Upper Airway Collapsibility During Rapid Eye Movement Sleep Is Associated With the Response to Upper Airway Surgery for Obstructive Sleep Apnea

Yung-An Tsou; Liang-Wen Hang; Eysteinn Finnsson; Jón S. Ágústsson; Scott A. Sands; Wan-Ju Cheng

doi:10.21053/ceo.2024.00246

Clinical and Experimental Otorhinolaryngology > Volume 18(1); 2025 > Article

Tsou, Hang, Finnsson, Ágústsson, Sands, and Cheng: Upper Airway Collapsibility During Rapid Eye Movement Sleep Is Associated With the Response to Upper Airway Surgery for Obstructive Sleep Apnea

Original Article

Clinical and Experimental Otorhinolaryngology 2025; 18(1): 50-56.

Published online: December 24, 2024

DOI: https://doi.org/10.21053/ceo.2024.00246

Upper Airway Collapsibility During Rapid Eye Movement Sleep Is Associated With the Response to Upper Airway Surgery for Obstructive Sleep Apnea

Yung-An Tsou^1,^2,³

, Liang-Wen Hang^1,⁴

, Eysteinn Finnsson⁵

, Jón S. Ágústsson⁵

, Scott A. Sands⁶

, Wan-Ju Cheng^7,^8,⁹

¹College of Medicine, China Medical University, Taichung, Taiwan

²Department of Otolaryngology, Head and Neck Surgery, China Medical University Hospital, Taichung, Taiwan

³Department of Audiology and Speech-Language Pathology, Asia University, Taichung, Taiwan

⁴Sleep Medicine Center, Department of Pulmonary and Critical Care Medicine, China Medical University Hospital, Taichung, Taiwan

⁵Nox Research, Nox Medical ehf, Reykjavík, Iceland

⁶Division of Sleep and Circadian Disorders, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA

⁷Department of Public Health, China Medical University, Taichung, Taiwan

⁸National Center for Geriatrics and Welfare Research, National Health Research Institutes, Miaoli County, Taiwan

⁹Department of Psychiatry, China Medical University Hospital, Taichung, Taiwan

Corresponding author: Wan-Ju Cheng National Center for Geriatrics and Welfare Research, National Health Research Institutes, 35 Keyan Rd, Zhunan, Miaoli County 350401, Taiwan
Tel: +886-5-6325080 ext. 12, Fax: +886-4-22361230 Email: s871056@nhri.edu.tw

Received September 2, 2024 Revised December 4, 2024 Accepted December 23, 2024

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Objectives.

Endotype-based interventions have shown promise in the treatment of patients with obstructive sleep apnea, and upper airway surgery is a key therapeutic option. However, responses to surgery vary among patients with obstructive sleep apnea. This study aims to examine changes in endotypic traits following upper airway surgery and to explore their association with surgical outcomes.

Methods.

We prospectively recruited 25 patients with obstructive sleep apnea who visited a single sleep center for upper airway surgery. These patients underwent polysomnographic studies both before and after surgical intervention. During non-rapid eye movement and rapid eye movement sleep, we estimated endotypic traits—including collapsibility (V_passive), arousal threshold, loop gain, and upper airway compensation—with the phenotyping using polysomnography method. Based on improvements in the apnea-hypopnea index, patients were classified as either responders or non-responders. We compared the preoperative endotypic traits between these groups using Mann-Whitney tests. Additionally, we compared changes in endotypic traits pre- and post-surgery between responders and non-responders using generalized linear mixed models.

Results.

We identified 12 responders and 13 non-responders. Compared to non-responders, responders exhibited improved collapsibility during rapid eye movement sleep (22.3 vs. −8.2%eupnea in V_passive, P=0.01), and their arousal threshold decreased during non-rapid eye movement sleep (−22.4%eupnea, P=0.02). No endotypic trait predicted surgical response; however, the apnea-hypopnea index during rapid eye movement sleep was higher among responders than non-responders (51.8/hr vs. 34.4/hr, P=0.05).

Conclusion.

Upper airway surgery significantly reduced collapsibility during rapid eye movement sleep in responders. The target pathology for upper airway surgery is a compromised upper airway during rapid eye movement sleep.

