients with OSAHS. Methods: Prospective randomized, controlled parallel study was performed. Seventy patients with OSAHS received autoadjusted CPAP treatment for 2 months and then received the standard manual titration (CPAP-pretreated group). The other 70 did not receive any CPAP treatment before receiving the standard manual titration (non-CPAP-pretreated group). Results: The CPAP-pretreated group had signiﬁcantly improved sleep eﬃciency and arousal index in non-rapid eye movement (NREM) sleep compared with the initial CPAP group at titration, whereas there were no signiﬁcant diﬀerences between the two groups in other sleep parameters. Eight patients in the non-CPAP-pretreated group discontinued CPAP treatment 9 months after the titration, whereas one patient in the CPAP-pretreated group discontinued treatment. Conclusions: A preceding CPAP treatment showed minimal eﬀects on sleep parameters on titration night and subsequent CPAP compliance rate, although it was speculated that this preceding treatment might be of beneﬁt for better compliance in some patients. Ó 2007 Elsevier B.V. All rights reserved. Keywords: Obstructive sleep apnea–hypopnea syndrome; Continuous positive airway pressure; Titration; Sleep eﬃciency; Arousals; Compliance
1. Introduction Obstructive sleep apnea–hypopnea syndrome (OSAHS) is caused by the periodic reduction or cessation of breath due to the narrowing or occlusion of the upper airway during sleep. Continuous positive airway pressure (CPAP) application is currently the noninvasive treatment of choice for the majority of patients with OSAHS. CPAP treatment markedly attenuates *
Corresponding author. Tel.: +81 3 3964 1211; fax: +81 3 3964 0659. E-mail address: [email protected]
(M. Suzuki). 1389-9457/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.sleep.2007.01.013
obstructive apneas, hypopneas, snoring, ﬂow limitations, desaturation and arousals and improves sleep fragmentation. It is currently recommended that the speciﬁc optimal pressure for each patient should be determined by manual titration performed during an attended overnight in-laboratory polysomnography (PSG) [1–3]. OSAHS patients encounter the CPAP apparatus for the ﬁrst time on titration night, and then the eﬀects of overnight CPAP treatment (i.e., titration parameters) are estimated immediately. However, not all patients subjectively sleep well in a CPAP titration setting because of inconvenience and intolerance of the initial phase of CPAP treatment itself. In contrast, auto-
M. Suzuki et al. / Sleep Medicine 8 (2007) 773–778
adjusted CPAP or formula-calculated ﬁxed CPAP treatment sometimes precedes the titration night to familiarize patients with the CPAP apparatus or to prevent a cardiovascular, cerebrovascular attack or sudden death [4–6]. Although research data on CPAP titration have been accumulated, few analyses of the eﬀect of the initial experience of the treatment itself (i.e., ﬁrst-night eﬀect of CPAP) on sleep parameters have been performed. The purpose of this study was to compare the eﬀects of CPAP pretreatment and non-pretreatment on sleep architecture and other descriptors on titration night and for subsequent CPAP compliance rate. 2. Patients and methods 2.1. Protocol A prospective randomized, controlled study design was employed. OSAHS patients in whom apnea–hypopnea index (AHI) was P20 events/h by overnight baseline PSG were divided into two groups using a random number table and sealed envelopes. Seventy OSAHS patients received autoadjusted CPAP treatment for 2 months and then received the standard manual titration (CPAP-pretreated group). The other 70 patients did not receive any CPAP treatment before receiving the standard manual titration (non-CPAP-pretreated group). For patients in the CPAP-pretreated group, technologists provided 30-min instructions on the CPAP apparatus, nasal masks, and how to adjust the autoadjusted CPAP, as well as information about symptoms indicating an incorrect CPAP setting. Technologists also suﬃciently performed hands-on intervention. Then a REMStar Auto (Respironics; Pittsburgh, PA, USA) autoadjusted CPAP machine was prescribed for 2 months after the diagnostic study until the standard CPAP manual titration night. The pressure range was set between 4.0 and 20.0 cmH2O before titration. There was a card with a built-in monitoring chip in the REMstar Auto CPAP machine for the collection and storage of CPAP pressure and usage data. The monitoring device provided the pressure at which the patient spent 90% of the night at or below (90% device pressure) for each patient. Titration was performed manually during an attended overnight in-laboratory sleep-in another 2 months after the baseline PSG night. For patients in the non-CPAP-pretreated group, the standard CPAP manual titration by the technologists was scheduled also in another 2 months after the baseline PSG night. Suﬃcient instruction on the CPAP apparatus, nasal masks, and how to adjust the autoadjusted CPAP, in addition to information on symptoms indicating an incorrect CPAP setting, were given by technologists. Hands-on intervention was also performed by the technologists on titration night. The techniques for the standard CPAP manual titration were the same both in the
