Clinical Scenario

A 55-year-old white male with a 20–pack-year history of smoking presented with syncope, a 2 month history of increasing exertional dyspnea, and a cough that had been refractory to bronchodilators and antibiotics. He was married and lived with his wife. The patient’s father and one brother had died of lung cancer. The emergency department physician believed that the patient was disabled, required special care and assistance, and warranted immediate hospitalization. The physical examination, however, was normal except for stridor heard over the trachea during forced expiration. The workup for syncope included an electrocardiogram, two-dimensional echocardiogram, bilateral carotid duplex ultrasonography, and computed tomography (CT) of the head, all of which were normal. A pulmonary consultant ordered a CT scan of the chest, which revealed a 1.5 cm right upper lung (RUL) nodule and a 4 cm right paratracheal mass invading the trachea (Figure 1).

Figure 1 A, Normal tracheal lumen proximal to the tumor. B, Right paratracheal mass compressing and invading the lower trachea below the aortic arch. C, The normal tracheal lumen is seen at the level of the main carina distal to the tumor. D, Right upper lobe spiculated pulmonary nodule.

Question 1: If this patient is proven to have primary lung malignancy, which of the following is the most accurate predictor of survival?

Answer: D

Estimated survival is an important factor for decision making in all disease processes. If this patient has primary lung cancer, he would be clinically staged IIIB because of tracheal involvement (T4 tumor). In such cases of advanced cancer, prognostic considerations are especially important because treatment goals nearing the end of life may change from efforts to prolong life at all costs, to those of palliating symptoms, preserving quality of life, and maintaining dignity. Estimating survival based on objective data is warranted because physicians’ subjective assessments of predicted survival are often incorrect, with the direction of error being usually optimistic.*¹ Patients, in most circumstances, want their doctors to be realistic when it comes to prognosis. Although exact numbers may not be requested by patients or their families, having knowledge of relatively accurate prognostic indicators helps physicians conduct meaningful and honest discussions.

A clinical symptom such as stridor, although a sign of severe laryngeal or tracheal obstruction,² does not delineate a benign or malignant disease process. In general, stridor and other symptoms of upper or central airway obstruction (CAO), including exertional dyspnea, wheezing, or even syncope, are nonspecific.³ In fact, results from analysis of 100 variables from several studies showed that only dyspnea, dysphagia, weight loss, xerostomia, anorexia, and cognitive impairment were strongly and independently associated with cancer patient survival. These signs and symptoms were outranked, however, by assessment of performance status,¹ which deservedly has become a “sixth vital sign” in clinical oncology.^† Because performance status is the strongest prognostic indicator of survival in patients with cancer, it is frequently used as an entry criterion and adjustment factor in clinical trials of anticancer treatment.¹ One commonly used measure of performance is the Karnofsky Performance Status score (range, 0 to 100 in 10 point increments, where 0 is death and 100 is perfect health), a general measure of functional impairment in which the lower the score, the worse is survival for most serious illnesses.¹

This patient was assigned a Karnofsky score of only 40. In the setting of lung cancer, known central airway obstruction, and Karnofsky scores below 50, a combination of interventional bronchoscopy and external beam radiation therapy (EBRT) to relieve the airway obstruction has been shown to be the therapeutic option of choice, often resulting in rapid restoration of airway patency, improved symptoms, improved performance status, and increased survival.⁴

The case continues

Several hours after admission, the patient lost consciousness while standing to go to the restroom. He was promptly intubated with a 7.5 mm endotracheal tube (ETT) and was placed on mechanical ventilation assist control–volume control mode (ACVC): tidal volume was 500 mL, respiratory rate 14/min, fraction of inspired oxygen (FiO₂) 1.0 and flow 80 L/min, and positive end-expiratory pressure (PEEP) 5 cm H₂O. On these settings, the peak airway pressure was 60 cm H₂O and plateau pressure was 20 cm H₂O. Flexible bronchoscopy was planned to determine the pattern of tracheal obstruction, to identify associated mucosal changes, and to potentially support the indication for a therapeutic airway procedure. The only flexible bronchoscope available had an outside diameter of 6 mm.

