Comparison of upper airway obstruction during zolpidem-induced sleep and propofol-induced sleep in patients with obstructive sleep apnea: a pilot study

Abstract

Study Objectives:

Drug-induced sleep endoscopy (DISE) using propofol is commonly used to identify the pharyngeal structure involved in collapse among patients with obstructive sleep apnea. DISE has never been compared with zolpidem-induced sleep endoscopy. We hypothesized that propofol at recommended sedation levels does not influence upper airway collapsibility nor the frequency of multilevel pharyngeal collapse as compared with zolpidem-induced sleep.

Methods:

Twenty-one patients with obstructive sleep apnea underwent polysomnography and sleep endoscopy during zolpidem-induced sleep and during DISE with propofol. A propofol target-controlled infusion was titrated to achieve a bispectral index between 50 and 70. Airway collapsibility was estimated and compared in both conditions by peak inspiratory flow and the magnitude of negative effort dependence. Respiratory drive was estimated by the difference between end-expiratory and peak-negative inspiratory pharyngeal pressure (driving pressure). Site and configuration of pharyngeal collapse during zolpidem-induced sleep and DISE with propofol were compared.

Results:

The frequency of multilevel collapse during zolpidem-induced sleep was similar to that observed during DISE with propofol (72% vs 86%, respectively; difference: 14%; 95% confidence interval: −12% to 40%; P = .453). The endoscopic classification of pharyngeal collapse during both conditions were similar. Peak inspiratory flow, respiratory drive (effect size: 0.05 and 0.03, respectively), and negative effort dependence (difference: −6%; 95% confidence interval: −16% to 4%) were also similar in both procedures.

Conclusions:

In this pilot study, recommended propofol doses did not significantly increase multilevel pharyngeal collapse or affect upper airway collapsibility and respiratory drive as compared with zolpidem-induced sleep.

Clinical Trial Registration:

Registry: clinicaltrials.gov; Name: Natural and Drug Sleep Endoscopy; URL: https://clinicaltrials.gov/ct2/show/study/NCT03004014; Identifier: NCT03004014.

Citation:

Ordones AB, Grad GF, Cahali MB, Lorenzi-Filho G, Sennes LU, Genta PR. Comparison of upper airway obstruction during zolpidem-induced sleep and propofol-induced sleep in patients with obstructive sleep apnea: a pilot study. J Clin Sleep Med. 2020;16(5):725–732.

BRIEF SUMMARY

Current Knowledge/Study Rationale: Drug-induced sleep endoscopy with propofol is used to identify the site of collapse of the airway in patients with obstructive sleep apnea. However, propofol may increase upper airway collapsibility.

Study Impact: Drug-induced sleep endoscopy with propofol has never been compared with zolpidem-induced sleep endoscopy; therefore, we performed endoscopy during zolpidem-induced sleep and drug-induced sleep endoscopy with propofol in patients with obstructive sleep apnea. Propofol used at the recommended sedation levels does not influence the results of sleep endoscopy as compared with zolpidem-induced sleep.

INTRODUCTION

Obstructive sleep apnea (OSA) is a common disorder characterized by repetitive pharyngeal collapse during sleep.^1,2 One or more pharyngeal structures may collapse during sleep in patients with OSA, including the soft palate, lateral pharyngeal walls, tongue base, and epiglottis. The recognition of the pharyngeal structure involved in collapse allows site-specific treatment for OSA.^3,4

Drug-induced sleep endoscopy (DISE) has been used since 1991 to identify the pharyngeal structures causing collapse.^5–9 The major applications of DISE are to guide the surgical approach for OSA, including the selection of candidates for hypoglossal nerve stimulation.^8,10–12 Although sleep induction using sedatives has been used to facilitate the endoscopic procedure, DISE has never been compared with zolpidem-induced sleep endoscopy.¹³ We have previously shown that low levels of midazolam did not increase upper airway collapsibility.¹⁴ Currently, DISE is most commonly performed under propofol sedation, frequently without sedation depth monitoring.¹⁵ Upper airway collapsibility, respiratory drive, and site and configuration of pharyngeal collapse change according to depth of sedation during incremental propofol dose.^16–20 Reports of DISE under propofol sedation have shown that up to 76% of individuals with OSA have >1 pharyngeal structure involved in airway obstruction.^7,9,21,22 The influence of propofol sedation in multisite pharyngeal collapse is not known.