CPAP-pretreated and non-CPAP-pretreated groups. The technologists increased CPAP pressure during sleep in a stepwise fashion every 0.2–1.0 cmH2O to abolish respiratory events and arousals. The pressure at which arousals through supine rapid eye movement (REM) sleep disappeared was basically determined as the optimal pressure. The CPAP apparatus used for titration was a REMStar Pro, ﬁxed-type CPAP machine (Respironics; Pittsburgh, PA, USA). REMStar Pro was prescribed for OSAHS patients in both groups at the optimal pressure for each patient as determined by titration study. Monthly follow-up for CPAP usage and problems was performed to establish eﬀective utilization patterns and to remediate problems. The study protocol is outlined in Fig. 1. The waiting time for sleep services (ﬁrst-night PSG and titration) is usually almost 2 months because of a shortage of sleep laboratories in Japan. The Institutional Review Board of our institute approved the study after a review by the Ethics Committee, and written informed consent was obtained from each patient before participation in the study. 2.2. Polysomnography (PSG) Overnight sleep studies, baseline PSG and CPAP titration, were carried out by digital PSG (Alice 4: Respironics; Pittsburgh, PA, USA) according to our previous method . Brieﬂy, electroencephalography (EEG, C4/A1, C3/A2), electrooculography (EOG), submental electromyography (EMG) and electrocardiography (ECG) using surface electrodes, measurement of air ﬂow at the nose and mouth using a thermistor, measurement of respiratory movements of the rib cage and abdomen by inductive plethysmography, and measurement of percutaneous arterial oxygen saturation (SpO2) using a ﬁnger pulse oximeter were simultaneously carried out. The predominant sleep stage was scored for each 30-s epoch according to the criteria established by Rechtshaﬀen and Kales (R&K) . Arousals were identiﬁed according to the criteria of
Fig. 1. Study protocol. CPAP-pretreated group, after baseline (ﬁrst night) PSG; subjects slept with CPAP device at home for 2 months and in subsequent standard manual titration study in the sleep laboratory (n = 70). In the non-CPAP-pretreated groups, patients did not use CPAP device for 2 months at home and received the standard manual titration (n = 70). AHI, apnea–hypopnea index; OSAHS, obstructive sleep apnea hypopnea syndrome.
M. Suzuki et al. / Sleep Medicine 8 (2007) 773–778
the American Sleep Disorders Association (ASDA) . Arousals preceded by the termination of breathing events or snoring were deﬁned as respiratory arousals. Spontaneous arousals were deﬁned as pure EEG arousals without clear changes in other physiological parameters or movement arousals, excluding periodic leg movement arousals according to the criteria of R&K by an EMG activation. Apneas were identiﬁed as a nearly ﬂat airﬂow (<10% of the baseline, at which the baseline amplitude was identiﬁed during the nearest preceding period of regular breathing with a stable oxygen saturation) for at least 10 s. Hypopneas were identiﬁed as an airﬂow or a thoracoabdominal excursion of approximately <70% of the baseline for at least 10 s associated with either an oxygen desaturation of >3% or an arousal. 2.3. Statistical analyses All descriptive statistics are presented as means ± standard error. Descriptive statistics were calculated for each variable. Unpaired variables were evaluated by the Mann–Whitney U-test, and paired variables by the Wilcoxon t-test. All p-value tests were two-tailed. A p-value of less than 0.05 was considered to indicate statistical signiﬁcance. The SPSS computerized statistical package (version 11.01) was used. 3. Results 3.1. Physical and baseline PSG parameters of CPApretreated and non-CPAP-pretreated groups
