american thoracic society documents · american thoracic society documents 290 american journal of...

AMERICAN THORACIC SOCIETYDOCUMENTS

Research Priorities in Pathophysiology for Sleep-disordered Breathingin Patients with Chronic Obstructive Pulmonary DiseaseAn Official American Thoracic Society Research StatementAtul Malhotra, Alan R. Schwartz, Hartmut Schneider, Robert L. Owens, Pamela DeYoung, MeiLan K. Han,Jadwiga A. Wedzicha, Nadia N. Hansel, Michelle R. Zeidler, Kevin C. Wilson, and M. Safwan Badr; on behalf of the ATSAssembly on Sleep and Respiratory Neurobiology

THIS OFFICIAL RESEARCH STATEMENT OF THE AMERICAN THORACIC SOCIETY WAS APPROVED DECEMBER 2017

Background: Obstructive sleep apnea (OSA) and chronicobstructive pulmonary disease (COPD) are common conditions;the co-occurrence of these diseases, called the overlap syndrome(OVS), has been associated with poor health outcomes.

Purpose: The purpose of this Official American Thoracic SocietyResearch Statement is to describe pathophysiology, epidemiology,outcomes, diagnostic metrics, and treatment of OVS, as well as toidentify important gaps in knowledge and make recommendationsfor future research.

Methods: Clinicians and researchers with expertise in sleepmedicine, pulmonary medicine, or both were invited to participate.Topics were divided among the participants according to theirinterest and expertise. A literature search was conducted; the search

was not a formal systematic review. Evidence was considered andsupplemented with the panelists’ nonsystematic clinicalobservations. Important knowledge gaps were identified.

Results: Recommendations for research to fill existing knowledgegaps were made. The recommendations were formulated bydiscussion and consensus.

Conclusions:Many important questions about OVS exist. ThisAmerican Thoracic Society Research Statement highlights the typesof research that leading clinicians and researchers believe willhave the greatest impact on better understanding the spectrum ofdisease, improving diagnosis, and optimizing therapy.

Keywords: lung; sleep; COPD; hypoxia; apnea

ContentsOverview

Key ConclusionsRecommendations

IntroductionMethodsSignificance of SDB in COPD

PathophysiologyDiagnosisEpidemiologyOutcomes

Appraisal of Existing Metrics

Is the AHI Useful in Patients withCOPD?

Is “Sleep Apnea” theAppropriate Term in Patientswith COPD?

Is REM-related OxygenDesaturation Predictive of PoorOutcomes in Patients withCOPD?

Is Capnometry Valuable inSleep Studies of Patients withCOPD?

Should SupplementalOxygen Be Withdrawnfor OvernightPolysomnography inPatients with SuspectedOVS?

Revisiting the Sleep StudyPolysomnographic Features ofSDB in COPD

SDB in COPDSample Cases

Conclusions

Correspondence and requests for reprints should be addressed to Atul Malhotra, M.D., University of California San Diego, Pulmonary, Critical Care, and SleepMedicine, 9300 Campus Point Drive, #7381, La Jolla, CA 92037. E-mail: [email protected].

This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.

Am J Respir Crit Care Med Vol 197, Iss 3, pp 289–299, Feb 1, 2018

Copyright © 2018 by the American Thoracic Society

DOI: 10.1164/rccm.201712-2510ST

Internet address: www.atsjournals.org

American Thoracic Society Documents 289

mailto:[email protected]

http://www.atsjournals.org/

http://dx.doi.org/10.1164/rccm.201712-2510ST

http://www.atsjournals.org

Overview

The overlap syndrome (OVS) is theco-occurrence of chronic obstructivepulmonary disease (COPD) and obstructivesleep apnea (OSA). This statementsummarizes the existing literature, identifiesknowledge gaps, and provides guidanceregarding future priorities in research forsleep-disordered breathing (SDB) inpatients with COPD and other lung diseases.

Key Conclusionsd OVS is common because COPD and

OSA are common. There is no definitiveevidence that patients with COPD are atincreased risk for developing OSA.

d Mortality associated with OVS is higherthan mortality associated with eitherCOPD alone or OSA alone. Otheroutcomes are likely worse too but havenot been adequately measured.

d The optimal criteria for confirmingOSA in patients with COPD are unclear.The usual criteria for OSA rely oncounting apneas and hypopneas, but thepredictive values of these respiratoryevents are unclear in COPD and,therefore, might lead to misseddiagnoses.

Recommendationsd We recommend investigations that

seek to understand mechanismsunderlying impaired respiratorymechanics during sleep and how thisunderstanding might guide therapyto prevent complications.

d We recommend studies that improve ourunderstanding of gas exchangeimpairment during sleep in COPD.

d We recommend studies that investigatewhether there is a direct correlationbetween the prevalence of OSA andthe severity of COPD.

d We recommend observational studiesthat compare clinical outcomes amongpatients with OVS to clinical outcomesamong patients with OSA alone orCOPD alone.

d We recommend randomized trials thatcompare clinical outcomes amongpatients with OVS whose OSA is treatedto clinical outcomes among patientswith OVS whose OSA is untreated.

d We recommend studies that compare theapnea–hypopnea index with othermeasures of OSA severity (e.g., duration

of hypoxemia, nadir of hypoxemia,and area under the oxygen desaturationcurve) as predictors of clinical outcomesin patients with COPD.

d We recommend that future researchevaluate sleep-disordered breathing(i.e., central sleep apnea, OSA, increasedupper airway resistance, hypoventilation,sleep fragmentation, etc.), rather thanOSA specifically, in patients with COPD.

d We recommend studies that seek todetermine the pathophysiology ofoxygen desaturation during REMsleep in patients with COPD.

d We recommend trials that comparebilevel positive airway pressure to bothcontinuous positive airway pressure andno positive airway pressure therapy inpatients with COPD who desaturateduring REM sleep.

d We recommend trials that comparevarious methods of measuring gasexchange in patients with OVS. Asexamples, trials that compare pulseoximetry plus transcutaneous CO2

measurement to pulse oximetry alone,pulse oximetry plus transcutaneous CO2

measurement to arterial blood gasmeasurement, and pulse oximetry plustranscutaneous CO2 measurement topulse oximetry plus end-tidal CO2

measurement.d We recommend trials that compare

supplemental oxygen to both continuouspositive airway pressure therapy andbilevel positive airway pressure inpatients with COPD who exhibitnocturnal oxygen desaturation.

d We recommend research to developdevices that can identify and characterizesleep without EEG. The goal is toimprove the assessment of the severity ofsleep-disordered breathing in patientswith disruptions in sleep architecture.

d We recommend studies that comparevarious methods to assess and quantifyventilation, the severity of inspiratoryand expiratory flow limitation, and thepresence of hyperinflation duringsleep (e.g., pneumotachography, nasalpressure signaling, and flow sensors) inpatients with COPD.