Keywords: Nasal Surgery; Palatoplasty; Tongue; Tonsil; Pharynx

INTRODUCTION

Obstructive sleep apnea (OSA) is characterized by complex pathological etiologies [1], including high upper airway collapsibility, poor muscle responsiveness, high loop gain, and low arousal threshold. Endotype-based treatment has been shown to improve OSA symptoms and severity [2]. Upper airway surgery, a key treatment option for OSA, demonstrates variable responses among patients.

Reduction of anatomical collapsibility has been identified as the primary mechanism underlying the effectiveness of upper airway surgery [3,4]. Studies utilizing passive pharyngeal critical pressure (Pcrit) have documented a significant decrease in collapsibility following uvulopalatopharyngoplasty (UPPP) [5,6]. Moreover, the presence of non-anatomical pathologies in OSA predicts poor surgical response. A low loop gain, indicative of diminished chemoreflex sensitivity to respiratory disturbances, has been linked to positive surgical outcomes [7,8]. In the aforementioned studies, endotypic traits were assessed during non-rapid eye movement (NREM) sleep; however, respiratory events during rapid eye movement (REM) sleep and NREM sleep are associated with distinct endotypic characteristics [9]. Evaluating endotypic traits across different sleep stages may offer further insights into the individual pathology of OSA.

The non-invasive phenotyping using polysomnography (PUP) method enables the simultaneous estimation of various endotypic traits across different sleep stages [10]. Access to information on endotypic traits allows sleep specialists to offer personalized surgical or non-surgical treatment recommendations for patients with OSA. This study aims to examine the changes in stage-specific endotypic traits following upper airway surgery between responders and non-responders, and any endotypic traits that can predict surgical outcomes in patients with OSA.

MATERIALS AND METHODS

This study was approved by the Institutional Review Board of China Medical University Hospital (No. CMUH109-REC3-018). Informed consent was obtained from all participants.

Participants

We prospectively recruited patients with OSA who were seeking upper airway surgery from a single sleep center in Taiwan between April 2020 and April 2023. Inclusion criteria were: (1) pre-surgery in-lab polysomnographic data showing an apnea-hypopnea index (AHI) >5/hr; (2) age 20 years or older; and (3) surgery was performed within 6 months following the polysomnographic study. A total of 31 patients consented to participate in this study, and 25 of them completed a follow-up polysomnographic study after the surgery and were included in the following analysis. The median delay from the initial polysomnographic study to surgery was 49 days (interquartile range [IQR], 34–94 days), and from surgery to the follow-up study was 212 days (IQR, 148–377 days). Age and sex were retrieved from medical records.

Polysomnographic study

Patients arrived at the sleep laboratory before 11:00 pm. Technicians then measured patients’ neck and waist circumferences, along with body weight and height, before the polysomnographic study, which was performed using a NOX A1 PSG system (Nox Medical). Respiratory events and sleep staging were manually scored by certified sleep technicians. For this study, obstructive apnea was defined as the cessation of airflow through the nose with paradoxical chest and abdominal movements; and hypopnea was defined as a ≥30% reduction in nasal pressure with paradoxical chest and abdominal movements, resulting in a desaturation of ≥4% in oxygen saturation [11].

Estimation of collapsibility from PSG studies

Endotypic traits were estimated using the PUP method [10,12] and as a validated cloud-based Python implementation (PUPpy) [13]. The PUP and PUPpy methods rely on an uncalibrated estimate of minute ventilation during sleep, which is derived from the flow signal in polysomnography. These methods use inverse modeling of the respiratory control system to estimate respiratory drive. The arousal threshold was defined as the estimated chemical drive just before scored electroencephalography arousal [14]. Loop gain was estimated by fitting polysomnographic data to a model of the chemical feedback system [10], and it was reported as the ventilatory response to a 1 cycle/min disturbance. Collapsibility is represented by ventilation at the eupneic drive (V_passive), while V_passive was further square-root transformed as 1-(1-V_passive/100)^0.5 [15]. Ventilation at the arousal threshold was recorded as V_active, and upper airway muscle compensation was calculated as V_active-V_passive (%eupnea). The estimations were performed during NREM and REM sleep separately.

