Central airway obstruction (CAO) is a potentially life-threatening condition that refers to an obstruction in the trachea or mainstem bronchi. While the incidence of CAO is unknown, it may be increasing as a result of the lung cancer epidemic as well and increased treatment options for patients with advanced stage disease.1 It is estimated that roughly 20% to 30% of patients with lung cancer will at some point develop CAO.2 In general, CAO is comprised of 4 types of anatomic lesions: purely endobronchial, purely extrinsic, mixed endobronchial and extrinsic, and tracheobronchomalacia. In addition to anatomic considerations, CAO is commonly split into malignant and nonmalignant categories, and both the anatomic and malignant status of a lesion will dictate which procedures can and should be considered.
Given the rising incidence of lung cancer, it is no surprise that the incidence of malignant CAO is greater than that of nonmalignant.3 Most commonly, malignant CAO is the result of metastasis of primary lung cancer as well as nonthoracic malignancies such as renal cell, breast, and thyroid cancers, and melanoma. Although rare, primary tracheobronchial tumors such as bronchial carcinoid, adenoid cystic carcinoma, and primary squamous cell and adenocarcinoma of the trachea may also occur.4,5 Nonmalignant CAO is most commonly caused by iatrogenic injury from prior endotracheal intubation or tracheostomy, and less commonly from connective tissue or inflammatory disorders. Table 1 lists common causes of malignant and nonmalignant CAO. Although less prevalent, the incidence of nonmalignant CAO may be increasing,6 possibly due to increased discovery because of increased imaging, greater availability of bronchoscopy, or an overall increase in endotracheal intubation and tracheostomy with resultant complications.
Chest radiography (CXR) has had limited utility in detecting the presence of CAO, due to low sensitivity and specificity. Multidetector computed tomography (MDCT) with multiplanar and 3-dimensional (3D) reconstructions has a superior resolution to depict anatomic detail of the central and small airways, improves lesion characterization, and more accurately assesses the extent of large airway pathology.7–9 MDCT also facilitates planning of potential invasive procedures.10 The ability to rapidly acquire MDCT images also enables dynamic assessment of the large airways using dedicated CT protocols such as end-inspiratory CT combined with either dynamic expiratory or static end-expiratory CT.11,12 MDCT thus provides a fast, noninvasive way to evaluate large airway diseases like tracheobronchomalacia and excessive dynamic airway collapse, which are recognized as an important cause of nonspecific respiratory abnormalities like cough, wheezing, increased sputum production, and infection.13
3D printing technology offers a fast and cost-efficient way to reconstruct 3D models of the large airways by using data acquired with MDCT. 3D reconstruction has become a valuable tool in evaluating airway malformations, tracheobronchomalacia, extrinsic compression from tumors, complications from airway interventions, and other disease processes. Personalized airway models can be used to create 3D stent models that can then be printed and converted into a silicone stent using injection molding14,15 (Figure Supplemental Content 1, http://links.lww.com/JTI/A129. Diagrammatic depiction of a simulated tracheal lesion and the approach to designing a stent with 3D printing).
Flexible bronchoscopy is a procedure in which a long, thin tube is passed through the mouth or nose and guided through the vocal cords into the lungs for inspection of the airways. Generally, flexible bronchoscopy is used for initial evaluation, although rigid bronchoscopy may also be selected when certain interventions are anticipated.16 Flexible bronchoscopy can typically be performed under moderate sedation, while general anesthesia is required for rigid bronchoscopy. The flexible bronchoscope has 3 parts: (1) a control handle that can flex or extend the distal tip of the scope and a suction port and opening for instruments to be inserted through the working channel; (2) a long, flexible shaft with lighting and imaging cables, and continuation of the working channel; and (3) a distal tip that contains a camera, light component, and the opening of the working channel. With the bronchoscope, the operator may suction out blood/secretions and pass small tools through the working channel, such as a transbronchial needle, forceps, a brush, or various thermal ablative catheters. A major limitation is that only one tool can be passed through the working channel at any given time. Variations of flexible bronchoscopes also exist, including some with specialized functions such as endobronchial ultrasound with an ultrasound probe at the distal tip, which allows the operator to visualize, and biopsy mediastinal and hilar structures that exist outside of the central airways.
