Abstract
Importance
Drug-induced sleep endoscopy is a diagnostic technique that allows dynamic evaluation of the upper airway during artificial sleep. The lack of a standardized procedure and the difficulties associated with direct visual detection of obstructive events result in poor intraobserver and interobserver reliability, especially when otolaryngology surgeons not experienced in the technique are involved.
Objectives
To describe a drug-induced sleep endoscopy technique implemented with simultaneous polygraphic monitoring of cardiorespiratory parameters (DISE-PG) in patients with a diagnosis of obstructive sleep apnea syndrome and discuss the technique’s possible advantages compared with the standard procedure.
Design, Setting, and Participants
This prospective cohort study included 50 consecutive patients with obstructive sleep apnea syndrome who underwent DISE-PG from March 1, 2013, to June 30, 2014. A standard protocol was adopted, and all the procedures were carried out in an operation room by an experienced otolaryngology surgeon under the supervision of an anesthesiologist. Endoscopic and polygraphic obstructive respiratory events were analyzed offline in a double-blind setting and randomized order.
Main Outcomes and Measures
The feasibility and safety of the DISE-PG technique, as well as its sensitivity in detecting respiratory events compared with that of the standard drug-induced sleep endoscopy procedure.
Results
All 50 patients (43 men and 7 women; mean [SD] age, 51.1 [12.1] years) underwent DISE-PG without technical problems or patient difficulties regarding the procedure. As expected, polygraphic scoring was more sensitive than endoscopic scoring in identifying obstructive events (mean [SD] total events, 13.3 [6.8] vs 5.3 [3.6]; mean [SD] difference, 8.8 [5.6]; 95% CI, 7.3 to 10.4; Cohen d, –1.5). This difference was most pronounced in patients with a higher apnea-hypopnea index (AHI) at baseline (mean [SD] difference for AHI >30, 27.1% [31.0%]; 95% CI, –36.2% to 90.4%; Cohen d, 0.2; for AH I >40, 76.0% [35.5%]; 95% CI, 4.6% to 147.4%; Cohen d, 0.5; for AHI >50, 92.2% [37.2%]; 95% CI, 17.3% to 167.1%; Cohen d, 0.6) and a high percentage of hypopneas (≥75% of all obstructive events) at baseline (mean [SD] difference, 20.2% [5.4%]; 95% CI, 9.2% to 31.3%; Cohen d, 1.1). No other anthropomorphic or polygraphic features at baseline were associated with the differences between the DISE-PG and baseline home sleep apnea test.
Conclusions and Relevance
The DISE-PG technique is feasible, safe, and more sensitive at detecting an obstructed breathing pattern than is drug-induced sleep endoscopy alone. The DISE-PG technique could be helpful for accurate comprehension of upper airway obstructive dynamics (ie, degree of obstruction and multilevel pattern) and a nonobstructive breathing pattern (ie, central apneas).
This cohort study describes a drug-induced sleep endoscopy technique implemented with simultaneous polygraphic monitoring of cardiorespiratory parameters in patients with obstructive sleep apnea syndrome.
Key Points
Question
Is it possible to improve the detection of respiratory events and better understand the characteristics of breathing pattern during drug-induced sleep endoscopy in patients with obstructive sleep apnea syndrome?
Findings
In this cohort study of 50 patients, online polygraphic cardiorespiratory monitoring during drug-induced sleep endoscopy was more effective than endoscopic analysis alone in defining the breathing pattern and in the detection of respiratory events, especially central apnea and hypopnea events.
Meaning
A combined endoscopic and polygraphic approach during drug-induced sleep endoscopy appears to be feasible and safe and allows a better definition of the breathing pattern during the procedure.
Introduction
Obstructive sleep apnea syndrome (OSAS) is a sleep breathing disorder characterized by periods of cessation (apnea) and/or reduction (hypopnea) of oronasal airflow during sleep accompanied by drops in the blood oxygen level caused by repetitive upper airway obstructions. Identifying the site of obstruction and the pattern of airway collapse during sleep are essential key points to guide nonventilatory treatment decision making processes for OSAS in adults. Several studies have shown that upper airway endoscopic findings from tests performed while the patient is awake do not correspond exactly to obstructive sleep events. First introduced by Croft and Pringle in 1991, drug-induced sleep endoscopy (DISE) is a diagnostic technique providing a dynamic anatomical analysis of the upper airways during the pharmacologic induction of artificial sleep. Direct observation of anatomical modifications, such as collapse, vibrations, apnea or hypopnea, and their localization, is provided by a flexible fiber optic nasopharyngoscope. The lack of standardization in the DISE technique, including sedation, operation procedure, and classification, has produced several variations in the performance of DISE. In particular, the detection of obstructive events depends on the direct visual observation of changes in upper airway caliber, causing a subjective interpretation of the test results and, consequently, poor intraobserver and interobserver reliability when otolaryngology surgeons not experienced in the technique are involved.