Surgical procedure and outcome evaluation

All patients reported having already tried continuous positive airway pressure (CPAP) and having poor tolerance to this; thus, they sought surgical treatment. Surgical procedure was determined according to airway obstruction level by Muller test, and reconfirmed by drug-induced sleep endoscopy if the surgeon deemed it necessary. Drug-induced sleep endoscopy was performed by the dexmedetomidine-based sedation protocol by total intravenous anesthesia, and each patient’s sleep depth was monitored under bispectral index around 50–70 [16]. The surgical procedure was individually tailored for each patient to address the specific level of airway obstruction, and most of the patients underwent multi-level surgery. Surgical procedures were categorized as (1) palatoplasty, including relocation pharyngoplasty for lateral pharyngeal airway collapse [17] and suspension palatoplasty for anterior-posterior palatal collapse [18]; (2) UPPP for oropharyngeal obstruction by palatine tonsils [19]; (3) tongue base coblation endoscopic lingual lightening for lingual tonsils hypertrophy, which causes retroglossal obstruction [20]; and (4) nasal surgery, including nasal laser or mini-invasive septoturbinoplasty for nasal airway obstruction [21]. This was an observational study and did not influence clinical decisions. Responders were identified as those with an improvement in AHI of ≥10 events/hr or 20%, while the remaining participants were classified as non-responders [22].

Statistical analysis

Baseline demographic, polysomnographic characteristics, and endotypic traits were compared between the responders and non-responders using Mann-Whitney tests. Wilcoxon signed-rank tests were used to examine the differences between pre- and post-surgery endotypic traits in the responder and non-responder groups, respectively. The interaction between groups and surgical state (pre- and post-) on endotypic traits was examined using generalized linear mixed models. Endotypic traits estimated from REM and NREM sleep were examined in separate models. Sensitivity analysis was performed, excluding participants who solely underwent nasal surgery (n=22). All statistical analyses were performed using SAS 9.4 (SAS Institute).

RESULTS

Among the 25 patients, only two were female, and both were classified as responders. The average age of the participants was 42.5 years (standard deviation [SD], 9.6 years), and the mean body mass index was 27.6 kg/m² (SD, 4.2 kg/m²). Responders and non-responders displayed similar demographic profiles (Table 1). Responders exhibited a higher overall AHI (42.4/hr) than non-responders (24.0/hr). Additionally, responders were characterized by a poor upper airway collapsibility profile and worse compensatory function than non-responders (Table 2). However, the P-values corresponding to the differences between the groups were greater than 0.05, which may be attributed to the small sample size. Notably, the AHI during REM sleep was higher in responders (51.8/hr vs. 34.4/hr, P=0.05).

Upon comparing pre-surgical and post-surgical endotypic traits within individuals (Table 2), we noted an improvement in collapsibility during REM sleep, with V_passive changing from 35.5 to 57.8%eupnea. The interaction effect between surgical status and group was significant for collapsibility (Fig. 1). Responders exhibited an improvement of 22.3%eupnea in V_passive, while non-responders experienced a decrease of 8.2%eupnea. During NREM sleep, the arousal threshold for responders decreased by 22.4%eupnea, whereas no significant change was observed in non-responders. In a sensitivity analysis that excluded participants who underwent nasal surgery without pharyngeal surgery (Supplementary Table 1), among responders, the improvement in collapsibility following surgery was significant for both REM and NREM sleep (△_REM=24.40 and △_NREM=14.07).

