ORIGINAL ARTICLE

Central sleep apnea and complex sleep apneain patients with epilepsy

Martina Vendrame & Stephanie Jackson & Sana Syed &

Sanjeev V. Kothare & Sanford H. Auerbach

Received: 4 April 2012 /Revised: 17 April 2013 /Accepted: 22 April 2013# Springer-Verlag Berlin Heidelberg 2013

AbstractPurpose We sought to examine the prevalence of centralsleep apnea (CSA) and complex sleep apnea (CompSA) inpatients with epilepsy and to examine their clinical profile,with respect to epilepsy type, etiology, medication use, andEEG abnormalities.Methods We undertook a retrospective analysis of 719consecutive patients with epilepsy who underwentpolysomnography (PSG) at our institution between2004 and 2011. Of the 458 patients with complete data,we excluded 42 patients with congestive heart failure orleft ventricular ejection fraction <40 %. Comparison ofclinical and PSG variables between the three groupswere conducted with Fisher exact test and analysis ofvariance.Results Out of 416 patients tested, 315 (75 %) had obstruc-tive sleep apnea (OSA), 16 (3.7 %) had CSA, 33 (7.9 %) hadCompSA. There were more males in the CSA and CompSAgroups than in the OSA group (81.2, 81.8, and 59.6 %,respectively, p=0.04). Focal seizures were more prevalentin patients with CSA than in patients OSA or CompSA(62.5, 265, and 21.1 %, respectively, p=0.02).Conclusion About 11 % of epilepsy patients have sleep-breathing disorders with central apneas, which is not higherthan that in a general population. These data should be ex-panded with future research investigating the role of interictal,ictal, and postictal central apneas in epileptogenesis andepilepsy.

Keywords Sleep-disordered breathing . Seizures .

Convulsions . Epilepsy . SUDEP

Introduction

Patients with epilepsy may be at higher risk of obstructivesleep apnea (OSA) than the general population [1, 2]. Pre-dictors for OSA in this population may be older age [1] andmale gender [1, 2]. Manni et al. also found that people withepilepsy and OSA had a higher body mass index (BMI), andwere more frequently sleepy than people with epilepsywithout OSA [2]. No association has been seen betweenseizure frequency and a higher risk of OSA in people withepilepsy [1, 2]. Further evidence supports the notion thattreatment of these patients with continuous positive airwaypressure (CPAP) may be associated with better seizure con-trol [3–7].

Despite this increasing concern regarding OSA in epilep-sy patients, there are very few to no studies on sleep-disordered breathing with central sleep apneas (CSA).While OSA and CSA results from different pathophysiolo-gy, it is increasingly accepted that OSA and CSA oftencoexist in the same patient; the presence of one can predis-pose to the other, and that overall OSA and CSA are not asdistinct as previously thought. Central apneas have beendescribed in patients with other neurological conditionssuch as stroke [8, 9], brain tumors [10–12], Parkinson’sdisease [13, 14], Arnold–Chiari malformation [15], Joubertsyndrome [16], and neurological infectious diseases [17,18]. In these patients, CSA is thought to be a result ofchemoreflex control instability, causing periods of inade-quate respiratory drive insufficient to trigger breathing [19].

In the present study, we sought to:

1) Assess the prevalence of CSA and complex sleep apnea(CompSA) in patients with epilepsy, and

M. Vendrame (*) : S. Jackson : S. Syed : S. H. AuerbachNeurology Department, C-3, Boston University,72 E Concord Street,Boston, MA 02118, USAe-mail: vendrame.martina@gmail.com

S. V. KothareDivision of Clinical Neurophysiology, Children’s Hospital Boston,Boston, MA 02115, USA

Sleep BreathDOI 10.1007/s11325-013-0858-8

2) Examine the clinical profile of these epilepsy patientsfor epilepsy type, etiology, medication use, and EEGabnormalities.

Methods

Patients

We undertook a retrospective analysis on 719 consecutiveadult patients with epilepsy (age >21 years) referred to theBoston University Sleep Disorders Center between 2004and 2011. We recorded data on clinical history, body massindex, and polysomnography (PSG) findings. We excludedall patients with incomplete PSG or clinical data. In order tominimize clinical difference between our patient groups, weexcluded patients with congestive heart failure or left ven-tricular ejection fraction <40 %.