Question 2: The next appropriate step is to:

Answer: D

In the absence of evidence that the endotracheal tube is kinked, or that the patient’s high peak airway pressures are due to mucus plugging or bronchospasm, it is likely that the obstructing tracheal mass itself is responsible for elevated peak airway and normal plateau pressures. Bronchodilators, therefore, would not be expected to improve respiratory status. Bronchoscopy should be rapidly performed, given that ventilation may be difficult because of a less than 2 mm difference between the size of the outer diameter of the bronchoscope and the internal lumen of the endotracheal tube.

Ideally, a No. 8 or larger ETT is preferred for bronchoscopy because it allows proper ventilation during the procedure and potentially prevents a further increase in peak airway pressure or development of auto-PEEP. These tubes provide at least a 2 mm difference between the scope and the ETT diameter, preventing critical alterations in ventilatory parameters.⁵ Changing the ETT, however, by extubation and repeat laryngoscopy or by use of a tube exchanger is hazardous in a patient with critical tracheal obstruction. Reintubation may be difficult, and should the airway be lost, ventilation cannot be ensured. A safer approach may be to perform inspection flexible bronchoscopy through the existing ETT while monitoring peak airway pressures, tidal volumes, heart rate, and oxygenation. In case of tachycardia, a rise in blood pressure, or oxygen desaturation during the procedure, the bronchoscope can be immediately withdrawn until stable vital signs return.

The case continues

Flexible bronchoscopy was performed after a bite block was inserted into the mouth and was secured around the existing ETT. A swivel adapter with a fitted rubber cap was attached, allowing bronchoscopy with minimal loss of tidal volume. The FiO₂ was increased to 1.0, starting 5 minutes before the procedure, and continued until after the procedure, then was titrated down to prebronchoscopy levels. PEEP was removed during bronchoscopy to avoid raising peak airway pressures by as much as 25 mm Hg.⁶ If discontinuation of PEEP had not been feasible, it would have been reduced by 50%. Volume control ventilatory mode was preferred so that the increasing airway resistance secondary to the bronchoscopy would not result in reduced tidal volume. If pressure-controlled ventilation had been used, the peak pressure setting would have been increased to compensate for the loss of tidal volume consequent to the increased resistance. Moderate sedation was achieved with 2 mg intravenous midazolam. Had the patient been unable to cooperate or otherwise tolerate the procedure comfortably, supplemental doses may have been needed in 1- to 2-minute increments and according to moderate sedation guidelines used in our institution. To remove airway secretions, bronchoscopic suction was applied using short, 3 second or less bursts, because as much as 200 to 300 cm³ of the patient’s tidal volume can be removed during each suction period.⁶

A mass was seen protruding from the right lateral and posterior walls of the lower trachea. The tracheal lumen was narrowed by more than 70%. A severe mixed extrinsic compression and endoluminal obstruction pattern was noted. The remaining airways were normal except for an enlarged carina (Figure 2). The procedure lasted only 1 minute, during which the peak airway pressure increased to 80 cm H₂O but without hypoxemia or hemodynamic instability.

Figure 2 A, Flexible bronchoscopic view of the mass showing 70% obstruction of the airway lumen. B, Bronchoscopic view distal to the tumor shows patent airways and normal main carina.

Question 3: Had this patient undergone physiologic testing during his initial workup, which of the following patterns would have been expected on the flow-volume loop?

Answer: C

In cases of extrathoracic tracheomalacia, laryngomalacia, or vocal cord processes, the flow-volume loop pattern may be that of flattening of the inspiratory curve (Figure 3, A). This is due to the fact that intraluminal pressure during inspiration (PL) is lower than extraluminal pressure (Pm = atmospheric pressure) for the extrathoracic airway. Therefore, any obstruction in this airway segment will be made worse during inspiration and will cause limitation of the inspiratory flow. During expiration, however, to have an expiratory flow, the intraluminal pressure (PL) is higher than the atmospheric pressure (Pm), the extrathoracic airway will dilate, and the flow will be normal.