DISCUSSION

This study examined changes in stage-specific endotypic traits of OSA among responders and non-responders to upper airway surgery. Improvements in upper airway collapsibility, particularly during REM sleep, emerged as the key factor associated with a favorable surgical outcome. However, pre-surgical endotypic traits did not predict the response to upper airway surgery in this study. Our findings suggest that upper airway surgery can reduce the severity of OSA by reducing upper airway collapsibility during REM sleep. Previous studies employing invasive methods have demonstrated a decrease in collapsibility, as measured by Pcrit, following upper airway surgery. This was evidenced by an increase in the size of the retropalatal airway [5,6]. However, these studies assessed Pcrit during NREM sleep. Nevertheless, greater collapsibility during REM sleep compared to NREM sleep has been noted as the primary contributing factor to REM sleeprelated OSA [9]. This, along with our observation that pre-surgical REM AHI was higher in surgical responders than in non-responders, suggests that correcting the collapsibility of the upper airway during REM sleep may be a crucial yet overlooked factor in the success of surgical intervention. Additionally, estimating V_passive during REM sleep may more reliably reflect the true passive pharyngeal collapsibility and better represent the benefits of surgery observed in responders, which are not evident in nonresponders. While adherence to CPAP therapy is lower among patients with REM-related OSA compared to those with NREM OSA [23], surgical intervention remains a preferred treatment option for patients with respiratory events that predominantly occur during REM sleep. Notably, three patients underwent isolated nasal surgery, two of whom were responders. Although isolated nasal surgery can increase nasal airflow and reduce mouth breathing, it has not been shown to significantly decrease AHI in review studies [24,25]. Future research is necessary to determine how the management of nasal obstruction could modify pathological traits in patients with OSA.

The observed decrease in the arousal threshold during NREM sleep following surgery among responders likely indicates an improvement in AHI. This effect of surgery on the arousal threshold was corroborated by a previous study [7]. Similarly, CPAP treatment reduces the arousal threshold, whereas its discontinuation results in an elevated threshold during NREM sleep [26,27]. In the absence of effective treatment, OSA may potentially cause the arousal threshold to rise over time. Nevertheless, despite a reduction in AHI during REM sleep, this threshold did not decrease significantly in responders. Respiratory events may contribute less to the increased arousal threshold during REM sleep than during NREM sleep. Furthermore, a low arousal threshold has been recognized as a characteristic endotypic trait of REM-predominant OSA [28], suggesting a different pathological role of the arousal threshold during REM versus NREM sleep.

Three cohorts have been used to explore the pre-surgical endotypic traits of patients with OSA that are associated with better outcomes. One study included 31 Chinese adult patients [29], while the other two comprised 46 and 23 Australian patients [4,7]. Two of these studies indicated an association between low baseline loop gain and a favorable surgical response during NREM sleep [7,29]. This has led to the view that a more stable ventilatory control system is a key non-anatomical predictor of the success of upper airway surgery [8]. However, the most recent study, which estimated endotypic traits using CPAP dial-down methods or polysomnographic signals [4], did not indicate a relationship between loop gain and surgical outcomes in NREM sleep. This finding was also observed in our study for both REM and NREM sleep. It is possible that the Chinese study demonstrated a positive relationship between loop gain and surgical success because the loop gain was high (0.7) in that sample [29]. In contrast, in the two studies with Australian patients, which reported loop gains similar to those in the present study (0.45), one indicated a positive relationship, while the other found no relationship. Ethnicity or baseline loop gain alone does not fully explain these inconsistent findings. Another factor associated with surgical success is the presence of REM-related OSA [30]. Our results corroborate this, as we observed an improvement in collapsibility during REM sleep following upper airway surgery.

The strength of this study lies in its evaluation of endotypic traits during natural sleep and correlation with surgical outcomes. Relative to previous studies, this research involved a shorter interval between the initial polysomnographic study and upper airway surgery, reducing the potential for bias due to physiological changes during this period. However, this study has certain limitations. First, the small sample size, consisting solely of patients of Asian ethnic origin, limits the generalizability of the findings. Nevertheless, based on previous research, our sample size is considered adequately powered to detect changes preand post-surgery (n≥18) [5], as well as differences between responders and non-responders (n≥16) [7]. In addition to ethnicity, our sample consisted predominantly of middle-aged men without obesity. Since endotypic traits vary with age and sex [31], the results may not extend to other demographic groups. Second, hypopnea was scored using a criterion of a ≥4% decrease in oxygen saturation. Loop gain estimated with this criterion is considered a poor predictor of surgical response compared to using ≥3% oxygen desaturation or arousal criteria [32]. However, previous studies have consistently applied a ≥3% oxygen desaturation criterion, and the relationship between loop gain and surgical outcomes has been inconsistent [4,7,29]. The criteria used do not fully explain the discrepancies between studies. Third, the criterion for surgical responders in our study differs from those in previous research, such as a ≥50% reduction in AHI to a final AHI of <20 events/hr [29] or an AHI reduction ≥50% with a treatment AHI <10 events/hr [7,32]. Under these criteria, the number of responders would be very low (four and two out of 25 participants, respectively), as patients with an AHI >40 events/hr are seldom classified as responders. In contrast, the criteria in the present study tend to classify patients with high baseline AHI as responders. Therefore, comparisons with other research should be approached with caution. Given that reduction in AHI has been questioned as the sole treatment goal for OSA, future studies should also consider other outcomes, such as hypoxic burden, excessive sleepiness, and autonomic function [33,34].