Rigid bronchoscopy is another tool used to identify and manage CAO. Rigid bronchoscopes are comprised of hollow metal tubes available in several different diameters and variable lengths, ranging from shorter scopes that end in the trachea to longer scopes that terminate in the right or left mainstem bronchi. Bronchoscopes that terminate in the mainstem bronchi contain fenestrations that enable mechanical ventilation of the contralateral lung. The distal end of the rigid bronchoscope is typically beveled to facilitate lifting of the epiglottis during intubation with the scope. The proximal portion of the scope contains a built in or attachable port for jet or conventional ventilation. The light source can either be attached to the barrel or, more commonly, directly to a rigid telescope that is passed into the bronchoscope. The proximal portion also has a large working channel, which allows insertion of multiple instruments including various sized rigid forceps, rigid and flexible dilators, multiple types of lasers, scissors, argon plasma, electrocautery, snares, baskets and loops for removal of foreign bodies, cryotherapy catheters, microdebriders, and stent deployment devices.17 The most common instrument passed through the working channel is a flexible bronchoscope, which may be advanced into segmental airways for lavage of blood and secretions, and for visualization of the airway distal to the obstruction before coring, excising, or stenting.18 While the large caliber of the rigid bronchoscope makes intubation more difficult, its greatest advantage is that larger equipment and multiple tools can fit through the scope and be used simultaneously. Compared with flexible bronchoscopy, rigid bronchoscopy allows the operator to perform higher risk procedures with the ability to suction out larger quantities of blood/secretions.
TECHNIQUES FOR TREATMENT OF CAO VIA FLEXIBLE AND RIGID BRONCHOSCOPY
In patients with tracheal stenosis who are good surgical candidates, benign CAO (most commonly after intubation/tracheostomy) is usually treated with tracheal resection (Fig. 1). However, patients who are poor surgical candidates, and who have shorter life expectancies, metastatic malignant disease, or inflammatory/rheumatologic conditions, generally are managed with bronchoscopic techniques. These can be performed with either flexible or rigid bronchoscopy. Sometimes rigid bronchoscopy is preferred due to the enhanced control of the airway, improved ventilation during the procedure, and larger working channel. Alternatively, flexible bronchoscopy may be used (with or without rigid bronchoscopy) for better visualization of distal airways. Multiple studies in the literature report symptomatic benefit from bronchoscopic intervention in patients with CAO. In one prospective study, 37 patients with malignant CAO underwent multimodality bronchoscopic intervention (mechanical debulking, dilatation, endobronchial electrocoagulation and argon plasma coagulation (APC), balloon bronchoplasty, and/or airway stenting) with significant improvement in exercise capacity, lung function, and dyspnea at 30 days.19 Another prospective study found that interventional bronchoscopic procedures resulted in improved airway diameter (most >80%) and improved dyspnea at 6 months.20 However, there are no prospective, randomized studies that have examined mortality benefit or compared the various possible endoscopic treatment modalities to each other or to external beam radiation therapy. This leaves the decision on treatment modality to the multidisciplinary team, which can include oncology, pulmonary medicine, thoracic surgery, radiation oncology, and rheumatology. If an endoscopic therapy is chosen, operator preference, location and size of the lesion, and the potential risk of bleeding and airway obstruction are typically the determining factors when selecting rigid versus flexible bronchoscopy, and tools for tumor excision/destruction. CT of the neck and chest is the main imaging modality used to determine location, size, extent, and vascularity of a lesion and is crucial in the preoperative setting, especially with large central lesions, which may impair ventilation. Chest CT may also help determine the length of an airway lesion, its proximity to large vascular structures, and the presence of an airway fistula to the esophagus or other mediastinal structures.