Polysomnography

PSGs were performed using a digital polygraph (Bio-logicSystems Corp., Mundelein, IL). Sleep staging and arousalswere scored by certified technicians according to standardmethod [20, 21]. Airflow was recorded with the Pro-Technasal pressure transducer (Pro-Tech Services, Inc., Mukilteo,WA) and with the respiratory impedance plethysmographyduring the diagnostic phase; when CPAP was applied, theflow channel from the CPAP pneumotachometer plus respira-tory impedance plethysmography was used to measure air-flow. Apneas were defined as cessation of airflow equal orlonger than 10 s, and hypopneas were defined as 30 % de-clines in airflow of at least 10-s duration with at least a 4 %oxyhemoglobin desaturation [21]. Apneas were scored ascentral when accompanied by evidence of absent respira-tory effort, and as obstructive when accompanied by re-spiratory effort. Events were scored as mixed apneas ifthey met apnea criteria and were associated with absentrespiratory effort in the initial portion of the event,followed by resumption of inspiratory effort in the secondportion of the event [21]. Arousals were classified asrespiratory-related arousals if they were preceded by areduction in airflow with an oxygen desaturation that didnot meet the criteria for a hypopnea [22].

When CPAP titration was performed with a “split-night”protocol, the diagnostic portion was initiated after a mini-mum of 2 h of sleep. CPAP titration was performed inaccordance to the American Academy of Sleep Medicineguidelines [23]. CPAP was started if there was an apnea–hypopnea index (AHI) ≥5 or 10 or more respiratory-relatedarousals per hour of sleep. CPAP titration was started at 4 to

5 cm H2O and increased in increments of 1–2 cm H2O inorder to eliminate respiratory sleep disturbances.

Definitions of sleep-disordered breathing

OSA was defined as AHI ≥5/h. We considered the patientsto have CSA when we observed a central apnea index ofmore than five events per hour and at least 50 % of the totalAHI was purely central in origin [21, 24]. CompSA wasdiagnosed when CPAP titration eliminated obstructiveevents defining OSA, but the residual central apnea indexwas equal or greater than ≥5/h or Cheyne–Stokes respiratorypattern became prominent [25]. Cheyne–Stokes breathingwas defined as at least three consecutive cycles of cyclicalcrescendo/decrescendo change in breathing amplitude withfive or more central apneas or hypopneas per hour of sleepand/or a cyclical crescendo and decrescendo change inbreathing amplitude with a duration of 10 min or more [21].

Statistical analysis

Clinical findings and PSG features were compared acrossthe three patient groups (OSA, CSA, and CompSA) usingthe Fisher exact test for categorical variables and analysis ofvariance for continuous variables (SPSS version 16.0,Microsoft Inc). For all analysis, p values of 0.05 or lesswere considered statistically significant.

Results

Prevalence of sleep-related breathing disorders in patientswith epilepsy

Of the 458 patients with complete data, we excluded 42patients with congestive heart failure or left ventricular ejec-tion fraction <40 %. Out of 416 patients tested, 315 (75 %)had OSA, 16 (3.7 %) had CSA, and 33 (7.9 %) had CompSA.

Demographics and clinical findings

Demographic and clinical data of patients with sleep-disorderedbreathing are summarized in Table 1. There were more malesin the CSA and CompSA groups than in the OSA group (81.2,81.8, and 59.6 %, respectively, p=0.04). CSA patients tendedto have higher BMI than OSA and CompSA patients, al-though no statistical significance was reached (p=0.06). Com-parison of epilepsy-related variables between groups showedthat 62.5 % of CSA patients had focal seizures, while 26 and21.1 % of patients with OSA and CompSA had focal seizures(p=0.02).

Sleep Breath

Diagnostic PSG findings

Findings of the diagnostic portion of the PSG are presentedin Table 2. Patients from all groups showed poor sleepefficiencies, with means ranging from 65.3 to 72.5 %, withno significant differences across groups. Patients had asimilar total arousal index (mean ranged from 19.6–22.6/h,p=0.7) and respiratory-related arousal index (mean rangedfrom 17.6–22.7/h, p=0.8). Although the total AHIs were notdifferent between the three groups (p=0.66), patients withCSA had higher total central apnea index (CAI) than pa-tients with OSA or CompSA (p <0.05 in each case), duringboth rapid eye movement (REM) and non-rapid eye move-ment (NREM) sleep (p=0.04 and p=0.008). Obstructiveapnea indices (OAIs) were higher in both patients withOSA and CompSA compared to patients with CSA (p=

0.03). Finally, there was no difference between the threegroups for minimum oxygen saturation.