In conclusion, our study demonstrated that collapsibility was significantly reduced from baseline following upper airway surgery in responders, but not in non-responders. We were unable to identify any baseline endotypic traits that could predict the response to surgery; however, we did find an association between higher baseline AHI during REM sleep and surgical success, representing a potential predictor. This study, along with available data, suggests that stage-specific endotypic traits may be valuable in assessing the impact of surgery on changes in pharyngeal collapsibility. Nonetheless, identifying endotypic predictors of surgical success remains challenging.

HIGHLIGHTS

▪ The response of obstructive sleep apnea to upper airway surgery varies based on the underlying pathology.

▪ This study investigated endotypic traits before and after surgery during rapid eye movement (REM) and non-rapid eye movement sleep.

▪ Surgical responders exhibited a significant decrease in upper airway collapsibility during REM sleep.

▪ The target pathology for upper airway surgery is a compromised upper airway during REM sleep.

CONFLICTS OF INTEREST

SAS received grant support from Apnimed, Prosomnus, and Dynaflex and has served as a consultant for Apnimed, Nox Medical, Inspire Medical Systems, Merck, Eli Lilly, Respicardia, LinguaFlex, and Forepont. His interactions with industry are managed by his institution. EF and JSÁ are employees of Nox Medical. No other potential conflicts of interest relevant to this article were reported.

ACKNOWLEDGMENTS

SAS received funding from the NIH NHLBI (grant No. R01HL 146697) and the AASM Foundation (grant No. 228-SR-20). LWH received funding from the National Science and Technology Council (grant No. NSTC 113-2314-B-039-063-MY3) and China Medical University Hospital (grant No. DMR-113-024). WJC received funding from the National Science and Technology Council (grant No. NSTC 113-2314-B-400-020). The funders had no role in the study design; collection, analysis, and interpretation of data; writing of the report; or decision to submit the article for publication.

AUTHOR CONTRIBUTIONS

Conceptualization: WJC. Methodology: SAS, EF, JSÁ. Formal analysis: WJC. Data curation: YAT, LWH. Visualization: WJC. Project administration: WJC. Funding acquisition: SAS. Writing–original draft: YAT, WJC. Writing–review & editing: LWH, EF, JSÁ, SAS. All authors read and agreed to the published version of the manuscript.

SUPPLEMENTARY MATERIALS

Supplementary materials can be found online at https://doi.org/10.21053/ceo.2024.00246.

Supplementary Table 1.

Endotypic traits before and after surgery among responders and non-responders following surgery, excluding participants who underwent only nasal surgery

ceo-2024-00246-Supplementary-Table-1.pdf

Fig. 1.

Pre- to post-surgery changes in endotypic traits: (A) arousal threshold, (B) collapsibility presented as V_passive, (C) loop gain, and (D) compensation, among responders and non-responders. The endotypic traits were assessed during rapid eye movement sleep. %eupnea, percentage of eupneic drive (for arousal threshold) or percentage of eupneic ventilation (for V_passive and compensation). P-values were calculated using Wilcoxon signed-rank tests to compare preoperative and postoperative data.

Table 1.