0.2 ± 0.1 and 7.3 ± 1.9 to 0.2 ± 0.1 (%), respectively (each p value was less than 0.001). These results show that the standard CPAP manual titration was so eﬀective in both groups that it permitted the normalization of respiratory parameters. 3.3. Diﬀerences in sleep parameters between CPAPpretreated and non-CPAP-pretreated groups Under an adequate attenuation of respiratory events, we compared sleep parameters between the two groups. Table 3 shows the comparison of changes in sleep parameters from the baseline to titration PSG. The average changes in sleep eﬃciency (SE) in the CPAP-pretreated and non-CPAP-pretreated groups were 8.3% ± 1.7 and 1.2 % ± 1.5, respectively, suggesting that the CPAP-pretreated group had a signiﬁcantly improved SE compared with the non-CPAP-pretreated group. There was a signiﬁcant diﬀerence also in the absolute ﬁgures of the SE between the two groups (p = 0.002). In contrast, there were no signiﬁcant diﬀerences between the two groups in changes in other parameters (Table 3). In the next series of this study, the eﬀects on arousal changes from the baseline to titration PSG between the two groups were elucidated. There was a signiﬁcant difference in changes in arousal index (Ar-I) in non-rapid eye movement (NREM) sleep. There was a signiﬁcant diﬀerence also in the absolute ﬁgures of the Ar-I in NREM between the two groups (p = 0.005). However, no signiﬁcant diﬀerences were observed in changes in other parameters between the two groups (Table 4). 3.4. CPAP and compliance
Three of 70 patients in the CPAP-pretreated group and two of 70 patients in the non-CPAP-pretreated group were women. There were no signiﬁcant diﬀerences in age and body mass index (BMI) between the CPAPpretreated and non-CPAP-pretreated groups. There were no signiﬁcant diﬀerences in baseline PSG parameters between the two groups (Table 1). The baseline arousals between the two groups are compared in Table 2. There were no signiﬁcant diﬀerences between the two groups (Table 2). 3.2. Eﬃcacy of standard CPAP titration in CPAPpretreated and non-CPAP-pretreated groups The standard CPAP manual titration could attenuate overall respiratory events both in the CPAP-pretreated and non-CPAP-pretreated groups from the baseline to titration: AHI, 58.9 ± 2.1 to 7.0 ± 0.5 and 57.9 ± 2.1 to 9.1 ± 0.7 (events/h); apnea indices, 29.0 ± 2.9 to 0.6 ± 0.1 and 28.8 ± 2.7 to 2.1 ± 0.3 (events/h); hypopnea indices, 30.0 ± 1.7 to 6.4 ± 0.5 and 29.1 ± 2.1 to 8.3 ± 0.6 (events/h); and cumulative percentage of time spent at saturation below 90 (CT 90), 9.3 ± 1.4 to
The average optimal pressures determined by titration PSG in the CPAP-pretreated and non-CPAP-pretreated groups were 11.3 ± 0.3 and 11.7 ± 0.4 cmH2O, respectively, which were not signiﬁcantly diﬀerent. The average 90% device pressure for autoadjusted CPAP during the two months before the titration night in the CPAP-pretreated group was 9.3 ± 0.3 cmH2O, which was lower than the average optimal pressure (11.3 ± 0.3 cmH2O) by 2.0 cmH2O (p < 0.001). In the CPAP-pretreated group, the compliance during the 2 months before titration was 4.1 ± 0.4 h. The average hours of daily use in another nine months from titration PSG in the CPAP-pretreated and non-CPAP-pretreated groups were 5.1 ± 0.5 and 4.9 ± 0.4 h, respectively; which were not signiﬁcant diﬀerent (p = 0.54). There was a signiﬁcant diﬀerence between the pretreatment compliance of 4.1 h and post-titration compliance of 5.1 h in the CPAP-pretreated group (p = 0.036). Eight patients in the non-CPAP-pretreated group discontinued CPAP treatment nine months after the titration night, whereas one patient in the CPAP-pretreated group discontinued. Yates v-test showed that there
M. Suzuki et al. / Sleep Medicine 8 (2007) 773–778
Table 1 Comparison of physical and ﬁrst-night PSG parameters between CPAP preceding and non-CPAP-pretreated groups
Age BMI (kg/m2) AHI (events/h) CT90 (%) TST (min) SE (%) %REM %Stage1 %Stage2 %SW Ar-I (events/h) Sleep latency (min) REM latency (min)
45.3 ± 1.2 28.4 ± 0.6 58.9 ± 2.1 9.3 ± 1.4 368 ± 7.8 78.7 ± 1.6 16.3 ± 0.7 43.5 ± 2.4 39.7 ± 2.3 0.5 ± 0.2 56.4 ± 2.0 17.3 ± 2.3 134.0 ± 8.5
47.5 ± 1.4 27.8 ± 0.5 57.9 ± 2.1 7.3 ± 1.2 362 ± 8.9 80.5 ± 1.1 16.2 ± 0.7 41.9 ± 2.3 40.9 ± 2.0 0.8 ± 0.2 54.2 ± 1.5 13.6 ± 1.3 135.2 ± 7.5
0.20 0.65 0.71 0.34 0.72 0.79 0.83 0.69 0.77 0.19 0.45 0.91 0.78
AHI, apnea–hypopnea index; CT 90, cumulative percentage of time spent at saturation below 90; TST, total sleep time; SE, sleep eﬃciency; %REM, percentage of rapid eye movement sleep time in TST; %stage 1, percentage of stage 1 sleep in TST; %stage 2, percentage of stage 2 sleep in TST; %SW, percentage of slow-wave sleep in TST; Ar-I, arousal index.
