Key points
- •
Drug-induced sleep endoscopy (DISE) is an upper airway evaluation technique with 3 key features: the use of pharmacologic agents to achieve sedation, the target depth of sedation as approximating natural sleep as much as possible, and the endoscopic evaluation of the upper airway.
- •
The VOTE Classification incorporates the 4 major structures that contribute to airway obstruction in most patients: Velum (palate), Oropharyngeal lateral walls, Tongue, and Epiglottis.
- •
DISE may improve treatment selection in sleep-disordered breathing, especially in patients with obstructive sleep apnea unable to tolerate positive airway pressure therapy.
Introduction
Because up to 50% of patients with obstructive sleep apnea (OSA) are unable to tolerate positive airway pressure therapy, alternative treatments, such as surgery, upper airway stimulation, or oral appliances, may be required. Comprehensive patient evaluation is a key to success for the latter group of treatments. The goal of evaluation is to determine the pattern of airway obstruction, with the ultimate aim of designing targeted, effective treatment.
The ideal evaluation technique would be an assessment of breathing, sleeping patients, as this would provide a real-time, dynamic assessment. It would also be safe, noninvasive, and low cost. The desire to directly visualize airway obstruction led some investigators to perform fiberoptic examination during natural sleep in the late 1970s and early 1980s. However, these efforts were generally abandoned due to the discomfort experience by patients, particularly with movement of the endoscope to view multiple areas of the pharyngeal airway.
Fiberoptic evaluation of the upper airway under conditions of sedation was developed in a number of centers in Europe in the late 1980s, and Croft and Pringle first described their technique of “sleep nasendoscopy” in 1991. The nomenclature was changed to “drug-induced sleep endoscopy” (DISE) by Kezirian and Hohenhorst (W. Hohenhorst, personal communication, 2005) to reflect the 3 key features of this method: the potential use of various pharmacologic agents to achieve sedation, the target depth of sedation as approximating natural sleep as much as possible, and the endoscopic evaluation of the upper airway. In contrast to other procedures that usually provide 2-dimensional assessments during wakefulness in the upright sitting position, DISE provides a 3-dimensional evaluation of the airway during unconscious sedation.
This article presents recommendations regarding DISE technique and reviews the evidence concerning the role of DISE in the evaluation of OSA.
Introduction
Because up to 50% of patients with obstructive sleep apnea (OSA) are unable to tolerate positive airway pressure therapy, alternative treatments, such as surgery, upper airway stimulation, or oral appliances, may be required. Comprehensive patient evaluation is a key to success for the latter group of treatments. The goal of evaluation is to determine the pattern of airway obstruction, with the ultimate aim of designing targeted, effective treatment.
The ideal evaluation technique would be an assessment of breathing, sleeping patients, as this would provide a real-time, dynamic assessment. It would also be safe, noninvasive, and low cost. The desire to directly visualize airway obstruction led some investigators to perform fiberoptic examination during natural sleep in the late 1970s and early 1980s. However, these efforts were generally abandoned due to the discomfort experience by patients, particularly with movement of the endoscope to view multiple areas of the pharyngeal airway.
Fiberoptic evaluation of the upper airway under conditions of sedation was developed in a number of centers in Europe in the late 1980s, and Croft and Pringle first described their technique of “sleep nasendoscopy” in 1991. The nomenclature was changed to “drug-induced sleep endoscopy” (DISE) by Kezirian and Hohenhorst (W. Hohenhorst, personal communication, 2005) to reflect the 3 key features of this method: the potential use of various pharmacologic agents to achieve sedation, the target depth of sedation as approximating natural sleep as much as possible, and the endoscopic evaluation of the upper airway. In contrast to other procedures that usually provide 2-dimensional assessments during wakefulness in the upright sitting position, DISE provides a 3-dimensional evaluation of the airway during unconscious sedation.
This article presents recommendations regarding DISE technique and reviews the evidence concerning the role of DISE in the evaluation of OSA.
Technique
Indications/Contraindications
Any diagnostic evaluation will be useful if the benefits outweigh the risks. For DISE, the benefits include potential value in treatment selection, and the risks are related to the sedative agent used and the potential for significant airway compromise. Indications and contraindications are listed in Box 1 .
