This may be the author’s version of a work that was submitted/acceptedfor publication in the following source:

Duce, Brett, Kulkas, Antti, Langton, Christian, Toyras, Juha, & Hukins,Craig(2017)Amsterdam positional OSA classification: the AASM 2012 recommendedhypopnoea criteria increases the number of positional therapy candidates.Sleep and Breathing, 21(2), pp. 411-417.

This file was downloaded from: https://eprints.qut.edu.au/102818/

c© 2016 Springer-Verlag Berlin Heidelberg

This is a post-peer-review, pre-copyedit version of an article published inSleep and Breathing. The final authenticated version is available online at:http://dx.doi.org/10.1007/s11325-016-1432-y

Notice: Please note that this document may not be the Version of Record(i.e. published version) of the work. Author manuscript versions (as Sub-mitted for peer review or as Accepted for publication after peer review) canbe identified by an absence of publisher branding and/or typeset appear-ance. If there is any doubt, please refer to the published source.

https://doi.org/10.1007/s11325-016-1432-y

Page 1 of 20

Title:

Amsterdam Positional OSA Classification: The AASM 2012 Recommended

Hypopnea Criteria Increases the Number of Positional Therapy Candidates

Authors: 5

Brett Duce1, 2, Antti Kulkas3,5, Christian Langton2, Juha Töyräs2,4,5, Craig Hukins1

Affiliations and addresses:

1Sleep Disorders Centre, Department of Respiratory & Sleep Medicine, Princess

Alexandra Hospital, Ipswich Rd, Woolloongabba, Qld, Australia 10

2Faculty of Science and Engineering, Queensland University of Technology

3Department of Clinical Neurophysiology, Seinäjoki Central Hospital, Seinäjoki,

Finland

4Diagnostic Imaging Center, Kuopio University Hospital, Kuopio, Finland

5Department of Applied Physics, University of Eastern Finland, Kuopio, Finland 15

Email address, telephone and fax number of corresponding author:

(E) brett.duce@health.qld.gov.au

(T) + 61 7 3176 5751

(F) + 61 7 3176 7096

20

Page 2 of 20

Abstract

Purpose: This study examined the effect of hypopnoea criteria on the prevalence of

positional obstructive sleep apnea (pOSA) identified under the Amsterdam Positional OSA 25

Classification (APOC) system.

Methods: 303 consecutive patients undertaking polysomnography (PSG) for the suspicion of

OSA were included in this retrospective investigation. PSGs were scored using both the

2007 American Academy of Sleep Medicine (AASM) recommended hypopnea criteria

(AASM2007Rec) and the 2012 AASM recommended hypopnea criteria (AASM2012Rec). For each 30

hypopnea criteria OSA patients were grouped according to the APOC categories (I, II or II) or

else deemed non-APOC if they did not meet the APOC criteria. Outcome measures, such as

Functional Outcomes of Sleep Questionnaire (FOSQ), MOS 36-item short-form health survey

(SF-36) and psychomotor vigilance task (PVT), were also compared between the groups.

Results: The AASM2012Rec increased the prevalence of OSA compared to AASM2007Rec. The 35

AASM2012Rec trebled the number of APOC I patients compared to AASM2007Rec (297%

increase) as well as increased the proportion of females in the APOC I group. AASM2012Rec

did not change the number of APOC II and APOC III patients. In fact the same patients were

present in these categories irrespective of hypopnea criteria. The proportion of non-APOC

patients proportionally decreased with the AASM2012Rec criteria. There were no differences in 40

outcome measures between the AASM2012Rec and AASM2007Rec groups.

Conclusions:

This study demonstrates that, compared to AASM2007Rec, AASM2012Rec increases the

prevalence of who could be successfully treated with positional therapy. The proportion of

females with pOSA also increases as a consequence of AASM2012Rec. 45

Keywords: Obstructive sleep apnea, hypopnea definition, positional OSA, Amsterdam

Positional OSA Classification, prevalence.

Page 3 of 20

INTRODUCTION

Obstructive sleep apnea (OSA) is a prevalent disorder characterized by repeated partial or 50

complete closure of the upper airway during sleep. The consequences of this disorder

include excessive daytime somnolence [1], reduced quality of life [2], neurocognitive deficits

[3] and cardiovascular disease [4]. Categorisation of OSA severity is based on the Apnea-

Hypopnea Index (AHI), which is the count of all apnoeas and hypopnoeas per hour of sleep

recorded. In some OSA patients the number and duration of apnoeas and hypopneas differ 55

significantly with body position during sleep [5]. These patients, described as having

positional OSA (pOSA), show greater upper airway collapsibility when lying supine opposed

to the lateral position [6]. The prevalence of pOSA varies according to how it is defined [7].

