This chapter will review the preoperative, intraoperative, and postoperative considerations for successfully carrying out laryngotracheal reconstruction. Key points of consideration to ensure success while avoiding potential pitfalls will be highlighted.
49 Laryngotracheal Reconstruction
The management of laryngotracheal stenosis in children is often challenging and best managed at a tertiary institution with expertise in pediatric airway disorders, requiring a high level of integration of multiple medical services, led by an otolaryngologist. In general, multidisciplinary expertise in anesthesia, surgery, pulmonology, intensive care management, and pediatric hospitalist medicine is requisite to adequately manage these patients. Furthermore, expertise in nursing and speech therapy is essential for education, counseling tracheotomy care instruction, and home care integration. Although there has been an emerging trend toward the endoscopic management of these pathologies with airway balloons, open airway reconstruction remains the definitive tool for the long-term repair of pediatric laryngotracheal stenosis. In the long term, severe stenosis in particular appears to do better with open approaches. This chapter will focus primarily on open surgical approaches for laryngotracheal stenosis, focusing specifically on laryngotracheal reconstruction (LTR), which is not a single technique but a collection of multiple approaches depending on the nature of the stenosis (Video 49.1).
49.2 Definition and Classification
The incidence of laryngotracheal stenosis in neonates remains around 0.1% to 1% with a majority caused by long-term intubation. In some cases, the child may be born with congenital laryngeal stenosis—due to thickening of the cricoid ring or lack of recanalization of the airway during development. These instances of congenital stenosis are more frequent as part of some syndromes such as Down syndrome. The grading scale most universally employed to categorize subglottic stenosis (SGS), based on endotracheal tube size, was proposed by Myer and Cotton in 1994. They described Grade 1 SGS as 0% to 50% narrowing, Grade 2 SGS as 50% to 75% narrowing, Grade 3 SGS as 75% to 99% narrowing, and Grade 4 SGS as no-identifiable lumen as shown in ▶ Fig. 49.1. This grading scheme is important in comparative outcomes assessment, for planning surgical approach, and for objectively classifying disease severity. In 2009 Moniere introduced a modification of the Myer Cotton Scale, whereby each level of stenosis is further subclassified (a) for isolated stenosis, (b) for presence of medical comorbidities, (c) for glottic involvement, and (d) for both comorbidities and glottic involvement. This modified scale can add objectivity and lend important information when one counsels families regarding expected outcomes, or inform surgical decisions such as electing between a single- and a double-stage approach.
49.3 Evaluation, Preoperative, and Anesthesia Considerations
The evaluation of children with suspicion of laryngotracheal stenosis should include a careful history and physical assessment. Preoperative radiographic imaging plays a limited role in the diagnosis of laryngotracheal stenosis except to help characterize and determine the length of a stenotic airway segment. Undoubtedly, the gold standard in the preoperative airway evaluation for SGS characterization is rigid direct microlaryngoscopy and bronchoscopy (DLB) under general anesthesia. The preferred anesthetic management method for this evaluation is with children under spontaneous ventilation anesthesia with oxygen insufflation using a combination of inhalational agents and propofol or dexmedetomidine. This approach allows for a dynamic assessment of the laryngotracheobronchial tree while manipulating the airway without needing intubation or ventilation through a rigid bronchoscope. The surgeon is able to simply utilize narrow rigid fiberoptic “naked” telescopes for diagnostic purposes, minimizing airway trauma associated with the larger diameter ventilating bronchoscope. Usage of the ventilating bronchoscope is then reserved for cases where therapeutic or interventional maneuvers have to be performed in the trachea and mainstem bronchi. After DLB is performed, in order to objectively determine the severity of the stenosis, the lumen of the stenotic airway is typically sized with endotracheal tubes (ETTs). Direct microlaryngoscopy and bronchoscopy is supplemented in select patients by flexible nasopharyngolaryngoscopy, both done under sleep and awake. Sleep flexible endoscopy can help assess the tongue base level of the airway along with the nasopharyngeal and hypopharyngeal levels of the airway without laryngeal manipulation/suspension. Flexible nasopharyngolaryngoscopy is important in nonintubated patients where one should carefully assess the nasopharynx, oropharynx, and particularly the vocal cord level. Abnormalities of glottic mobility due to neurologic problems, scarring of the glottis, or involvement of the cricoarytenoid joints complicate surgical therapy. Flexible laryngobronchoscopy under anesthesia is also best accomplished without intubation, under spontaneous ventilation with oxygen insufflation through the side port of the flexible scope, most efficiently using a combination of dexmedetomidine and Propofol. This is particularly useful in patients with severe micrognathia to assess the degree of possible obstruction at the tongue base. Awake flexible laryngoscopy is important to rule out vocal cord immobility. Most infants with acquired SGS will have had a history of neonatal intubation. For neonates presenting with inflammatory SGS and multiple failures to extubate, close cooperation needs to occur among the neonatology and otolaryngology teams. Close attention should be paid to the patient’s medical condition, with emphasis on cardiopulmonary status, ventilation status, oxygen requirements, and the details of previous extubation failures. Possible concomitant presence of gastroesophageal reflux (GERD) should be investigated and treated, as many have reported a correlation between the presence of GERD and SGS and that GERD may affect surgical healing after LTR. Patients with dysphagia, or severe laryngeal and hypopharyngeal inflammation should be considered for eosinophilic esophagitis, as this emerging disorder has also recently been shown to be associated with SGS and influence negative healing and outcomes after LTR.