CENTRAL AIRWAY DILATION
Central airway dilation is the act of using a tool to circumferentially enlarge a stenotic airway. This can be accomplished by (1) using the barrel of a rigid bronchoscope itself as a rigid dilator, or (2) using a controlled radial expansion balloon through the working channel of a flexible bronchoscope or over a wire placed into the airway under bronchoscopic and fluoroscopic visualization. Rigid dilation involves careful advancement of the beveled end of a rigid scope through a stenotic airway lesion, causing controlled dilation and often mild tearing of the stenosis. Extreme care and experience is needed during these dilations, as large tracheal/bronchial tears may occur if the operator fails to stay in the correct plane of the airway, which may lead to tracheal-esophageal or tracheobronchial-mediastinal fistula formation or trauma to surrounding mediastinal structures. A controlled radial expansion balloon dilation is performed by guiding a narrow catheter with a surrounding balloon through the working channel of a bronchoscope and through the stenotic airway. After the catheter is properly positioned, the balloon is inflated with saline to the desired diameter. Finally, a wire may also be placed into the stenotic lesion under direct scope visualization, with the removal of the bronchoscope over the wire, followed by dilation of a balloon with contrast-enhanced saline under direct fluoroscopic guidance. The bronchoscope is then reinserted to inspect the operative site for bleeding or other complications. While these modalities are often immediately effective, certain cases require placement of an airway stent to maintain patency, typically when the lesion is purely extrinsic and simple dilation causes very transient airway opening. A CXR is typically obtained after these procedures to evaluate for airway injury or barotrauma, which can manifest as pneumomediastinum, pneumopericardium, or pneumothorax.
Balloon dilation of the airways was first performed back in the 1980s21 and is best suited for circumferential stenosis of the central airways. One of the most common causes is postintubation/posttracheostomy tracheal stenosis, but can also be a sequela of inflammatory, infectious, or granulomatous conditions such as sarcoidosis, granulomatosis with polyangiitis (previously referred to as Wegener granulomatosis), posttransplant anastomotic stenosis, smoke or toxic substance inhalation, and trauma. Bronchoscopic intervention is indicated when a symptomatic patient with airway stenosis has dyspnea, shortness of breath, or postobstructive pneumonia and is not a candidate for a definitive surgical tracheal resection. In a retrospective study evaluating patients who had bronchoscopic balloon dilatation performed between 1997 and 2002, 24 patients underwent 59 procedures with half requiring subsequent stenting. All procedures resulted in immediate, postprocedure improvement in stenosis with no serious complications.22
MECHANICAL DEBRIDEMENT OF CAO
Mechanical debridement is often indicated in endobronchial and mixed central airway lesions (Figs. 2, 3) Flexible forceps may be used through the working channel of a flexible bronchoscope and can be opened and closed during debridement to remove small pieces of the tumor. Because of their small size, forceps have limited effectiveness at removing larger tumors and are often used in conjunction with other thermal and cryoablative technologies, rigid forceps, and coring techniques. Several different types of larger rigid forceps are available for use through a rigid bronchoscope, with an increased tumor or lesion resection per bite. Forceps come in various sizes and lengths with blunt or cutting edges and may have several teeth or 2 large teeth (referred to as “rat tooth forceps”).
When a large lesion is predominantly endoluminal, the operator, using the barrel of the rigid bronchoscope for visualization of the airway and lesion, may opt to “core out” the lesion. This allows rapid recannulation of the lesion and may be necessary if the patient is difficult to ventilate due to tracheal obstruction and if immediate patency is required. The operator must be careful to stay in the parallel plane of the airway to avoid perforation of the large airways, which can lead to various fistulae formations.
Finally, a newer means of mechanical debridement, referred to as a microdebrider, was originally used by ENT surgeons for sinus and glottic surgery and is now used in central airway interventional procedures. Microdebriders are powered instruments that consist of a hollow shaft with a small opening at the distal end that has an internal rotating blade and suction. Airway debulking is accomplished by simultaneously shaving and automatically suctioning removed tissue with visual guidance through a rigid bronchoscope. The main advantage of the microdebrider is the ability to rapidly excise tissue while maintaining a clear working field. There are no significant prospective data on the use of microdebriders in CAO. In a study by Lunn and colleagues, 16 patients with CAO were treated with the microdebrider, with successful recannulation of the airways in all patients. There was only mild bleeding, which was easily controlled with standard means, and there were no procedure-related or long-term complications reported.23 Postoperative chest imaging is not always necessary for small endobronchial lesions that are resected using mechanical means. However, if a lesion extends into an airway wall and an extensive mechanical debulking or coring procedure was performed, a CXR should be ordered to assess for complications. CT is rarely indicated in the immediate postprocedural period unless there are signs of a large air leak, airway trauma, or fistula formation.