Using a strict supine-independent OSA (siOSA) definition of pOSA, where the total AHI is

>5/hr, the supine AHI:non-supine AHI ratio is ≥ 2:1 and the non-supine AHI of <5/hr, the 60

prevalence is approximately 30% [8]. A prevalence of approximately 60% occurs when a

more lenient supine-predominant OSA (spOSA) definition of total AHI is >5/hr and supine

AHI:non-supine AHI ratio is ≥ 2:1 [8].

In recent years, the creation of new positional sleeping devices has renewed interest in 65

pOSA. As a result, there have been a number of studies that have examined the efficacy and

compliance associated with positional therapy (PT) [9][10]. In addition to these PT trials,

some attention has directed towards identifying patients that would most benefit from PT.

Frank and colleagues [11] recently developed the Amsterdam Positional OSA Classification

(APOC) to provide clinicians with a framework based on the diagnostic polysomnogram 70

(PSG) results. In their study, they posit that multifactorial OSA patients (whose OSA severity

is influenced in part by sleeping position) as well as the true positional OSA patient can

benefit from position therapy.

The APOC system requires patients to spend a minimum of 10% of their time in their best 75

sleep position (BSP) and worst sleeping position (WSP), and then compares the AHI in these

positions. There are three different classification groups in the APOC system. APOC I is

Page 4 of 20

where an OSA patient has an AHI less than 5 events per hour in their BSP. APOC II requires

the AHI in their BSP to be greater than 5 events per hour but in a lower AHI severity category

than their WSP. Lastly, an APOC III requires an overall AHI greater than 40 events per hour 80

and achieve a 25% reduction compared to overall AHI when in their BSP.

In their follow up study, this same group characterised the prevalence of pOSA patients

according to their APOC system [12]. This study showed prevalence amongst OSA patients

to be 36%, 28%, 4% for APOC groups I, II and III, respectively. In this prevalence study, they 85

used the American Academy of Sleep Medicine’s (AASM) recommended hypopnoea criteria

from the 2007 Manual for the Scoring of Sleep and Associated Events (AASM2007Rec). To

score a hypopnoea, the AASM2007Rec requires a ≥30% decrease in nasal pressure amplitude

with at least 4% SpO2 desaturation. This recommended hypopnea criteria has since been

replaced. The new recommended hypopnoea criteria (AASM2012Rec) is now defined as a 90

≥30% decrease in nasal pressure amplitude with at least 3% SpO2 desaturation or

associated EEG arousal. This change in hypopnoea criteria has been shown to not only

increase the AHI in OSA patients [13] but also increases the number of patients diagnosed

with OSA.

95

The effect of the AASM2012Rec hypopnea criteria upon prevalence within this APOC system is

unknown. The aim of this study was therefore to investigate the influence that hypopnea

criteria has on the categorisation of pOSA patients according to the APOC system.

Page 5 of 20

METHODS 100

Patient Selection

All patients undertaking diagnostic PSG at the Sleep Disorders Centre, Princess Alexandra

Hospital for the clinical suspicion of OSA during the period of January 2015 to December

2015 were included in this study. Patients were excluded (a total of 207 patients were

excluded) if less than 2 hours of sleep were recorded, a split night treatment protocol 105

(diagnostic to PAP therapy) was implemented, a primary PSG channel (nasal pressure,

pulse oximetry, all EEG, respiratory effort, body position) contained too much artefact for

reliable analysis, supplemental oxygen was given during the course of the PSG or if another

sleep disorder (eg sleep hypoventilation) was the primary suspicion.

110

Polysomnography

PSG’s were recorded with the Compumedics Grael acquisition devices and Profusion 4.1

software (Abbotsford, Australia). The recording montage comprised of EEG (F4-M1, C4-M1,

O2-M1), left and right EOG (recommended derivation: E1-M2, E2-M2), chin electromyogram

(EMG, mental/submental positioning), modified lead II ECG, nasal pressure (DC amplified), 115

oronasal thermocouple, body position, thoracic and abdominal effort (inductive

plethysmography), pulse oximetry, left and right leg movement (anterior tibialis EMG) and

sound pressure (dBA meter: Tecpel 332). Body position was recorded using a triaxial

accelerometer (Compumedics, Abbotsford, Australia) and attached at the midline of the

lower sternum. Reliability of the body position recording was confirmed with synchronized 120

digital video recordings and corrected where necessary during the PSG scoring process.

