Airway Collapsibility during Drug-Induced Sleep Endoscopy Relates with Clinical Features of OSA

Abstract

Introduction:

Increased pharyngeal collapsibility leads to obstructive sleep apnea (OSA). Positive airway pressure titration during drug-induced sleep endoscopy (DISE-PAP) provides objective collapsibility metrics, the pharyngeal opening pressure (PhOP) and active pharyngeal critical pressure (Pcrit_A). We examined the interrelationships between risk factors of OSA, airway collapsibility measures, and clinical manifestations of the disease.

Methods:

This is a cross-sectional analysis of consecutive OSA patients undergoing DISE-PAP. Nasal PAP was increased stepwise until inspiratory flow limitation was abolished, signifying PhOP. Pcrit_A was derived from the resulting titration pressure-flow relationships. Clinical data including demographics, anthropometrics, sleep studies and patient-symptom questionnaires were obtained from the electronic medical record. Multivariate regression was used to evaluate the relationship between risk factors, airway collapsibility, and clinical data.

Results:

On average, the 164 patients meeting inclusion criteria were middle-aged (54.2±14.7 years), overweight-obese (BMI 29.9±4.5kg/m²), male (72.6%), White (79.3%) and had severe OSA (AHI 32.0±20.5 events/hr). Mean PhOP was 7.5±3.3 cm H₂O and mean Pcrit_A was 0.80±3.70 cm H₂O. Younger age (Standardized β=−0.191,p=.015) and higher BMI (Standardized β=0.176,p=0.028) were associated with higher PhOP, but not Pcrit_A. PhOP and Pcrit_A were both associated with AHI, supine AHI, and SpO₂ nadir. Higher PhOP was associated with higher snoring scores (Standardized β=0.246,p=0.008), but not other patient-reported outcomes.

Conclusion:

Objective assessment of passive and active airway mechanics during DISE relates with clinical risk factors for OSA. Quantitative measures of collapsibility provide accessible and meaningful data, enhancing the standard sleep surgery evaluation.

Level of Evidence:

4

Lay Summary

Pharyngeal opening pressure (PhOP) and active pharyngeal critical pressure (Pcrit_A) are objective measures of upper airway collapsibility obtained during drug-induced sleep endoscopy with positive airway pressure (DISE-PAP) titration and are related to clinical features of obstructive sleep apnea. Our data support the inclusion of airway collapsibility metrics as a complementary component of the sleep surgery evaluation.

Introduction

Obstructive sleep apnea (OSA) is a sleep-related breathing disorder characterized by repetitive episodes of pharyngeal collapse resulting in airflow obstruction, recurrent arousals and intermittent hypoxemia.¹ Untreated OSA contributes to increased prevalence and severity of cardiovascular disease, neurocognitive deficits, and metabolic dysfunction.^2,3,4,5 Continuous positive airway pressure (CPAP) tolerability is relatively low, prompting patients to seek alternatives to PAP therapy such as upper airway surgery.^6,7 As PAP alternative therapy outcomes are mixed, there is a need to better select PAP alternative patients for the appropriate surgical or non-surgical interventions.^8,9,10

To provide context for a patient’s disease burden, OSA clinicians commonly obtain three categories of patient data: anthropometrics/demographics, polysomnography (PSG), and patient-reported outcome measures (PROMs).¹¹ In current clinical practice, OSA severity is defined by the apnea-hypopnea index (AHI) and treatment response by reductions in AHI.¹² However, AHI does not consistently correlate with symptomatology, fully characterize disease burden, or reliably predict treatment response.^13,14,15 Validated patient-reported outcome measures (PROMs) such as the Epworth Sleepiness Scale (ESS) characterize symptoms amongst patients with OSA.^{16,17,18,19,20,21} Despite the availability of these diverse data, their value in directing the PAP-alternative treatment pathway remains limited.