PSG findings on CPAP

Findings of the PSG during CPAP titration are shown inTable 3. Of the 336 patients who underwent CPAP titration,275 had a “split-night” study and 61 had a “full-night”titration study. There were no significant differences in re-sults between the two types of study. Similar to the baselineportion of the PSGs, patients had similar total arousal index(mean ranged from 22.5–28.2/h, p=0.8) and respiratory-related arousal index (mean ranged from 8.5–12.5/h, p=0.7). For patients with OSA, CPAP led to clear improvementin AHI, whereas in patients with CSA and CompSA, theoverall AHI during CPAP remained elevated (mean AHI=

Table 1 Clinical variables ofpatients with epilepsy and sleep-disordered breathing

Data are presented as mean ± SDor number (%)

BMI body mass index, ESSEpworth Sleepiness Scale, AEDantiepileptic drugs, SSRIserotonin-reuptake inhibitors,EEG electroencephalogram,MRI magnetic resonanceimaging

*OSA vs CSA p <0.05

**OSA vs CompSA p <0.05

***CSA vs CompSA p <0.05

OSA (n=315) CSA (n=16) CompSA (n=33) p value

Age 53.6±11 54.3±12.1 54.3±13.3 0.56

Gender (men) 188 (59.6 %) 13 (81.2 %) 27 (81.8 %) 0.04*, **

BMI (kg/m2) 33.9±6.5 27.1±5.9 32.9±8.9 0.06

ESS 13.0±5.2 13.2±8.9 12.3±7.3 0.75

Insomnia complaints 204 (64.7 %) 12 (75 %) 28 (84 %) 0.88

Known epilepsy etiology 192 (60.9 %) 10 (62.5 %) 19 (57.5 %) 0.83

AED polytherapy 133 (42.2 %) 4 (25 %) 7 (21 %) 0.57

SSRI therapy 189 (60 %) 7 (43.7 %) 18 (54.5 %) 0.64

Narcotic therapy 51 (16.1 %) 4 (25 %) 7 (21.1 %) 0.9

Abnormal EEG 90 (28.4 %) 1 (6 %) 7 (21.1 %) 0.8

Abnormal brain MRI 65 (20.6 %) 2 (12.5 %) 5 (15.1 %) 0.85

Seizure frequency (/month) 1.0±3.0 1.0±5.2 1.0±8.0 0.9

Focal seizures 82 (26 %) 10 (62.5 %) 7 (21.1 %) 0.02*, ***

Nocturnal seizures 22 (6.9 %) 1 (6.2 %) 2 (6 %) 1.0

Table 2 PSG feature of diag-nostic portion of testing

Data are presented as mean ± SD

REM rapid eye movement sleep,AHI Apnea–hypopnea index,CAI central apnea index, NREMnon-rapid eye movement sleep,OAI obstructive apnea index,HI hypopnea index

*OSA vs CSA p <0.05

**CSA vs CompSA p <0.05

OSA (n=315) CSA (n=16) CompSA (n=33) p value

Total sleep time (min) 185±86 172±99 177±82 0.85

Sleep efficiency (%) 72.5±52.3 65.3±30.1 69.6±49.3 0.99

Total arousal index (/h) 22.6±18.2 19.6±13.3 24.3±18.8 0.7

Respiratory-related arousal index (/h) 17.6±15.8 18.1±13.2 22.7±21.5 0.8

AHI (/h) 26.5±19.6 30.7±32.2 28.6±25.4 0.66

CAI (/h) 2.2±1.22 21.2±18.12 0.9±1.01 0.007*, **

REM CAI (/h) 1.3±1.09 9.3±7.78 0.7±0.92 0.04*, **

NREM CAI (/h) 2.9±1.44 25.4±18.3 0.9±1.2 0.008*, **

OAI (/h) 15±16.5 4.1±7.1 19.5±21.1 0.03*, **

HI (/h) 9.3±10.2 5.4±4.4 8.2±5.5 0.2

Minimum oxygen desaturation (%) 83±10 80±11 81±10 1.0

Sleep Breath

2.2, 24.5, 19.4/h, respectively, p=0.04). Both patients withCompSA and CSA had a higher CAI on CPAP than didthose with OSA (p=0.003). Although CompSA patients hadlower AHI than those with CSA (means 14.6 and 19.8,respectively), no statistical significance was reached.