Introduction

Chronic obstructive pulmonary disease(COPD) is a major source of morbidity,mortality, and healthcare costs. It is

increasing in epidemic proportions becauseof the aging population, indoor andoutdoor air pollution, historical smokingtrends, and other factors (1). Efforts toslow the progression of COPD have failedin the majority of patients, and researchseeks to identify new therapeuticapproaches to improve COPD-relatedoutcomes. Like COPD, obstructive sleepapnea (OSA) has a high prevalenceand is associated with adverse healthconsequences (2–4). Patients with bothOSA and COPD are considered to havethe overlap syndrome (OVS), which isassociated with particularly poor outcomes,including high mortality (5, 6). Thisobservation has led to an increased focuson sleep as an important factor in the careof the patient with COPD.

There exist many important questionsabout sleep-disordered breathing (SDB) inCOPD. Is the pathophysiology of SDBunique/special in COPD? How prevalent isSDB in COPD? Does SDB affect clinicaloutcomes in COPD? Should differentmetrics be used to identify SDB inCOPD? Are there polysomnographiccharacteristics of SDB that are unique inCOPD? How should positive airwaypressure therapy be used to treat SDB inCOPD, given the conflicting evidence(7–10)? How should supplemental oxygenbe used in patients with SDB and COPD,given the uncertain benefits in patientswith COPD and either exertional ornocturnal hypoxemia (11)?

The purpose of this American ThoracicSociety research statement is to summarizeexisting evidence, identify knowledge gaps,and make specific recommendations thatpertain to SDB in patients with COPD.

Methods

This research statement was developedaccording to the rules of the AmericanThoracic Society. Details are provided in theonline supplement.

Significance of SDB in COPD

PathophysiologyCOPD is associated with abnormalities inrespiratory mechanics and gas exchange(12). As a result, sleep is not a period of restfor the respiratory system, but rather aphysiological challenge for patients with

AMERICAN THORACIC SOCIETY DOCUMENTS

290 American Journal of Respiratory and Critical Care Medicine Volume 197 Number 3 | February 1 2018

impaired pulmonary mechanics. Duringsleep, the drive to breathe that is normallyprovided by wakefulness is lost (Figure 1).This change is associated with impairedload compensation, increased airflowresistance, and hypoventilation (13), evenamong healthy persons. Individuals withCOPD have even greater increases inairflow resistance and elastic loading duringsleep because of intrathoracic airwaynarrowing and hyperinflation, respectively(16), similar to obese patients with anarrow upper airway (14, 15). Althoughhealthy humans are able to preserve stablesleep and adequate alveolar ventilationdespite such changes, individuals withabnormal respiratory mechanics may beunable to tolerate the changes and, as aresult, may deteriorate during sleep (16).

Patients with COPD may be unable tocompensate for added loads. The usualcompensatory physiologic responses tosustained resistive loading include elevationof the respiratory rate or a shortening of theexpiratory time. In patients with COPD,however, these responses worsen dynamic

hyperinflation and increase the work ofbreathing. Patients with COPD have aphysiological need to lengthen theirexpiratory time, but this mechanismcompromises ventilation during sleep,particularly in severe COPD (17).

Marked disturbances of the sleep-wakecycle are seen in patients with COPDanalogous to those observed in patientswith primary sleep disorders, such as OSA,restless legs syndrome, periodic limbmovements, and psychiatric disorders (18).A large proportion of the sleep-wakedisturbances exhibited by patients withCOPD are not well described by existingclassification schemes. Sleep disruptionmay be related to impaired lung functionand hyperinflation or, alternatively, tochanges in sleep-related behaviorsassociated with worsening disease(e.g., increased time in bed or daytimenapping). The mechanism for thesedisturbances remains unclear. Alterationsin the sleep-wake cycle are associated withelevations in inflammatory markers, yet thelong-term effects remain unknown (19).

Patients with COPD may manifestsubtle oscillations in ventilation, sleeparchitecture, and oxygenation that do notmeet the 4% detection threshold forhypopnea (20). The physiological effectsof resultant chronic intermittenthypoxia-reoxygenation warrant furtherinvestigation. Whether these cyclicepisodes contribute to adverse outcomesin patients with COPD is yet tobe determined. Moreover, there isconsiderable complexity underlying thebiology of sustained versus intermittenthypoxemia and the two in combination,which requires further study.

RECOMMENDATION: We recommendinvestigations that seek to understandmechanisms underlying impairedrespiratory function during sleep and howthese understandings might guide therapyto prevent complications.

DiagnosisOVS represents the additive or synergisticeffects of COPD and OSA. The diagnosisof OVS is made by full, in-laboratorypolysomnography performed on a patientwith COPD who presents with sleep-relatedcomplaints. The typical polysomnographicfeatures are recurrent episodes of hypopneaalternating with hyperpnea, arousals, and avariable degree of oxygen desaturation.These findings are presumed consequencesof upper airway narrowing or obstruction.Although this pattern is also seen in patientswho have classic sleep apnea/hypopneasyndrome (i.e., whose risk factors includeunfavorable upper airway anatomy and/oran unstable ventilatory control system),in OVS the cycle of instability may beinitiated by triggers such as worseningairflow obstruction, hyperinflation, orhypoxia (17, 21). The presence and severityof upper airway narrowing during thehypopnea phase is unclear, and methodsto identify the underlying pathophysiologyof a hypopnea have not been validated. Asa result, SDB in patients with COPD hasbeen classified under the rubric “OSA,”owing to the limitation of qualitativediagnostic tools to identify the cause thatinitiated the cycle of instability.

For patients with mild to moderateCOPD, the panel has had good clinicalexperience using home sleep testing in theirclinical practices, particularly for patientswithout comorbid insomnia. Althoughthe detection of apnea is reasonablystraightforward, hypopneas, as defined by

Decreased I/EAirflow

↓PaO2↑PaCO2

ChemoreceptorStimulation

Arousal

Expiratory UANarrowing

DiaphragmDysfunction

DynamicHyperinflation

AlveolarHypoventilation

InspiratoryUA Narrowing

VentilatoryOvershoot

Increased work ofBreathing

Decreased VentilatoryMotor Output

Worsening AirflowObstruction

Upper AirwayMuscle Activity↓

CaudalTraction

Sleep + Circadian Factors + Supine Position

Figure 1. Sleep cycle instability in chronic obstructive pulmonary disease. I/E = inspiratory/expiratory;UA = upper airway.



a transient reduction in alveolar ventilationof at least two breaths, are more challengingto assess definitively with home sleeptesting, because detection methods relyon the consequences of decreasedventilation, namely oxygen desaturationand arousals. However, it is unclear if themagnitude of oxygen desaturation may bedampened by hyperinflation and/or thepresence of carboxyhemoglobin (byshifting the oxyhemoglobin dissociationcurve to the left) (22, 23).

RECOMMENDATION: Werecommend studies that improve ourunderstanding of gas exchangeimpairment during sleep in COPD. Suchstudies could help guide definitions ofvarious respiratory events as well ascriteria to judge severity of disease inpatients with COPD and OSA.