Patients with OSA may undergo drug-induced sleep endoscopy (DISE) to identify the site, pattern, and degree of upper airway collapse, but subjectivity and reproducibility are known limitations.^{22,23,24,25,26,27,28} DISE with administration of PAP (DISE-PAP) combines endoscopic evaluation with objective quantification of upper airway collapsibility via the pharyngeal critical closing pressure (Pcrit) and pharyngeal opening pressure (PhOP).^29,30,31 Upper airway collapsibility is a known pathogenic factor in OSA, and has primarily been studied in natural sleep with the pharyngeal critical pressure (Pcrit).³² Thus, understanding collapsibility measures obtained during DISE and their relationships to known risk factors and clinical manifestations of OSA is paramount. Previous data have shown therapeutic PAP level to predict airway collapsibility, and therapeutic PAP level to associate with AHI.^11,33 As collapsibility can be quantified with DISE-PAP, there is a need to understand how its values associate with other disease features of OSA. Understanding DISE-obtained measures of collapsibility may facilitate their inclusion in the evaluation of the PAP-alternative patient and may modify surgical management of OSA.

Our primary aim was to investigate the relationships between traditional anthropometric and demographic risk factors (age, race, sex, and BMI) and upper airway collapsibility. We hypothesized that the presence of risk factors would associate with greater airway collapsibility. Our secondary aim was to investigate the relationships between airway collapsibility and clinical manifestations of OSA. We hypothesized greater collapsibility was reflected in clinical data suggestive of more severe OSA. Finally, we explored the relationship between upper airway collapsibility and PROMs. To examine these questions, we obtained demographic and anthropometric data, PSG data, PROMs, and collapsibility metrics during DISE-PAP in PAP alternative OSA patients (Figure 1).

Figure 1:

Schematic of the primary and secondary aims of this investigation. Risk factors defined as age, body mass index (BMI), race, and sex. Upper airway collapsibility is measured as the active pharyngeal critical pressure (Pcrit_A) and pharyngeal opening pressure (PhOP). PSG data includes total, supine, and non-supine apnea-hypopnea indices (AHI), SpO₂ nadir, and total sleep time spent under 90% SpO₂. Patient reported outcome measures (PROMs) include the Epworth Sleepiness Scale (ESS), Functional Outcomes of Sleepiness (FOSQ), Insomnia Severity Index (ISI), Nasal Obstructive Symptom Evaluation (NOSE), and visual analog scores (VAS) of snoring and sleep quality.

Methods

Participants

We performed an analysis of a cohort of OSA patients seeking PAP alternatives. All patients underwent DISE-PAP at the Hospital of the University of Pennsylvania. The study was approved by the Institutional Review Board at the Hospital of the University of Pennsylvania (IRB #s 833511, 849542, and 850115). Inclusion criteria were as follows: all patients 18 years or older diagnosed with OSA (AHI > 5) via prior sleep study either by home sleep apnea testing or overnight PSG. Exclusion criteria were as follows: primary central sleep apnea, protocol deviations (mouth tape during DISE-PAP, different mask types), signal quality issues, and patients that were unable to be titrated due to non-flow limited breathing during DISE.

Clinical Data

Demographic and anthropometric data:

Patient demographic and anthropometric data (age, sex, race/ethnicity, body mass index [BMI]) were recorded at the initial clinic visit and recorded from the electronic medical record (EMR).

Polysomnographic data:

All sleep studies were scored based on the American Academy of Sleep Medicine (AASM) 2012 guidelines.³⁴ Given diverse referral sources, patients included had either both hypopnea definitions: 4% desaturation or 3% desaturation and/or arousal. For patients with multiple PSGs, the data closest in time to the DISE-PAP procedure were recorded from the EMR.

Patient-Reported Outcome Measures (PROMs):

Patient-reported outcome surveys were administered during each patient’s initial clinic visit. PROMs included the Epworth Sleepiness Scale (ESS), Insomnia Severity Index (ISI), Functional Outcomes of Sleep Questionnaire-10 (FOSQ-10), Nasal Obstruction Symptom Evaluation (NOSE), and visual analog scale (VAS) of sleep quality and snoring. The data were then recorded from the EMR.