Preoperative characteristics of study participants (n=25)

Variable	Responder (n=12)	Non-responder (n=13)	P-value
Demographics
Age (yr)	45.08±8.94	40.15±9.90	0.252
Body mass index (kg/m²)	28.58±5.32	26.71±2.74	0.744
Neck circumference (cm)	38.68±4.57	38.27±3.00	0.429
Preoperative polysomnographic characteristics
AHI (/hr)	42.42±28.26	24.02±11.42	0.192
Time with SpO₂ ≤90% (%)	17.79±25.05	5.46±4.36	0.384
AHI (/hr), NREM	38.68±30.62	20.95±12.89	0.231
AHI (/hr), REM	51.78±24.15	34.44±16.84	0.050
Arousal threshold (%eupnea), NREM	157.89±45.94	137.45±26.81	0.277
Arousal threshold (%eupnea), REM	178.65±46.92	162.22±43.24	0.425
V_passive (%eupnea), NREM	58.63±23.88	71.24±13.46	0.211
V_passive (%eupnea), REM	35.47±22.64	53.84±29.02	0.184
Loop gain, NREM	0.55±0.19	0.48±0.12	0.328
Loop gain, REM	0.48±0.20	0.42±0.10	0.676
Compensation (%eupnea), NREM	−11.61±19.41	−4.83±22.48	0.328
Compensation (%eupnea), REM	4.78±25.08	0.31±10.95	0.970
Surgery type
Tongue base coblation	2 (16.7)	4 (30.8)	0.645
Palatoplasty	1 (8.3)	6 (46.2)	0.073
Nasal surgery	6 (50.0)	5 (38.5)	0.695
UPPP	8 (66.7)	5 (38.5)	0.238

Values are presented as mean±standard deviation or number (%).

AHI, apnea-hypopnea index; SpO₂, oxygen saturation; NREM, non-rapid eye movement; REM, rapid eye movement; %eupnea, percentage of eupneic drive (for arousal threshold) or percentage of eupneic ventilation (for V_passive and compensation); V_passive, ventilation at eupneic drive; UPPP, uvulopalatopharyngoplasty.

Table 2.

Endotypic traits before and after surgery among responders and non-responders to surgery

Variable	Responder (n=12)			Non-responder (n=13)			P-value^b)
Variable	Pre-surgery	Post-surgery	P-value^a)	Pre-surgery	Post-surgery	P-value^a)	P-value^b)
REM sleep
AHI	51.78±24.15	35.83±25.19	0.043	34.44±16.84	37.58±26.26	0.455	0.041
Arousal threshold	178.65±46.92	161.81±46.61	0.320	162.22±43.24	172.16±46.14	0.496	0.366
V_passive	35.47±22.64	57.77±24.31	0.007	53.84±29.02	45.62±23.04	0.301	0.021
Loop gain	0.48±0.20	0.49±0.13	0.966	0.42±0.10	0.50±0.13	0.055	0.407
Compensation	4.78±25.08	5.46±17.52	0.898	0.31±10.95	−0.57±13.33	1.000	0.893
NREM sleep
AHI	38.68±30.62	20.44±23.12	0.001	20.95±12.89	24.50±20.16	0.635	0.004
Arousal threshold	157.89±45.94	135.45±27.89	0.021	137.45±26.81	144.36±34.13	0.636	0.076
V_passive	58.63±23.88	69.90±18.95	0.092	71.24±13.46	66.27±25.60	0.893	0.078
Loop gain	0.55±0.19	0.47±0.11	0.129	0.48±0.12	0.46±0.15	0.735	0.325
Compensation	−11.61±19.41	−2.96±17.93	0.380	−4.83±22.48	−11.59±18.08	0.068	0.063

Values are presented as mean±standard deviation.

REM, rapid eye movement; AHI, apnea-hypopnea index; V_passive, ventilation at eupneic drive; NREM, non-rapid eye movement.

^a) Pre-surgery and post-surgery endotypic traits were compared between groups (responders vs. non-responders) using Wilcoxon signed-rank tests.

^b) The P-value for interaction between surgical state (preoperative and postoperative) and group was generated using generalized linear mixed models.