Figure 3 A, Dynamic (i.e., variable) extrathoracic obstruction with flow limitation and flattening of the inspiratory limb of the loop. B, Dynamic (i.e., variable) intrathoracic obstruction with flow limitation and flattening of the expiratory limb of the loop. C, Fixed upper/central airway obstruction with flow limitation and flattening of both inspiratory and expiratory limbs of the flow-volume loop. D, In the “airway collapse” pattern, maximal flow is reached quickly after expiration of a small volume of air. Following maximal flow, a large fall in flow occurs, although only a small volume is exhaled. In a subsequent phase, the flow rate falls very little during the remainder of expiration. This phase is responsible for the long plateau. PL, Intraluminal pressure; Pm, atmospheric pressure; Ppl, pleural pressure.

The opposite is true in cases of variable intrathoracic obstruction such as intrathoracic tracheomalacia. In this case, the obstruction is worsened during expiration when the pleural pressures (Ppl) exceed the intraluminal pressure (PL) (Figure 3, B), and there will be flattening of the expiratory curve of the flow-volume loop.

Our patient had a tracheal mass limiting airflow both during inspiration and during exhalation. The pattern on the flow-volume loop would likely have been that of blunting of both inspiratory and expiratory curves (Figure 3, C), the so-called square pattern. This pattern may be seen before spirometry yields abnormal results but may not be appreciated until the airway diameter is already narrowed to about 8 to 10 mm and is often masked in severe COPD.

The airway collapse pattern (Figure 3, D) is usually seen in severe COPD and may suggest excessive dynamic airway collapse. This flow-volume loop pattern suggests compression of the mainstem bronchi and trachea. The sudden drop in flow velocity occurs early during expiration when intrathoracic pressure is still increasing, as documented by esophageal pressure measurements. Because there is no reduction of force expelling air from the lungs, the sudden fall in velocity can be explained only by a sudden increase in resistance to airflow, which occurs because of collapse of the central airways. The small, relatively constant flow (i.e., plateau) that occurs after the sudden drop in velocity is consistent with air being forced through a narrow airway.

The case continues

This patient was married and had good social support. He had previously expressed his desire for diagnosis and was ready to consider all available treatment options, including mechanical ventilation, surgery, and other treatment modalities. He very likely had lung cancer and thus required tissue diagnosis. Alternatives presented to the patient’s wife included rigid bronchoscopy, flexible bronchoscopy with biopsy, and percutaneous needle aspiration of the pulmonary nodule.

Question 4: Which of the following reasons for recommending rigid bronchoscopy should be given during the informed consent process?

Answer: D

Patients with respiratory failure and malignant CAO treated by bronchoscopic intervention, who subsequently underwent definitive systemic therapy, were shown to have improved dyspnea and longer survival (median, 38.2 months; range, 1.7 to 57.0 months) compared with those who did not undergo bronchoscopic intervention (median, 6.2 months; range, 0.1 to 33.7 months; P < .001).⁷ Restoring airway patency also improves exercise capacity and dyspnea.⁷ In this study, a significant improvement in symptoms was noted in 94% of patients undergoing airway stent insertion, although ultimately 40% of patients required multiple bronchoscopic procedures. This raises concerns about the cost per quality of life-years gained in patients with advanced disease.⁸ The expense of palliative interventional procedures such as bronchoscopic laser resection and stent insertion, however, is probably small compared with the costs of repeated external beam radiation therapy or prolonged hospitalization in the intensive care unit. Rapid tissue diagnosis, restoration of airway patency, subsequent extubation and removal from mechanical ventilation, and discharge from the intensive care unit to a ward or even outpatient setting because of this single procedure likely result in considerable savings that positively impact health care costs and resource utilization.

Evidence that removal from mechanical ventilation in critically ill patients with airway obstruction and respiratory failure is possible because of interventional bronchoscopic procedures is limited but reproducible. Lo et al. successfully extubated 7 of 7 patients who were on mechanical ventilation for respiratory failure from malignant obstruction.⁹ Colt and Harrell successfully removed ventilatory support in 10 of 19 patients (52.6%) with malignant and benign obstruction,¹⁰ suggesting that when procedures were unsuccessful, immediate referral to hospice and comfort care was justified. Schaffer and Allen inserted metal stents in 6 patients with malignant CAO and in 2 with benign CAO, successfully extubating 6 of them 2 to 11 days later.¹¹ Furthermore, successful removal from mechanical ventilation has been associated with prolonged survival when systemic treatment is also administered (98 vs. 8.5 days).¹²