Table 2 Comparison of ﬁrst-night arousal indices between CPAP preceding and non-CPAP-pretreated groups
REM Ar-I NREM Ar-I Respir Ar-I in REM Spont Ar-I in REM Respir Ar-I in NREM Spont Ar-I in NREM
48.1 ± 2.1 60.4 ± 2.0 44.5 ± 2.2 3.9 ± 0.4 53.1 ± 2.3 7.3 ± 0.6
45.7 ± 1.5 60.0 ± 2.2 41.0 ± 1.6 4.8 ± 0.6 52.5 ± 2.5 7.5 ± 0.7
0.53 0.95 0.25 0.72 0.87 0.97
Ar-I, arousal index; NREM, non-REM; Respir Ar-I, respiratory arousal index; spont Ar-I, spontaneous arousal index.
Table 3 Comparison of changes in sleep parameters from baseline to titration PSG between CPAP preceding and non-CPAP-pretreated groups CPAPpretreated (70) Changes in SE (%) Changes in %REM Changes in %Stage1 Changes in %Stage Changes in %SW Changes in Ar-I (events/h) Changes in Sleep latency (min) Changes in REM latency (min) *
8.3 ± 1.7 6.8 ± 0.9 26.1 ± 2.3 2 17.2 ± 2.0 2.0 ± 0.5 33.8 ± 2.2
1.2 ± 1.5 5.3 ± 1.1 20.3 ± 2.7 14.3 ± 1.9 1.5 ± 0.5 27.7 ± 2.0
0.006* 0.430 0.075 0.530 0.835 0.063
5.2 ± 3.2
1.7 ± 2.0
46.7 ± 9.8
30.1 ± 9.9
Table 4 Comparison of changes in arousal indices from baseline to titration PSG in CPAP preceding and non-CPAP-pretreated groups
Changes in Changes in Changes in REM Changes in REM Changes in NREM Changes in NREM *
REM Ar-I NREM Ar-I respir Ar-I in
27.3 ± 2.4 37.5 ± 2.3 39.2 ± 2.2
26.9 ± 1.9 30.7 ± 2.7 35.2 ± 1.8
0.560 0.036* 0.266
spont Ar-I in
9.9 ± 1.1
7.9 ± 0.9
respir Ar-I in
46.6 ± 2.3
42.4 ± 2.5
spont Ar-I in
9.0 ± 1.0
11.2 ± 1.3
was a signiﬁcant diﬀerence between the two groups in CPAP adherence (p = 0.039). 4. Discussion There are ﬁndings in this study that may contribute to the understanding of the therapeutic theory on CPAP for patients with OSAHS. Pretreatment increased the improvement in SE from baseline and also resulted in a decrease from baseline in the total Ar-I during NREM sleep. Few other changes were signiﬁcantly diﬀerent between the groups. Pretreatment aﬀects subsequent adherence to CPAP. 4.1. Sleep eﬃciency in titration night CPAP application is currently the most frequently used medical treatment for patients with OSAHS. This treatment requires the determination of the speciﬁc pressure for each patient in initiating. It is recommended that this should be accomplished by manual titration performed during an attended overnight in-laboratory PSG [1–3]. Titration endpoints for ‘‘optimal’’ CPAP can include values that lead to the elimination of apneas–hypopneas, snoring, inspiratory ﬂow limitation and arousals. CPAP induced the attenuation of arousals and light sleep, and slow wave (SW, stages 3 and 4) sleep enhancement [3,10]. In contrast, titration night PSG does not always improve SE compared with baseline PSG because of the inconvenience and intolerance of the initial phase of CPAP treatment itself (ﬁrst-night eﬀect of CPAP). SE on titration night is important because of its strong correlation with a change in score on the Epworth Sleepiness Scale at follow-up and with CPAP compliance from 6 to 12 months . In addition, Drake et al. reported that OSAHS patients whose SE improved during CPAP titration used their CPAP machines for an average of more than 2.03 h per night after the mean follow-up days of 46.9 as compared with individuals whose SE did not improve during CPAP
M. Suzuki et al. / Sleep Medicine 8 (2007) 773–778