REFERENCES

1. Carberry JC, Amatoury J, Eckert DJ. Personalized management approach for OSA. Chest. 2018 Mar;153(3):744-55.

2. Messineo L, Magri R, Corda L, Pini L, Taranto-Montemurro L, Tantucci C. Phenotyping-based treatment improves obstructive sleep apnea symptoms and severity: a pilot study. Sleep Breath. 2017 Dec;21(4):861-8.

3. Bosi M, Incerti Parenti S, Sanna A, Plazzi G, De Vito A, Alessandri-Bonetti G. Non-continuous positive airway pressure treatment options in obstructive sleep apnoea: a pathophysiological perspective. Sleep Med Rev. 2021 Dec;60:101521.

4. Wong AM, Landry SA, Joosten SA, Thomson LD, Turton A, Stonehouse J, et al. Examining the impact of multilevel upper airway surgery on the obstructive sleep apnoea endotypes and their utility in predicting surgical outcomes. Respirology. 2022 Oct;27(10):890-9.

5. Schwartz AR, Schubert N, Rothman W, Godley F, Marsh B, Eisele D, et al. Effect of uvulopalatopharyngoplasty on upper airway collapsibility in obstructive sleep apnea. Am Rev Respir Dis. 1992 Mar;145(3):527-32.

6. Tucker Woodson B. Acute effects of palatopharyngoplasty on airway collapsibility. Otolaryngol Head Neck Surg. 1999 Jul;121(1):82-6.

7. Joosten SA, Leong P, Landry SA, Sands SA, Terrill PI, Mann D, et al. Loop gain predicts the response to upper airway surgery in patients with obstructive sleep apnea. Sleep. 2017 Jul;40(7):zsx094.

8. Li Y, Ye J, Han D, Zhao D, Cao X, Orr J, et al. The effect of upper airway surgery on loop gain in obstructive sleep apnea. J Clin Sleep Med. 2019 Jun;15(6):907-13.

9. Joosten SA, Landry SA, Wong AM, Mann DL, Terrill PI, Sands SA, et al. Assessing the physiologic endotypes responsible for REM- and NREM-based OSA. Chest. 2021 May;159(5):1998-2007.

10. Sands SA, Edwards BA, Terrill PI, Taranto-Montemurro L, Azarbarzin A, Marques M, et al. Phenotyping pharyngeal pathophysiology using polysomnography in patients with obstructive sleep apnea. Am J Respir Crit Care Med. 2018 May;197(9):1187-97.

11. Thornton AT, Singh P, Ruehland WR, Rochford PD. AASM criteria for scoring respiratory events: interaction between apnea sensor and hypopnea definition. Sleep. 2012 Mar;35(3):425-32.

12. Terrill PI, Edwards BA, Nemati S, Butler JP, Owens RL, Eckert DJ, et al. Quantifying the ventilatory control contribution to sleep apnoea using polysomnography. Eur Respir J. 2015 Feb;45(2):408-18.

13. Finnsson E, Olafsdottir GH, Loftsdottir DL, Jonsson SA, Helgadottir H, Agustsson JS, et al. A scalable method of determining physiological endotypes of sleep apnea from a polysomnographic sleep study. Sleep. 2021 Jan;44(1):zsaa168.

14. Sands SA, Terrill PI, Edwards BA, Taranto Montemurro L, Azarbarzin A, Marques M, et al. Quantifying the arousal threshold using polysomnography in obstructive sleep apnea. Sleep. 2018 Jan;41(1):zsx183.

15. Sands SA, Alex RM, Mann D, Vena D, Terrill PI, Gell LK, et al. Pathophysiology underlying demographic and obesity determinants of sleep apnea severity. Ann Am Thorac Soc. 2023 Mar;20(3):440-9.

16. Capasso R, Rosa T, Tsou DY, Nekhendzy V, Drover D, Collins J, et al. Variable findings for drug-induced sleep endoscopy in obstructive sleep apnea with propofol versus dexmedetomidine. Otolaryngol Head Neck Surg. 2016 Apr;154(4):765-70.

17. Li HY, Lee LA. Relocation pharyngoplasty for obstructive sleep apnea. Laryngoscope. 2009 Dec;119(12):2472-7.

18. Li HY, Lee LA, Kezirian EJ, Nakayama M. Suspension palatoplasty for obstructive sleep apnea: a preliminary study. Sci Rep. 2018 Mar;8(1):4224.