Indications
Patients with obstructive sleep apnea (OSA) (or snoring, in some countries)
Unable to tolerate positive airway pressure (in countries in which positive airway pressure is the first-line treatment modality for OSA)
Consideration of surgery, oral appliances, positional therapy, or combination approaches
Contraindications (relative)
Allergy to sedative agents
Pregnancy
Significant medical comorbidities
Optional contraindications used by some surgeons:
Markedly severe OSA (AHI >70 events per hour)
Obesity (body mass index >35 kg/m 2 )
Sedative Agent
Control of the depth of sedation is essential. The sedative agent will generally be administered intravenously at the minimum dose to achieve the target depth of sedation: the loss of consciousness, defined as loss of response to verbal stimulation at a normal conversational volume, similar to a modified Ramsay score of 5. Multiple sedative agents have been used as effective agents in the performance of DISE. The commonly used sedative agents included propofol, midazolam, propofol with midazolam, and dexmedetomidine. Pharmacologic properties of these sedative agents is shown in Table 1 .
| Propofol | Midazolam | Dexmedetomidine | |
|---|---|---|---|
| Sedative agents | 2-6-diisopropylphenol | Benzodiazepine | Alpha-2 adrenergic receptor agonist |
| Functional half-life | 4–6 min | 45 min | 6 min |
| Elimination half-life | 3 h | 150 min | 2 h |
| Accumulation | Inactive metabolite (no accumulation) | Active metabolite (alpha-hydroxymidazolam) | Inactive metabolite (no accumulation) |
| Therapeutic range | Small | Large | Small |
| Respiratory side effects (potential) | Respiratory depression and hypopharyngeal reflex depression | Respiratory depression | None |
| Cardiovascular side effects (potential) | Hypotension | Hypotension | Fluctuation of blood pressure and heart rate |
The choice of sedative agent, to some extent, depends on its ability to reproduce some of the changes that occur in natural sleep. During sleep, upper airway patency relies on pharyngeal dilator muscle tone and changes in lung volume that counteract collapsing forces, principally intraluminal negative pressure generated during inspiration and anatomic narrowing of the airway. Patients with OSA maintain pharyngeal patency with greater dilator (genioglossus) muscle tone during wakefulness, but sleep onset results in marked decreases in muscle tone due to the loss of the wakefulness stimulus and decreases in negative pressure reflex activity and lung volume. Rapid eye movement (REM) sleep has greater reductions in muscle tone and lung volumes than non-REM (NREM) sleep. The continuum of sedation ranges from wakefulness to conscious sedation to unconscious sedation to general anesthesia, in which arousability to all stimulation is lost. Deeper levels of sedation are associated with progressive decreases in upper airway dilator muscle tone and neuromuscular reflex activation, both of which increase airway collapsibility.
Unconscious sedation represents a lesser degree of neural depression than anesthesia and may be a closer approximation to natural sleep. The interest in the transition from wakefulness to unconscious sedation is based on the concept of a thalamocortical switch determining consciousness or unconsciousness (no response to verbal stimulation) that may be common to natural sleep and sedation.
Propofol can be administered with target-controlled infusion, continuous infusion, or using small boluses. The most-detailed study of changes in upper airway physiology associated with propofol sedation had results consistent with the thalamocortical switch described previously, as changes in upper airway collapsibility (passive Pcrit), Bispectral Index Score (based on frontal electroencephalogram activity), and muscle tone occurred at this transition from consciousness to unconsciousness disproportionate to changes in propofol concentration. During propofol unconscious sedation, healthy individuals have decreases in genioglossus tone to 10% of maximum awake activity, which is one-half to one-third of the level in NREM sleep in healthy individuals, but greater than during REM sleep in healthy individuals and patients with OSA.
To achieve this depth of sedation, target-controlled infusion with the Diprifusor (Astra-Zeneca, Inc, London, United Kingdom) technology is available in many countries, using a proprietary algorithm to achieve effect site (brain) concentration using a 3-compartment pharmacokinetic model. This technology is not available in the United States and many other countries, so some surgeons use a protocol involving continuous infusion (usually unconscious sedation requires 100–150 μg/kg per minute), with possible additional boluses (eg, 20–50 mg). Other surgeons prefer using a continuous infusion without administration of boluses. It is important that different individuals can require markedly different dosages of propofol to reach the target depth of sedation.
Dexmedetomidine has been used extensively in anesthesia for a wide range of procedures and for prolonged periods of sedation, with comparable performance during general anesthesia. As is true for midazolam, there has been no study evaluating the impact of dexmedetomidine on upper airway muscle tone and airway collapsibility.
Dexmedetomidine may achieve a depth of anesthesia comparable to propofol for general surgical procedures, but in one study, half of the patients required supplemental propofol during drug-induced sleep endoscopy because they did not achieve adequate sedation despite a maximum dose of dexmedetomidine (1.4 μg/kg/h). The senior author (Dr Kezirian) has seen a number of patients who underwent DISE using dexmedetomidine at other institutions in which there was no airway obstruction visualized despite apparent loss of consciousness; this may reflect a lesser decrease in upper airway muscle tone with dexmedetomidine compared with propofol and midazolam, but there are no data evaluating this to date.
Preoperative Preparation
Preoperative anesthetics other than the sedative agent of choice should be avoided. To prevent regurgitation and aspiration, patients should remain nil per os (NPO) before the procedure. Anticholinergic agents, such as atropine or glycopyrrolate, can be administered intravenously 30 minutes before the procedure to reduce secretions, leading to better visualization and avoidance of coughing due to aspiration during the procedure; mild tachycardia is commonly observed with these anticholinergic agents.
A topical vasoconstrictor is generally applied to 1 or both nostrils, and topical anesthetic is applied to 1 nostril. The dose of topical anesthetic is minimized to avoid excessive pharyngeal anesthesia that can lead to coughing during the procedure (again, due to aspiration of secretions) or blunting of muscular reflexes that promote upper airway patency. Awake fiberoptic examination performed before initiating sedation can confirm adequate topical anesthesia. This awake examination also allows the anesthesia team the opportunity to visualize the airway before administration of sedation to a patient with sleep-disordered breathing, often providing some additional level of comfort.
Surgeon preferences regarding patient positioning vary widely. Some prefer to use the supine position in all patients, as this body position is used by many patients for at least a portion of the night and may reflect the body position that is most problematic for OSA. Other surgeons prefer to use the patient’s natural body position during sleep. Still others evaluate the patient in both the lateral and supine positions or turn the head to the side while the trunk is in the supine position. There is no single best approach, but if possible the surgeon should examine the body (or head) position that will provide the greatest amount of information to use in treatment selection.
Setting and Monitoring
DISE is usually performed in the operating room or procedure suite, but it is also possible to perform DISE in various outpatient settings, depending on the availability of personnel and appropriate equipment for administering sedation safely. The temperature of the room should be set as comfortable as possible. Lights should be dimmed, and the room noise should be minimized.
Oxygen saturation, cardiac rhythm, and blood pressure monitoring are required during the procedure. Supplemental oxygen may not be necessary, but it should be available for potential use. Some surgeons prefer routine administration of oxygen via nasal cannula or face mask placed on the upper chest in a “blow-by” fashion.
The most important evaluation of the depth of sedation is clinical, through the onset of snoring, disordered breathing events, and the loss of consciousness. Bispectral index (BIS) score may provide additional monitoring of sedation. BIS score is associated with the depth of sedation, with a potential target BIS score of 55 to 70, based on changes in muscle tone and upper airway collapsibility. Greater depth of sedation may produce greater loss of muscle tone and increased airway collapsibility. One study of DISE using propofol showed an increase in airway obstruction (both severity and contribution of the palate and tongue) at BIS scores of 50 to 60 versus 65 to 75.
Risks and Disadvantages
No catastrophic events have been reported with DISE. Endotracheal intubation is extremely rare, and cricothyrotomy and tracheostomy also have not been reported. Oversedation that may lead to airway compromise or central apnea can be prevented by titrating the sedative agent to the lowest level that maintains the target of sedation. Supplemental oxygen or concurrent positive airway pressure administration may be an option to prevent complications in high-risk patients, such as those with morbid obesity or greater medical comorbidity. Other DISE risks include local pain on intravenous infusion of propofol and allergic reaction to medications.
Diagnosis
Airway Evaluation: The VOTE Classification
DISE is principally an examination of the pharynx, and during the examination, the flexible fiberoptic laryngoscope is moved a number of times to evaluate the entire length of the pharynx across a number of cycles of airway obstruction and normal breathing. A surgeon must recognize that they are often visualizing only a portion of the pharynx at a single point in time, necessitating evaluation and reevaluation of different regions to understand, to the extent possible, the source(s) of airway obstruction over a period of time in different body positions, during various maneuvers, and so forth.
A variety of classification schemes have been described to characterize DISE findings. There are at least 7 schemes reported in the literature, with a wide range of complexity. The VOTE Classification was proposed as a standard for DISE scoring because it incorporates the 4 major structures that contribute to airway obstruction in most patients: Velum (palate), Oropharyngeal lateral walls, Tongue, and Epiglottis. The hope is that widespread adoption of the VOTE Classification would lead to the sharing of findings and results across centers, enhancing clinical and research communication and collaboration.
The Structures of the VOTE Acronym
The VOTE Classification allows a surgeon to characterize the structures that contribute to pharyngeal obstruction in a patient, incorporating the degree and configuration of airway narrowing related to these structures that are each composed of multiple components.
Velum
Velum-related obstruction is that related to the palate and occurs due to the soft palate, uvula, or lateral pharyngeal wall tissue at the level of the velopharynx. Airway closure related to the velum can occur with collapse in an anteroposterior ( Fig. 1 ), concentric, or, less commonly, lateral configuration. Because it is not always possible to distinguish between the soft palate, uvula, and lateral pharyngeal walls at the level of the velopharynx on DISE, the VOTE Classification groups them under the umbrella of the Velum. The lateral pharyngeal walls at the level of the velopharynx have some interaction with the remainder of the lateral wall tissues, but in the VOTE Classification, there is an attempt to describe these separately.