Outcome Questionnaires and Vigilance Testing

Prior to undertaking the diagnostic PSG, patients completed the Epworth Sleepiness Scale

(ESS) [14], the Functional Outcomes of Sleep Questionnaire (FOSQ) [15] and the Short 125

Form-36 quality of life questionnaire (SF-36) [16]. Patients also completed the 10 minute

version of the PEBL Psychomotor Vigilance Task (PVT) [17][18] on an ASUS Transformer

Pad with attached keyboard.

Page 6 of 20

Polysomnogram Scoring Protocol 130

PSGs were de-identified and all previous respiratory event scoring was removed. PSGs were

then scored in random order using the AASM2012 criteria. Randomization of PSG’s was

performed using the freely accessible Randomizer website [19]. PSGs were scored with

Compumedics Profusion 4.0 (Build 410) software. The PSG scorer (BD) has over 15 years’

experience in scoring PSGs and participates regularly in intra- and inter-laboratory scoring 135

concordance activities.

APOC Criteria

Patients who met the International Classification of Sleep Disorders 3rd edition (ICSD-3) [20]

criteria for OSA were sorted into the relevant APOC groups. All PSGs must record at least 140

10% of total sleep time (TST) in their best sleeping position (BSP) and worst sleeping

position (WSP). To be classed as APOC I, patients had to have an AHI less than 5 events

per hour in their BSP. APOC II required the AHI in their BSP to be greater than 5 events per

hour but in a lower AHI severity category than their WSP. Lastly, an APOC III patient has to

have an overall AHI greater than 40 events per hour and achieved a ≥ 25% reduction in their 145

AHI when in their BSP. Patients who did not meet any of the APOC criteria were classified as

non-APOC patients.

Data and Statistical Analysis

Statistical analyses were performed using GraphPad Prism 6.05 (GraphPad Software, La 150

Jolla, CA). Normality in the distribution of data collected was determined by the D’Agostino-

Pearson omnibus K2 test [21]. Data are presented as mean ± standard deviation or median

and interquartile range for normally distributed and non-normally distributed data,

respectively. AASM2007Rec and AASM2012Rec group data were compared using either an

unpaired t-test or Mann-Whitney test for normally distributed and non-normally distributed 155

data, respectively. The proportion of male:female in each group was compared using a Chi

square test. Comparisons between APOC groups were performed using one-way ANOVA

Page 7 of 20

with Holm-Sidak multiple comparisons test or Kruskal Wallis with Dunn’s multiple

comparisons test for normally distributed and non-normally distributed data, respectively. A

p<0.05 was considered statistically significant. 160

Page 8 of 20

RESULTS

Patient Cohort

A total of 303 patients were included in this study. Demographic and PSG data for OSA

groups according to scoring criteria are presented in Table 1. Fifty-three percent of the 165

patient cohort had an AHI of ≥5/h when respiratory events were scored according to the

AASM2007Rec, whereas 84% of the same group achieved the same threshold when respiratory

events were scored with AASM2012Rec. The AASM2012Rec added another 94 patients above the

AHI ≥5/h threshold. OSA patients according to AASM2007Rec were slightly more obese

(p=0.03) and had a greater incidence of hyperlipidemia (p=0.02) and type 2 diabetes 170

(p<0.001) than OSA patients diagnosed according to AASM2012Rec. The PSG of OSA patients

according to AASM2007Rec had more N1 sleep (p=0.02) and more EEG arousals (p<0.001)

than their AASM2012Rec counterparts.

APOC Frequency 175

The distribution of OSA patients into their respective APOC groups is shown in Figure 1. The

number of OSA patients grouped into APOC I increased three-fold with AASM2012Rec (p=0.02).

As a proportion of the OSA population in the cohort, this translated to an 85% increase. The

number of OSA patients grouped into APOC II and APOC III did not change between

AASM2007Rec and AASM2012Rec. In fact, both AASM2007Rec and AASM2012Rec groups had exactly 180

the same patients in APOC II and APOC III categories. These two groups did not show an

appreciable change as a proportion of OSA patients. The non-APOC group, those OSA

patients that did not meet any of the APOC criteria, increased in number with AASM2012Rec,

but slightly decreased as a proportion of all OSA patients. This decrease was not statistically

significant. 185

APOC & non-APOC Group Characteristics

The characteristics of each APOC group and the non-APOC group are shown in Table 2.