Discussion

In our retrospective study, about 11 % of patients withepilepsy were affected by sleep-breathing disorders withcentral apneas (CSA in 3.7 % and CompSA in 7.9 %).The presence of focal seizures was associated with CSA,and this was the only epilepsy-related variable associatedwith a specific breathing disorder.

CSAwas found in about 3.7 % patients with epilepsy, withan overall higher CAI than patients with other sleep-breathingdisorders. Consistent with prior reports in the general popula-tion [26], male gender was more common in the CSA groupthan in the OSA group (81.2 vs 59.6 %). However, we couldnot identify any presenting features associated with CSA orother sleep-breathing disorder, including differences in BMI,sleepiness (Epworth Sleepiness Scale scores), or insomniacomplaints. This, in contrast to prior reports on a generalpopulation, showing higher prevalence of insomnia com-plaints and higher BMIs in patients with OSA compared topatients with CSA [24, 27]. Similar to findings on epilepsypatients with OSA [1, 2], seizures frequency was not associ-ated with higher risk of sleep-breathing disorder. Furthermore,we did not find an association between nocturnal seizures andspecific sleep-breathing disorders. Contrarily, in a sample ofmedically refractory epilepsy patients, higher prevalence ofOSA was seen in patients with seizures during sleep [1].Additional analysis on higher numbers of epilepsy patientsmay help elucidate a clinical profile of patients with epilepsyand CSA.

Central sleep apnea can also be seen during titration withCPAP in patients with OSA. Although this entity, called

“CPAP-emergent CSA” or CompSA has received increasingattention, its prevalence and natural history are not wellestablished [28–30]. In the present study, 33 of the 419tested epilepsy patients (7.9 %) had CompSA, a prevalencelikely comparable to the one of the general population withOSA on CPAP. In a large retrospective study of patientswith OSA undergoing CPAP titration, CompSA was foundin about 6.5 % of the 1,286 patients studied [31]. In aretrospective study of 223 patients, Morgenthaler et al.reported CompSA in 15 % of patients [25]. Dernaika et al.reported CompSA in 20 % of 116 patients with split-nightstudies [32]. In another study of 99 patients with OSA,Lehman et al. reported a prevalence of about 13 % [33].

As expected, patients with CompSA had higher obstruc-tive apnea index than CSA patients during the baselinestudy, but higher CAI during CPAP titration (median 14.6in CompSA vs 1.2 in OSA). In contrast, no significantdifferences between CAI were seen between patients withCompSA and CSA during titration studies. The mechanismunderlying CompSA is unclear but may be related to therelief of upper-airway obstruction resistance during CPAP,resulting in higher CO2 excretion and hypocapnia, andconsequent cessation in respiratory effort [34, 35]. ChronicCPAP therapy is not a recognized risk factor for CompSA.CompSA generally resolves with ongoing CPAP therapy,with about 10 % of patients experiencing resolution ofrespiratory events with ongoing therapy [31, 32, 36]. Oneof the limitations of the present investigation is related to thefact that data on long-term CPAP use on patients withepilepsy and CompSA is lacking.

Beside a higher proportion of male gender, there waslittle to distinguish patients with CompSA from those withOSA on clinical grounds (Table 1), and the overall clinicalprofile of patients with CompSA resembled that of patientswith OSA, rather than that of patients with CSA. This maysuggest that patients with CompSA might not be patientswith CSA who had accumulated risk factors for OSA, butrather patients with OSAwho responds differently to CPAP.

Table 3 PSG features duringCPAP titration

Data shown are summarized overthe entire CPAP-titration period(i.e., not only at optimal pressure).Data are presented as mean ± SD

REM rapid eye movement sleep,AHI apnea–hypopnea index, CAIcentral apnea index, NREM non-rapid eye movement sleep, OAIobstructive apnea index,HI hypopnea index