Laser therapy is a commonly used method of treating intraluminal tumors, particularly those associated with bleeding. This modality is most amenable to short, localized lesions in which the distal region beyond the obstruction is visible. Contraindications, as with any thermal ablative technology, include patients requiring a high amount of oxygen (FiO2 >40%), those who cannot tolerate rigid bronchoscopy, the presence of a silicone or hybrid airway stent, and distal airway involvement. A major advantage of laser therapy is that it is immediate-acting and has great debulking capacity (Figs. 4, 5). Another distinct advantage is that the laser coagulates while cutting, which limits bleeding. While its onset of action is immediate, it may not be long-lasting, depending on the etiology of the underlying lesion. In the case of rapidly enlarging lesions or recurrent disease, it can be performed in conjunction with additional therapies that can maintain airway patency, such as placement of an endoluminal stent. Chest CT before a laser procedure may help to measure the length of the lesion and assess proximity to large vessels, as lasers can have a deeper level of penetration compared with other endoscopic thermal technologies. Postoperative chest imaging is not always indicated when the patient is asymptomatic and the lesion is endobronchial and small; however, a CXR may be performed following extensive debulking to assess potential complications and reexpansion of postobstructive atelectasis.
APC is a noncontact thermoablation technique, wherein argon gas is expelled from a probe that has been passed through the working channel of a bronchoscope. An electric current is applied at the distal end of the probe where the gas is being expelled, causing the argon gas to become ionized and to conduct a monopolar electric current that “grounds” itself to the nearest target lesion.24 Because of heat production, the tissue is destroyed and coagulated in the target airway, with denaturation of proteins and evaporation of intracellular water. Compared with laser therapy, APC carries less risk for airway perforation given its modest target depth of 1 to 2 mm, but can take longer to desiccate and debulk central lesions. Because of this minimal depth of penetration, postoperative imaging is generally not needed for simple APC ablation when the patient is asymptomatic. Argon gas embolism has been reported as a rare complication and typically is a clinical diagnosis due to hemodynamic instability from embolization of gas into the coronary arteries or neurologic sequelae due to cerebral gas embolization. Following APC, intracardiac air on intraoperative transesophageal echocardiography and intracerebral air on head CT have been reported. These complications can be avoided by keeping the gas flow rate below 2 L/min.
Electrocautery is a procedure that utilizes contact electrical thermal injury in the treatment of CAO. It can be performed via a flexible bronchoscope with a small flexible catheter or via rigid bronchoscopy using a larger rigid electrocautery probe. It is an immediate-acting, superficial, thermal therapy, which like APC carries a lower risk of perforation, as compared with laser. However, the risk of airway fire is still present as with all other thermal technologies we have discussed, and thus requires lower FiO2 (<40%) during use.
Brachytherapy, while expensive, is another option to treat CAO, although its use has declined in recent years with the advent of alternative, more cost-effective options. Endobronchial brachytherapy is generally reserved for palliative management of locally advanced non–small cell lung cancer causing CAO (Fig. 6), although it has also been used in nonmalignant cases such as lung transplant anastomotic stenosis. A catheter is passed through the working channel of a flexible bronchoscope and placed in the desired position within the airway under direct and fluoroscopic visualization. The bronchoscope is removed, and the catheter is secured in place typically through the nasopharynx. Finally, radioactive beads are delivered through the catheter, abutting the desired lesion, with the goal of decreasing lesion size and improving symptoms. This procedure is not curative but may be a good option when the patient has failed prior dilations, stenting, or thermal ablative technologies.
When airway dilation and debridement alone are unable to sufficiently relieve CAO, stenting may need to be performed, of which there are a myriad of options. Airway stenting was originally conceived in 1965 by Dr William Montgomery when he developed a silicone T-tube. However, it was not until 1990 that Dumon introduced the first completely endoluminal stent. Endobronchial stents remain the only nonsurgical tool available to combat extrinsic airway compression and are frequently used in intrinsic or mixed disease in conjunction with other therapies.