EpidemiologyThe epidemiology of SDB in patients withrespiratory disorders is not well established,with inconsistent reports in the literature.The inconsistent reports may be due, inpart, to changes in diagnostic criteria andimproved testing technology during the past20 years. For example, the diagnosis of OSAcan now be made based on hypoxemia,which may exaggerate the association of thetwo disorders, because patients with COPDreadily desaturate in the presence of mildflow limitation.

Both OSA and COPD are commondisorders. More than 5% of the UnitedStates population (i.e., at least 13.7 millionpeople) and about 5–10% of the worldpopulation is burdened by COPD (2, 3),which is a leading cause of morbidity andmortality. Its high prevalence is attributableto high rates of smoking and environmentalpollution (41). Among middle-aged(i.e., 50–70 yr) men, the prevalence ofmoderate to severe OSA is roughly 17%,whereas among middle-aged women, theprevalence is approximately 9% (24). Theseprevalences exist throughout most of thedeveloped world (25).

The most recent epidemiological studieslooking at patients with both OSA andCOPD have estimated that OSA and COPDcoincide at rates predicted by theepidemiology of each disorder alone (26, 27).About 1 in 10 patients with one disorder willhave the other disorder by chance alone;there is no clear evidence that having onedisorder predisposes someone to acquiringthe other, despite many theories of potential

pathophysiological links. An importantlimitation of these studies, however, is thatmost have restricted assessment to cohortswith only mild COPD. Therefore, it isunknown whether more severe COPDpredisposes to OSA.

The prevalence of SDB in COPD maybe higher than typically reported if allseverities of COPD are considered, assuggested by a study that demonstrated a66% prevalence of OSA among patients withmoderate to severe COPD (2). Possiblepathophysiological explanations for ahigher prevalence of OSA among patientswith moderate to severe COPD includeupper airway narrowing due to edema fluidshifting from the lower extremities to theneck in those with cor pulmonale andupper airway myopathy due to eitherCOPD itself (which can cause a skeletalmuscle myopathy) or inhaled or systemicsteroid use (28). Arguing against arelationship between COPD severity andOSA is that dynamic hyperinflation mayreduce upper airway collapsibility due totracheal traction (17, 29).

RECOMMENDATION: Werecommend studies that investigatewhether there is a direct correlationbetween the prevalence of OSA and theseverity of COPD.

Outcomes

Mortality. Patients with OVS have beenlong recognized as having a unique andsevere phenotype; however, survival datahave only recently become available. Onestudy found that individuals with COPDplus untreated OSA have decreased survivalcompared with individuals with COPDalone or COPD plus OSA treated withcontinuous positive airway pressure(CPAP) (5). These findings have beenconfirmed in other cohorts (30) and makeintuitive sense; because OSA and COPDare each known to increase cardiovascularrisk (31–33), one would expect theabsence of one (i.e., COPD alone) or thetreatment of one (i.e., COPD plus treatedOSA) to be associated with better survival.

There are several reasons that thesurvival estimates may be incorrect, however.First, the data do not derive from randomizedtrials and, therefore, it is possible that theyare due to unmeasured confounders. Asexamples, individuals who used CPAP inthe study (2) differed from those who didnot in terms of their financial means and

willingness to pursue CPAP therapy. Thepossibility that such factors may beresponsible for some of the differencescannot be excluded (34). Second, patientswith OVS tend to die from cardiovascularevents that are attributed to cardiac disease,with little acknowledgment that underlyingOSA or COPD may have contributed to thecardiovascular event (5). Thus, the impact ofOVS, COPD, and OSA on survival may beunderestimated. The precise mechanisms ofincreased cardiovascular mortality in OVShave proven elusive.

Few studies have been performed thatevaluated the impact of treatment onsurvival. One observational study of morethan 1,000 patients with COPD reportedthat the duration of CPAP use directlycorrelated with improved survival (30).Such observational data justify randomizedtrials to evaluate the treatment of OSA inpatients with COPD.

Myocardial abnormalities. Intermittenthypoxemia due to OSA with subsequentpulmonary vasoconstriction confers severalundesirable consequences in patients withCOPD. The increased right ventricularload created by hypoxemia-inducedvasoconstriction can result in hypertrophy,which is capable of doubling the rightventricular mass, as demonstrated by a studythat compared patients with OVS to patientswith severity-matched COPD alone (35).Myocardial fibrosis can also be induced byhypoxemia. A cardiac magnetic resonanceimaging study used a novel, gadolinium-enhancement technique to measure theextracellular matrix of the ventricle andfound increased myocardial fibrosis thatcorrelated with worsening hypoxemia (36).Studies looking at the impact of treatmentof OSA on right ventricular mass andmyocardial fibrosis have not been performedin patients with COPD.

Blood gases. Patients with OVS tendto have worse blood gases than thosewith either COPD alone or OSA alone,as illustrated by the following clinicalobservations of the panel. Patients with OVSdevelop daytime hypercapnia at a relativelypreserved lung function, whereas patientswith COPD alone develop daytimehypercapnia once lung function becomesseverely compromised. Patients with OVSalso tend to develop daytime hypercapniaat a lower body mass index andapnea–hypopnea index (AHI) than patientswith OSA alone. Hypoxemia is moreprolonged and profound in patients with



OVS than in those with COPD alone orOSA alone and, perhaps as a result, patientswith OVS more frequently have pulmonaryhypertension than those with OSA alone(35, 37, 38).

Inflammation and metabolism.Patients with OVS have increased chronicinflammation and, possibly, impairedmetabolism. In patients with OSA, sleepfragmentation and chronic intermittenthypoxia facilitate oxidative injury andchronic inflammation. In patients withCOPD, impaired gas exchange similarlyfacilitates oxidative injury and chronicinflammation, which may contribute toextrapulmonary complications of COPD(i.e., loss of skeletal muscle mass, decreasedbone mineral density, atherosclerosis, andinsulin resistance). The pathways mediatingoxidative injury and inflammation areprobably similar in COPD and OSA,because the profiles of elevated markers ofthese processes are similar among patientswith these conditions.

Data from the Sleep Heart Health Study(39) demonstrated that sleep-relatedrespiratory events accompanied byoxyhemoglobin oxygen desaturation greaterthan or equal to 2% are associated withfasting hyperglycemia. This suggests that

mild degrees of chronic intermittenthypoxia may predispose individuals toadverse metabolic outcomes. Such mildintermittent hypoxia is common in patientswith OVS and can be undetected inconventional polysomnographic analyses.

RECOMMENDATIONS:1. We recommend observational studies

that compare clinical outcomes amongpatients with OVS to clinical outcomesamong patients with OSA alone orCOPD alone.

2. We recommend randomized trials thatcompare clinical outcomes amongpatients with OVS whose OSA is treatedto clinical outcomes among patientswith OVS whose OSA is untreated.

3. Clinical outcomes in the precedingrecommendations include mortality,cardiovascular-specific mortality,nonfatal cardiovascular events, qualityof life, and excessive daytime sleepiness.