*OSA vs CSA p <0.05

**OSA vs CompSA p <0.05

OSA (n=306) CSA (n=15) CompSA (n=33) p value

Total sleep time (min) 210±74 256±63 221±71 0.89

Sleep efficiency (%) 89.3±43.7 93±55.3 92.4±48.5 0.9

Total arousal index (/h) 28.2±18.5 22.5±18.3 24.8±16.9 0.8

Respiratory-related arousal index (/h) 8.5±2.2 12.5±11.7 11.8±8 0.7

AHI (/h) 2.2±2.1 24.5±18.8 19.4±17.3 0.04*,**

CAI (/h) 1.2±0.7 19.8±16.9 14.6±15.6 0.003*,**

REM CAI (/h) 2.6±1.9 7.1±6.8 2.5±1.8 0.6

NREM CAI (/h) 4.4±3.5 26.6±19.5 16.6±15.2 0.009*,**

OAI (/h) 0.5±0.9 1.3±1.1 1.9±1.3 0.1

HI (/h) 0.5±0.3 3.4±2.2 2.9±2.1 0.3

Best CPAP pressure (cm H2O) 9.0±3.2 8.8±2.5 8.7±3.9 0.9

Sleep Breath

Further studies need to further characterize the clinical pro-file of CompSA patients.

The presence of central apneas in patients with epilepsyis of particular interest given its potential relation to thepathogenesis of sudden unexpected death in epilepsy(SUDEP). SUDEP is the most common cause of deathdirectly related to epilepsy, and clinical evidence suggeststhat it occurs preferentially during sleep [37]. Dysregulationin respiratory physiology and dysfunction in cardiac andcerebral physiologies have been implicated in the pathogen-esis of SUDEP. [38–41]. Risk factors for SUDEP includepoorly controlled seizures, chronic epilepsy, generalizedseizures, antiepileptic drug polytherapy, male gender, andsymptomatic etiology [38, 42]. In our study, comparison ofepilepsy-related variables between groups showed thatabout two thirds of CSA patients had focal seizures, whileless than one third of patients with OSA or CompSA hadfocal seizures (p=0.02), suggesting that focal seizures maybe a risk factor for CSA. While these findings may be incontrast with prior data indicating a relation betweenSUDEP and generalized seizures rather than focal [42],there is evidence that ictal apneas may occur often in pa-tients with localization-related epilepsy [43, 44].

The present study included the analysis of central apneaoccurring during interictal periods, as none of the patients hadseizures during PSG. Further investigations of apnea occur-ring ictally and postictally will elucidate the potential role ofcentral apneas in epileptogenesis, epilepsy, and SUDEP. Fur-thermore, the retrospective nature of the present study haslimited data availability on some patients, especially regardingdetail information on semiology and etiology of seizures. Theuse of certain antiepileptic medications, especially benzodiaz-epines, may affect the occurrence and nature of apneic events,and data on dosages, length of usage, and compliance waslacking. The use of split-night studies limited our capability toexamine differences in sleep architecture and may have led tosome potential bias regarding the distribution of events acrossdifferent sleep stages and sleep cycles. Furthermore, by de-sign, this study concerned itself with the presentation ofpatients with sleep-disordered breathing, and did not includea control population without sleep-disordered breathing. An-other limitation of this study is related to selection bias, giventhat we included only patients referred to the sleep center.Because CSA tends to present with a more unobtrusive symp-tomatology than OSA, epilepsy patients with CSA may havenot come to the attention of the sleep center and thus notincluded in the study. Future prospective studies might revealfurther insight on the possible impact of central apneas onepilepsy patients. Studies of the mechanisms involved inrespiratory control and the clinical response to CPAP therapyover time should be conducted in order to learn more about thepotential role of these disorders and their treatment in thispatient population.

Acknowledgments We thank Walter Lehan RtPSG for his help inanalyzing PSG data.

References

1. Malow BA, Levy K, Maturen K, Bowes R (2000) Obstructivesleep apnea is common in medically refractory epilepsy patients.Neurology 55:1002–1007

2. Manni R, Terzaghi M, Arbasino C, Sartori I, Galimberti CA,Tartara A (2003) Obstructive sleep apnea in a clinical series ofadult epilepsy patients: frequency and features of the comorbidity.Epilepsia 44:836–840

3. Devinsky O, Ehrenberg B, Barthlen GM, Abramson HS, LucianoD (1994) Epilepsy and sleep apnea syndrome. Neurology44:2060–2064

4. Malow BA, Foldvary-Schaefer N, Vaughn BV, Selwa LM, ChervinRD, Weatherwax KJ, Wang L, Song Y (2008) Treating obstructivesleep apnea in adults with epilepsy: a randomized pilot trial.Neurology 71:572–577