titration . Patient compliance may be correlated with the degree of improvement in SE that patients experience during their initial titration night. Hukins et al. randomly assigned 91 patients to commence CPAP either at an arbitrary pressure determined on the basis of BMI before titration or at an optimal pressure after a CPAP titration, and demonstrated that there was a signiﬁcantly higher SE at arbitrary pressure on titration night in agreement with our results . In the present study, the CPAP-pretreated group had a signiﬁcantly better improvement in SE. A preceding CPAP treatment would be of beneﬁt for better compliance of patients whose SE is low at baseline PSG.
ture. However, titration protocol in this study (e.g., starting pressure, rate of titration up or down) was the same across subjects. There were statistical limitations because the power calculation was not done before performing the study, which might have reduced the value of the present study. In summary, a preceding CPAP treatment showed minimal eﬀects on sleep parameters on titration night and subsequent CPAP compliance rate when a suﬃcient intervention for CPAP treatment was performed on patients with OSAHS, although it was speculated that this preceding treatment might be of beneﬁt for better compliance in patients whose SE is low or arousal index in NREM is high.
4.2. Respiratory arousals Nocturnal arousals are a prominent feature and the essential cause of a disturbed sleep structure in patients with OSAHS. Respiratory eﬀort produces arousals by mechanoreceptors, and these arousals induce upper airway muscle tonus activity and apnea termination coincident with postapneic hyperventilation . These respiratory arousals are associated with sleep-related breathing events registered in PSG tracing and might be linked to undetected increased ventilatory eﬀorts . Detecting arousals would lead to the detection of sleep-related breathing events. There may be diﬀerences in arousability between sleep stages. Ar-I in NREM sleep is higher than that in REM sleep for patients with severe OSAHS . In the baseline PSG, the highest respiratory Ar-I was observed in stage 1, which is significantly diﬀerent from the frequencies in stage 2, SW and REM, and these respiratory Ar-Is decreased signiﬁcantly from the baseline PSG to the ﬁrst CPAP titration night [17,18]. In REM sleep, patients are not entirely dependent on arousal or on chemical and mechanoreceptor feedback to increase ﬂow . The arousal threshold during REM sleep state might be higher than that during NREM sleep [17,18]. Thus, arousals in NREM sleep might be much more correlated with respiratory events than those in REM sleep. 4.3. Limitations in this study The most ideal study to compare arousals should have other micro-architectural sleep parameters, such as cyclic alternating pattern, microarousals (less than 3 s) or delta power . If these parameters were analyzed, other ﬁndings regarding arousal might be obtained. A single-night titration study design might have aﬀected our results . It is diﬃcult to compare a full overnight study to a CPAP titration study in terms of sleep architecture. Sleep architecture may vary due to the eﬀects of CPAP titration, and the time the eﬀective CPAP pressure was achieved might vary by patient factors, circumstances or their circadian rhythm and, therefore, may have aﬀected sleep architec-
Acknowledgements The authors express their deepest appreciation to Mr. Kiyoshi Moriyama PSGT, Sleep Respiratory Center, Komagamine Hospital, Tokyo; Mr. Yoichiro Takei RPSGT and Miss Manami Yahashi RPSGT, both of Division of Sleep Medicine, Setagaya Sleep Center, Tokyo, for their PSG scoring and data analyses. This work was supported by Research Grant No. 17591802 to M.S. from the Ministry of Education, Culture, Sports, Science and Technology, Japan. References  American Thoracic Society. Indications and standards for use of nasal continuous positive pressure (CPAP) in sleep apnea syndromes. Am J Respir Crit Care Med 1994;150:1738–1745.  