Supraglottoplasty and Epiglottopexy for Sleep-Variant Laryngomalacia

Procedure: Drug induced sleep endoscopy (DISE), direct laryngoscopy, rigid bronchoscopy, epiglottopexy, and supraglottoplasty for sleep-variant laryngomalacia. Introduction: Laryngomalacia is the most common cause of stridor in infants but can also rarely affect school-aged children. In these patients, symptoms of obstruction are usually limited to nighttime and occur as a form of sleep-disordered breathing, so called sleep-variant laryngomalacia. Surgical correction of the supraglottic collapse can lead to resolution of sleep disordered breathing in this population. Materials and Methods: Patients with clinical symptoms suspicious for sleep-variant laryngomalacia should undergo in-office flexible laryngoscopy and drug induced sleep endoscopy for the purposes of diagnosis. First, the epiglottopexy is performed. A CoblatorTM (Smith+Nephew, Andover, MA, USA) is used to foreshorten the median glossoepiglottic fold, deepen the vallecular sulcus, and denude the mucosa. The lingual surface of the epiglottis is sutured to the base of tongue with a PDS suture. After epiglottopexy is complete, the supraglottoplasty is performed. The aryepiglottic folds are divided, and the redundant supra-arytenoid tissue is truncated with a cold-steel technique. The patient is kept intubated for approximately 24 hours. Flexible laryngoscopy is performed two weeks post-operatively to assess healing. Results: A 6-year-old female with nighttime stridor and sleep disordered breathing despite prior adenotonsillectomy was found to have sleep-variant laryngomalacia on sleep endoscopy. She underwent endoscopic supraglottoplasty and epiglottopexy. Surgery and postoperative care were uncomplicated, she healed well, and her sleep symptoms resolved. Conclusion: While less common than traditional laryngomalacia, a diagnosis of sleep-variant laryngomalacia should be considered in older children with stridor during sleep. Surgical cure can be achieved with endoscopic supraglottoplasty and epiglottopexy.

Laryngomalacia is the most common cause of stridor in infants and the most common congenital abnormality of the larynx. When laryngomalacia was first described 80 years ago, it was used to described infants with inspiratory stridor [1]. Now, the term has expanded to describe the inward collapse of the supraglottic structures, typically during inspiration, resulting in obstruction of the airway. The terminology is used despite the age of the patient [2]. Flexible laryngoscopy is critical in the diagnosis of laryngomalacia; the type laryngomalacia is generally classified by its laryngoscopic appearance. The most commonly used classification, proposed by Olney et al., is as follows: type 1—prolapse of the mucosa overlying the arytenoid cartilages, type 2—foreshortened aryepiglottic folds, and type 3—posterior displacement of the epiglottis [3]. However, severity of laryngomalacia is determined by clinical presentation and the presence or absence of concerning features such as failure to thrive, aspiration, cyanotic episodes, pectus excavatum, and/or cor pulmonale. In 2008, Richter et al. described a cohort of patients with sleep-disordered laryngomalacia. While these patients exhibited inspiratory stridor related to dynamic prolapse of the supra-arytenoid tissue, they were school-age children rather than infants. A majority of the patients had adenotonsillectomy for obstructive sleep apnea (OSA), however they continued to have persistent stridor during sleep. Sleep endoscopy demonstrated dynamic prolapse of the supra-arytenoid tissue and surgical correction of the larynx ultimately led to symptom resolution [2]. Here we present a school-age child with sleep-variant laryngomalacia, whom we successfully treated with epiglottopexy and supraglottoplasty.