For patients with laryngotracheal stenosis without an existing tracheotomy tube, further and definitive reconstructive management is based upon the clinical picture and the severity of the stenotic segment. Presence of chronic pulmonary disease, often represented by baseline oxygen requirement in the setting of bronchopulmonary dysplasia and poor pulmonary functional reserve, is a contraindication for single-stage LTR, necessitating placement of a tracheotomy prior to or during the LTR. This is because reconstructive laryngotracheal surgery requires lung function adequate enough to withstand not only the surgery but also the postoperative course in ICU and subsequent extubation. Indeed, those children with significant pulmonary disease should undergo consultation by a pediatric pulmonologist. In general, it is inadvisable to perform LTR in children with anything more than a mild nighttime oxygen requirement; in selected cases, continued oxygen administration may be possible by nasal prongs after decannulation.
49.3.1 Timing of Reconstruction
The ideal timing of LTR surgery remains somewhat ill-defined. Although some have demonstrated that children younger than 24 months have higher rates of reconstruction failure despite lesser degrees of stenotic pathology when compared to older children, other series have suggested that although younger children have a higher rate of re-intubation after single-stage procedures, age alone may not be a stand-alone predictor for reconstructive failure (defined as failure to decannulate or avoid tracheotomy). In children with existing tracheotomies, any LTR timing decisions must consider the fact that severe SGS managed with tracheotomy, where formal LTR is deferred, is potentially life-threatening as yearly tracheotomy-specific mortality in children due to tracheotomy tube obstruction is purportedly in the 1% to 3.4% range. Associated tracheotomy tube morbidity also includes the need for comprehensive nursing care and monitoring, delayed speech and language development, feeding difficulties, and infection. Therefore, now many authors propose that reconstruction as early as possible is recommended so as to avoid tracheotomy-related complications.
49.4 Surgical Management
49.4.1 Role for Endoscopic Treatment
In general, endoscopic treatment is limited to acquired (and not congenital) airway stenoses. Classically, endoscopic treatment has taken the form of laser ablation of narrowing lesions, but is only useful for nonmature, noncircumferential, short soft lesions that comprise mild Grade 1 or 2 stenoses. Recent case series have also described the usage of balloon dilating catheters as potential tools that may successfully treat some patients with SGS, even if severe, but larger confirmatory studies are necessary to validate this approach. Comparative, retrospective studies have concluded that for severe SGS (grade 3 and 4), endoscopic balloon dilation has limited application compared with LTR in terms of achieving a long-term adequate airway lumen, and in some cases failed balloon dilation is perhaps even detrimental, increasing the risk of unplanned urgent interventions compared with LTR. 1 In any circumstance, dilation may certainly help temporize obstructive symptoms. Multiple, serial repeated dilations may eventually weaken the airway lateral walls, effectively making the pathology worse.
49.5 Open Surgical Techniques
49.5.1 Anterior Cricoid Split
The anterior cricoid split (ACS) procedure was introduced by Cotton and Seid in 1980 as an alternative approach to tracheotomy in the failing to extubate premature neonate with healthy lungs but laryngeal obstruction due to edema and early stenosis. In order to qualify for this procedure, the only reason for extubation failure must be laryngeal obstruction, and neonate should have grown to 1.5 kg, required no assisted ventilatory support for 10 days, have no supplemental oxygen need greater than an FiO2 of 35%, and have no evidence of congestive heart failure. The procedure consists of making an anterior vertical split through the first tracheal ring, cricoid cartilage, and lower thyroid cartilage followed by nasotracheal intubation for 10 to 14 days in the NICU. If criteria are strictly and carefully followed, case series have demonstrated ACS to be successful in avoiding tracheotomy in neonates. During ACS, placement of a small piece of thyroid ala cartilage into the vertical split (▶ Fig. 49.2) has also been described and may improve the success of the surgery.
49.5.2 LTR with Cartilage Grafting
LTR with interposition of cartilage graft was introduced by Fearon and Cotton in 1972 as a means to expand an otherwise narrowed subglottic airway segment. The principle of the procedure is to distract the cricoid cartilage either anteriorly and/or posteriorly by placing cartilaginous grafts over a luminal, appropriately sized stent. This procedure can be performed in either a single-stage (without a tracheotomy tube in place) or in a double-stage approach (where a tracheotomy tube is left in place postoperatively).
If carefully selected, the reported overall success rates of LTR in preventing tracheotomy or decannulation ranges from 81% to 100%. In general, the more severe the stenosis, the lower the likelihood of successful outcomes with LTR expansion techniques. For grade 3 and 4 stenoses, success rates are reportedly in the 75% to 85% range. Indeed, in these severe cases, when an adequate distance (at least >3 mm) exists between the lower margin of the vocal cords and the stenotic segment of the airway a consideration should be given to partial airway resection techniques, such as cricotracheal resection (CTR), as these reportedly have higher success rates than LTR for severe grade 3 and 4 stenoses, in the 90% to 95% range. 2 , 3 However, as opposed to CTR, when LTR procedures fail, there are more options for revision surgery—where LTR with re-grafting can be performed multiple times, resection techniques such as CTR can only be performed once.
Cartilage is the most frequently employed graft material to expand the laryngotracheal lumen during LTR procedures. While the use of thyroid alar and auricular cartilage grafts has been proposed, costal cartilage grafts remain the workhorse for LTR. 4 This is due to their availability, access, well-matched thickness, and generally robust rigidity requisite for reconstructing the laryngotracheal framework. Thyroid alar cartilage may be used for isolated anterior grafting, as a spacer graft during anterior cricoid splits in neonates or in select grade II or III stenosis, but its thin profile renders it difficult to carve/mold with flanges and inadequate for support of the posterior cricoid lamina. Hyoid bone has also been used with variable success in adult patients; however, its thinness limits expansive potential, bone is also difficult to carve to specification, and ossification may limit integration and neo-epithelialization along with significant resorption.