Appraisal of Existing Metrics

Is the AHI Useful in Patients withCOPD?The AHI—the sum of apneas (cessations inbreathing) plus hypopneas (reductions in

breathing) —was designed to assess theseverity of sleep apnea, generally in obesemen without underlying lung disease(40–42). However, the AHI has manylimitations, some of which are particularlyrelevant to patients with COPD and mayaffect the detection of OSA.

The definition of hypopnea has variedover time and across organizations, butmost definitions include oxyhemoglobinoxygen desaturation as a criterion. Therehave been calls for reductions in airflowwithelectrocortical arousal to be consideredhypopneas even without major oxygendesaturation. However, EEG is usuallynot performed during home sleep studiesand, therefore, the presence or absence ofoxygen desaturation remains a key criterionfor evaluating hypopneas, as recommendedby the United States’ major healthcarepayer, Medicare (43). Existing definitionsof hypopnea have important flaws. Oneproblem is that the degree and durationof oxygen desaturation are not considered.Thus, an individual with 20 minutes ofsustained hypoxemia to the 70–80% rangedue to hypoventilation would only becounted as having had a single hypopnea(Figure 2). A second problem is thatrecurrent arousals for any reason (i.e., poor

–100

100

0

–100

100

0

–100

100–0.50

–0.25

0.00

0.25

0.10

0.05

0.00

100

0

–100

100

100

95

90

0

–100

–0.05

0

–250

250

0

18630 18640 18650 18660 18670 18680 18690 18700 18710 18720 18730 18740 18750 18760 18770 18780 18790 18800 18810 18820 18830 18840 18850 18860 18870 18880

s

4

3

5

2

6

11

13

12

8

SpO

2%

Tho

ruV

Abd

ouV

EC

Gm

VE

EG

2uV

EM

GuV

EO

G-L

uVE

OG

-RuV

Flo

wcm

H2O

Figure 2. Prolonged respiratory events. A representative polysomnograph recording in a patient with sleep-disordered breathing and obstructive lungdisease demonstrates how sustained desaturation due to prolonged hypoventilation would be counted as a single respiratory event.



sleep quality, hyperinflation, andnocturnal cough) may result in an irregularbreathing pattern that could be interpretedas recurrent hypopneas. Conversely,frequent arousals due to repetitiverespiratory events may not exhibit oxygendesaturations and, therefore, would not bescored as hypopneas. Such patients maybenefit from therapy, even though they maynot meet traditional criteria on the basis ofAHI. Finally, smokers may haveconsiderable carbon monoxide circulatingin their blood, which can shift theoxyhemoglobin curve to the left, leading tohigher saturations for any given PaO2

(22),potentially masking clinically importanthypoxemia. Each of these limitations mayaffect the detection of OSA among patientswith COPD.

The issue of sustained oxygendesaturation leading to underestimationof the AHI (and, therefore, underestimationof the severity of the condition) isparticularly pertinent to patients withCOPD, because one of the major findings onpolysomnography in patients with COPD isnocturnal oxygen desaturation withsustained hypoxemia, without frequentapneas and hypopneas. There are avariety of reasons for the oxygendesaturation other than upper airwayobstruction, including a drop in ventilationthat occurs during sleep (especially REMsleep) and ventilation–perfusionmismatch that worsens during sleep. Themagnitude of oxygen desaturation may beparticularly severe in patients with COPDbecause of the sigmoidal shape of theoxyhemoglobin oxygen desaturation curve.Those with mild baseline hypoxemia(e.g., oxygen saturation of 90–93%) areuniquely positioned near the steep portionof the oxyhemoglobin dissociation curve,where small changes in PaO2

lead to largechanges in oxygen saturation. Thus, evenwith mild perturbations, sustained andintermittent oxygen desaturationscommonly occur.

Taking these observations together, it isclear that the quantification of hypopneasand apneas alone may be insufficient toidentify and characterize OSA in patientswith COPD. As an example, a patient withpartial upper airway obstruction(hypopneas) plus severe COPD and apatient with complete upper airwayobstruction (apneas) plus mild COPD willboth be diagnosed as having OVS. However,the different pathogenesis of OSA in the

patients suggests that different treatmentstrategies be deployed.

It is unknown whether the AHIpredicts clinical outcomes in patients withCOPD, including symptoms, cardiovascularconsequences, and metabolic complications.In theory, the area under the oxygendesaturation curve or the total hypoxemicload may have better predictive value thanthe AHI in patients with COPD, as may thenadir oxygen saturation, but this has notbeen evaluated.

RECOMMENDATION: We recommendstudies that compare the AHI with othermeasures of obstructive sleep apnea severity(e.g., duration of hypoxemia, nadir ofhypoxemia, arterial tonometry, and areaunder the oxygen desaturation curve) aspredictors of clinical outcomes in patientswith COPD. Various definitions of severityof OVS could be assessed for their predictivevalue using hard clinical outcomes.

Is “Sleep Apnea” the AppropriateTerm in Patients with COPD?A large proportion of patients with COPDexperience some deterioration in gasexchange and respiratory mechanicsduring sleep. Non-REM sleep ischaracterized by increased upper airwayresistance, impaired load compensation, andblunted chemosensitivity, all of which maydisturb gas exchange. REM sleep mayfurther compromise respiratory functionbecause of a propensity for upper airwaycollapse and skeletal muscle atonia inaccessory muscles of respiration(e.g., rib cage intercostal muscles) (44).

These sleep-related changes probablyall contribute to the observation that a largeproportion of patients with COPDdesaturate during sleep, even those who arenot hypoxemic during wakefulness. Thiswas illustrated by a study that found that38% of patients with moderate to severeCOPD who do not qualify for home oxygentherapy on the basis of their daytime PaO2

have nocturnal oxygen desaturationwithout evidence of sleep apnea. Therefore,the assessment of gas exchange during sleepis critical in patients with COPD (45).

SDB and OSA are sometimes conflated.However, we believe that SDB betterdescribes the pathophysiology that occurs inpatients with COPD. SDB encompasses allrespiratory dysfunction during sleep,including central and obstructive sleepapnea, worsening airflow obstruction,hypoventilation, oxygen desaturation, sleep

fragmentation, and other abnormalitiesthat are not necessarily implied bytraditional terminology.

RECOMMENDATION: We recommend thatfuture research evaluate SDB (i.e., centralsleep apnea, OSA, increased upper airwayresistance, hypoventilation, sleepfragmentation, etc.), rather than OSAspecifically, in patients with COPD.Mechanistic research could help identifyendotypes amenable to targetedintervention rather than simply describingdifferent airflow patterns.

Is REM-related Oxygen DesaturationPredictive of Poor Outcomes inPatients with COPD?REM sleep is defined by skeletal muscleatonia that affects most muscles of the body,with some notable exceptions, including thediaphragm (46). It is characterized byvariability in the EEG and otherphysiological signals. Phasic eyemovements during REM sleep have beenassociated with suppressed upper airwaymotor tone and reductions in ventilation.Patients with sleep apnea often have morerespiratory events during REM sleep.Therefore, oxygen desaturation tends toworsen during REM in patients with OVS.