Based on hypopnoea criteria, there was little difference in functional outcomes between the

two APOC I groups. However, there was gender redistribution with an increase in the 190

Page 9 of 20

number of females (p=0.001) meeting the APOC I criteria when the AASM2012Rec hypopnea

criteria was applied. As described before, there were no differences between APOC II and

APOC III groups since they contained the exact same patients. The two non-APOC groups

(AASM2007Rec non-APOC and AASM2012Rec non-APOC) did not show any differences with

respect to physical characteristics or functional outcomes between them. Both non-APOC 195

groups were - more obese than their APOC group counterparts were (p<0.0001 for

AASM2012Rec non-APOC vs AASM2012Rec APOC I, p=0.005 for AASM2007Rec non-APOC vs

AASM2007Rec APOC I and p=0.02 for AASM2007Rec non-APOC vs AASM2007Rec APOC II).

Page 10 of 20

DISCUSSION 200

In this study, we examined the effect of hypopnea criteria on the categorisation of pOSA

patients according to the APOC system. Our data shows that AASM2012Rec increases the

number and proportion of patients meeting the criteria for APOC I compared to AASM2007Rec.

Furthermore, the proportion of females meeting the criteria for APOC I also increased with

AASM2012Rec. The AASM2012Rec hypopnea criteria also did not change the number of patients 205

that were classified as APOC II and APOC III under the AASM2007Rec hypopnea criteria. In

fact, exactly the same patients were classified to these groups regardless of the hypopnea

criteria. Lastly, we find that the AASM2012Rec hypopnea criteria increases the number of

patients that do not meet the APOC criteria (non-APOC patients) but decreases the

proportion of non-APOC patients within the OSA patient group. 210

The APOC system was developed to aid the identification of true pOSA patients,

multifactorial OSA patients (where position plays some role in their OSA severity) and non-

positional OSA patients. Since it is a relatively new system, there is little data to describe the

prevalence of these APOC groupings. The developers of APOC did provide some 215

prevalence data as a follow up to their initial APOC development publication [12] albeit using

the AASM2007Rec hypopnea criteria. Our prevalence study shows some similarities and

differences with respect to theirs. As a proportion of the patient cohort, our study showed

less patients meeting the APOC criteria. This particular feature was present in our cohort

irrespective of the hypopnea criteria used. It is likely that differences in patient demographics 220

could explain these differences in APOC proportions. Although overall cohort body mass

index (BMI) was not provided by Ravesloot and colleagues [12], we suspect that the overall

BMI in our cohort is higher and thus could make more patients less likely to meet the APOC

criteria. This inverse relationship between BMI and prevalence of pOSA is clearly

demonstrated in the literature [7]. 225

The proportion of patients within each APOC category were however similar to those of

Ravesloot and colleagues [12] when the AASM2007Rec criteria was implemented. Once

Page 11 of 20

AASM2012Rec criteria were implemented to this cohort, we saw that the APOC proportions

changed significantly with the proportion of APOC I patients increasing significantly. Under 230

the APOC system, patients classified as APOC I are considered to be true positional OSA

patients. That is, in these patients the implementation of PT should be adequate to manage

their OSA disorder. The prevalence of APOC I patients and its changes with respect to

hypopnea criteria is interesting for a number of reasons. Under the AASM2007Rec criteria we

can see that the APOC I prevalence is similar to studies that used definitions of siOSA [7]. 235

This definition is somewhat similar in that it presumes that PT alone should be adequate to

treat these patients. When we applied the AASM2012Rec criteria to this cohort, we trebled the

number of patients classified as APOC I and thus presumably trebled the number of patients

that could treated successfully with PT. Therefore, in a financially-strained health system PT

could be of great benefit to managing the extra OSA patients diagnosed according to the 240

AASM2012Rec criteria as it is relatively inexpensive compared to PAP therapy. Interestingly,

trebling of APOC I numbers with AASM2012Rec does not correspond with the changes

observed in siOSA patient numbers in our previous study [8]. This would suggest subtle

differences exist between the definitions of APOC I and siOSA.

245

The prevalence of APOC II patients in our study was somewhat less than that of Ravesloot

and colleagues [12]. Indeed, it was interesting that the number of APOC II and APOC III

patients did not change with hypopnea criteria whatsoever. This was an unexpected finding.