Preoperative management of patients may include reflux medication, polysomnogram, in-office flexible laryngoscopy, and sleep endoscopy. Our patient was previously diagnosed with OSA and underwent prior intracapsular tonsillectomy and adenoidectomy. She presented three years later with a chief . She had no daytime respiratory symptoms. In-office flexible laryngoscopy did not demonstrate any obvious obstruction, but she did have inward prolapse of the arytenoid towers with rapid breathing, suggestive of possible laryngomalacia. She subsequently underwent sleep endoscopy, laryngoscopy, and rigid bronchoscopy, which were notable for complete prolapse of the arytenoids into the airway. This video demonstrates our epiglottopexy and supraglottoplasty technique for laryngomalacia. Patients receive intra-operative steroids and antibiotics. Nasotracheal intubation is performed with an uncuffed age-appropriate endotracheal tube. Nasal intubation is preferred to orotracheal intubation because the tube positioning is more favorable; there is more room for the laryngoscope and the tube is anchored more posteriorly, which facilitates the epiglottopexy and helps to place the aryepiglottic folds on tension during the supraglottoplasty. A leak check is performed to ensure that the tube is appropriately sized. The tube is secured around the head with a twill tie. A Benjamin-Lindholm suspension laryngoscope is inserted and the patient is placed in suspension. A ProciseTM (Smith+Nephew, Andover, MA, USA) laryngeal wand coblator is used with the ablation and coagulation set to 5 and 3 respectively. The coblator is used to foreshorten the median glossoepiglottic fold, also known as the suspensory ligament of the epiglottis. This is typically done with the coagulation setting because ablation is more powerful and can damage the underlying cartilage of the epiglottis or create an unwanted connection to the paraglottic space. Some surgeons denude the mucosa of the suspensory ligament with a carbon dioxide (CO2) laser; However, we prefer the coblator technique because the ligament visibly contracts and the epiglottis is significantly lifted with this method. As coblation continues, mucosa is removed from the base of tongue and lingual surface of the epiglottis until a rhomboid shape is achieved. It is important to divide the central portion of the suspensory ligament—at the deepest aspect of the vallecula—enough to allow this area to fold in on itself during the pexy. The ablation setting, or a cold-steel technique, can be used carefully during this portion of the procedure. A 5 millimeter laparoscopic needle driver (Karl Storz, El Segundo, CA) is then used to place a mattress suture from the tongue base to the lingual surface of the epiglottis. Five to six knots are tied down with the assistance of a knot pusher. During this step, an assistant holds tension on the left suture tail and the surgeon holds tension on the right. The surgeon slides the knot pusher at an angle along the laryngoscope to avoid coiling of the suture and ties the knot down securely. After the epiglottopexy is complete, attention is turned to the supraglottoplasty. The arytenoid is grasped and retracted postero-medially. A curved scissor is used to divide the aryepiglottic fold with care taken not to violate the lateral pharyngoepiglottic folds. The arytenoid is then reflected anteriorly and the underlying cartilages are palpated. The overlying redundant mucosa is truncated with sharp dissection. The same steps are carried out on the contralateral side. Patients may or may not be extubated in the operating room depending on medical co-morbidities, pulmonary status, age, and presence of post-surgical edema. If remaining intubated, flexible bronchoscopy is performed to check the position of the nasotracheal tube above the carina and to suction any bronchial secretions. The tube is then secured, and patients are transferred to the pediatric intensive care unit (PICU) for monitoring. They are extubated the following day if they meet extubation criteria and there are no concerning findings, such as pneumonia or new consolidation, on chest x-ray. Nebulized racemic epinephrine is administered immediately after extubation. Patients are started on dual acid suppression and receive a short course of peri-operative steroids, generally one to two additional doses. Post-operative swallowing is evaluated by a speech and language pathologist (SLP). Antibiotics are continued for one week. The antibiotic course is extended to two weeks if there is exposed arytenoid cartilage after the supraglottoplasty. Flexible laryngoscopy is performed in the office two weeks post-operatively to evaluate healing. Post-operative polysomnography is usually scheduled for 6-8 weeks after surgery.

This video demonstrates successful supraglottoplasty and epiglottopexy for sleep-variant laryngomalacia in a 6-year-old girl. Surveillance flexible laryngoscopy is performed two weeks post-operatively and demonstrates a well-healed larynx with the epiglottis nicely opposed to the base of tongue. The patient is currently tolerating a full oral diet with resolution of sleep symptoms.