The case continues

During our dialogue with the patient’s wife, it was believed that the risks of further airway compromise and clinical deterioration were outweighed by the benefit of restoring airway patency with an ability to immediately diagnose the abnormality, wean the patient from mechanical ventilation, and offer systemic therapy. Although we planned to make the diagnosis on frozen section of the tumor removed from the airway, we could have also chosen to perform endobronchial needle aspiration (EBNA), brushing, or endobronchial forceps biopsy. In our experience, EBNA has a lower risk of bleeding than brushing or biopsy, and it provides immediate diagnosis when on-site cytologic examination is available. On-site cytology is not yet the standard of practice in many institutions; however, it may increase procedure-related costs, and, contrary to endobronchial biopsy, tissue architecture is not seen.¹³

Question 5: Had rigid bronchoscopic resection not been offered, which procedure would you have chosen to make a diagnosis in this patient?

Answer: E

Rather than approaching the tracheal mass, some might address the pulmonary nodule for diagnosis. CT-guided percutaneous needle aspiration of the right upper lobe nodule has a high diagnostic rate (91%) but would not provide information on staging, and it carries an increased risk for pneumothorax, especially in patients on positive-pressure ventilation (5% to 60%).¹⁴ Endobronchial ultrasound (EBUS) radial probe-guided transbronchial biopsy has a diagnostic yield of 70%¹⁵ but requires special training for image acquisition and interpretation and is not universally available. Electromagnetic navigation (EMN) in addition to EBUS radial probe biopsy increases the yield to 88%. Pneumothorax occurs in about 5%¹⁶ of patients, but this rate might be higher (14%) in mechanically ventilated patients.¹⁷

Video-assisted thoracic surgery (VATS) or thoracotomy has excellent diagnostic yields for solitary pulmonary nodules, especially when curative surgery is subsequently warranted. This patient’s mediastinal adenopathy and tracheal involvement made him ineligible for curative surgery. Furthermore, these surgical procedures require general anesthesia with placement of a double-lumen tube to allow independent lung ventilation, which would not have been feasible until tracheal patency was restored.

In patients with visible endobronchial lesions, endobronchial needle aspiration (EBNA) with on-site cytology examination has a diagnostic yield of 96% when combined with biopsy, washings, and brushing, compared with about 76% using any of these procedures alone.¹³ Therefore EBNA at the bedside in the ICU is a reasonable alternative, particularly if diagnosis is desired before consideration for rigid or flexible bronchoscopic resection.¹⁸

The case continues

Rigid bronchoscopy with Nd : YAG laser standby was immediately scheduled to restore airway patency and provide diagnosis.

Question 6: If rigid bronchoscopy had not been available, what therapeutic alternative would you have chosen to restore airway patency in this patient?

Answer: E

Systemic therapies such as EBRT and systemic chemotherapy might be appropriate treatment strategies once a tissue diagnosis has been obtained. In case of non–small cell lung carcinoma (NSCLC), this tumor would be a clinical stage IIIB with an estimated median survival of 10 months and a 5 year survival of 7%.¹⁹ Current evidence shows that for individuals with unresectable disease, good performance score, and minimal weight loss, treatment with combined chemoradiotherapy results in better survival than radiotherapy alone. Concurrent chemoradiotherapy seems to be associated with improved survival compared with sequential chemoradiotherapy.²⁰ Restoration of airway patency should improve performance score and accelerate appropriate initiation of systemic therapy. Therefore even in the absence of rigid bronchoscopy, restoration of airway patency is warranted.

Brachytherapy is an endobronchial therapy with proven efficacy in patients with endoluminal tumor and a substantial extrabronchial component. A Cochrane meta-analysis concluded, however, that for palliation of NSCLC symptoms, endobronchial brachytherapy (EBB) alone was less effective than EBRT. For patients previously treated by EBRT who are symptomatic from recurrent endobronchial central obstruction, EBB may be considered.²¹ Tissue effects are delayed and necrotic debris after radiation can worsen airway obstruction.