5. Vaughn BV, D'Cruz OF, Beach R, Messenheimer JA (1996)Improvement of epileptic seizure control with treatment of ob-structive sleep apnoea. Seizure 5:73–78

6. Beran RG, Plunkett MJ, Holland GJ (1999) Interface of epilepsyand sleep disorders. Seizure 8:97–102

7. Vendrame M, Auerbach S, Loddenkemper T, Kothare S,Montouris G (2011) Effect of continuous positive airway pressuretreatment on seizure control in patients with obstructive sleepapnea and epilepsy. Epilepsia 52:e168–e171

8. Johnson KG, Johnson DC (2010) Frequency of sleep apnea instroke and TIA patients: a meta-analysis. J Clin Sleep Med6:131–137

9. Nezu T, Naganuma M, Shono Y, Toyoda K, Minematsu K (2010)Central apnoea associated with subcortical haemorrhage in the lefttemporal lobe. J Neurol Neurosurg Psychiatry 81:299–301

10. Rosen GM, Bendel AE, Neglia JP, Moertel CL, Mahowald M(2003) Sleep in children with neoplasms of the central ner-vous system: case review of 14 children. Pediatrics 112:e46–e54

11. Mandrell BN, Wise M, Schoumacher RA, Pritchard M, West N,Ness KK, Crabtree VM, Merchant TE, Morris B (2011) Excessivedaytime sleepiness and sleep-disordered breathing disturbances insurvivors of childhood central nervous system tumors. PediatrBlood Cancer 58:746–751

12. Pollak L, Shpirer I, Rabey JM, Klein C, Schiffer J (2004)Polysomnography in patients with intracranial tumors before andafter operation. Acta Neurol Scand 109:56–60

13. Diederich NJ, Vaillant M, Leischen M, Mancuso G, Golinval S,Nati R, Schlesser M (2005) Sleep apnea syndrome in Parkinson'sdisease. A case–control study in 49 patients. Mov Disord 20:1413–1418

14. Maria B, Sophia S, Michalis M, Charalampos L, Andreas P,John ME, Nikolaos SM (2003) Sleep breathing disorders inpatients with idiopathic Parkinson's disease. Respir Med97:1151–1157

15. Spence J, Pasterkamp H, McDonald PJ (2010) Isolated centralsleep apnea in type I Chiari malformation: improvement aftersurgery. Pediatr Pulmonol 45(11):1141–1144

16. Wolfe L, Lakadamyali H, Mutlu GM (2010) Joubert syndromeassociated with severe central sleep apnea. J Clin Sleep Med6:384–388

17. Habek M, Brinar VV (2009) Central sleep apnea and ataxia causedby brainstem lesion due to chronic neuroleptospirosis. Neurology73:1923–1924

Sleep Breath

18. Gascon-Bayarri J, Campdelacreu J, Monasterio C, Prats E, Ferrer I,Rene R (2010) Central apnea in a sporadic case of Creutzfeldt–Jakob disease with brainstem involvement. Acta Neurol Belg110:113–115

19. Malhotra A, Owens RL (2010) What is central sleep apnea? RespirCare 55:1168–1178

20. ASDA Atlas Task Force (1992) EEG arousals: scoring rules andexamples: a preliminary report from the Sleep Disorders AtlasTask Force of the American Sleep Disorders Association. Sleep15:173–184

21. Iber C, Ancoli-Israel S, Chesson A, Quan SF, American Academyof Sleep Medicine (2007) The AASM manual for the scoring ofsleep and associated events: rules, terminology and technical spec-ifications, 1st edn. American Academy of Sleep Medicine,Westchester, Ill

22. Ayappa I, Norman RG, Krieger AC, Rosen A, O'Malley RL,Rapoport DM (2000) Non-Invasive detection of respiratoryeffort-related arousals (REras) by a nasal cannula/pressure trans-ducer system. Sleep 23:763–771

23. Kushida CA, Chediak A, Berry RB, Brown LK, Gozal D, Iber C,Parthasarathy S, Quan SF, Rowley JA (2008) Clinical guidelinesfor the manual titration of positive airway pressure in patients withobstructive sleep apnea. J Clin Sleep Med 4:157–171

24. Sin DD, Fitzgerald F, Parker JD, Newton G, Floras JS, Bradley TD(1999) Risk factors for central and obstructive sleep apnea in 450men and women with congestive heart failure. Am J Respir CritCare Med 160:1101–1106