The amplitude of inspiratory flow as measured by peak inspiratory flow during inspiratory flow limitation has been shown to be associated with airway collapsibility and airway dimensions.²³ Negative effort dependence (NED), defined as the percentage reduction from peak to midinspiratory flow, reflects the interaction between respiratory effort and the dynamic compliance of the structure responsible for collapse.²⁴ Sedatives could possibly influence NED by either decreasing respiratory effort or increasing the compliance of the pharyngeal structure involved in collapse.²⁴

We hypothesized that propofol at recommended sedation levels (bispectral index [BIS]: 50 to 70), which corresponds to loss of consciousness, does not increase the number of pharyngeal structures involved in collapse as compared with zolpidem-induced sleep endoscopy.^15,17 We also hypothesized that DISE under recommended levels of propofol sedation does not increase upper airway collapsibility and dynamic compliance of pharyngeal structures. To test these hypotheses, individuals with OSA underwent zolpidem and propofol-induced sleep endoscopy.

METHODS

Participants

This pilot study was conducted at the Clinics Hospital, University of São Paulo, Brazil. Patients scheduled for pharyngeal surgery to treat OSA aged between 18 and 70 years were invited to participate. OSA was defined by an apnea-hypopnea index (AHI) of >5 events/h. Patients were excluded if they had nasal obstruction, significantly deviated nasal septum, nasal turbinate hypertrophy, or nasal polyps. Patients with uncontrolled heart failure, renal insufficiency, diabetes, or hyperthyroidism were also excluded. Patients were also excluded if taking medications that could affect upper airway muscle function, such as benzodiazepines and muscle relaxants.²⁵ All participants gave written consent before study entry. The study was approved by the Clinics Hospital Ethics Committee and registered at clinicaltrials.gov (NCT03004014).

Protocol

Each participant underwent a detailed clinical evaluation that included height, weight, and neck circumference measurements. All patients underwent a full polysomnography (PSG). During the following night, patients underwent zolpidem-induced sleep endoscopy. DISE with propofol was performed in the operating room, 2 days after zolpidem-induced sleep endoscopy.

PSG

PSG included electroencephalogram, electrooculogram, chin and tibial electromyography, electrocardiogram, oximetry, oronasal airflow (thermistor and pressure cannula), and measurements of rib cage and abdominal movements, snoring, and body position (Embla N7000; Natus, Inc., Middleton, WI). Apnea was defined as a reduction of ≥90% of the thermistor signal amplitude for ≥10 seconds. Hypopnea was defined as a reduction of ≥30% of nasal pressure amplitude for ≥10 seconds followed by oxygen desaturation ≥3% or arousal.²⁶

Instrumentation during both sleep endoscopies

All participants were asked to sleep in a supine position with a comfortable pillow, with the head in a forward neutral position. Participants were fitted with a nasal mask, attached to a heated pneumotachometer (Hans-Rudolph, Kansas City, MO) and a differential pressure transducer (Validyne, Northridge, CA) for airflow measurement. Mask pressure was measured by another pressure transducer (Validyne, Northridge, CA). After nasal topical application of a decongestant (oxymetazoline 0.025%) and anesthetic (lidocaine 10%), a 1.9-mm-diameter pressure catheter (Millar, Houston, TX) and a 2.8-mm pediatric bronchoscope (BFXP-160F; Olympus, Tokyo, Japan) were inserted through the left and right nostrils, respectively. Both devices were passed through sealed ports of the nasal mask. The pharyngeal pressure catheter was placed at the level of the epiglottis. Initially, the bronchoscope’s tip was placed at the nasopharynx above the palate. Several sequences of flow-limited breaths, defined as a lack of increase in flow despite decreasing pharyngeal pressure,²⁷ were recorded. If the patient presented repetitive obstructive apneas, low levels of continuous positive airway pressure (CPAP) was used to stabilize the airway and allow periods of stable flow limitation. After enough data were recorded at the velopharynx, the bronchoscope’s tip was moved forward to evaluate oropharyngeal and hypopharyngeal structures. All signals were captured at a sampling frequency of 200 Hz. Endoscopic images were sampled at 30 frames per second. BIS was used to assess the level of sedation during DISE (see below). During DISE, BIS and pulse oximetry monitors were recorded from an external camera. Physiologic signals, endoscopic image and the external camera image were synchronized and recorded on a personal computer using a data acquisition system (Power 1401; Cambridge Electronic Design, Cambridge, England). All sleep endoscopy procedures were performed by the same experienced medical team.