Drug Induced Sleep Endoscopy with Positive Airway Pressure Titration

The DISE-PAP evaluation was performed on an integrated recording platform designed to acquire clinically relevant anatomic and physiologic characteristics of each patient’s upper airway. A detailed description of the DISE-PAP setup has been previously published.³⁵ In brief, patients are positioned supine with the neck in neutral position and mouth in its natural position. Propofol anesthesia was administered to achieve sedation using a probability ramp infusion system as previously described, with a target bispectral index range of 50 – 70.³⁶ The nasolaryngoscope was passed through a custom-fitted mask into the nasal cavity. A pressure-sensitive catheter (Mikro-cath^™, Millar, Houston, TX) was passed intra-nasally to the retro-epiglottic space to measure respiratory effort.³⁷ CPAP (S9 VPAP, ResMed Inc., San Diego CA) was applied and titrated using a modified nasal mask (Pulmodyne, Indianapolis, IN) attached to a pneumotachometer for airflow monitoring. Pulse oximetry was monitored, and a dual lumen oral cannula provided supplemental O₂ (2–4 L/min) and measured end-tidal CO₂ to detect mouth breathing. After respiratory events were observed at atmospheric pressure, PAP administration began. PAP was administered at 2 cm H2O, 4 cm H2O, and then increased by increments of 1 cm H2O until flow limitation was abolished. After the patient reached a threshold where they no longer cycled between arousals, hypopneas, and apneas, a minimum of six breaths were evaluated before increasing PAP. Once flow limitation was abolished for greater than 50% of breaths at a given level for at least 6 breaths, PhOP was achieved. Pcrit_A were determined by evaluating pressure-flow relationships from DISE-PAP titration. At each pressure level, three breaths with minimum tidal volume and flow-maximum were selected for analysis and plotted along a pressure-flow curve. After calculation, the pressure level at which flow intersected the x-axis was determined to be Pcrit_A.

Definition of Collapsibility Measures

• Pharyngeal Opening Pressure (PhOP): the DISE-PAP titration level at which inspiratory flow limitation was abolished for >50% of breaths.

  • Passive airway collapsibility: Prolonged PAP administration diminishes neuromuscular responses in the upper airway.
• Pharyngeal critical pressure (Pcrit_A): the DISE-PAP level at which airflow initially commenced.

  • Active Pharyngeal critical Pressure (Pcrit_A): As neuromuscular responses to flow limitation are still present, this is considered an active Pcrit_A.

Statistical Analyses

Continuous variables were reported as means and standard deviations or means and ranges, as appropriate, and categorical variables using frequencies and percentages. To examine associations between patient characteristics (age, sex [male vs. female], race/ethnicity [White vs. not White], BMI) and collapsibility, we utilized linear regression with PhOP and Pcrit_A as continuous outcomes. To examine associations between collapsibility and polysomnography data (AHI, supine and non-supine AHI, SpO₂ nadir, TST <90% SpO₂) and PROMs (ESS, ISI, FOSQ-10, NOSE, Sleep quality VAS, Snoring VAS), similar modeling was used with the PSG/PROM data as continuous outcomes and PhOP and Pcrit_A as predictors. Modeling was performed adjusted for age, sex, BMI, and race. Results are reported as the model beta-coefficient (equal to the change in outcome for 1 unit increase in predictor) and standard error (SE) and associated p-value. A standardized beta, equal to the SD change in outcome for a 1 SD increase in predictor, was also calculated. Median split analysis was performed, categorizing patients by age and BMI, and mean (SD) Pcrit_A and PhOP were calculated. Unpaired, two-sided t-tests were used to compare means. Mean (SD) PhOP and Pcrit_A were also categorized by AHI severity (mild [5–15 events/hour], moderate [15–30 events/hour], and severe [> 30 events/hour]). Single-factor ANOVA was used to compare collapsibility measures across OSA severities. Reflecting the descriptive nature of the analyses, statistical significance was based on a P value of < 0.05. Analyses were performed using Stata/SE v14.3 (StataCorp, College Station, TX) and SAS Version 9.4 (SAS Institute, Cary, NC).