Controversy exists regarding theclinical importance of respiratory eventsand gas exchange alterations during REMsleep (47), with most believing that theimportance likely depends on theunderlying cause of the hypoxemia.Fluctuations in tidal volume during REMsleep are expected, and hypoxemia due tosuch fluctuations is unlikely to be indicativeof major pathology or require therapy (48).This is supported by a study that showedno major symptoms associated withrespiratory events occurring during REMsleep (49). However, occasional patientsexhibit profound oxygen desaturationsduring REM sleep that are likely a result ofatonia in the accessory muscles ofrespiration (48). Such derangements in gasexchange are unlikely to be physiologic and,thus, provide rationale for therapy. Thisview is supported by a study that foundimportant consequences, such as arterialhypertension, associated with REM-relatedrespiratory events (47).

In patients with COPD, data duringREM sleep remain sparse, but the panel’sclinical experience suggests that somepatients with severe COPD will experiencemajor oxygen desaturations during REM



sleep, likely as a result of hypoventilation.Such patients may benefit from bilevelpositive airway pressure therapy, althoughoutcome data are lacking. Similarly, volume-assured devices are becoming available buthave been minimally studied in this context.

RECOMMENDATIONS:1. We recommend studies that seek to

determine the pathophysiology ofoxygen desaturation during REM sleepin patients with COPD.

2. We recommend trials that comparebilevel positive airway pressure to bothCPAP and no positive airway pressuretherapy in patients with COPD whodesaturate during REM sleep.

Is Capnometry Valuable in SleepStudies of Patients with COPD?Monitoring gas exchange is critical to athorough evaluation of sleep and breathing.Traditionally, oxygen saturation has beenused as a surrogate for gas exchange underthe assumption that most importantdisturbances would lead to oxygendesaturation, and because of high cost andpoor availability of reliable transcutaneoussystems. However, oxygen saturation is animperfect measure, as an occurrence ofoxygen desaturation could in theory berelated to multiple mechanisms, including

hypoventilation, upper airway collapse,ventilation–perfusion mismatch, shunting,and other factors.

The addition of capnometry could behelpful for several reasons. First,capnometry plus pulse oximetry is moresensitive than pulse oximetry alone. Becauseof the shape of the oxyhemoglobindissociation curve, importanthypoventilation may occur with only minorchanges in SaO2

. For example, a 10–mm Hgincrease in PaCO2

may drop PaO2from

80 mm Hg to roughly 70 mm Hg, whichwould not lead to any major change in thesaturation. Moreover, patients who arereceiving supplemental oxygen mayexperience marked hypoventilation thatcould be overlooked because of lack ofoxygen desaturation. Capnometry has thepotential to detect hypoventilation thatpulse oximetry may miss in these situations.Second, pulse oximetry does not provideinformation about the mechanismunderlying oxygen desaturation, butcapnometry may in some circumstances.For example, a decrease in oxygensaturation accompanied by an increase inPaCO2

is suggestive of hypoventilation as theunderlying mechanism rather thanventilation–perfusion mismatch or otherfactors. Mechanistic information mighthelp to guide the choice of therapy, such as

whether bilevel positive airway pressure orCPAP therapy is more appropriate.

In our clinical practices, we prefertranscutaneous CO2 monitoring over end-tidal CO2 in patients with COPD,particularly if supplemental oxygen is beingprovided. In patients with unstable breathingpatterns, particularly if tidal volume is low,end-tidal PCO2 can be unreliable because ofthe absence of an alveolar plateau.Transcutaneous CO2 is particularly helpfulsemiquantitatively, because it is directionallyaccurate regarding disturbances in gasexchange. Depending on the device beingused, however, transcutaneous CO2 changescan be delayed because of slow timeconstants, making abrupt dynamic changesdifficult to assess. Other methods ofmeasuring gas exchange, such as arterialblood gases, have been scarcely studied.

RECOMMENDATION: We recommendtrials that compare various methods ofmeasuring gas exchange in patients withOVS, for example, trials that comparepulse oximetry plus transcutaneous CO2

measurement to pulse oximetryalone, pulse oximetry plus transcutaneousCO2 measurement to arterial blood gasmeasurement, and pulse oximetry plustranscutaneous CO2 measurement topulse oximetry plus end-tidal CO2

measurement.

50

0

–50

50

0

0

–25

–50

50

0

–50

50

0

–50

50

0

–50

50

0

0.2

0.0

0.5

0.0

95

90

85

.672 21674 21676 21678 21680 21682 21684 21686 21688 21690 21692 21694 21696 21698 21700 21

–0.5

–50

SaO

2%7

Flo

wL/

sE

KG

uVO

2-A

1uV

C4-

A1

uVF

3-A

2uV

RO

CuV

LOC

uVC

HIN

uVC

3-A

2uV

1

Figure 3. Alpha intrusion. A representative polysomnograph recording in a patient with sleep-disordered breathing and obstructive lung diseasedemonstrates an increase in alpha intrusion and illustrates how it can be difficult to detect hypopneas terminating in arousals from sleep.



Should Supplemental Oxygen BeWithdrawn for OvernightPolysomnography in Patients withSuspected OVS?Many patients with COPD are givensupplemental oxygen on a long-term basisbecause of demonstrable abnormal gasexchange or for comfort. However,supplemental oxygen may mask hypopneasduring overnight polysomnography byblunting oxygen desaturation due tohypoventilation. For this reason, we routinelyremove supplemental oxygen beforepolysomnography, unless the oxygen saturationduring relaxed wakefulness is less than 88%.

It is important that oxygen desaturationbe detected and the response to positiveairway pressure titration be assessed duringpolysomnography, because positive airwaypressure alone may eliminate oxygendesaturation, eliminating the need fornocturnal supplemental oxygen with itsexpense and risks. Supplemental oxygen aloneis generally considered ineffective for thetreatment of OSA and canworsen hypercapnia(11, 50, 51), although the patients enrolled in

these studies differed from our patients ofinterest (i.e., rather than COPD withconfirmed nocturnal oxygen desaturation, thestudies enrolled patients with COPD withexercise-induced oxygen desaturation orpatients with OSA with or without COPD).

RECOMMENDATION: We recommendtrials that compare supplemental oxygento both CPAP therapy and bilevel positiveairway pressure in patients with COPDwho exhibit nocturnal oxygendesaturation.

Revisiting the Sleep Study

Polysomnographic Features of SDB inCOPDPolysomnography in patients with COPD isadequate to categorize sleep apnea andsleep-related respiratory events (e.g., centralversus obstructive sleep apnea, hypopneasversus apneas), as well as to quantify theseverity of sleep apnea (e.g., number ofevents per hour, duration of events, degreeof hypoxia with each event, and percentage

of events with and without arousal). Thisapproach allows the detection of SDB inmost cases. However, the severity of theSDB may be underestimated bypolysomnography in patients with COPD,particularly those who require supplementaloxygen or who have disturbances in sleeparchitecture due to depression or insomnia(both increase alpha intrusion and make itdifficult to detect hypopneas terminating inarousals from sleep) (Figure 3).

Polysomnography is not intended todetect COPD-specific respiratorydisturbances, such as expiratory snoringwith associated flow limitation (52, 53)(Figure 4) or dynamic hyperinflation (seeFigures E2–E4 in the online supplement).In one study, expiratory snoring had11-times higher odds of being associatedwith airflow obstruction (FEV1/FVC, 70%)than the absence of expiratory snoring.Additional measurements that quantifyventilation and/or respiratory effort arerequired to detect SDB in COPD. Forexample, a pneumotachograph affixed to amask would best serve to quantify

EOG

Chin

109, 2’ 110, 2’

EEG

Right legLeft leg

P-Snoring

Snoring

P-Flow

Rib cage

ECG

O2 saturation(%)

94 94 94 94 94 94 94 94 94 94 94 94 94 94 94 94 9494 94 94 94 94 94 94 9494 94 94 94 94 94 94 9494 94 94 94 94 94 949493 93 93 93 93 9393

Inspiratory

Expiratory

10 Seconds

Figure 4. Expiratory flow limitation and expiratory snoring in chronic obstructive pulmonary disease. A representative polysomnograph recording duringnon-REM sleep in a patient with sleep-disordered breathing and obstructive lung disease. Arrows point to the expiratory flow limitation and expiratorysnoring. Chin = chin electromyogram; EOG= electrooculogram; P-Flow = pressure flow; P-Snoring = snoring tracing obtained by pressure flow sensor;snoring = snoring tracing obtained by chin microphone (53).



ventilation, the severity of inspiratory andexpiratory flow limitation, and the presenceof hyperinflation during sleep. However,conventional pneumotachographs are oflimited use in patients with COPD duringsleep, because they are too cumbersomeand impose excessive dead space andresistive loads on the respiratory system(54). Alternative flow surrogates include anunfiltered nasal pressure signal at a highsampling rate (preferably .200 Hz) orflow sensors with negligible resistanceand dead space, both of which can bereadily integrated into conventionalpolysomnographic units. Similarly,esophageal catheters would be best suited toquantify respiratory loads but are rarelyused in clinical sleep studies. Inductiveplethysmography or intercostal EMGactivity might be good surrogates fordetecting changes in inspiratory orexpiratory effort.

RECOMMENDATIONS:1. We recommend research to develop

and/or validate devices that canidentify and characterize sleep withoutEEG. The goal is to improve theassessment of the severity of SDB inpatients with disruptions in sleeparchitecture.

2. We recommend studies that comparevarious methods to quantifyventilation, the severity of inspiratoryand expiratory flow limitation, andthe presence of hyperinflation duringsleep (e.g., pneumotachograph, nasalpressure signaling, and flow sensors) inpatients with COPD.

SDB in COPDCOPD is associated with alterations inrespiratory mechanics and gas exchange,leading to dynamic hyperinflation and/orhypoventilation (55). During wakefulness,patients compensate for expiratory flowobstruction by adopting specificcompensatory mechanisms to minimizehyperinflation (e.g., by prolongingexpiratory time or pursed lip breathing)and preserve alveolar ventilation. Sleep isassociated with changes in respiratorycontrol that interfere with thesecompensatory mechanisms (Figure 5) andlead to a cascade of events (Figure 1). Mostnotably, loss of the ventilatory drive leads tosignificant reductions in tidal volume. Anincrease in inspiratory time wouldseemingly be an effective compensatorymechanism to mitigate the reduction intidal volume. However, increased

inspiratory time reduces expiratory time,which may lead to dynamic hyperinflationin patients with COPD. Similarly, anincrease in the respiratory rate wouldseemingly be another compensatorymechanism, but this also reduces expiratorytime. In either case, respiratory patternsduring sleep in COPD may worsendynamic hyperinflation by shortening theexpiratory time. In addition, an inspiratoryflow limitation is commonly observedduring sleep, and even mild degrees ofinspiratory flow limitation lengtheninspiration and further compromiseexpiratory time. Thus, normal sleep-induced alterations in breathing pattern canbe detrimental in patients with COPD.

Sample CasesWe use a series of cases to illustrate thespectrum of novel COPD-specific events forwhich both pneumotachograph andesophageal pressure signals are required todetect SDB events. Although these methodshave not been widely adopted for diagnosticpurposes in most sleep centers, these signalscan illuminate breathing disturbancesduring sleep that can be easily detected bycurrently used surrogate signals. The casesare provided in the online supplement.

Pathophysiology

PolysomnographicFeatures in COPD

Clinical Outcomes

Daytime symptoms• Morning fatigue• Daytime sleepiness• Decline in executive functions• Depression• Chronic fatigue syndrome• PTSD/pain management problems

Cardiopulmonary findings• Endothelial dysfunction• Arterial and pulmonary hypertension• Hypercapnia and hypoxia• Increase in mortality

• Atypical obstructive apnea/hypopnea• Expiratory flow limitation• Inspiratory flow limitation with dynamic hyperinflation• REM-related hypoxia/hypercapnia• Tachypnea (rapid shallow breathing)• Sustained hypoxia and hypercapnia (CO2 retention)

Sleep Disturbances• Disturbance in sleep architecture• Frequent arousals• Sympathetic/parasympathetic imbalance• Alpha intrusion

Abnormal BreathingPatterns

SleepDisturbances

Wakefulness

Sleep

Compensatory MechanismsRespiratory loads

Impaired CompensatoryMechanisms

Figure 5. Sleep-disordered breathing in chronic obstructive pulmonary disease (COPD). PTSD = post-traumatic stress disorder.



Conclusions

Polysomnographic features in COPD are aconsequence of sleep-induced respiratoryloads. Patients with COPDmay respond toincreased respiratory loads with markedsleep disturbances or, alternatively,they may have minimal sleep disturbancebut a highly abnormal breathing pattern

with or without severe alterations in gasexchange. The same principles describedin this statement for COPDmay also applyto other respiratory diseases characterizedby lower airway obstruction, such asasthma and cystic fibrosis. Depending onwhich polysomnographic featurespredominate, patients may present with avariety of daytime symptoms and clinical

outcomes, ranging from insomnia tohypersomnia with and withoutcardiopulmonary complications. Westrongly advocate for further research intoSDB in patients with COPD, tounderstand better the spectrum of disease,optimize the definitions and equipmentused to assess patients, and determine theoptimal mode of therapy. n

This official research statement was prepared by an ad hoc subcommittee of the ATS Assembly on Sleep and Respiratory Neurobiology.

Members of the subcommittee are as follows:ATUL MALHOTRA, M.D. (Chair)M. SAFWAN BADR, M.D.PAMELA DEYOUNG, R.P.S.C.-T.MEILAN K. HAN, M.D.NADIA N. HANSEL, M.D., M.P.H.ROBERT L. OWENS, M.D.HARTMUT SCHNEIDER, M.D.ALAN R. SCHWARTZ, M.D.JADWIGA A. WEDZICHA, M.D.KEVIN C. WILSON, M.D.MICHELLE R. ZEIDLER, M.D.