We presumed that with the generalised increase in AHI associated with AASM2012Rec criteria

[13], some of the APOC I patients may move into a different APOC category. This obviously 250

did not occur and we can only suggest patient cohort differences as a reasonable

explanation. The import of this data however suggests that a relatively small number of

patients in our cohort, according to the APOC system, would be suitable for using PT as an

adjunct to other therapies (such as CPAP). The decrease in the proportion of patients that

did not meet any of the APOC criteria (non-APOC) together with the increase in the 255

proportion of APOC I patients suggests that the AASM2012Rec preferentially increases the

Page 12 of 20

frequency of events in the supine position. It remains to be seen whether these results will

increase the prescription of PT by sleep centres.

Another finding of this study was that the AASM2012Rec hypopnea criteria increases the 260

number of female patients classed as APOC I. These findings are congruent with our

previous study [8] into the effect of hypopnea criteria on the prevalence of pOSA. Indeed, we

found that the AASM2012Rec hypopnea criteria increased the proportion of females in both the

spOSA category and the siOSA category. As discussed in our previous study, there are

specific differences between males and females with OSA and how they present [22]. 265

Females are less likely to be present with apneas and have less severe SpO2 desaturations

[23, 24]. Our study results therefore suggest more females could be successfully treated with

PT alone. This is important for clinicians as females with minimally symptomatic OSA are

less likely to continue with CPAP in the long term [25]. This is an notable avenue for further

exploration. 270

Despite these gender differences between AASM2007Rec and AASM2012Rec hypopnea criteria,

there was very little to separate these APOC groups in terms of outcome data (ESS, SF-36,

FOSQ and PVT). There are two possible explanations for these similar outcomes. One

explanation is that the outcome measures are not appropriate for this group. This is probably 275

the least plausible explanation as the SF-36 and the FOSQ questionnaires have been used

in various sleep disorders research studies and demonstrate reasonable differences between

OSA and the normal population [15, 26] and improve with CPAP therapy [27, 28]. The other

explanation could be that there actually is no difference between the groups. The

AASM2007Rec hypopnea criteria, with its exclusion of EEG arousal as a consequence, has 280

been shown to lack sensitivity in symptomatic lean patients [29]. Thus it is possible that the

AASM2007Rec hypopnea criteria overlooked these pOSA patients despite their symptoms and

reduced quality of life.

Page 13 of 20

Our study is not without its limitations. These PSGs were examined from the one sleep 285

centre and may be subject to a demographic bias. This was evident when we examine our

APOC prevalence compared to the previous prevalence study by the developers of the

APOC system. Another limitation to this study was the standard use of a position sensor

attached to the trunk only. The measurement of body position from the trunk means that

patients could have their neck either oriented in the supine position or rotated into a lateral 290

position during that period. Evidence from van Kesteren and colleagues [30] have

demonstrated there can be vast differences in AHI magnitude with respect to head position

relative to trunk position. Thus, it is possible that our position sensors will incorrectly

categorise pOSA patients occasionally. At this moment in time, there is no position sensor or

sensors made to provide simultaneous position of the trunk and neck. 295

Conclusion

In conclusion, our study has shown that, according the APOC scheme, utilisation of the

AASM2012Rec hypopnea criteria will increase the number of patients classed as APOC I.

These results would suggest that an increased number of patients could be managed 300

successfully with PT. The change associated with AASM2012Rec hypopnea criteria is limited to

the APOC I category as the patients categorised in APOC II and APOC III were exactly the

same irrespective of hypopnea criteria. Thus there is no change in the number of patients

that could successfully use PT as an adjunct to their usual therapy. Studies to examine the

efficacy of the APOC system are still required. 305

Financial support

Antti Kulkas is supported by the Emil Aaltonen and Tampere Tuberculosis Foundations. The

sponsors had no role in the design or conduct of this research.

310

Conflict of Interest

Page 14 of 20

All authors certify that they have no affiliations with or involvement in any organization or

entity with any financial interest in the subject matter or materials discussed in this

manuscript.

315

Ethical Approval

The Institutional Human Research Ethics Committee of the Princess Alexandra Hospital

approved this study (HREC/16/QPAH/021). All procedures performed in studies involving

human participants were in accordance with the ethical standards of the institutional and/or

national research committee and with the 1964 Helsinki declaration and its later amendments 320

or comparable ethical standards. For this type of study formal consent by the patients was

not required.

Page 15 of 20

REFERENCES 325

1. Gottlieb DJ, Whitney CW, Bonekat WH, et al (1999) Relation of sleepiness to respiratory

disturbance index: The sleep heart health study. Am J Respir Crit Care Med 159:502–507. doi: 10.1164/ajrccm.159.2.9804051

2. Lopes C, Esteves a M, Bittencourt LR a, et al (2008) Relationship between the quality of life 330 and the severity of obstructive sleep apnea syndrome. Brazilian J Med Biol Res 41:908–913.

3. Olaithe M, Bucks RS (2013) Executive dysfunction in OSA before and after treatment: a meta-analysis. Sleep 36:1297–305. doi: 10.5665/sleep.2950

4. Young T, Peppard PE, Gottlieb DJ (2002) Epidemiology of Obstructive Sleep Apnea. Am J Respir Crit Care Med 165:1217–1239. doi: 10.1164/rccm.2109080 335

5. Cartwright RD (1984) Effect of sleep position on sleep apnea severity. Sleep 7:110–114. doi: 10.1378/chest.90.3.424

6. Ong JSL, Touyz G, Tanner S, et al (2011) Variability of human upper airway collapsibility during sleep and the influence of body posture and sleep stage. J Sleep Res 20:533–537. doi: 10.1111/j.1365-2869.2011.00925.x 340

7. Joosten SA, O’Driscoll DM, Berger PJ, Hamilton GS (2014) Supine position related obstructive sleep apnea in adults: Pathogenesis and treatment. Sleep Med Rev 18:7–17. doi: 10.1016/j.smrv.2013.01.005

8. Duce B, Kulkas A, Langton C, et al (2016) The AASM 2012 Recommended Hypopnea Criteria Increases the Incidence of Obstructive Sleep Apnea but not the Proportion of Positional 345 Obstructive Sleep Apnea. Sleep Med. doi: http://dx.doi.org/10.1016/j.sleep.2016.07.013

9. Bignold JJ, Mercer JD, Antic NA, et al (2011) Accurate position monitoring and improved supine-dependent obstructive sleep apnea with a new position recording and supine avoidance device. J Clin Sleep Med 7:376–383. doi: 10.5664/JCSM.1194

10. de Vries GE, Hoekema A, Doff MH, et al (2015) Usage of positional therapy in adults with 350 obstructive sleep apnea. J Clin Sleep Med 11:131–137. doi: 10.5664/jcsm.4458

11. Frank MH, Ravesloot MJL, van Maanen JP, et al (2014) Positional OSA part 1: towards a clinical classification system for position-dependent obstructive sleep apnoea. Sleep Breath. doi: 10.1007/s11325-014-1022-9

12. Ravesloot MJL, Frank MH, van Maanen JP, et al (2015) Positional OSA part 2: retrospective 355 cohort analysis with a new classification system (APOC). Sleep Breath. doi: 10.1007/s11325-015-1206-y

13. Duce B, Milosavljevic J, Hukins C (2015) The 2012 AASM Respiratory Event Criteria Increase the Incidence of Hypopneas in an Adult Sleep Center Population. J Clin Sleep Med 11:1425–1431. doi: 10.5664/jcsm.5280 360

14. Johns MW (1991) A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep 14:540–545. doi: 10.1016/j.sleep.2007.08.004

15. Weaver TE, Laizner AM, Evans LK, et al (1997) An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep 20:835–43.

16. Ware JEJ, Sherbourne CD (1992) The MOS 36-item short-form health survey (SF-36). I. 365 Conceptual framework and item selection. Med Care 30:473–483.

17. Mueller ST, Piper BJ (2014) The Psychology Experiment Building Language (PEBL) and PEBL Test Battery. J Neurosci Methods 222:250–9. doi: 10.1016/j.jneumeth.2013.10.024

18. Wilkinson RT, Houghton D (1982) Field test of arousal: a portable reaction timer with data storage. Hum Factors 24:487–493. 370

19. Urbaniak, G.C., and Plous S (2011) Research Randomizer (Version 3.0). https://www.randomizer.org. Accessed 14 Jan 2016

20. American Academy of Sleep Medicine (2014) International classification of sleep disorders, 3rd ed. American Academy of Sleep Medicine, Darien, IL

Page 16 of 20

21. D’Agostino R, Pearson ES (1973) Tests for Departure from Normality. Empirical Results for the 375 Distributions of b 2 and √b 1. Biometrika 60:613. doi: 10.2307/2335012

22. Shepertycky MR, Banno K, Kryger MH (2005) Differences between men and women in the clinical presentation of patients diagnosed with obstructive sleep apnea syndrome. Sleep 28:309–14.