Results

Participant Characteristics

Between June 2020 and July 2022, 202 patients underwent DISE-PAP at the Hospital of the University of Pennsylvania. Of these, 164 met inclusion criteria and 38 were excluded: 8 (21%) had technical issues, 9 (24%) had protocol deviations, 16 (42%) had poor flow signal, 3 (8%) had non-flow limited breathing before PAP administration, and 2 (5%) patients did not achieve PhOP after maximal titration (20 cm H₂O). See Figure 3 for CONSORT diagram. Our patients had heterogenous sleep studies, with 94 sleep studies scored with 4% desaturation criteria, 10 had sleep studies scored with 3% desaturation criteria, and 60 sleep studies with unverifiable scoring criteria due to absence of original report. Finally, 59 patients had overnight polysomnography, while the remaining 105 had home sleep apnea testing.

Figure 3:

CONSORT diagram for patients included in this study.

Baseline clinical data and collapsibility measures of our study sample are summarized in Table 1. On average, patients were middle-aged (54.2±14.7 years) and overweight/obese (BMI 29.9±4.5 kg/m²), and a majority were male (72.6%) and White (79.3%). The group, on average, had severe OSA (AHI 32.0±20.5 events/hour), with mean PhOP of 7.5±3.3 cm H₂O (n=164) and mean Pcrit_A of 0.8±3.7 cm H₂O (n=104).

Table 1:

Describes the baseline demographics/anthropometrics, polysomnographic data (PSG), and patient-reported outcomes (PROMs) of the patients in our cohort. AHI = Apnea-hypopnea index. TST < 90% SpO₂ = total sleep time spent under 90% SpO₂. VAS = Visual Analog Scale. NOSE = Nasal obstruction symptom evaluation. ESS = Epworth Sleepiness Scale. ISI = Insomnia Severity Index. FOSQ-10 = Functional Outcomes of Sleepiness. Pcrit_A = active pharyngeal closing pressure. PhOP = pharyngeal opening pressure.

	Sample Size	Mean	SD
Anthropometric and Demographic Risk Factors
Age (years)	164	54.16	14.67
BMI (kg/m2)	164	29.7	4.54
White vs Non-White %	130	79.3%
Male (%)	119	72.6%
PSG Data
AHI (events/hr)	164	31.98	20.42
Supine AHI (events/hr)	133	43.79	26.49
Non-Supine AHI (events/hr)	111	19.70	18.18
SpO2 Nadir (%)	163	78.89	8.84
TST <90% O2 (min)	140	12.70	16.12
PROMs
Sleep Quality VAS	144	66.68	21.19
Snoring VAS	125	58.68	33.59
NOSE	147	39.56	24.56
ESS	153	9.57	5.39
ISI	144	16.14	5.22
FOSQ-10	145	2.92	0.65
Collapsibility Measures
PcritA (cm H2O)	104	0.80	3.70
PhOP (cm H 2 O)	164	7.49	3.28

PhOP and Pcrit_A were compared above and below the median age (55 years) and BMI (30 kg/m²) to better understand these collapsibility measures. Further, PhOP and Pcrit_A were compared by OSA severity. The mean PhOP and Pcrit_A in patients over 55 years old were 6.7±2.4 and 0.3±3.0 cm H₂O, respectively. The mean PhOP and Pcrit_A in patients under 55 years old were 8.1±3.8 and 1.3±4.3 cm H₂O, respectively. The mean PhOP and Pcrit_A in patients with BMI of 30 kg/m² or more were 7.9±3.7 and 1.2±3.6 cm H₂O, respectively. The mean PhOP and Pcrit_A in patients with BMI less than 30 kg/m² were 7.0±2.8 and 0.4±3.8 cm H₂O, respectively. Finally, the mean PhOP and Pcrit_A in patients with mild, moderate, and severe OSA were 6.2±2.9 and −0.2±2.6 cm H₂O, 7.5±2.8 and 0.8±4.4 cm H₂O, and 8.0±3.6 and 1.4±3.5 cm H₂O.