25. Morgenthaler TI, Kagramanov V, Hanak V, Decker PA (2006)Complex sleep apnea syndrome: is it a unique clinical syndrome?Sleep 29:1203–1209

26. Hume KI, Van F, Watson A (1998) A field study of age and genderdifferences in habitual adult sleep. J Sleep Res 7:85–94

27. Javaheri S, Parker TJ, Liming JD, Corbett WS, Nishiyama H,Wexler L, Roselle GA (1998) Sleep apnea in 81 ambulatory malepatients with stable heart failure. Types and their prevalences,consequences, and presentations. Circulation 97:2154–2159

28. Westhoff M, Arzt M, Litterst P (2012) Prevalence and treatment ofcentral sleep apnoea emerging after initiation of continuous posi-tive airway pressure in patients with obstructive sleep apnoeawithout evidence of heart failure. Sleep Breath 16:71–78

29. Kuzniar TJ, Kovacevic-Ristanovic R, Freedom T (2011) Complexsleep apnea unmasked by the use of a mandibular advancementdevice. Sleep Breath 15:249–252

30. Nakazaki C, Noda A, Yasuda Y, Nakata S, Koike Y, Yasuma F,Murohara T, Nakashima T (2012) Continuous positive airwaypressure intolerance associated with elevated nasal resistance is

possible mechanism of complex sleep apnea syndrome. SleepBreath 16(3):747–752. doi:10.1007/s11325-011-0570-5

31. Javaheri S, Smith J, Chung E (2009) The prevalence and naturalhistory of complex sleep apnea. J Clin Sleep Med 5:205–211

32. Dernaika T, Tawk M, Nazir S, Younis W, Kinasewitz GT (2007)The significance and outcome of continuous positive airwaypressure-related central sleep apnea during split-night sleep stud-ies. Chest 132:81–87

33. Lehman S, Antic NA, Thompson C, Catcheside PG, Mercer J,McEvoy RD (2007) Central sleep apnea on commencement ofcontinuous positive airway pressure in patients with a primarydiagnosis of obstructive sleep apnea-hypopnea. J Clin Sleep Med3:462–466

34. Skatrud JB, Dempsey JA, Badr S, Begle RL (1988) Effect ofairway impedance on CO2 retention and respiratory muscle activ-ity during NREM sleep. J Appl Physiol 65:1676–1685

35. Dempsey JA (2005) Crossing the apnoeic threshold: causes andconsequences. Exp Physiol 90:13–24

36. Kuzniar TJ, Pusalavidyasagar S, Gay PC, Morgenthaler TI (2008)Natural course of complex sleep apnea—a retrospective study.Sleep Breath 12:135–139

37. Nobili L, Proserpio P, Rubboli G, Montano N, Didato G, TassinariCA (2011) Sudden unexpected death in epilepsy (SUDEP) andsleep. Sleep Med Rev 15:237–246

38. Surges R, Thijs RD, Tan HL, Sander JW (2009) Sudden unexpect-ed death in epilepsy: risk factors and potential pathomechanisms.Nat Rev Neurol 5:492–504

39. Bateman LM, Spitz M, Seyal M (2010) Ictal hypoventilationcontributes to cardiac arrhythmia and SUDEP: report on twodeaths in video-EEG-monitored patients. Epilepsia 51:916–920

40. Seyal M, Pascual F, Lee CY, Li CS, Bateman LM (2011) Seizure-related cardiac repolarization abnormalities are associated withictal hypoxemia. Epilepsia 52:2105–2111

41. Seyal M, Bateman LM, Albertson TE, Lin TC, Li CS (2010)Respiratory changes with seizures in localization-related epilepsy:analysis of periictal hypercapnia and airflow patterns. Epilepsia51:1359–1364

42. Tellez-Zenteno JF, Ronquillo LH, Wiebe S (2005) Sudden unex-pected death in epilepsy: evidence-based analysis of incidence andrisk factors. Epilepsy Res 65:101–115

43. Bateman LM, Li CS, Seyal M (2008) Ictal hypoxemia inlocalization-related epilepsy: analysis of incidence, severity, andrisk factors. Brain 131:3239–3245

44. Seyal M, Bateman LM (2009) Ictal apnea linked to contralateralspread of temporal lobe seizures: intracranial EEG recordings inrefractory temporal lobe epilepsy. Epilepsia 50:2557–2562

Sleep Breath