Author Disclosures: A.M. practices atUniversity of California San Diego, which

received a philanthropic donation fromResMed in support of their sleep center.M.K.H. served as a consultant forAstraZeneca, Boehringer Ingelheim,GlaxoSmithKline, Novartis, and Sunovion; andreceived research support and served as aspeaker for Novartis. N.N.H. received researchsupport and served on an advisory committeefor AstraZeneca and GlaxoSmithKline; andreceived research support from BoehringerIngelheim. R.L.O. received honorarium andtravel reimbursement from ResMed andItamar; and served on a scientific advisoryboard for Novartis. H.S. holds a patent (U.S.200800922898) issued to RespEQ for adisposable sleep and breathing monitor.A.R.S. served as a consultant for inSleepHealth and Respicardia; received research

support from ImThera Medical, inSleepHealth, and Maxis; and holds stock, stockoptions, or other ownership interests inDiscover Medical and RespEQ. J.A.W.served as a speaker and on an advisorycommittee for AstraZeneca, BoehringerIngelheim, GlaxoSmithKline, and Novartis;received research support fromGlaxoSmithKline, Johnson and Johnson,Novartis, OM Pharma, Takeda, and ViforPharma; served on an advisory committeefor Johnson and Johnson; served as aspeaker for Chiesi; and had meetingexpenses covered by AstraZeneca, BoehringerIngelheim, GlaxoSmithKline, Novartis, Pfizer,and Teva. M.S.B., P.D., K.C.W., and M.R.Z.reported no relationships with relevantcommercial interests.

References

1. Rice MB, Malhotra A. The air we breathe and lung disease. J Thorac Dis2015;7:E245–E247.

2. Soler X, Gaio E, Powell FL, Ramsdell JW, Loredo JS, Malhotra A, et al.High prevalence of obstructive sleep apnea in patients with moderateto severe chronic obstructive pulmonary disease. Ann Am Thorac Soc2015;12:1219–1225.

3. Kohli P, Sarmiento K, Malhotra A. Update in sleep medicine 2012. Am JRespir Crit Care Med 2013;187:1056–1060.

4. Ayas NT, Owens RL, Kheirandish-Gozal L. Update in sleep medicine2014. Am J Respir Crit Care Med 2015;192:415–420.

5. Marin JM, Soriano JB, Carrizo SJ, Boldova A, Celli BR. Outcomes inpatients with chronic obstructive pulmonary disease and obstructivesleep apnea: the overlap syndrome. Am J Respir Crit Care Med 2010;182:325–331.

6. Jen R, Li Y, Owens RL, Malhotra A. Sleep in chronic obstructivepulmonary disease: evidence gaps and challenges. Can Respir J2016;2016:7947198.

7. McEvoy RD, Pierce RJ, Hillman D, Esterman A, Ellis EE, Catcheside PG,et al.; Australian trial of non-invasive Ventilation in Chronic AirflowLimitation (AVCAL) Study Group. Nocturnal non-invasive nasalventilation in stable hypercapnic COPD: a randomised controlled trial.Thorax 2009;64:561–566.

8. Meecham Jones DJ, Paul EA, Jones PW, Wedzicha JA. Nasal pressuresupport ventilation plus oxygen compared with oxygen therapy alone inhypercapnic COPD. Am J Respir Crit Care Med 1995;152:538–544.

9. Gay PC, Hubmayr RD, Stroetz RW. Efficacy of nocturnal nasalventilation in stable, severe chronic obstructive pulmonary diseaseduring a 3-month controlled trial. Mayo Clin Proc 1996;71:533–542.

10. Struik FM, Sprooten RT, Kerstjens HA, Bladder G, Zijnen M, Asin J,et al. Nocturnal non-invasive ventilation in COPD patients withprolonged hypercapnia after ventilatory support for acute respiratoryfailure: a randomised, controlled, parallel-group study. Thorax 2014;69:826–834.

11. Albert RK, Au DH, Blackford AL, Casaburi R, Cooper JA Jr, Criner GJ,et al.; Long-Term Oxygen Treatment Trial Research Group. Arandomized trial of long-term oxygen for COPD with moderatedesaturation. N Engl J Med 2016;375:1617–1627.

12. Loring SH, Garcia-Jacques M, Malhotra A. Pulmonary characteristics inCOPD and mechanisms of increased work of breathing. J ApplPhysiol (1985) 2009;107:309–314.

13. Orem J. The nature of the wakefulness stimulus for breathing. Prog ClinBiol Res 1990;345:23–30. [Discussion p. 31.]

14. Smith PL, Gold AR, Meyers DA, Haponik EF, Bleecker ER. Weight lossin mildly to moderately obese patients with obstructive sleep apnea.Ann Intern Med 1985;103:850–855.

15. Sands SA, Eckert DJ, Jordan AS, Edwards BA, Owens RL, Butler JP,et al. Enhanced upper-airway muscle responsiveness is a distinctfeature of overweight/obese individuals without sleep apnea. Am JRespir Crit Care Med 2014;190:930–937.

16. Skatrud JB, Dempsey JA, Badr S, Begle RL. Effect of airwayimpedance on CO2 retention and respiratory muscle activity duringNREM sleep. J Appl Physiol (1985) 1988;65:1676–1685.

17. Biselli P, Grossman PR, Kirkness JP, Patil SP, Smith PL, Schwartz AR,et al. The effect of increased lung volume in chronic obstructivepulmonary disease on upper airway obstruction during sleep. J ApplPhysiol (1985) 2015;119:266–271.

18. Soler X, Diaz-Piedra C, Ries AL. Pulmonary rehabilitationimproves sleep quality in chronic lung disease. COPD 2013;10:156–163.

19. Kohli P, Pinto-Plata V, Divo M, Malhotra A, Harris RS, Lazaar A, et al.Functional capacity, health status, and inflammatory biomarkerprofile in a cohort of patients with chronic obstructive pulmonarydisease. J Cardiopulm Rehabil Prev 2015;35:348–355.

20. Ruehland WR, Rochford PD, O’Donoghue FJ, Pierce RJ, Singh P,Thornton AT. The new AASM criteria for scoring hypopneas: impacton the apnea hypopnea index. Sleep 2009;32:150–157.

21. Gleeson K, Zwillich CW, White DP. The influence of increasingventilatory effort on arousal from sleep. Am Rev Respir Dis 1990;142:295–300.



22. Kambam JR, Chen LH, Hyman SA. Effect of short-term smoking halt oncarboxyhemoglobin levels and P50 values. Anesth Analg 1986;65:1186–1188.

23. Chowdhuri S, Quan SF, Almeida F, Ayappa I, Batool-Anwar S,Budhiraja R, et al.; ATS Ad Hoc Committee on Mild ObstructiveSleep Apnea. An Official American Thoracic society researchstatement: impact of mild obstructive sleep apnea in adults. Am JRespir Crit Care Med 2016;193:e37–e54.

24. Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM.Increased prevalence of sleep-disordered breathing in adults. Am JEpidemiol 2013;177:1006–1014.