Associations between Demographic and Anthropometric Risk Factors and PhOP/Pcrit_A

Adjusted associations between demographic and anthropometric variables and PhOP/Pcrit_A are shown in Table 2. In multivariate analyses, younger age (Standardized β = −0.191; p=0.015) and higher BMI (Standardized β = 0.176; p=0.028) were independently associated with PhOP. Every 1-year increase in age was associated with a −0.043 cm H₂O decrement in PhOP, and every 1-unit increase in BMI was associated with a 0.127 cm H₂O increase in PhOP. Age and BMI did not associate with Pcrit_A. Race and gender were not associated with either upper airway collapsibility measure obtained during DISE.

Table 2:

Table showing the associations between clinical risk factors of OSA and measures of airway collapsibility obtained during DISE-PAP. BMI = Body mass index. Pcrit_A = Active pharyngeal critical pressure. PhOP = Pharyngeal opening pressure.

	PcritA			PhOP
	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p
Demo/Anthro
Age (years)	−0.031 ± 0.027	−0.121	0.24	−0.043 ± 0.017	−0.191	0.015
BMI (kg/m2)	0.013 ± 0.078	0.017	0.86	0.127 ± 0.057	0.176	0.028
Male gender (vs Female)	0.786 ± 0.833	0.096	0.35	0.741 ± 0.579	0.101	0.20
White Race (vs non-White)	0.253 ± 1.754	0.027	0.89	1.154 ± 1.183	0.143	0.33

^*Expected change in Pcrit_A/PhOP for a 1-unit change in demographic/anthropometric variable.

^**Expected change in Pcrit_A/PhOP for a 1-unit change in demographic/anthropometric variable.

Associations between PhOP/Pcrit_A and OSA Manifestations

Polysomnographic Data

Adjusted associations between PhOP/Pcrit_A are shown in Table 3. In multivariate analyses, higher PhOP and Pcrit_A was associated with higher AHI (Std. β = 0.293; p<0.0001, Std. β = 0.273; p<0.0001, respectively). PhOP and Pcrit_A were associated with higher supine AHI (Std. β = 0.255; p=0.002, Std. β = 0.282; p=0.002, respectively), but not non-supine AHI. PhOP and Pcrit_A were associated with lower SpO2 nadir (Std. β = −0.166; p=0.029, Std. β = −0.334; p<0.0001, respectively), but not total sleep time spent under <90% SpO₂.

Table 3:

This table describes the associations between measures of airway collapsibility and polysomnographic data.

	AHI			Supine AHI			Non-supine AHI			SaO2 Nadir (%)			TST < 90%
	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p
PhOP	1.8 ± 0.45	0.29	8.9E-05	2.1 ± 0.67	0.26	0.0018	0.63 ± 0.56	0.097	0.27	−0.44 ± 0.20	−0.17	0.029	0.49 ± 0.42	0.095	0.25
PcritA	1.6 ± 0.5	0.27	0.0018	2.0 ± 0.74	0.28	0.0072	0.92 ± 0.78	0.15	0.24	−0.79 ± 0.2	−0.33	0.00014	0.35 ± 0.48	0.072	0.47

^*Changes in PSG value based on a 1-unit change in Pcrit_A/PhOP.

^**Changes in PSG value based on a 1-unit change in Pcrit_A/PhOP.

Patient-Reported Outcome Measures (PROMs)

The relationships between PhOP/Pcrit_A on PROMs is described in Table 4. In adjusted analyses, higher PhOP, but not Pcrit_A was associated with greater snoring VAS (Standardized β = 0.246; p=0.008). PhOP and Pcrit_A were not associated with Sleep Quality VAS, NOSE, ESS, ISI, or FOSQ-10.

Table 4:

This table describes the associations between measures of airway collapsibility and patient reported outcome measures.