Zolpidem-induced sleep endoscopy

Sleep endoscopy during zolpidem-induced sleep was performed at the Sleep Laboratory during the night. All participants received zolpidem hemitartrate 10 mg orally 30 minutes prior to the start of the study. During zolpidem-induced sleep, participants were monitored using the same PSG montage, except for the nasal cannula and thermistor that were changed for a nasal mask and pneumotach as described above.

Propofol-induced sleep endoscopy

DISE was performed in the operating room during the day, just before the pharyngeal surgical procedure. Supplemental oxygen was not used before or during DISE with propofol. A target-controlled infusion system pump (Diprifusor; AstraZeneca) was used, starting with a target effect-site concentration of 1.5 μg/mL. The dose was increased by 0.1–0.2 μg/mL to achieve the recommended sedation depth: BIS level between 50 and 70 and Ramsay sedation scale level 5 (sluggish response to light glabellar tap or loud auditory stimulus).^15,17 BIS levels were selected to be in accordance with BIS levels of sleep stages N2 and N3.²⁸

Data analysis

The pharyngeal structures involved in collapse were classified according to the VOTE Classification System²⁹ during non-CPAP sleep. This classification considers the pharyngeal structure involved (velum, oropharynx, tongue base, and epiglottis) as well as the configuration of collapse (anteroposterior, lateral, or concentric) (Figure 1). Epiglottic collapse was considered both when the epiglottis was pushed by the tongue or if it collapsed independently. Partial airway collapse was not considered in the present study due to our previous experience with sleep endoscopy with simultaneous airflow recording suggesting that partial airway collapse does not influence airflow. Video image files from both sleep endoscopy studies were unidentified before being interpreted independently by 2 experienced investigators. A third investigator resolved any discrepancies. Only images from N2–N3 sleep during zolpidem-induced sleep and from BIS level of 50–70 and Ramsay sedation scale level 5 during DISE were included in the deidentified videos. The airway collapse patterns (VOTE endoscopic classification) were evaluated without CPAP use.

Figure 1. Representative examples taken during zolpidem- and propofol-induced sleep endoscopies showing agreement in the pattern of pharyngeal collapse.

Patient A: Endoscopic view of concentric velum collapse; patient B: tongue base collapse; patient C: lateral pharyngeal wall collapse.

Peak inspiratory flow, NED, and minimum peripheral oxygen saturation

In order to compare upper airway collapsibility and functional pharyngeal size during zolpidem-induced sleep endoscopy and DISE with propofol, inspiratory peak flow and NED from flow-limited breaths were compared (Figure 2).^30,31 Only breaths that were free of artifacts and arousals, taken from either atmospheric pressure or at the same CPAP level in each endoscopic procedure (zolpidem and propofol-induced sleep), were considered for analysis. Flow limitation was defined by the lack of increase in flow despite decreasing pharyngeal pressure.²⁷ NED was calculated as follows: ((peak inspiratory flow – flow at minimum pharyngeal pressure)/peak inspiratory flow) × 100.³¹ Respiratory drive was estimated by the driving pressure (DP), calculated as: (pharyngeal pressure at end-expiration − minimum pharyngeal pressure). Since sedation could lead to a reduction in respiratory drive and upper airway narrowing/obstruction, a more pronounced desaturation during respiratory events could occur. Minimum peripheral oxygen saturation (SpO₂) values during zolpidem- and propofol-induced sleep endoscopies were also compared.

Figure 2. Raw data recording of a sequence of flow-limited breaths in the same patient.

A: peak inspiratory flow; B: flow at minimum pharyngeal pressure; C: pharyngeal pressure at end-expiration; D: minimum pharyngeal pressure. NED = ([A – B]/A) × 100. DP = C – D. DP = driving pressure; NED = negative effort dependence.