23. Ware JC, McBrayer RH, Scott J a (2000) Influence of sex and age on duration and frequency 380 of sleep apnea events. Sleep 23:165–70.

24. Mohsenin V (2001) Gender Differences in the Expression of Sleep-Disordered Breathing. Chest 120:1442–1447.

25. Turnbull CD, Bratton DJ, Craig SE, et al (2016) In patients with minimally symptomatic OSA can baseline characteristics and early patterns of CPAP usage predict those who are likely to 385 be longer-term users of CPAP. J Thorac Dis 8:276–281. doi: 10.3978/j.issn.2072-1439.2016.01.54

26. Smith, Shneerson (1995) Is the SF 36 sensitive to sleep disruption? A study in subjects with sleep apnoea. J Sleep Res 4:183–188. doi: jsr004003183 [pii]

27. Bennett LS, Langford BA, Stradling JR, Davies RJ (1998) Sleep fragmentation indices as 390 predictors of daytime sleepiness and nCPAP response in obstructive sleep apnea. Am J Respir Crit Care Med 158:778–86. doi: 10.1164/ajrccm.158.3.9711033

28. Antic NA, Catcheside P, Buchan C, et al (2011) The effect of CPAP in normalizing daytime sleepiness, quality of life, and neurocognitive function in patients with moderate to severe OSA. Sleep 34:111–9. doi: 10.1016/j.yneu.2011.05.056 395

29. Guilleminault C, Hagen CC, Huynh NT (2009) Comparison of hypopnea definitions in lean patients with known obstructive sleep apnea hypopnea syndrome ( OSAHS ). Sleep Breath 13:341–347. doi: 10.1007/s11325-009-0253-7

30. van Kesteren ER, van Maanen JP, Hilgevoord A a J, et al (2011) Quantitative effects of trunk and head position on the apnea hypopnea index in obstructive sleep apnea. Sleep 34:1075–400 81. doi: 10.5665/SLEEP.1164

Page 17 of 20

Table 1 Patient characteristics of OSA patients (AHI ≥ 5/h) according to hypopnea criteria. 405

Parameter AASM2007Rec AASM2012Rec p value

Number 161 255 Age, yrs 61 (50, 68) 58 (47, 67) 0.15 BMI, kg/m

2 36.6 (31.1, 42.5) 34.2 (29.7, 40.6) 0.03

ESS 9 (5, 13) 9 (5, 13) 0.75 Gender, M:F 103:58 149:106 0.22 Co-morbidities, % of patients Hypertension 53 46 0.16 Hyperlipidemia 37 27 0.02 Type 2 Diabetes 32 19 > 0.001 GERD 23 20 0.45 Depression 23 18 0.19 Ischemic heart disease 17 12 0.12 TST, min 276 ± 94 292 ± 91 0.07 Sleep Efficiency, % 64.5 (47.9, 76.3) 67.4 (53.9, 79.9) 0.06 Sleep Stage Proportions, % of TST NREM Total 84.7 ± 8.0 83.7 ± 8.1 0.26 N1 16.9 (10.3, 28.3) 13.4 (7.9, 23.1) 0.02 N2 47.2 ± 13.2 49.1 ± 13.0 0.17 N3 14.7 (4.9, 25.2) 16.6 (7.1, 25.3) 0.67 R 15.3 ± 8.0 16.3 ± 8.0 0.22 AHI, events/hr 15.9 (9.6, 37.9) 21.0 (11.5, 41.8) 0.04 ArI, events/hr 31.1 (19.8, 46.8) 23.4 (15.4, 37.5) > 0.001 Proportion of TST in WSP, % 42.0 (7.5, 72.5) 39.0 (12.0, 69.0) 0.81

Values are presented as median (inter-quartile range). BMI; body mass index, ESS; Epworth sleepiness scale, GERD; gastroesophageal reflux disease, TST; total sleep time, NREM; non-REM sleep, N1; stage 1 sleep, N2; stage 2 sleep, N3; stage 3 sleep, R; REM sleep, AHI; apnea-hypopnea index, AASM2012Rec; 2012 AASM 410 recommended hypopnea definitions, ArI; arousal index, SpO2; oxygen saturation, WSP; worst sleeping position.