	ESS			ISI			FOSQ			NOSE			SQVAS			Snoring VAS
	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p	β ± SE*	Std. β**	p
PhOP	0.2 ± 0.1	0.11	0.2	−0.06 ± 0.1	−0.04	0.7	0.02 ± 0.02	0.1	0.2	0.4 ± 0.6	0.06	0.5	−0.3 ± 0.6	−0.05	0.6	2.6 ± 1.0	0.3	0.008
PcritA	0.1 ± 0.1	0.1	0.3	−0.09 ± 0.2	−0.06	0.6	0.006 ± 0.02	0.04	0.7	0.6 ± 0.6	0.09	0.4	−0.2 ± 0.6	−0.03	0.8	−1.3 ± 1.0	−0.2	0.2

^*Changes in PROM score based on a 1-unit change in Pcrit_A/PhOP.

^**Expected change in SD of PROM score based on 1-SD change in Pcrit_A/PhOP.

Discussion

In this study, we examined the relationships between risk factors for OSA, objective measures of airway collapsibility obtained during DISE, and clinical and subjective manifestations of OSA. Our work demonstrates several relationships between these clinical data. First, younger age and higher BMI were associated with higher PhOP values, but not higher Pcrit_A. Next, both PhOP and Pcrit_A were associated with total AHI, supine AHI, and SpO₂ nadir, but neither correlated with non-supine AHI or TST < 90%. Additionally, PhOP was associated with higher patient-reported snoring, but not with other patient-reported outcomes.

In our analyses, higher BMI and younger age were associated with higher PhOP, but not Pcrit_A. First, the association between higher BMI and PhOP supports the construct of PhOP as a collapsibility measure given the strong relationship between obesity and OSA.^38,39 Next, the association between younger age and higher PhOP provides context for PAP alternative patient presentation. Despite existing literature suggesting OSA severity increases with age,^40,41 younger patients in our sample tended to have higher PhOP, a passive collapsibility measure. Older patients presenting to our PAP alternatives clinic tend to have lower PhOP, which is supported by evidence reporting lower CPAP requirements in elderly patients compared to younger patients, matched for disease severity, BMI, and neck circumference.⁴² As older patients tend to have higher AHI both in published literature and our sample, non-anatomic factors^1,43,44 such as arousal threshold, respiratory control, and neuromuscular capability may play a greater role in their OSA pathogenesis (Figure 4). Finally, we note significant associations of BMI and age with PhOP, but not Pcrit_A. As Pcrit_A is an “active” measure of airway collapsibility (with intact neuromuscular activity), it is likely that differing levels of neuromuscular tone complicate its relationships with these established risk factors. Our data support the additive value of obtaining objective measures of airway collapsibility in the workup of patients seeking PAP alternative therapies.

Figure 4:

Presentation of a PAP Alternative OSA Patient: Drives of Pathogenesis. Anatomic and non-anatomic factors contribute to OSA severity, measured by AHI. The anatomic load is defined as passive pharyngeal collapsibility (PhOP). The non-anatomic contributors are arousal threshold, ventilatory control instability, and neuromuscular responsiveness. In younger patients, anatomic load may play a larger role in OSA pathogenesis versus in older patients.

In our cohort, PhOP and Pcrit_A were both associated with higher AHI, supine AHI, and O₂ nadir, but not with non-supine AHI or TST < 90% O₂. PhOP also correlated with higher snoring scores. The associations with supine, but not non-supine AHI can be explained by patient positioning during DISE-PAP. DISE-PAP is performed with the patient in supine position, suggesting PhOP most directly reflects supine sleep physiology.²⁹ The correlations between Pcrit_A, PhOP and lower SpO₂ nadir suggest worse airway collapsibility associates with greater desaturations during sleep. Finally, the lack of association with total sleep time spent < 90% SpO₂ could be due to the presence of comorbid cardiopulmonary disease, obesity hypoventilation syndrome, or non-anatomic features such as loop gain.^45,46 The relationship between PhOP and snoring lend support to the construct of PhOP as a passive collapsibility measure. Snoring results from soft tissue flutter induced by respiration in combination with insufficient neuromuscular control during sleep and may be driven by increased muscle thickness or tonsillar hyperplasia.^47,48,49 Pcrit_A, as an active measure of collapsibility, may therefore relate less with snoring. Taken together, our data support Pcrit_A and PhOP to be reflective of the physiologic manifestation of the disease (e.g. PSG data) but less so, the symptomatic manifestations (e.g. PROMs).