Duration of zolpidem- and propofol-induced sleep endoscopies

Zolpidem-induced sleep endoscopy started when the lights were turned off with the patient positioned with instrumentation in place and ended when the endoscope was removed from the patient. DISE with propofol started at the beginning of the infusion of the sedative and ended when the endoscope was removed.

Statistical analysis

Data for continuous variables are presented as means ± SDs or medians (first–third quartiles). Normality was assessed using the Kolmogorov-Smirnov test. A P value <.05 was considered significant. A paired t test or the nonparametric Wilcoxon signed-rank test was used to compare continuous variables (peak flow, NED, DP, minimum SpO₂, duration of the sleep endoscopies, number of respiratory cycles analyzed, and pharyngeal pressure during NED assessment). McNemar’s test for paired proportions was used to test for changes between the number of pharyngeal structures involved in collapse. For each pharyngeal structure involved in collapse (velum, oropharynx, tongue, or epiglottis) 1 point was attributed. Changes in the VOTE classification between zolpidem- and propofol-induced sleep endoscopies were also tested using McNemar’s test. These statistical tests allow performing a comparison between data from zolpidem- and propofol-induced sleep in the same patient. Independent t test was used to compare the sleep apnea severity between the group without CPAP treatment and the group that needed CPAP treatment to induce airflow limitation. Effect size for the comparison of parametric data was described as the difference (95% confidence interval). For nonparametric data, effect size was calculated by dividing Z value (the number of SDs from the mean for a data point) by the square root of the number of tests performed.³² Statistical analyses were conducted using SPSS 18 (SPSS Inc., Chicago, IL) and Stata 11.0 (StataCorp, College Station, TX).

RESULTS

Twenty-one patients with OSA were studied and their characteristics are described in Table 1. The number of pharyngeal structures involved in collapse was similar during zolpidem-induced sleep and DISE with propofol (Table 2). Collapses of the velum, lateral pharyngeal wall, tongue, and epiglottis during zolpidem- and propofol-induced sleep are compared in Table 3 and Figure 3.

Table 1.

Participant characteristics.

Characteristics	Values
Sex, % males	67
Age, years	44 ± 9
Neck circumference, cm	42 ± 3
BMI, kg/m2	31 ± 4
Tonsil size, %
Grade 1	53
Grade 2	33
Grade 3	14
Grade 4	0
Mallampati classification, %
I	10
II	33
III	57
IV	0
Epworth Sleepiness Scale	14 ± 7
AHI, events/h	44 ± 30
AI, events/h	20 ± 29
HI, events/h	24 ± 14
Minimum SpO2,%	75 ± 7

Data are presented as means ± SDs unless otherwise indicated. AHI = apnea-hypopnea index; AI = apnea index; BMI = body mass index; HI = hypopnea index; SpO₂ = peripheral oxygen saturation.

Table 2.

Number of upper airway structures with complete collapse.

	Zolpidem-Induced Sleep, n (%)	Propofol-Induced Sleep, n (%)	Difference (95% CI)	P*
Number of structures			—
1	6 (28)	3 (14)
2	12 (57)	15 (71)
3	2 (9.5)	1 (5)
4	1 (5.0)	2 (9)
Multilevel collapse	15 (72)	18 (86)	14% (−12% to 40%)	.453

CI = confidence interval. *McNemar’s test for paired proportions.

Table 3.

Site of pharyngeal collapse observed during zolpidem- and propofol-induced sleep endoscopies.

Airway Level	Zolpidem-Induced Sleep	Propofol-Induced Sleep	Difference (95% CI)	P
Velum, %				.135*
Anteroposterior	14	10	4 (−4 to 14)
Concentric	48	38	10 (−9 to 28)
Lateral	38	52	−14 (−29 to 1)
Oropharyngeal lateral walls, %	57	57	0 (−23 to 23)	1.000
Tongue base, %	14	19	−5 (−14 to 4)	1.000
Epiglottis, %	19	33	−14 (−29 to 1)	.250

CI = confidence interval. *The 3 different configurations of vellum collapse were analyzed together.

Figure 3. Venn diagrams comparing the number of participants showing collapse at each upper airway structure and multilevel collapse during zolpidem- and propofol-induced sleep endoscopies.

The mean duration of the zolpidem-induced sleep endoscopy was higher than during DISE with propofol (78.3 vs 30.8 minutes; P < .001). Mean propofol target concentration to reach the appropriate sedation level was 1.8 ± 0.3 μg/mL. Mean total dose of propofol was 322 ± 84 mg. The minimum SpO₂ values during zolpidem-induced sleep and DISE with propofol were similar (75% and 78%, respectively; P = 0.09).

Peak flow, NED, and DP analysis were performed in 17 participants (Table 4). Four participants were excluded from these analyses due to technical reasons. One participant was excluded because the pharyngeal pressure transducer failed. Three participants did not achieve stable flow limitation during comparable CPAP levels in both endoscopic studies. Five patients had stable flow limitation at atmospheric pressure, and no CPAP treatment was necessary. In 12 patients, CPAP treatment was necessary to induce stable flow limitation due to repetitive obstructive apneas. The group of patients who needed CPAP treatment had more severe sleep apnea (AHI: 54.8 vs 15.4 events/h; P = .002). The number of respiratory cycles analyzed was 987. On average, 33 ± 19 breaths were analyzed per patient during zolpidem-induced sleep endoscopy studies and 25 ± 14 breaths per patient during DISE with propofol. Propofol did not increase the peak inspiratory flow, NED, or the respiratory drive during sleep endoscopy (Table 4).

Table 4.

Peak flow, NED, and DP analysis.

	Zolpidem-Induced Sleep	Propofol-Induced Sleep	Effect Size or Difference (95% CI)	P
Peak flow, L/s	0.326 (0.250–0.359)	0.309 (0.265–0.358)	0.05	.758
NED, %	36.9 ± 20.5	42.9 ± 18	−6.0 (−16 to 4)	.223
DP, cm H2O	17.1 (9.4–22.3)	18.3 (9.6–22.6)	0.03	.831
Breaths analyzed, n	33 ± 20	25 ± 14	8 (−1 to 16)	.078
CPAP level, cm H2O	3.4 ± 2.8	3.3 ± 2.7	0.1 (−0.1 to 0.3)	.154

Data are presented as means ± SDs or medians (first–third quartile). Effect sizes were calculated for nonparametric data. Differences (95% CIs) were calculated for parametric data. CI = confidence interval; CPAP, continuous positive airway pressure; DP = driving pressure; NED = negative effort dependence.

Nasal pressures during peak flow, NED, and DP analysis were 3.4 ± 2.8 cm H₂O and 3.3 ± 2.7 cm H₂O during zolpidem-induced sleep and DISE with propofol, respectively, reflecting low levels of CPAP used for these analyses.

DISCUSSION

The major findings of this pilot study are as follows: (1) propofol titrated to reach the strict recommended sedation levels did not increase the proportion of multilevel pharyngeal collapse as compared with zolpidem-induced sleep; (2) propofol did not increase pharyngeal collapsibility nor the compliance of the pharyngeal structures (estimated by peak inspiratory flow and NED) as compared with zolpidem-induced sleep; (3) propofol did not impair respiratory drive as compared with zolpidem-induced sleep.

Sleep endoscopy

DISE is a practical approach to detect the pharyngeal structures involved in pharyngeal collapse. Multilevel pharyngeal collapse may have implications for the surgical treatment of OSA.¹² Vroegop et al²² observed multilevel collapse in 68% of patients with OSA who underwent DISE. In another study, Ravesloot and de Vries⁷ found multilevel collapse in 76% of the patients. The most common structure involved in collapse in both previous studies was the soft palate (81% and 83%), followed by the tongue base (47% and 56%).^7,22 We observed multilevel collapse in 86% of patients during DISE with propofol. The most common structure involved in collapse in the present study was the soft palate (100%), followed by the lateral walls (57%). One possible explanation for the higher proportion of lateral wall involvement in our study is that we studied patients selected for lateral pharyngoplasty who frequently had prominent lateral wall muscles. A nonsignificant higher proportion of tongue and epiglottic collapse during propofol-induced sleep was observed. The findings of the present study confirm that multilevel collapse is common. Larger studies to address the influence of sedation on individual pharyngeal structures are warranted.