25. Heinzer R, Vat S, Marques-Vidal P, Marti-Soler H, Andries D,Tobback N, et al. Prevalence of sleep-disordered breathing in thegeneral population: the HypnoLaus study. Lancet Respir Med 2015;3:310–318.

26. Sharma B, Feinsilver S, Owens RL, Malhotra A, McSharry D, Karbowitz S.Obstructive airway disease and obstructive sleep apnea: effect ofpulmonary function. Hai 2011;189:37–41.

27. Budhiraja R, Siddiqi TA, Quan SF. Sleep disorders in chronic obstructivepulmonary disease: etiology, impact, and management. J ClinSleep Med 2015;11:259–270.

28. Teodorescu M, Xie A, Sorkness CA, Robbins J, Reeder S, Gong Y, et al.Effects of inhaled fluticasone on upper airway during sleep andwakefulness in asthma: a pilot study. J Clin Sleep Med 2014;10:183–193.

29. Begle RL, Badr S, Skatrud JB, Dempsey JA. Effect of lung inflation onpulmonary resistance during NREM sleep. Am Rev Respir Dis 1990;141:854–860.

30. Stanchina ML, Welicky LM, Donat W, Lee D, Corrao W, Malhotra A.Impact of CPAP use and age on mortality in patients with combinedCOPD and obstructive sleep apnea: the overlap syndrome. J ClinSleep Med 2013;9:767–772.

31. Chowdhuri S, Crook ED, Taylor HA Jr, Badr MS. Cardiovascularcomplications of respiratory diseases. Am J Med Sci 2007;334:361–380.

32. Seemungal TA, Hurst JR, Wedzicha JA. Exacerbation rate, healthstatus and mortality in COPD–a review of potential interventions.Int J Chron Obstruct Pulmon Dis 2009;4:203–223.

33. Sin DD, Man SF. Why are patients with chronic obstructive pulmonarydisease at increased risk of cardiovascular diseases? The potentialrole of systemic inflammation in chronic obstructive pulmonarydisease. Circulation 2003;107:1514–1519.

34. Platt AB, Kuna ST, Field SH, Chen Z, Gupta R, Roche DF, et al.Adherence to sleep apnea therapy and use of lipid-lowering drugs: astudy of the healthy-user effect. Chest 2010;137:102–108.

35. Sharma B, Neilan TG, Kwong RY, Mandry D, Owens RL, McSharry D,et al. Evaluation of right ventricular remodeling using cardiacmagnetic resonance imaging in co-existent chronic obstructivepulmonary disease and obstructive sleep apnea. COPD 2013;10:4–10.

36. Neilan TG, Bakker JP, Sharma B, Owens RL, Farhad H, Shah RV, et al.T1 measurements for detection of expansion of the myocardialextracellular volume in chronic obstructive pulmonary disease. Can JCardiol 2014;30:1668–1675.

37. Resta O, Foschino BarbaroMP, Brindicci C, Nocerino MC, Caratozzolo G,Carbonara M. Hypercapnia in overlap syndrome: possibledeterminant factors. Sleep Breath 2002;6:11–18.

38. Sajkov D, Wang T, Saunders NA, Bune AJ, Neill AM, Douglas Mcevoy R.Daytime pulmonary hemodynamics in patients with obstructive sleepapnea without lung disease. Am J Respir Crit Care Med 1999;159:1518–1526.

39. Stamatakis K, Sanders MH, Caffo B, Resnick HE, Gottlieb DJ, Mehra R,et al. Fasting glycemia in sleep disordered breathing: lowering thethreshold on oxyhemoglobin desaturation. Sleep 2008;31:1018–1024.

40. Sleep-related breathing disorders in adults: recommendations forsyndrome definition and measurement techniques in clinicalresearch. The report of an American Academy of Sleep Medicinetask force. Sleep 1999;22:667–689.

41. American Academy of Sleep Medicine. International classification ofsleep disorders, 2nd ed.: diagnostic and coding manual.Westchester, IL: American Academy of Sleep Medicine; 2005.

42. Rechtschaffen A, Kales A. A manual of standardized terminology,techniques and scoring system for sleep stages of human subjects.Los Angeles, CA: Brain Information Service/Brain Research Institute,UCLA; 1968. NIH Publication 204.

43. Pillar G, Bar A, Shlitner A, Schnall R, Shefy J, Lavie P. Autonomicarousal index: an automated detection based on peripheral arterialtonometry. Sleep 2002;25:543–549.

44. Orem J. Central respiratory activity in rapid eye movement sleep:augmenting and late inspiratory cells [corrected]. Sleep 1994;17:665–673. [Published erratum appears in Sleep 18:137.]

45. Lacasse Y, Series F, Vujovic-Zotovic N, Goldstein R, Bourbeau J,Lecours R, et al. Evaluating nocturnal oxygen desaturation inCOPD--revised. Respir Med 2011;105:1331–1337.

46. Wiegand L, Zwillich CW, Wiegand D, White DP. Changes in upperairway muscle activation and ventilation during phasic REM sleep innormal men. J Appl Physiol (1985) 1991;71:488–497.

47. Mokhlesi B, Finn LA, Hagen EW, Young T, Hla KM, Van Cauter E, et al.Obstructive sleep apnea during REM sleep and hypertension.results of the Wisconsin Sleep Cohort. Am J Respir Crit Care Med2014;190:1158–1167.

48. Lu J, Sherman D, Devor M, Saper CB. A putative flip-flop switch forcontrol of REM sleep. Nature 2006;441:589–594.

49. Chami HA, Baldwin CM, Silverman A, Zhang Y, Rapoport D, Punjabi NM,et al. Sleepiness, quality of life, and sleep maintenance in REM versusnon-REM sleep-disordered breathing. Am J Respir Crit Care Med2010;181:997–1002.

50. Gottlieb DJ, Punjabi NM, Mehra R, Patel SR, Quan SF, Babineau DC,et al. CPAP versus oxygen in obstructive sleep apnea. N Engl J Med2014;370:2276–2285.

51. Malhotra A, Schwartz DR, Ayas N, Stanchina M, White DP. Treatmentof oxygen-induced hypercapnia. Lancet 2001;357:884–885.

52. Alchakaki A, Riehani A, Shikh-Hamdon M, Mina N, Badr MS, Sankari A.Expiratory snoring predicts obstructive pulmonary disease in patientswith sleep-disordered breathing. Ann Am Thorac Soc 2016;13:86–92.

53. Owens RL, Edwards BA, Malhotra A, Wellman A. Expiratory resistanceincreases end-expiratory lung volume during sleep. Am J Respir CritCare Med 2012;185:e10–e11.

54. Johnson PL, Edwards N, Burgess KR, Sullivan CE. Detection ofincreased upper airway resistance during overnightpolysomnography. Sleep 2005;28:85–90.

55. Verbraecken J, McNicholas WT. Respiratory mechanics and ventilatorycontrol in overlap syndrome and obesity hypoventilation. Respir Res2013;14:132.