DISE evaluation has traditionally relied on subjective evaluation of airway collapse pattern, most commonly the VOTE classification system.²⁸ A previous investigation relating clinical variables with VOTE scoring has shown multilevel collapse to associate with higher AHI and ESS, but not with BMI.⁵⁰ Further, the collapse pattern reported during DISE does not inform of collapsibility, the anatomic pathogenic factor in OSA. Herein, we demonstrated airway collapsibility measures, PhOP and Pcrit_A, associate with clinical features of OSA as well. These measures enable objective characterization of airway collapsibility and may complement the information provided by VOTE scoring during DISE.

Notably, Pcrit_A and PhOP bear key differences in terms of acquisition and clinical implications. Given that Pcrit_A values are often sub-atmospheric, pressure-flow curves must be constructed using data from pneumotachometer and recording platform during DISE. PhOP, on the other hand, occurs with positive pressure values that can be gleaned solely from a PAP machine with ramp feature.^51,52 A previous investigation by our group has shown opening pressures during DISE to predict response to hypoglossal nerve stimulation (HGNS).²⁹ Opening pressures < 8 cm H₂O had positive predictive value for response to HGNS. On average, younger patients (< 55 years) and patients with high BMI ( >30 kg/m²) had PhOP of 8 cm H₂O (8.14 and 7.94 cm H₂O, respectively). Similarly, patients with severe OSA had mean PhOP 8.02 cm H₂O. Taken together, these data suggest PhOP values greater than 8 cm H₂O may be suggestive of high collapsibility, though more work is needed to understand these values and their variability. Compared with PhOP, the composition, influences, and prognostic value of Pcrit_A during DISE are not well understood, particularly given depressant effects of propofol on neuromechanical airway control.⁵³ Further studies should aim to investigate the relationship between Pcrit_A and other active features of the airway such as respiratory effort indices during obstruction and genioglossus electromyography. Overall, Pcrit_A and PhOP may be useful clinical measures to predict response to airway surgery and assess changes in anatomic collapsibility after surgery for sleep apnea.

This study has some limitations. Given the cross-sectional study design, our data do not address how collapsibility changes over time within individual patients. With regards to PSG data, patients presented with heterogeneous sleep studies, resulting in varied AHI scoring criteria. With regards to DISE, instability of sedation plane can impact respiratory function and may confound these collapsibility measures.⁵⁴ Finally, the test-retest reliability of PhOP and Pcrit_A obtained during DISE has not yet been evaluated.

This study also has some notable strengths. First, this was a large cohort (n=164) and all patients received care at the same institution. Our DISE platform offers a controlled environment which enables synchronized evaluation of airway anatomy and physiology. Finally, our DISE-PAP protocol standardizes patient positioning, mitigating the influence of confounding features such as neck extension on collapsibility measures.

Conclusion

Objective assessment of airway collapsibility during DISE relates with risk factors and clinical manifestations of OSA. Our study suggests additive value of characterizing collapsibility during DISE, as younger patients presenting for PAP-alternatives tend to have higher collapsibility than older patients. As PhOP and Pcrit_A convey passive and active mechanics of the airway, these measures of collapsibility provide accessible and meaningful data which enhance the standard sleep surgery evaluation.

Figure 2:

Pressure-flow curve of DISE-PAP. No flow was seen at atmospheric pressure. At nasal pressure level 1 cm H₂O, endoscopy represents complete velopharyngeal obstruction and is reflected in complete apnea on nasal flow. After this point, airflow commences, demarcating 1 cm H₂O to be the active pharyngeal critical pressure (Pcrit_A). As PAP titration occurs, flow limitation is eventually abolished and the velopharynx is opened, signifying the pharyngeal opening pressure (PhOP) at 10 cm H₂O.