In the United States, there are 2 categories of endobronchial stents that are commercially available: metal and silicone25 (Fig. 7) (Table Supplemental Content 2, http://links.lww.com/JTI/A130. Types of airway stents, with advantages and disadvantages). Both maintain their position by exerting radial pressure and friction on the airway. The main advantages of metal stents are that they are less likely to migrate and may be placed via flexible bronchoscopy under moderate sedation, without endotracheal intubation and deep sedation (Figs. 8, 9). The main disadvantages of a metal stent are the increased potential for granulation tissue formation and difficulty with repositioning or removal if left in place for longer periods of time. These risks were greater with bare metal stents and have been somewhat mitigated with the advent of covered, self-expanding metal stents. Generally, metal stents should be avoided in patients with nonmalignant disease and an extended life expectancy and should be reserved for those with malignant CAO with a worse prognosis. Finally, perhaps the greatest disadvantage of metal stents is the frequency with which CAO eventually recurs. One study evaluated patients receiving metal stents and found that all 15 recipients developed stricture and granulation due to their stent.26 Other disadvantages include greater perforation risk and greater expense (Fig. 8C).
The second major category for airway stenting is silicone stents. Most of the commercially available silicone stents are based on the original Dumon stent, which is a silicone tube with external studs to decrease migration. They come in various shapes and sizes and are available in both straight and Y configurations. Their material is firm but compressible to allow deployment. Silicone stents are less likely to become incorporated into the tracheal and bronchial mucosa and are typically easier to remove than metal stents. Thus, silicone stents are more favorable for nonmalignant disease in patients with longer life expectancies.
Rigid bronchoscopy is required for silicone stent placement due to their large size and inability to be advanced or be adjusted using the smaller, flexible bronchoscope. The desired length and diameter of a stent may be estimated by preprocedure chest CT, combined with bronchoscopic measurements to ensure accuracy. The major disadvantage of silicone stents is their propensity to migrate. In a 2007 study, among 15 patients who received a silicone stent for central airway collapse, there were 9 stent migrations.27 While less likely, granulation tissue may also form at the end of silicone stents, similar to self-expanding metallic stents. The other disadvantage of silicone stents is their propensity to become occluded with mucus, especially when a longer length is used.
The Polyflex (Boston Scientific) is another type of silicone stent that is commercially available (Fig. 10). Made of polyethylene threads embedded in a layer of silicone, these stents do not have studding on the outside and, instead, have a thin wall resulting in a better inner to outer diameter ratio. Because they lack the studding of Dumon stents, they may be even more prone to migration. They are embedded with tungsten, making them radio-opaque, and are deployed out of a semirigid tube inserted down a rigid bronchoscope.
Finally, dynamic Y tracheobronchial airway stents were initially described by Freitag and colleagues in 199428 and remain commercially available for use (Figs. 11, 12). These are composed of U-shaped metallic rings in the tracheal limb anteriorly, with a posterior silicone membrane. This membrane is pliable and can be dynamically compressed during cough to physiologically mimic the human trachea, resulting in better mucus clearance. These are large stents, which are cumbersome to place, and typically do not even fit through a rigid bronchoscope. The Y stents must be deployed via direct laryngoscopy, with the folded stent placed through the glottis using a rigid deployment device, followed by rigid intubation and further adjustment of the stent’s location.
When evaluating a patient with CAO, who may be a candidate for stent placement, preoperative chest CT may help to determine the length, diameter, and type of stent (Y or linear) and to assess the expected proximity of the stent to important mediastinal structures. Postoperative CXR is recommended to provide a baseline of stent location, if it is radio-opaque, and possible subsequent migration. However, chest CT provides superior anatomic detail in assessing stent location and migration.
CAO is a growing problem that encompasses both malignant and benign lesions and comes in a variety of forms. A growing list of possible treatment modalities is at the clinician’s disposal, yet there is limited literature comparing outcomes and indications for the available options. As such, a multidisciplinary approach is encouraged, weighing individual operator comfort, available modalities, and the specific disease causing the CAO. MDCT and chest radiography are crucial in the clinician’s evaluation of obstruction and follow-up after the intervention.
1. Ernst A, Feller-Kopman D, Becker HD, et al. Central airway obstruction
. Am J Respir Crit Care Med. 2004;169:1278–1297.
2. Ginsberg RJ, Vokes EE, Ruben ADevita VT, Hellman S, Rosenberg SA. Non-small cell lung cancer. Cancer: Principles and Practice of Oncology, 5th ed. Philadelphia, PA: Lippincott-Raven; 1997:858–911.
3. Lortet-Tieulent J, Soerjomataram I, Ferlay J, et al. International trends in lung cancer incidence by histological subtype: adenocarcinoma stabilizing in men but still increasing in women. Lung Cancer. 2014;84:13–22.