Upper airway function

Peak inspiratory flow has been used to describe upper airway function and collapsibility. Marques et al³³ used peak inspiratory flow during flow limitation to show improvement in upper airway patency at the lateral sleeping position as compared with the supine position in patients with OSA. Azarbarzin et al³⁴ showed that peak inspiratory flow can estimate active critical closing pressure. NED is a marker of the dynamic compliance of the pharyngeal structure involved in collapse.^24,35 In the present study, peak inspiratory flow and NED remained unchanged during zolpidem- and propofol-induced sleep, suggesting that upper airway patency and collapsibility were unaffected by propofol.

Sedation

Hillman et al¹⁷ observed loss of consciousness with propofol concentrations between 1.5 and 4 μg/mL. The average BIS values observed at loss of consciousness were between 50 and 70. A slight increase in genioglossus muscle tonus was observed at these propofol concentrations, but a sudden reduction in tonus at higher concentrations was reported.¹⁷ Eastwood et al¹⁸ and Hillman et al¹⁷ found good correlation between BIS levels and propofol effect site concentration: BIS significantly decreased with increasing the propofol effect site concentration. ‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬

Rabelo et al¹⁶ compared OSA severity determined during natural sleep during the day and during propofol-induced sleep. The authors reported that propofol did not affect OSA severity but led to a lower minimum oxygen saturation as compared with natural sleep. However, a deeper sedation level may have been reached since a higher mean propofol target concentration was administered (2.3 ± 0.6 μg/mL). Using lower propofol doses (1.8 ± 0.3 μg/mL) that were able to achieve the recommended sedation depth, we observed similar minimum oxygen saturation as well as unaltered upper airway patency and collapsibility during zolpidem- and propofol-induced sleep in the present study. These results are in line with our previous observations that low doses of midazolam did not impair upper airway collapsibility determined by critical pharyngeal closing pressure (Pcrit), as compared with natural sleep.¹⁴

Our study has several limitations. First, due to the instrumentation used in both studies, the sample size was small, especially to compare collapse at individual pharyngeal levels. The small sample size reduced the ability of this study to detect clinically important differences between zolpidem- and propofol-induced sleep endoscopies. However, this pilot study will be valuable to design a larger study, especially to detect the effect of sedation on the individual levels and configurations of pharyngeal collapse. Second, the pharyngeal catheter and endoscope used may have affected the site of collapse and airflow. A previous study has found that upper airway collapsibility was not altered by the presence of a catheter in the upper airway.³⁶ Although there were 2 instruments in the airway, the endoscope was placed above the palate most of the time. Third, data acquired during zolpidem-induced sleep stages N2–N3 were compared with a BIS level between 50 and 70 during propofol-induced sleep because PSG was not feasible in the operating room. OSA is often more severe during rapid eye movement (REM) sleep. The exclusion of REM sleep from the analysis limits the conclusions of our findings to non-REM sleep only. Fourth, zolpidem could potentially have influenced sleep endoscopy outcomes. Upper airway collapsibility is not affected by zolpidem.³⁷ However, zolpidem may increase upper airway responsiveness by unknown mechanisms. Although we cannot determine the exact influence of zolpidem in our findings, if present, it would have increased the differences between natural and propofol-induced sleep. Fifth, we have not assessed the reproducibility of pharyngeal collapse pattern classification. However, data from other studies confirmed good reproducibility and good interrater reliability of DISE.^38–40 Last, the sample of patients with OSA studied were carefully selected for upper airway surgery and are not necessarily representative of the entire population with OSA.

In this pilot study, propofol used to reach the recommended sedation levels did not significantly increase multilevel pharyngeal collapse or affect upper airway collapsibility and respiratory drive. Future studies with larger sample sizes should address the potential influence of sedation on upper airway function and individual levels of pharyngeal collapse.

DISCLOSURE STATEMENT

All authors have seen and approved the manuscript. Work for this study was performed at the Universidade de São Paulo, São Paulo, Brazil. A.B.O. was supported by CNPQ; P.R.G. was supported by FAPESP and CNPQ. The authors report no conflicts of interest.

ABBREVIATIONS

AHI

apnea-hypopnea index

BIS

bispectral index

CPAP

continuous positive airway pressure

DISE

drug-induced sleep endoscopy

DP

driving pressure

NED

negative effort dependence

OSA

obstructive sleep apnea

PSG

polysomnography

REM

rapid eye movement

SpO₂

peripheral oxygen saturation