Page 18 of 20

Table 2 Characteristics of patients within each Amsterdam Positional OSA Classification (APOC) category according to hypopnea criteria.

Parameter APOC I APOC II APOC III Non-APOC AASM2007Rec AASM2012Rec AASM2007/2012Rec AASM2007/2012Rec AASM2007Rec AASM2012Rec

Number 34 102* 23 1 103 130

Age 58 ± 13 54 ± 13 60 ± 14 72 58 ± 13 58 ± 13

Gender (M:F) 23:11 54:47* 13:10 1:0 66:37 81:49

BMI 33.7 (28.8, 36.8) 32.4 (28.2, 36.2) 33.9 ± 7.3 30.9 38.8 (32.7, 43.0)*^ 38.2 (31.7, 42.6)§

ESS 11 (4, 13) 10 (5, 14) 9 ± 6 5 9 ± 5 9 ± 5

FOSQ 12 (10, 17) 13 (10, 17) 14 ± 4 18 14 (11, 17) 14 (11, 17)

SF-36 MCS 36.7 ± 13.4 36.3 ± 13.5 38.3 ± 10.8 36.2 36.9 ± 11.7 37.2 ± 11.9

SF-36 PCS 37.4 (34.4, 46.3) 40.7 (32.8, 46.7) 41.7 ± 10.5 61.0 36.9 ± 11.9 37.2 ± 11.4

1/RT 2.5 ± 0.6 2.5 ± 0.5 2.6 ± 0.4 3.0 2.7 (2.3, 2.9) 2.6 (2.3, 2.9)

PVT Lapses 12 (4, 40) 11 (3, 29) 8 (2, 15) 3 8 (2, 21) 9 (2, 21)

1/Slowest 10% 1.5 (1.0, 2.1) 1.6 (1.0, 2.1) 1.8 (1.5, 2.3) 2.1 1.6 (1.0, 2.1) 1.7 (1.1, 2.1)

Values presented as mean ± standard deviation or median (interquartile range) where appropriate. Since APOC II and APOC III groups contain exactly the same patients for each hypopnea criteria, these are presented as a single column. APOC, Amsterdam positional OSA classification; AASM2012Rec, AASM 2012 recommended hypopnea criteria; AASM2007Rec, AASM 2007 recommended hypopnea criteria; BMI; body mass index, ESS; Epworth sleepiness scale, FOSQ; functional outcomes of sleep questionnaire, SF-36 MCS; short-form 36 quality of life questionnaire mental component score, SF-36 PCS; short-form 36 quality of life questionnaire physical component score, 1/RT; PVT reciprocal response time, 1/Slowest 10%; the slowest 10% of the PVT reciprocal response time. * denotes p<0.05 compared to APOC I AASM2007Rec. ^ denotes p<0.05 compared to APOC II AASM2007Rec. § denotes p<0.05 compared to APOC I AASM2012Rec. .

Page 19 of 20

Figure 1 Distribution of OSA patients according to the Amsterdam Positional OSA Classification (APOC) for each hypopnea criteria.

AASM2012Rec, AASM 2012 recommended hypopnea criteria; AASM2007Rec, AASM 2007 recommended hypopnea criteria; non-APOC, OSA patients that did not meet any of the APOC criteria. **; denotes p < 0.01. All other comparisons were not statistically significant.

Page 20 of 20

Abbreviations

AASM, American Academy of Sleep Medicine;

AASM2012Rec, 2012 AASM recommended hypopnea definitions;

AASM2007Rec, 2007 AASM recommended hypopnea definitions;

AHI, Apnea-Hypopnea Index;

APOC, Amsterdam Positional OSA Classification;

ArI, EEG Arousal Index;

BMI, body mass index;

BSP, best sleeping position;

ECG, electrocardiogram;

EEG, electroencephalogram;

EOG, electrooculogram;

EMG, electromyogram;

HI, Hypopnea Index;

LCD, liquid crystal display;

NREM, non-rapid eye movement sleep;

N1, stage 1 sleep;

N2, stage 2 sleep;

N3, stage 3 sleep;

Non-APOC, patients not meeting any of the APOC criteria

OSA, Obstructive Sleep Apnea;

pOSA, positional obstructive sleep apnea;

PSG, polysomnography;

PT, positional therapy;

R, rapid eye movement sleep;

siOSA, supine independent OSA;

spOSA, supine predominant OSA;

WSP, worst sleeping position;