4. Hemminki K, Li X. Incidence trends and risk factors of carcinoid tumors: a nationwide epidemiologic study from Sweden. Cancer. 2001;92:2204–2210.
5. Gaissert HA, Grillo HC, Shadmehr MB, et al. Uncommon primary tracheal tumors. Ann Thorac Surg. 2006;82:268–272; discussion 272.
6. Grillo HC. The history of tracheal surgery. Chest Surg Clin N Am. 2003;13:175–189.
7. Ngo A-VH, Walker CM, Chung JH, et al. Tumors and tumorlike conditions of the large airways. AJR Am J Roentgenol. 2013;201:301–313.
8. Kang E-Y. Large airway diseases. J Thorac Imaging. 2011;26:249–262.
9. Boiselle PM, Lee KS, Ernst A. Multidetector CT of the central airways. J Thorac Imaging. 2005;20:186–195.
10. LoCicero J, Costello P, Campos CT, et al. Spiral CT with multiplanar and three-dimensional reconstructions accurately predicts tracheobronchial pathology. Ann Thorac Surg. 1996;62:811–817.
11. Boiselle PM. Imaging of the large airways. Clin Chest Med. 2008;29:181–193; vii.
12. Lee KS, Boiselle PM. Update on multidetector computed tomography
imaging of the airways. J Thorac Imaging. 2010;25:112–124.
13. Ridge CA, O’donnell CR, Lee EY, et al. Tracheobronchomalacia: current concepts and controversies. J Thorac Imaging. 2011;26:278–289.
14. Chepelev L, Souza C, Althobaity W, et al. Pre-operative planning and tracheal stent design in thoracic surgery: a primer for the 2017 Radiologic Society of North American (RSNA) hands-on course in 3D printing. 3D Print Med. 2017;3:14.
15. Cheng GZ, Folch E, Wilson A, et al. 3D printing and personalized airway stents. Pulm Ther. 2017;3:59–66.
16. Colt HG, Prakash UB, Offord KP. Bronchoscopy
in North America: Survey by the American Association for Bronchology, 1999. J Bronchology Interv Pulmonol. 2000;7:8–25.
17. Yarmus L, Ernst A, Feller-Kopman D. Emerging technologies for the thorax: indications, management and complications. Respirology. 2010;15:208–219.
18. Semaan R, Yarmus L. Rigid bronchoscopy
and silicone stents in the management of central airway obstruction
. J Thorac Dis. 2015;7:S352–S362.
19. Oviatt PL, Stather DR, Michaud G, et al. Exercise capacity, lung function, and quality of life after interventional bronchoscopy
. J Thorac Oncol. 2011;6:38–42.
20. Amjadi K, Voduc N, Cruysberghs Y, et al. Impact of interventional bronchoscopy
on quality of life in malignant airway obstruction. Respiration. 2008;76:421–428.
21. Cohen MD, Weber TR, Rao CC. Balloon dilatation of tracheal and bronchial stenosis. AJR Am J Roentgenol. 1984;142:477–478.
22. Mayse ML, Greenheck J, Friedman M, et al. Successful bronchoscopic balloon dilation of nonmalignant tracheobronchial obstruction without fluoroscopy. Chest. 2004;126:634–637.
23. Lunn W, Garland R, Ashiku S, et al. Microdebrider bronchoscopy
: a new tool for the interventional bronchoscopist. Ann Thorac Surg. 2005;80:1485–1488.
24. Platt RC. Argon plasma electrosurgical coagulation. Biomed Sci Instrum. 1997;34:332–337.
25. Godoy MC, Saldana DA, Rao PP, et al. Multidetector CT evaluation of airway stents: what the radiologist should know. RadioGraphics. 2014;34:1793–1806.
26. Gaissert HA, Grillo HC, Wright CD, et al. Complication of benign tracheobronchial strictures by self-expanding metal stents. J Thorac Cardiovasc Surg. 2003;126:744–747.
27. Murgu SD, Colt HG. Complications of silicone stent insertion in patients with expiratory central airway collapse. Ann Thorac Surg. 2007;84:1870–1877.
28. Freitag L, Eicker R, Linz B, et al. Theoretical and experimental basis for the development of a dynamic airway stent
. Eur Respir J. 1994;7:2038–2045.
central airway obstruction; bronchoscopy; airway stent; computed tomography
Supplemental Digital Content
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved