Obstructive sleep apnea is a heterogenous disorder characterized by a number of distinct phenotypes of upper airway collapse.
Evaluating dynamic pharyngeal collapse in the sleeping patient is clinically challenging.
Drug-induced sleep endoscopy is a safe, effective outpatient procedure for reliably assessing upper airway obstruction in the sleeping patient.
Although numerous reporting methods exist for drug-induced sleep endoscopy findings, the VOTE (Velum, Oropharynx, Tongue base and Epiglottis) classification is the most widely used.
Video content accompanies this article at www.oralmaxsurgeryatlas.theclinics.com .
Introduction: nature of the problem
Obstructive sleep apnea (OSA) is a heterogeneous disorder represented by a wide spectrum of pharyngeal dysfunction. Modern surgical techniques are used to address the various clinical phenotypes. However, assessment of the site, extent, and severity of the obstructive event in the sleeping patient is a clinical challenge.
Early methods of assessment included Mueller’s maneuver (forced inspiration against a closed glottis during awake endoscopy), lateral cephalometry, and computed tomography. However, these methods provided only static observations of the awake patient and could not identify dynamic, sleep-related obstructions. In 1991, Croft and Pringle first proposed drug-induced sleep endoscopy (DISE) as a novel, outpatient technique for evaluating OSA. The technique described the use of medication-induced sleep in conjunction with fiberoptic endoscopy of the upper airway to develop a personalized airway approach for the sleep apneic patient. Before the introduction of DISE, a region-based approach to care was used by most sleep surgeons by categorizing patients according to the Fujita classification system.
Owing to the need for a more targeted and clinically reliable means of characterizing patterns of upper airway obstruction, DISE has seen widespread adoption as one of the tools in clinical evaluation of the OSA patient and consideration of tailored surgical techniques. Although multiple classification systems have been proposed, the most widely adopted is the VOTE classification proposed by Kezirian and colleagues in 2011. This classification is used to characterize DISE findings based on the specific structures (Velum, Oropharynx, Tongue base and Epiglottis) that most commonly contribute to upper airway obstruction. The VOTE classification has greatly improved the accuracy with which we can manage specific OSA phenotypes.
DISE is a necessary tool in the armamentarium of the sleep surgeon, allowing thorough and targeted evaluation of dynamic pharyngeal obstruction in the sleeping patient.
After undergoing thorough clinical assessment including a complete head and neck, and oral–maxillofacial examination suitable candidates can be considered for DISE.
Recumbent patient bed
Distal chip-tip scope (author’s preferred: Olympus ENF-VH High Definition Scope; Olympus, Tokyo, Japan)
Lidocaine jelly on a cotton-tipped applicator
Preparation and patient positioning
The patient should be positioned supine and comfortable in a semidark and silent operating or endoscopy suite. In our practice, we do not provide oxygen supplementation to the patient during the procedure, in an attempt to correlate the endoscopy findings observed during patient’s desaturation corresponding to the lowest nocturnal SpO 2 . Standard patient monitoring includes pulse oximetry, electrocardiography, and noninvasive blood pressure measurement. Standard resuscitation equipment should be available. Lidocaine jelly applied to the nare can be applied for comfort before initiating endoscopy.
Anesthetic approach/sedation agent
Current data on sedative agents used in DISE remain limited, with limited study comparing obstruction to sleep staging with encephalography. Based on cardiac and pulmonary parameters, dexmedetomidine shows a more stable and potentially safer profile than propofol infusion.
Data from DISE patients induced with propofol have shown a greater degree of obstruction and desaturation observed than those induced with dexmedetomidine (Precedex). Propofol has been shown to demonstrate an oxygen nadir most representative of overnight polysomnography. Pharmacologically, propofol also has a shorter half-life and quicker onset. However, the sedation with dexmedetomidine, a highly selective, centrally acting α2-adrenoreceptor agonist, affords a larger therapeutic window for respiratory depression/apnea compared with propofol.
A sedation with dexmedetomidine resembles natural sleep owing to a reduction of norepinephrine release by the locus coeruleus and subsequent release of γ-aminobutyric acid and galanin by the ventrolateral preoptic nucleus of the hypothalamus. A more gradual onset of dexmedetomidine-induced sedation/sleep also allows for ample time to observe dynamic airway behavior.
The prospective comparison studies on the use of different anesthetic techniques for DISE are lacking, and the choice of drug is often dictated by institutional preference. Yet, the paramount objective of allowing the surgeon to observe gradual onset of the upper airway collapse must be fulfilled with any technique. The sedation sequence is most commonly accomplished by administering an intravenous (IV) bolus of the selected drug (a loading dose), followed by an infusion.
The use of target-controlled infusions and other mathematical pharmacokinetic models, such as probability ramp control approach for administering propofol during DISE, has been shown to improve accuracy of achieving the required effect-site propofol concentration and reliability of the observational window for the obstructive events. The predicted mean effect-site concentration for propofol during DISE ranges between 2.0 and 4.8 μg/mL, with the loss of upper airway tone occurring at 2.94 ± 0.97 μg/mL, and upper airway obstruction at 4.2 ± 1.3 μg/mL.
Because target-controlled infusions are not currently available in the United States, in our practice we rely on manual-controlled propofol infusions. Our typical approach involves administering the background infusion of IV propofol 150 to 180 μg/kg/min with the superimposed small IV propofol boluses 0.1 to 0.2 mg/kg every 30 seconds. In our experience, this sequence assures a reliable and gradual upper airway collapse within 5 to 7 minutes.
For a dexmedetomidine DISE sequence, we administer dexmedetomidine IV bolus 1.5 μg/kg over 10 minutes, followed by infusion of 1.5 μg/kg/h. A reliable induction of sleep and upper airway collapse is observed in the vast majority of patients within 15 minutes, with occasional top-up dexmedetomidine rapid IV boluses 0.25 μg/kg required after the administration of the loading dose had been completed.
As sedation is administered, the surgeon places the distal chip-tip scope into the nare identifying the side with least obstruction. The scope is passed through the middle meatus, above the inferior turbinate, and directly inferior to the middle turbinate and passed posteriorly until the nasopharynx is reached. From this position, the scope is positioned in a rostral view (facing toward the larynx) to best visualize the velum as sedation is infused.
After observation at the level of the velum is complete and the degree of obstruction is noted, the scope is passed distally to observe the airway at the level of the oropharynx. Subsequently, the scope is passed to observe the tongue base and epiglottis. Observed findings are recorded in a table as described in Fig. 1 .