Photodynamic therapy (PDT) has resulted in successful weaning from mechanical ventilation of a small number of patients with malignant CAO and respiratory failure.²² The therapeutic effect of PDT is delayed, however, for at least 48 hours, and this might actually worsen airway obstruction during the initial post-treatment period because of sloughing of airway mucosa and retained tumor debris. PDT usually is not indicated in patients with critical tracheal narrowing but appears to be most effective when there is more than 50% narrowing from lobar or segmental bronchial mucosal disease.²³

Cryotherapy results in thrombosis and necrosis of tumor tissues and would address only the exophytic, endoluminal component of the obstructing airway lesion. Cryotherapy has been shown to be most effective when performed in combination with EBRT.²⁴ Similar to PDT and EBB, the effect is delayed and initially might worsen airway obstruction.

Electrocautery and argon plasma coagulation (APC) remove the exophytic component and allow superficial cauterization (3 to 6 mm), which may not suffice to stop large airway bleeding in cases of obstructing tracheal tumor. As gas is forced into the airway wall, it can collect in a blood vessel, causing embolism or perforation. Argon gas is heavy, inert, and 17 times less soluble in the body than carbon dioxide. Thus, it may also pass into the systemic circulation.²⁵ Although the tracheal wall is fairly dense, erosion from tumor presents a risk to major vessels. Systemic, life-threatening gas embolism occurring as a complication of bronchoscopic application of APC has been reported²⁶ and might be of particular risk in view of the highly vascular pattern of this patient’s particular tracheal lesion.

Fully or partially covered metal stent insertion is a plausible alternative in cases of malignant CAO. This can be performed through the ETT in the ICU.²⁷ It is also a suitable alternative in patients with significant comorbidities precluding general anesthesia. These stents are more costly than silicone stents and can be difficult to remove. Because the bulk of the obstruction resulted from malignant extrinsic compression, the need for excessive debulking was unlikely,²⁸ and stent insertion alone could probably restore airway patency. These stents can be placed with or without fluoroscopic guidance.^27,29

The case continues

The patient’s family had good insight into his disease. Expectations were realistic, and we had learned from the patient’s wife that they had cared for his dying father and brother, both of whom had lung cancer. After they had been advised of all of the alternatives, the family elected to proceed with rigid bronchoscopy under general anesthesia. They were informed of the possibility of failure to restore airway patency and were told about the risks for bleeding, perforation, worsening respiratory distress, temporary or prolonged mechanical ventilation, and death.

Question 7: Which type of anesthesia method would you suggest for this patient’s rigid bronchoscopy?

Answer: A

Concern for loss of airway during induction exists because of critical tracheal narrowing.³⁰ Induction should be cautious and preferably should be conducted without neuromuscular blockers; use of these might result in complete loss of muscular tone of the posterior membrane, possibly worsening an already critically narrowed airway.³¹ Spontaneous-assisted ventilation with intravenous anesthetics has been shown to have a good cardiac safety profile in patients undergoing rigid bronchoscopy under general anesthesia. Spontaneous-assisted ventilation is provided through an open circuit during the rigid bronchoscopic procedure.³² This usually is done without the use of muscle relaxants and requires an Ambu bag with a high-flow oxygen source connected to the side port of the rigid bronchoscope. In case of an air leak resulting in an inability to ventilate and oxygenate satisfactorily, the mouth and nose can be packed with gauze and Vaseline nasal packing.

The delivery of volatile anesthetics may be problematic when rigid bronchoscopy is performed using an open system with spontaneous-assisted ventilation because the quantity of anesthetic gas reaching the patient is uncertain, and operating room personnel and the operator are at risk for pollution from escaped anesthetics.³³

Positive-pressure ventilation will increase flow velocity and turbulent flow past the region of stenosis. Subsequently, a laminar flow pattern cannot be resumed, potentially resulting in ineffective ventilation of the distal airways and loss of effective gas exchange.³⁴

Jet ventilation can be considered for severe tracheal obstruction.³⁵ High-frequency jet ventilation (HFJV) or intermittent low-frequency jet ventilation can be delivered through a rigid or flexible bronchoscope. HFJV through catheters with internal diameters as small as 2.0 mm placed beyond the tracheal obstruction allows safe levels of oxygenation, provided an adequate expiratory passage is present.³⁶ Jet ventilation poses a risk for barotrauma and pneumothorax and therefore was not our method of choice for this patient.