This study implemented the DISE technique with simultaneous online polygraphic monitoring of cardiorespiratory parameters (DISE-PG) to better define the respiratory pattern and identify and objectify the respiratory events, especially subobstructive ones, that might be overlooked by visual endoscopic analysis.
The study aims were to describe our DISE-PG procedure, compare the detection of obstructive respiratory events during DISE (endoscopic scoring) with a simultaneous polygraphic scoring method, and discuss the possible advantages of DISE-PG compared with the standard DISE procedure.
Methods
This prospective cohort study was performed in patients with OSAS referred to the Otolaryngology Unit of Sant’Orsola-Malpighi University Hospital in Bologna, Italy, from March 1, 2013, to June 30, 2014. It consisted of a double-blind observational assessment of obstructive respiratory events detected both endoscopically and with simultaneous online cardiorespiratory monitoring during 50 consecutive DISE procedures. Since the study was part of our routine clinical practice, the protocol did not undergo approval of the local ethics board. Nevertheless, the study was conducted according to University of Bologna ethics guidelines and the Declaration of Helsinki. All patients provided written informed consent before each procedure.
All patients were 18 years or older. Obstructive sleep apnea syndrome was diagnosed according to standard criteria by clinical evaluation and an 8-channel home sleep apnea test (HSAT, type III polygraphy; Embletta Gold, Embla System Inc). Before the DISE procedure, each patient underwent a comprehensive otolaryngology clinical evaluation, upper airway fiberoptic endoscopy with the Muller maneuver in a sitting position, and lateral skull radiography for cephalometry. For DISE we selected patients who refused continuous positive airway pressure therapy.
DISE-PG Protocol
A proposed standard protocol was adopted (Figure) and DISE was carried out in an operating room by an experienced otolaryngology surgeon with the supervision of an anesthesiologist for the sedation protocol. Sedation was achieved in all patients by intravenous administration of propofol with the aid of target-controlled infusion (which provides a different dosage for each patient). The depth of sedation was monitored by bispectral index technology and maintained between 50 and 70 during the procedure.
Figure. Flowchart of Drug-Induced Sleep Endoscopy Technique With Simultaneous Polygraphic Monitoring of Cardiorespiratory Parameters.
The patient is attached to polysomnographic sensors outside the operating room (area of timeline above the blue dotted line). In the operating room (area below the blue dotted line), the endoscopic video recording is integrated in the polygraphic file, and the surgeon may visualize the polygraphic traces and the endoscopic images at the same time. During the sedation, multiple anatomical regions (velum, oropharynx, hypopharynx, tongue, and epiglottis) are evaluated. The endoscopic study was also performed during Esmarch and right-to-left head rotation maneuvers at each anatomical level.
A desirable observation window during DISE was defined as a period with 2 or more respiratory cycles considered a complete stable sequence of snoring, obstructive hypopnea or apnea, oxygen desaturation, and rebreathing. Observation was repeated at different levels of the upper airway according to the VOTE (velum, oropharynx, tongue, epiglottis) classification, as well as at the hypopharynx, defined as the level starting below the plane of the palatine tonsils toward the pyriform sinus and involving the lateral pharyngeal walls. Observation was also repeated during the Esmarch maneuver and with right-to-left head rotation; these latter results will be included in further ongoing research. The total duration of the DISE procedure was about 20 minutes.
Before starting the DISE procedure the patient was connected to polysomnographic sensors attached to a type III polygraphic system (the same as that used for HSAT) outside the operating room. The device was connected to a personal computer for online monitoring of the patient. During the DISE-PG procedure, the following cardiorespiratory parameters were recorded and measured according to standard guidelines: airflow with nasal pressure transducer and nasal-mouth thermistor, chest and abdominal efforts with respiratory inductive plethysmographic belts, blood oxygen saturation with pulse oximetry, snoring from nasal pressure transducer, electrocardiography by a single modified electrocardiograph lead II, and body position. The endoscopic video recording was integrated in the polygraphic file to visualize the polygraphic traces and endoscopic images at the same time. The respiratory events were assessed offline in a double-blind setting of DISE and polygraphy. At endoscopic evaluation, an endoscopic obstructive respiratory event was classified as an airway collapse of more than 75% lasting at least 10 seconds and associated with oxygen desaturation, according to Soares et al. This definition was chosen because semiquantitative scoring criteria were considered more reproducible than a qualitative analysis (ie, the VOTE classification) and most authors consider an obstruction of more than 75% to be significant enough to warrant surgical intervention.
For more accurate scoring of endoscopic obstructive events, DISE videos without polygraphic tracing (DISE video group) were shown offline in a randomized order to 2 experienced otolaryngologists (R.G. and L.C.) and the obstructive respiratory events scored by each otolaryngologist were noted. Thereafter, the 2 otolaryngologists reviewed all DISE videos together, discussing any ambiguous and discordant events to find greater agreement; only concordant events were considered for calculation and analyses (ie, absolute number of endoscopic obstructive events).
Drug-induced sleep endoscopy polygraphic tracings without endoscopic video support (DISE-PG group), thus conducted in the same setting as HSAT, were analyzed offline in a randomized order, and each respiratory event was scored by an experienced sleep physician (S.B.). Standard guidelines were adopted for each test to calculate the number of obstructive, mixed, and central apneas; number of hypopneas; the apnea-hypopnea index (AHI); 3% oxygen desaturation index; mean oxygen saturation; lowest oxygen saturation; and cumulative time with oxygen saturation below 90%.
Statistical Analysis
All data were summarized with descriptive statistics (mean and SD and frequency for categorical data) for the whole sample. The absolute total number of endoscopic obstructive respiratory events (DISE video group) was compared with the absolute total number of polygraphic events (DISE-PG group) in each examination using the paired sample 2-tailed t test. Absolute and relative differences among these groups were compared with baseline polygraphic and endoscopic features using a paired sample 2-tailed t test and Fisher exact test as appropriate. To identify possible factors associated with the difference between DISE-PG and DISE, differences such as absolute number of total obstructive events and obstructive apneas between the 2 measures were included in a multiple linear regression model with baseline BMI (calculated as weight in kilograms divided by height in meters squared), AHI, distance from mandibular plane to hyoid, percentage of hypopnea at baseline, and number of obstructive sites as the dependent variable. Finally, baseline HSAT features were compared with those obtained from the type III polygraphy during DISE using a paired sample 2-tailed t test. For all continuous data the standardized mean difference and its relative 95% CI were calculated. The effect size (Cohen d) was calculated for all the comparisons between the standardized mean difference. All analyses were performed using IBM SPSS Statistics, version 20 (IBM Inc).
Results
Fifty patients (43 men and 7 women) with a mean (SD) age of 51.1 (12.1) years (range, 30-76 years) and mean (SD) BMI of 27.9 (3.7) (range, 19.8-36.1) with mild to severe OSAS underwent the DISE-PG protocol. According to the World Health Organization classification of obesity, 27 patients (54%) were overweight (BMI, 25.1-29.9), 11 patients (22%) were obese (BMI, ≥30.0), and 12 patients (24%) had a normal BMI (≤25.0). The baseline HSAT results are reported in Table 1. According to the clinical guidelines of the American Academy of Sleep Medicine, 28 patients (56%) were diagnosed with severe OSAS, while 14 (28%) and 8 (16%) patients had moderate and mild OSAS, respectively. Anthropometric and otolaryngologic findings are summarized in Table 2 and eFigures 1-3 in the Supplement. According to the anamnestic and clinical examination, 35 patients (70%) had nasal obstruction (septal deviation or inferior turbinate hypertrophy). Fifteen of 47 patients (31.9%) had had tonsillectomy in childhood for chronic tonsillitis, while none had undergone a previous intervention for OSAS. Thirty-two of 47 patients (68%) showed hypertrophic tonsils (grades 1-4) and 15 of 47 had grade 0 tonsils (32%) (Table 2).
Table 1. Comparison of Polygraphic and Oximetric Results of Baseline HSAT vs DISE-PG.
| Characteristic | Baseline HSAT Result, Mean (SD) | DISE-PG Result, Mean (SD) | Difference HSAT vs DISE-PG | ||
|---|---|---|---|---|---|
| Mean (SD) | 95% CI | Cohen d | |||
| AHI | 37.2 (23.1) | 46.7 (21.0) | −9.5 (28.2) | −17.6 to 1.4 | −0.4 |
| AHI while supine | 49.3 (26.9) | 46.7 (21.0) | 2.5 (30.6) | −6.8 to 11.8 | 0.1 |
| Apneas, % | |||||
| Obstructive | 61.7 (30.4) | 56.3 (26.1) | 5.4 (37.1) | −7.5 to 18.4 | 0.2 |
| Mixed apneas | 3.6 (30.4) | 2.2 (26.1) | 5.4 (37.1) | −7.5 to 18.4 | 0.2 |
| Central apneas | 1.9 (3.9) | 6.5 (11.8) | −4.6 (13.3) | −9.2 to 0.1 | −0.5 |
| Hypopneas, % | 35.2 (28.2) | 30.9 (19.8) | 4.3 (31.5) | −7.7 to 16.2 | 0.2 |
| ODI | 34.4 (25.3) | 48.8 (22.6) | −14.3 (29.1) | −22.8 to −5.9 | −0.6 |
| LOS, % | 78.9 (7.7) | 72.5 (7.6) | 6.4 (8.3) | −3.8 to 9.0 | 0.8 |
| Oxygen saturation, mean, % | 93.2 (2.7) | 88.5 (3.8) | 4.8 (4.1) | −3.4 to 6.1 | 1.4 |
| CT90, % | 13.6 (18.8) | 43.7 (23.2) | −30.1 (25.7) | −38.1 to −22.1 | −1.4 |
Abbreviations: AHI, apnea-hypopnea index; CT90, cumulative time spent with oxygen saturation <90%; DISE-PG, drug-induced sleep endoscopy–polygraphy; HSAT, home sleep apnea test; LOS, lowest oxygen saturation; ODI, oxygen desaturation index.
Table 2. Characteristics of the Study Populationa.
| Characteristic | No. (%) |
|---|---|
| Severity of obstructive sleep apnea syndrome | |
| Mild | 8 (16.0) |
| Moderate | 14 (28.0) |
| Severe | 28 (56.0) |
| Examination of pharynx (n = 46) | |
| Mallampati class | |
| 1 | 3 (6.5) |
| 2 | 19 (41.3) |
| 3 | 22 (47.8) |
| 4 | 4 (8.7) |
| Tonsil grade (n = 47) | |
| Size 0 | 15 (31.9) |
| Size 1 | 15 (31.9) |
| Size 2 | 11 (23.4) |
| Size 3 | 3 (6.4) |
| Size 4 | 3 (6.4) |
| Muller maneuver in retropalatal area (n = 48) | |
| Obstruction | |
| Lateral | 21 (43.8) |
| Anteroposterior | 3 (6.3) |
| Circular | 8 (16.7) |
| None | 16 (33.3) |
| Muller maneuver in retrolingual area (n = 48) | |
| Obstruction | |
| Lateral | 10 (20.8) |
| Anteroposterior | 8 (16.7) |
| Circular | 6 (12.5) |
| None | 24 (50.0) |
| Mandibular plane to hyoid, mm (n = 36)b | |
| ≥20 | 26 (72.2) |
| <20 | 10 (27.8) |
| Previous otolaryngologic surgery (n = 47) | |
| Tonsillectomy | 15 (31.9) |
N = 50.
Mean (SD), 25.7 (7.5) mm.
Results of the Muller maneuver were positive in the oropharyngeal region in 32 of 48 patients (67%), showing a lateral pattern of collapse in 21 patients (44%), a circular pattern in 8 (17%), and an anteroposterior pattern in 3 patients (6%) (Table 2). Results of the Muller maneuver were negative in the hypoglottic region in 24 of 48 patients (50%), and the most frequent pattern of collapse was lateral (10 [21%]), followed by anteroposterior (8 [17%]) and circular (6 [13%]) patterns. Cephalometric evaluation showed a mean (SD) distance from the mandibular plane to hyoid of 25.7 (7.5) mm (range, 12.0-42.4 mm), with a distance from the mandibular plane to hyoid of 20 mm or more in 26 of 36 patients (72%) and less than 20 mm in 10 of 36 patients (28%).
The mean (SD) time calculated for DISE was 24.4 (7.2) minutes (range, 11.0-40.0). No technical problems during the procedures were reported. Patients did not report any discomfort. During the DISE procedure, all patients exhibited snoring, but no obstruction was documented in 2 patients. Most patients had palatal obstruction: lateral collapse of the oropharyngeal walls was observed in 41 patients (82%) and velar obstruction in 33 (66%). Tongue collapse occurred in about half of the patients (29 [58%]) (eFigure 4A in the Supplement). A single obstruction site was observed in 7 patients (14%), while most patients (41 [82%]) had obstructions at multiple levels (eFigure 4B in the Supplement). Considering the overall number of obstructive events detected during the procedure in each patient, the DISE video analysis was less sensitive than was the DISE-PG scoring (mean [SD] score, 5.3 [3.6] vs 13.3 [6.8]; mean [SD] difference, 8.8 [5.6]; 95% CI, 7.3-10.4, Cohen d, –1.5). This difference was more pronounced in patients with predominant hypopneas, because a partial obstruction is not identified by the criteria by Soares et al for the endoscopic detection of obstructive events at DISE (mean [SD] hypopnea number in the polygraphic group, 5.0 [4.4]). However, when comparing only DISE-PG obstructive apnea with the overall number of endoscopic obstructive events in the DISE video group, the DISE-PG analysis once again was more sensitive than DISE (mean [SD] score, 5.3 [3.6] vs 8.1 [5]; mean [SD] difference, 2.8 [4.7]; 95% CI, 1.5-4.1; Cohen d, –0.6). This difference was more marked for patients with a higher baseline AHI (baseline AHI >30: mean [SD] difference, 27.1% [31.0%]; 95% CI, –36.2% to 90.4%; Cohen d, 0.2; baseline AHI >40: mean [SD] difference, 76.0% [35.5%]; 95% CI, 4.6% to 147.4%; Cohen d, 0.5; baseline AHI >50: mean [SD] difference, 92.2% [37.2%]; 95% CI, 17.3% to 167.1%; Cohen d, 0.6) and a high percentage of hypopneas (≥75% of all obstructive events; mean [SD] difference, 20.2% [5.4%]; 95% CI, 9.2% to 31.3%; Cohen d, 1.1) on baseline HSAT (Table 3 and eTable in the Supplement). No other anthropomorphic or polygraphic features at baseline included in the multiple linear regression model were associated with the absolute differences between the 2 groups (β: AHI, –0.02; 95% CI, 0.63 to 0.60; percentage of hypopnea, –0.03; 95% CI, –0.41 to 0.40; BMI, 0.70; 95% CI, –2.50 to 4.0; MPH, 0.10; 95% CI, –1.35 to 1.62; NOS, –5.80; 95% CI, –14.89 to 3.37) (Table 4). Owing to the small number of cases, no difference in groups with different BMI or number of obstructive sites could be defined. Moreover, no specific obstructive site was associated with differences in the detection of obstructive events between the DISE video group and the DISE-PG group.
Table 3. Differences in the Detection of Obstructive Events Between the DISE Video and DISE-PG Groups.
| Event | No. | DISE Video, Mean (SD) |
DISE-PG, Mean (SD) |
Absolute Difference in Detection of Obstructive Events by DISE Video vs DISE-PG | ||
|---|---|---|---|---|---|---|
| Difference, Mean (SD) [Range] | 95% CI | Cohen d | ||||
| Apnea | 50 | 5.3 (3.6) | 13.3 (6.8) | 8.8 (5.6) [−4 to 23] | 7.3-10.4 | −1.5 |
| Obstructive apnea | 50 | 5.3 (3.6) | 8.1 (5) | 2.8 (4.7) [−8 to 13] | 1.5-4.1 | −0.6 |
Abbreviations: DISE, drug-induced sleep endoscopy; DISE-PG, drug-induced sleep endoscopy–polygraphy.
Table 4. Multiple Regression Analysis of Factors Influencing the Absolute Difference Between Respiratory Events Detected on DISE vs DISE-PG.
| Variable | β (95% CI) |
|---|---|
| Absolute difference in all obstructive events | |
| AHI | −0.02 (−0.63 to 0.60) |
| Hypopnea, % | −0.03 (−0.41 to 0.40) |
| BMI | 0.70 (−2.50 to 4.00) |
| MPH | 0.10 (−1.35 to 1.62) |
| NOS | −5.80 (−14.89 to 3.37) |
| Absolute difference in obstructive apnea events | |
| AHI | −0.19 (−1.27 to 0.70) |
| Hypopnea, % | −0.08 (−0.75 to 0.54) |
| BMI | 0.22 (−3.39 to 7.04) |
| MPH | 0.12 (−1.86 to 2.92) |
| NOS | −0.10 (−17.30 to 12) |
Abbreviations: AHI, apnea-hypopnea index; BMI, body mass index; DISE-PG, drug-induced sleep endoscopy; DISE-PG, drug-induced sleep endoscopy–polygraphy; MPH, distance from mandibular plane to hyoid; NOS, number of obstructive sites detected on DISE.
To establish whether the upper airway respiratory pattern observed during drug-induced sleep during the DISE procedure could reproduce the respiratory pattern identified during spontaneous nocturnal sleep at baseline with HSAT, the HSAT and DISE-PG findings were compared (Table 1). Differences were seen in all oximetric parameters and overall mean (SD) AHI (HSAT, 37.2 [23.1]; DISE-PG, 46.7 [21.0]), but mean (SD) supine AHI showed no difference at baseline (49.3 [26.9]) or during DISE-PG (46.7 [21.0]). Central apneas were observed during DISE-PG in about one-fourth of patients (13 [26%]), with more central apneas during DISE-PG compared with baseline (mean [SD], 6.5 [11.8] vs 1.9 [3.9]; mean [SD] difference, 4.6 [13.3]; 95% CI, 0.1-9.2, Cohen d, –0.5).
Discussion
The efficacy of nonventilatory treatments in patients with OSAS is often associated with adequate identification of the site and pattern of upper airway obstruction, and application of the appropriate individualized treatment. A standardized DISE protocol could help improve this therapeutic choice, providing direct dynamic and anatomical observation of the upper airway during pharmacologic-induced sleep. The standard DISE procedure has some limits, but they could be partially overcome by the implementation of DISE-PG.
The 2014 European position paper on DISE proposed standardization of most procedural aspects, but it did not reach a consensus on the scoring and classification system of endoscopic obstructive events, stating that “there are several DISE scoring systems, varying from perhaps too simple, to perhaps too complex or comprehensive.”(p461) Our study adopted the semiquantitative criterion suggested by Soares et al (obstructive event >75% reduction of upper airway caliber) since most authors consider an obstruction of more than 75% to be significant enough to warrant surgical intervention at that particular level of the airway. However, this system identifies only the obstructive events and does not distinguish between partial (hypopnea) and complete (apnea) obstruction, thereby leading to possible underestimation of the severity of OSAS during DISE. Our implementation of DISE-PG proved useful for a more sensitive identification of patients with obstructive patterns, particularly those with predominant hypopnea. To reduce the subjective variability and improve the endoscopic detection of obstructions, DISE recordings were blindly evaluated by 2 independent experienced otolaryngology surgeons who reached a final agreement on the identification of all endoscopic obstructive events. Nevertheless, DISE-PG proved more sensitive in identifying not only hypopneas but also obstructive apneas, as it allows the standardized and objective polygraphic scoring of obstructive events.
Another intrinsic limit of the DISE technique is that endoscopic evaluation provides only a single level of observation of obstructive sites, preventing a comprehensive analysis of upper airway functioning, whereas DISE-PG evaluation of the respiratory pattern throughout the procedure will identify obstructive events occurring at an airway level different from the level observed endoscopically. A further limitation is the condition of drug-induced sleep and the short observation period during the procedure. These factors suggest that the breathing pattern observed during DISE-PG may not be comparable to that observed during spontaneous nocturnal sleep (baseline polygraphic findings). Rabelo et al compared diurnal polysomnograms (90-120 minutes of recording) with and without sedation using propofol. They reported only statistical differences in the lowest oxygen saturation, while AHI, 3% oxygen desaturation index, and mean oxygen saturation were similar in the 2 conditions. Our study compared DISE-PG (using propofol for sedation) with baseline polygraphic findings (spontaneous night sleep). The AHI during DISE-PG was higher than baseline polygraphic findings, but this difference is owing to the forced supine position during the DISE procedure, as AHI during the DISE-PG procedure and supine AHI at baseline did not differ in our sample. Like Rabelo et al, we found that oxygen saturation findings deteriorated during DISE-PG compared with baseline polygraphy: not only the lowest oxygen saturation but all oximeter findings (3% oxygen desaturation index, mean oxygen saturation, and cumulative time spent below 90% oxygen saturation) were worse during drug-induced sleep than during spontaneous sleep. The worsening of oxygen saturation during the DISE procedure may be explained by the higher degree of drug-induced muscle relaxation and reduced respiratory drive. The differences with the findings of Rabelo et al could be explained by the difference in clinical features of the 2 groups of patients: our sample had more severe OSAS, more overweight patients, and forced supine position during the procedure. A series of central apneas were observed during sleep induction at the beginning of the DISE procedure in some patients. Babar-Craig et al also observed the same respiratory pattern and ascribed these events to the rapid biodistribution of the sedation drug. In our experience, the use of target-controlled infusion cannot prevent central apneas in all patients, although it reduces the effect of rapid induction such as a manual bolus. Online monitoring of polygraphic findings during the DISE procedure may help identify central events, especially in the case of otolaryngologists not experienced in the technique, thereby avoiding misleading evaluations of upper airway obstructive patterns changing during DISE. Heo et al also observed a time-related changing pattern during DISE, but in association with changed levels of obstruction. The DISE-PG procedure may be also useful to identify the best window of endoscopic observation, when the breathing pattern is most regular and stable, avoiding the misleading occurrence of central apneas.
From a practical point of view, the DISE-PG procedure is feasible and easily reproducible. The procedure is safe and does not cause any discomfort to the patient. The mean length of DISE-PG is comparable with that of the standard DISE procedure. Our protocol is time saving, as patients were connected to polygraphic sensors by physicians, nurses, or technicians outside the operating room while the previous patient who underwent DISE-PG was recovering from anesthesia. Because the polygraphic system used during the DISE-PG procedure is the same as the baseline procedure, there are no additional costs if the institution owns a basic HSAT system for diagnostic purposes. During the DISE procedure, the thermistor (that is placed outside the nose) and nasal cannula did not alter the feasibility of fibroscopy: the otolaryngology surgeons who performed DISE did not report any additional technical difficulty in our sample of adults, but the procedure may be more complex in children and adolescents. Knowledge of sleep medicine is required for the concomitant evaluation of endoscopic and polygraphic recordings during DISE-PG. However, once the DISE-PG procedure is concluded, otolaryngology surgeons not experienced in the technique can join other sleep specialists (ie, neurologists and pneumologists) for offline evaluation.
Limitations
The critical issue of our study is the arbitrary choice of endoscopic obstructive event criteria (obstruction >75%), but this reflects the uncertainty and lack of agreement on scoring systems highlighted by the European position paper. In addition, cessation of airway flow might persist even if the upper airway is not collapsed for a full 10 seconds (eg, collapse occurs during the inspiratory cycle to prevent inhalation and then partly relaxes during noninspiratory effort), which might explain the greater sensitivity of the DISE-PG procedure compared with DISE.
Conclusions
This study demonstrated the feasibility and safety of a combined (endoscopic and polygraphic) approach during the DISE procedure. As expected, DISE-PG proved more sensitive than standard DISE in the identification of obstructive respiratory events, especially hypopneas. The better definition of the breathing pattern during DISE may not be crucial if the procedure aims solely to identify the obstructive site for surgical correction. However, the DISE-PG procedure may establish the relevance of an obstruction found during DISE and provide a better overview of nonobstructive respiratory patterns (ie, central apneas). Therefore, the DISE-PG procedure might play an essential role in clarifying the different endoscopic scoring systems proposed for DISE. Further studies should define the endoscopic features occurring when no polygraphic obstructions are identified by visual analysis, and the possible role of procedural interference influencing DISE results (ie, the endoscope might alter the obstruction pattern).
eFigure 1. Examination of Pharynx According to Mallampati Classification
eFigure 2. Tonsil Size According to Friedman Grading
eFigure 3. Obstructive Patterns at Muller Maneuver During Awake Upper Airway Fiberoptic Endoscopy in the Sitting Position
eFigure 4. Endoscopic Results of Drug-Induced Sleep Endoscopy
eTable. Percentage Differences in the Detection of Obstructive Events Between the DISE Video and DISE-PG Groups in Subgroups With Different AHI and Percentage of Hypopneas at Baseline HSAT
References
- 1.American Academy of Sleep Medicine The International Classification of Sleep Disorders. 3rd ed Darien, IL: American Academy of Sleep Medicine Publishing; 2014. [Google Scholar]
- 2.Campanini A, Canzi P, De Vito A, Dallan I, Montevecchi F, Vicini C. Awake versus sleep endoscopy: personal experience in 250 OSAHS patients. Acta Otorhinolaryngol Ital. 2010;30(2):73-77. [PMC free article] [PubMed] [Google Scholar]
- 3.Soares D, Folbe AJ, Yoo G, Badr MS, Rowley JA, Lin HS. Drug-induced sleep endoscopy vs awake Müller’s maneuver in the diagnosis of severe upper airway obstruction. Otolaryngol Head Neck Surg. 2013;148(1):151-156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Croft CB, Pringle M. Sleep nasendoscopy: a technique of assessment in snoring and obstructive sleep apnoea. Clin Otolaryngol Allied Sci. 1991;16(5):504-509. [DOI] [PubMed] [Google Scholar]
- 5.van Maanen JP, Ravesloot MJL, Safiruddin F, de Vries N. The utility of sleep endoscopy in adults with obstructive sleep apnea: a review of the literature. Curr Otorhinolaryngol Rep. 2013;1(1):1-7. doi: 10.1007/s40136-012-0005-2 [DOI] [Google Scholar]
- 6.Vroegop AV, Vanderveken OM, Wouters K, et al. . Observer variation in drug-induced sleep endoscopy: experienced versus nonexperienced ear, nose, and throat surgeons. Sleep. 2013;36(6):947-953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.World Medical Association World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. [DOI] [PubMed] [Google Scholar]
- 8.Collop NA, Anderson WM, Boehlecke B, et al. ; Portable Monitoring Task Force of the American Academy of Sleep Medicine . Clinical guidelines for the use of unattended portable monitors in the diagnosis of obstructive sleep apnea in adult patients. J Clin Sleep Med. 2007;3(7):737-747. [PMC free article] [PubMed] [Google Scholar]
- 9.Rabelo FA, Braga A, Küpper DS, et al. . Propofol-induced sleep: polysomnographic evaluation of patients with obstructive sleep apnea and controls. Otolaryngol Head Neck Surg. 2010;142(2):218-224. [DOI] [PubMed] [Google Scholar]
- 10.De Vito A, Agnoletti V, Berrettini S, et al. . Drug-induced sleep endoscopy: conventional versus target controlled infusion techniques—a randomized controlled study. Eur Arch Otorhinolaryngol. 2011;268(3):457-462. [DOI] [PubMed] [Google Scholar]
- 11.Kezirian EJ, Hohenhorst W, de Vries N. Drug-induced sleep endoscopy: the VOTE classification. Eur Arch Otorhinolaryngol. 2011;268(8):1233-1236. [DOI] [PubMed] [Google Scholar]
- 12.Berry RB, Brooks R, Gamaldo CE, et al. ; American Academy of Sleep Medicine . The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications, Version 2.1. Darien, IL: American Academy of Sleep Medicine; 2015. [Google Scholar]
- 13.Ravesloot MJ, de Vries N. One hundred consecutive patients undergoing drug-induced sleep endoscopy: results and evaluation. Laryngoscope. 2011;121(12):2710-2716. [DOI] [PubMed] [Google Scholar]
- 14.World Health Organization BMI classification. http://apps.who.int/bmi/index.jsp?introPage=intro_3.html. Updated January 9, 2017. Accessed January 9, 2017.
- 15.Epstein LJ, Kristo D, Strollo PJ Jr, et al. ; Adult Obstructive Sleep Apnea Task Force of the American Academy of Sleep Medicine . Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med. 2009;5(3):263-276. [PMC free article] [PubMed] [Google Scholar]
- 16.Certal VF, Pratas R, Guimarães L, et al. . Awake examination versus DISE for surgical decision making in patients with OSA: a systematic review. Laryngoscope. 2016;126(3):768-774. [DOI] [PubMed] [Google Scholar]
- 17.De Vito A, Carrasco Llatas M, Vanni A, et al. . European position paper on drug-induced sedation endoscopy (DISE). Sleep Breath. 2014;18(3):453-465. [DOI] [PubMed] [Google Scholar]
- 18.Rodriguez-Bruno K, Goldberg AN, McCulloch CE, Kezirian EJ. Test-retest reliability of drug-induced sleep endoscopy. Otolaryngol Head Neck Surg. 2009;140(5):646-651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kezirian EJ, White DP, Malhotra A, Ma W, McCulloch CE, Goldberg AN. Interrater reliability of drug-induced sleep endoscopy. Arch Otolaryngol Head Neck Surg. 2010;136(4):393-397. [DOI] [PubMed] [Google Scholar]
- 20.Babar-Craig H, Rajani NK, Bailey P, Kotecha BT. Validation of sleep nasendoscopy for assessment of snoring with bispectral index monitoring. Eur Arch Otorhinolaryngol. 2012;269(4):1277-1279. [DOI] [PubMed] [Google Scholar]
- 21.Heo SJ, Park CM, Kim JS. Time-dependent changes in the obstruction pattern during drug-induced sleep endoscopy. Am J Otolaryngol. 2014;35(1):42-47. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure 1. Examination of Pharynx According to Mallampati Classification
eFigure 2. Tonsil Size According to Friedman Grading
eFigure 3. Obstructive Patterns at Muller Maneuver During Awake Upper Airway Fiberoptic Endoscopy in the Sitting Position
eFigure 4. Endoscopic Results of Drug-Induced Sleep Endoscopy
eTable. Percentage Differences in the Detection of Obstructive Events Between the DISE Video and DISE-PG Groups in Subgroups With Different AHI and Percentage of Hypopneas at Baseline HSAT