Our case is representative of sleep-variant laryngomalacia, which can be successfully treated with endoscopic supraglottoplasty and epiglottopexy. There are several unique aspects of our technique, which are important to highlight. First, we recommend performing the epiglottopexy prior to the supraglottoplasty. The epiglottopexy places tension on the aryepiglottic folds when the epiglottis is suspended. By securing the epiglottis to the tongue base first, the new position of the supraglottic structures can be accurately assessed. This allows for a sufficient cut to be made in the aryepiglottic fold without having to account for subsequent repositioning of the epiglottis. This prevents over- or under-resection of tissue. The same concept is exhibited in reconstructive surgery; by convention, corrections are made in a top-down approach for the purpose of tissue conservation. As described above, we foreshorten the suspensory ligament and denude the vallecular mucosa with the coblator. However, some surgeons perform this step with a CO2 laser. As mentioned above, we prefer the coblator since it visibly shortens the ligament and the epiglottis can be seen elevating toward the base of tongue. Depending on the size of the child, coblation of the ligament alone can be sufficient and a stitch may not be necessary. With this technique, the area is left to heal by secondary intention. However, this usually necessitates a longer intubation. When possible, we prefer to use a suture to ensure long-lasting approximation of the epiglottis to the base of tongue. In general, we avoid performing concurrent lingual tonsillectomy and epiglottopexy as this can cause wide scarring of the epiglottis to the base of tongue, which may cause epiglottic flattening, altered flow around the epiglottis during swallowing, and aspiration. These concepts and surgical techniques are applicable to supraglottoplasty and epiglottopexy performed for any indication. The underlying etiology of sleep-variant laryngomalacia in otherwise normal and neurologically intact school-aged children is poorly understood. One proposed explanation is that mild upper airway obstruction combined with increased respiratory effort causes negative intrathoracic pressure potentiating an exacerbation of gastroesophageal reflux disease (GERD) during sleep. This can result in increased supraglottic edema, which may explain the redundant supra-arytenoid tissue and airway obstruction seen during a sleep-like state [2]. Our patients are placed on acid-suppression therapy prior to surgery and continued on the medications for several months following the procedure.  It is imperative to obtain a thorough history that includes sleep stridor to arrive at the correct diagnosis.  Keep in mind that the flexible laryngoscopy exam may appear unremarkable or demonstrate a subtle finding of “tall arytenoid towers.” Asking the child to run up and down the hallway prior to exam or to hyperventilate during the exam will often result in abnormal positioning of the arytenoids in sync with the phase of inspiration. Another potential mechanism of sleep-variant laryngomalacia may be related to the impact of hypoxia and hypercarbia on laryngeal function. Hypoxia helps to regulate the respiratory cycle and laryngeal tone. Hypoventilation, hypercapnia, and hypoxia may lead to decreased laryngeal tone through decreased neuromuscular transmission between the supraglottic larynx and centers for respiration in the brainstem [2, 4, 5]. In theory, surgical correction of sleep-variant laryngomalacia relieves the hypoxia and hypercarbia caused by obstruction and allows laryngeal tone to improve over time.

None

Thank you to Vidal Maurrasse for the voiceover.

1. Jackson, CJC. Diseases and Injuries of the Larynx. New York, NY: MacMillan 1942; 63. 2. Richter GT, Rutter MJ, deAlarcon A, Orvidas LJ, Thompson DM. Late-onset laryngomalacia: a variant of disease. Arch Otolaryngol Head Neck Surg. 2008 Jan;134(1):75-80. 3. Olney DR, Greinwald JH, Smith RJ, Bauman NM. Laryngomalacia and its treatment. Laryngoscope. 1999 Nov;109(11):1770-5. 4. McCray PB, Crockett DM, Wagener JS, Thies DJ. Hypoxia and hypercapnia in infants with mild laryngomalacia. Am J Dis Child. 1988 Aug;142(8):896-9. 5. Lanier B, Richardson MA, Cummings C. Effect of hypoxia on laryngeal reflex apnea: implications for sudden infant death. Otolaryngol Head Neck Surg 1983;91 (6) 597- 604