19. Li HY. Palatal surgery for obstructive sleep apnea: from ablation to reconstruction. Sleep Med Clin. 2019 Mar;14(1):51-8.

20. Li HY, Lee LA, Kezirian EJ. Efficacy of coblation endoscopic lingual lightening in multilevel surgery for obstructive sleep apnea. JAMA Otolaryngol Head Neck Surg. 2016 May;142(5):438-43.

21. Li HY, Lee LA, Hsin LJ, Fang TJ, Lin WN, Chen HC, et al. Intrapharyngeal surgery with integrated treatment for obstructive sleep apnea. Biomed J. 2019 Apr;42(2):84-92.

22. Kinouchi T, Terada J, Sakao S, Koshikawa K, Sasaki T, Sugiyama A, et al. Effects of the combination of atomoxetine and oxybutynin in Japanese patients with obstructive sleep apnoea: a randomized controlled crossover trial. Respirology. 2023 Mar;28(3):273-80.

23. Almeneessier AS, Almousa Y, Hammad O, Olaish AH, ALAnbay ET, BaHammam AS. Long-term adherence to continuous positive airway pressure in patients with rapid eye movement-only obstructive sleep apnea: a prospective cohort study. J Thorac Dis. 2017 Oct;9(10):3755-65.

24. Ishii L, Roxbury C, Godoy A, Ishman S, Ishii M. Does nasal surgery improve OSA in patients with nasal obstruction and OSA?: a meta-analysis. Otolaryngol Head Neck Surg. 2015 Sep;153(3):326-33.

25. Schoustra E, van Maanen P, den Haan C, Ravesloot MJ, de Vries N. The role of isolated nasal surgery in obstructive sleep apnea therapy: a systematic review. Brain Sci. 2022 Oct;12(11):1446.

26. Berry RB, Kouchi KG, Der DE, Dickel MJ, Light RW. Sleep apnea impairs the arousal response to airway occlusion. Chest. 1996 Jun;109(6):1490-6.

27. Haba-Rubio J, Sforza E, Weiss T, Schroder C, Krieger J. Effect of CPAP treatment on inspiratory arousal threshold during NREM sleep in OSAS. Sleep Breath. 2005 Mar;9(1):12-9.

28. Hoshino T, Sasanabe R, Murotani K, Hori R, Mano M, Nomura A, et al. Estimated respiratory arousal threshold in patients with rapid eye movement obstructive sleep apnea. Sleep Breath. 2022 Mar;26(1):347-53.

29. Li Y, Ye J, Han D, Cao X, Ding X, Zhang Y, et al. Physiology-based modeling may predict surgical treatment outcome for obstructive sleep apnea. J Clin Sleep Med. 2017 Sep;13(9):1029-37.

30. Eun YG, Kwon KH, Shin SY, Lee KH, Byun JY, Kim SW. Multilevel surgery in patients with rapid eye movement-related obstructive sleep apnea. Otolaryngol Head Neck Surg. 2009 Apr;140(4):536-41.

31. Hang LW, Tsai YC, Finnsson E, Agustsson JS, Sands SA, Cheng WJ. Sex-specific age-related worsening of pathological endotypic traits in patients with obstructive sleep apnea. Sleep. 2025 Jan;48(1):zsae185.

32. Landry SA, Joosten SA, Thomson LD, Turton A, Wong AM, Leong P, et al. Effect of hypopnea scoring criteria on noninvasive assessment of loop gain and surgical outcome prediction. Ann Am Thorac Soc. 2020 Apr;17(4):484-91.

33. Labarca G, Dreyse J, Salas C, Letelier F, Jorquera J. A validation study of four different cluster analyses of OSA and the incidence of cardiovascular mortality in a Hispanic population. Chest. 2021 Dec;160(6):2266-74.

34. Bock JM, Vungarala S, Karim S, Somers VK. Obstructive sleep apnea as a cardiovascular risk factor-beyond CPAP. Can J Cardiol. 2021 May;37(5):756-65.