Transoral robotic surgery for obstructive sleep apnea surgery

Introduction

Obstructive sleep apnea (OSA) is a common medical condition characterized by repeated nocturnal episodes of complete or partial airflow reduction. This leads to nocturnal desaturations, sympathetic overdrive, and cortical arousals. In Australia it affects approximately 10% of middle-aged males and 5% of middle-aged females. OSA is an independent risk factor for hypertension, ischemic heart and cerebrovascular disease. OSA is associated with excessive sleepiness and a two- to sevenfold increase in road and industrial accidents. It is a serious and costly public health problem.

While patients can be palliated with continuous positive airway pressure (CPAP), patient compliance remains a significant issue, with reported usage at the 5-year mark only 17%. There is a growing body of evidence to support surgical management of OSA, including the first multilevel randomized controlled trial in patients with moderate-severe OSA (SAMS trial). Excluded from the trial were patients with large obstructing lingual tonsils and supraglottic collapse.

Lingual tonsillar hypertrophy (LTH) is most commonly seen as a result of base of tongue lymphoid tissue commensurate growth in the years following a pediatric tonsillectomy. Although less commonly seen, epiglottic instability compromising the nocturnal airway has been demonstrated as part of multilevel airway collapse in 12% of OSA patients.

Traditionally surgical management of LTH or the epiglottis has been hampered by access issues, poor visualization, and uniplanar instrumentation. Not only has this made primary surgery difficult, but the management of primary and secondary hemorrhages highly morbid.

The advent of TORS with magnified 3D optics on 30-degree cameras, filtration of tremor, and multiplanar angled instrumentation with scaled intuitive movements, has allowed the sleep surgeon to finally tackle the final frontier of retrolingual OSA pathology in a safe and controlled environment.

Preoperative assessment

The preoperative workup includes a comprehensive assessment of the individual’s severity of symptoms, understanding of the level of anatomical obstruction, and candidacy for TORS surgery. Appropriate patient selection is at the cornerstone of achieving treatment success and minimizing treatment-associated morbidity.

All patients should have a baseline sleep study indicating at least moderate OSA (AHI >15) and have failed treatment with conventional nonsurgical options. A baseline Epworth Sleepiness Scale should be performed and a score of greater than 10 (indicates pathological sleepiness) may be considered a reasonable minimum score to consider intervention, thus indicating there is potential for significant quality of life improvement from surgery. A number of studies have shown an association with BMI and treatment success, ^, those with a BMI less than 30 are at the greatest likelihood to benefit from surgery, and, for patients with a BMI greater than 40, surgery should be avoided and treatment should focus on managing the obesity first.

Drug-induced sleep endoscopy (DISE) has been shown to be beneficial in nonresponders to pharyngeal surgery, and can lead to change in treatment decisions when used as part of the preoperative work-up. ^, However, when the level of obstruction is clear from clinical examination with awake nasendoscopy and Muller’s maneuver, we do not consider it mandatory in our practice. Classification of pharyngeal tongue hypertrophy using the Cormack-Lehane classification and oral tongue with the Friedman classification are also helpful tools to quantify anatomical issues. While cross-sectional imaging such as MRI or CT may identify tongue base hypertrophy, volumetric changes posttreatment have not been shown to correlate well with treatment effect to-date and it is not our current practice to perform baseline or postoperative imaging. Individuals with lingual tonsil or tongue base musculature hypertrophy or epiglottic collapse are those most likely to benefit from TORS surgery.

Once the severity of OSA and site of obstruction have been determined, attention should then turn to determining the patient’s general fitness for surgery and candidacy for TORS. Patients with OSA are known to have an increased incidence of cardiovascular disease, which must be stable and optimized before considering a general anesthetic. Due to the inherent bleeding risk with TORS surgery, it is also imperative to ensure the patient does not have any bleeding diathesis and that it would be safe to pause any preexisting anticoagulant or antiplatelet treatment. It is our practice to stop this for the 2-week period after surgery where there is an increased risk of postoperative hemorrhage. Adequate visualization is paramount to success in TORS surgery and patients need to be carefully examined for factors that may impede intraoperative access. These considerations are summarized in Table 45.1 . Aside from ensuring adequate mouth opening, neck circumference, mandibular body height, and hyoid-mental distance have been shown to be particularly important for adequate visualization of the tongue base and supraglottis.

Theater and instrument setup

Our preference is for nasotracheal intubation in order to provide unobstructed access to the tongue base. The patient is then positioned supine with the neck flexed and head extended using a head ring and/or shoulder roll as needed. The head should be positioned as close to the end of the operating table as possible ( Fig. 45.1 ). It is part of our standard protocol (see also postoperative care) to give 1 g of intravenous (IV) tranexamic acid on induction along with IV cefazolin, metronidazole, and dexamethasone. While some authors discuss the use of tracheostomy for some cases, it has been our experience in 12 years that this has not been necessary for such procedures, with no episodes of airway embarrassment encountered in the postoperative period to date.

We place a midline vertical mattress suture with 0-0 silk attached to a hemostat clip to aid tongue retraction and positioning. A thermoplastic nasal splint is placed over the lower dentition to protect both the teeth and ventral surface of the tongue from compression injury secondary to the gag. It is our usual practice to use a Boyle-Davis gag in dentate patients and a Dingman’s gag for the edentulous patient as this is at lower risk of slipping off the upper alveolar ridge than a Boyle-Davis gag. The shortest possible tongue blade is used to allow herniation of the tongue base into the surgical field while still giving adequate visualization of the epiglottis. Having the distal end of the blade at the level of the circumvallate papillae will give this optimal position. The tongue should be symmetrically retracted to minimize distortion of tongue base anatomy. Finally, a panoramic cheek retractor is placed laterally to protect the oral commissure and further improve access. An example of this is demonstrated in ( Fig. 45.2 ). While some authors advocate use of the Feyh-Kastenbauer (FK) retractor for TORS, we personally find the approach described above to be quicker and equally as effective as the FK retractor.

The da Vinci robot should then be docked. We normally have the anesthetist and anesthetic machine on the contralateral side to minimize clashing with the robot, and the cart is deployed at 45 degrees to the patient ( Fig. 45.3 ). The instrument trays should be stowed at the periphery of the operating room during docking to allow maximal space for safe movement and setup. Three arms are used for TORS with a 12 mm 30-degree videoendoscope facing upward in the central trocar to optimize the view for tongue base and epiglottic surgery. The lateral arms should then have a spatula monopolar cautery and fenestrated bipolar grasper, orientated to surgeon’s preference or side of surgery with the trocars sitting just inside the lateral corners of the mouth, to avoid the instruments clashing with the lips and teeth and causing iatrogenic injury. The camera trocar should be near vertical in orientation with the instrument trocars at a 20- to 30-degree angle toward horizontal relative to the camera. This aids with space and decreasing intraprocedural clashing of the instruments with the camera ( Fig. 45.4 ). It is also important to ensure there is adequate space between the arms before commencing surgery. An assistant should then sit at the head end of the patient with two Yankauer suckers to aid in retraction, suctioning the surgical field, and smoke evacuation. In addition, surgical clips and clip applicators should be readily available to the assistant to aid in hemostasis where required. A summary of equipment is listed in Table 45.2 .