Kushida C, Littner MR, Hirshkowitz M, et al. Practice parameters for the use of continuous and bilevel positive airway pressure devices to treat adult patients with sleep-related breathing disorders. An American Academy of Sleep Medicne Report. Sleep 2006;29:375–80.  Gay P, Weaver T, Loube D, Iber C. Evaluation of positive airway pressure treatment for sleep related breathing disorders in adults. A review by the positive airway pressure task force of the standards of practice committee of the American Academy of Sleep Medicne. Sleep 2006;29:381–401.  Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes in men with obstructive sleep apnoea–hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet 2005;365:1046–53.  Diomedi M, Placidi F, Cupini LM, Bernardi G, Silvestrini M. Cerebral hemodynamic changes in sleep apnea syndrome and eﬀect of continuous positive airway pressure treatment. Neurology 1998;51:1051–6.  Gami AS, Howard DE, Olson EJ, Somers VK. Day–night pattern of sudden death in obstructive sleep apnea. N Eng J Med 2005;352:1206–14.  Suzuki M, Saigusa H, Chiba S, Yagi T, Shibasaki K, Hayashi M, Moriyama K, Kodera K. Discrepancy in polysomnography scoring for a patient with obstructive sleep apnea hypopnea syndrome. Tohoku J Exp Med 2005;206:353–360.  Rechtschaﬀen A, Kales A. A manual of standardized terminology, technique and scoring system for sleep stages of human sleep. Brain Information Service, Los Angeles, 1968.
M. Suzuki et al. / Sleep Medicine 8 (2007) 773–778
 American Sleep Disorders Association EEG arousals: scoring rules and examples. ASDA Report. Sleep 1992;15:173–184.  Issa FG, Sullivan CE. The immediate eﬀects of nasal continuous positive airway pressure treatment on sleep pattern in patients with obstructive sleep apnea syndrome. Electroencephalogr Clin Neurophysiol 1986;63:10–7.  McArdle N, Douglas NJ. Eﬀect of continuous positive airway pressure on sleep architecture in the sleep apnea–hypopnea syndrome. A randomized controlled trial. Am J Resp Crit Care Med 2001;164:1459–63.  Drake CL, Day R, Hudgel D, Stefadu Y, Parks M, Syron ML, et al. Sleep during titration predicts continuous positive airway pressure compliance. Sleep 2003;26:308–11.  Hukins CA. Arbitrary-pressure continuous positive airway pressure for obstructive sleep apnea syndrome. Am J Resp Crit Care Med 2004;171:500–5.  Kubin L, Davis RO. Mechanisms of airway hypotonia. In: Pack AI, editor. Sleep apnea pathogenesis, diagnosis, and treatment. NewYork: Marcel Dekker; 2002. p. 99–154.
 Berry RB, Gleeson K. Respiratory arousal from sleep: mechanisms and signiﬁcance. Sleep 1997;20:654–75.  Siddiqui F, Walters AS, Goldstein D, Lahey M, Desai H. Half of patients with obstructive sleep apnea have a higher NREM AHI than REM AHI. Sleep Med 2006;7:281–5.  Fietze I, Quispe-Bravo S, Hansch T, Rottig J, Baumann G, Witt CH. Arousals and sleep stages in patients with obstructive sleep apnoea syndrome: changes under nCPAP treatment. J Sleep Res 1998;6:128–33.  Dingli K, Fietze I, Assimakopoulos T, Quispe-Bravo S, Witt C, Douglas NJ. Arousability in sleep apnoea/hypopnoea syndrome patients. Eur Respir J 2002;20:733–40.  Younes M. Contributions of upper airway mechanics and control mechanisms to severity of obstructive apnea. Am J Respir Crit Care Med. 2003;168:645–58.  Thomas RJ, Terzano MG, Parrino L, Weiss JW. Obstructive sleep-disordered breathing with a dominant cyclic alternating pattern – a recognizable polysomnographic variant with practical clinical implications. Sleep 2004;27:229–34.