The case continues

Rigid bronchoscopy with spontaneous ventilation was performed. The patient was extubated and under direct visualization was intubated with a 12 mm Efer-Dumon rigid nonventilating bronchoscope (Bryan Corp., Woburn, Mass). Neodymium-doped yttrium aluminum garnet (Nd : YAG) laser was available on standby in case of bleeding or for tumor coagulation. Various types and sizes of silicone stents were also readily available (Figure 4).

Figure 4 A, Efer-Dumon rigid bronchoscopes of different sizes with a rigid telescope (second instrument from the left). B, Two straight studded silicone stents (12 × 20 mm and 14 × 40 mm, respectively).

Potential dangers during bronchoscopic resection of a lesion in this location involve the superior vena cava and azygos vein located adjacent to the anterior and lateral wall of the lower trachea. The aorta is anterior (Figure 5).

Figure 5 Fused bronchoscopic and computed tomography (CT) image at the level of the tumor, showing the adjacent superior vena cava (SVC) and aorta.

Moderate bleeding occurred when the tumor was probed with the suction catheter (Figure 6). This stopped after instillation of 60 mL of saline. The tumor was debulked using the beveled edge of the rigid bronchoscope. Laser was not necessary. A 14 × 50 mm straight studded silicone stent was inserted into the lower trachea so that the distal aspect of the stent was located 1 cm above the main carina (see Figure 6). Preliminary tissue diagnosis was non–small cell carcinoma.

Figure 6 A, Hemorrhagic secretions after probing with a suction catheter. B, A 14 × 50 mm straight studded silicone stent in the lower trachea restored airway patency.

The case lasted 40 minutes. No additional intravenous anesthesia was given 10 minutes before rigid extubation, after which the patient was breathing well spontaneously. Reintubation with an endotracheal tube was not necessary. The patient was transferred to the ICU for 24 hours, during which no postoperative complications were noted. He was discharged home 48 hours after the rigid bronchoscopic intervention. Final diagnosis was squamous cell carcinoma. He returned to work and was fully ambulatory when seen in our clinic 1 week later.

Question 8: What is the most appropriate next step?

Answer: A

Complete staging should be performed to better assess prognosis and treatment strategy. This includes a whole body positron emission tomography (PET)/computed tomography (CT) scan and brain magnetic resonance imaging (MRI), which in this patient showed no extrathoracic disease. Because this patient had stage IIIB NSCLC (T4 tumor due to tracheal involvement), he was not a surgical candidate. Optimal treatment for stage IIIB NSCLC depends on several variables, including extent of disease, age, comorbidities, performance scores, and weight loss.²⁰ This patient’s Karnofsky performance status score post procedure had increased to 90; therefore referral for palliative or hospice care was considered premature. Concurrent chemoradiotherapy, which prolongs survival and is at the present time considered to be the standard of care for inoperable locally advanced NSCLC,³⁷ is usually restricted to healthy cancer patients who maintain a good performance status. Because this patient’s status had significantly improved,³⁸ concurrent chemoradiotherapy was offered following discussion in our multidisciplinary chest conference (with representatives from cardiothoracic surgery, oncology, critical care medicine, interventional pulmonology, and radiation oncology).

The case continues

While on systemic therapy (1 month after the procedure), the patient developed acute shortness of breath and an inability to raise secretions.

Question 9: The most appropriate next step is to:

Answer: A

Acute onset of shortness of breath and difficulty raising secretions are likely due to a stent-related complication, rather than to progression of disease or pulmonary embolism. Stent migration should be expected after radiation therapy, especially in cases of tumor shrinkage, causing loss of intimate contact between the stent and the airway wall. In one report, indwelling airway stents could be removed in 50% of patients with malignant CAO at a median of 31.7 days after successful systemic therapy.³⁹ The risk of stent migration has an estimated incidence of 1% to 5%.²⁸ Other stent-related complications include kinking, obstruction by tumor or granulation tissue overgrowth, bacterial colonization, and impaired mucociliary clearance causing retained secretions and stent obstruction. Repeat flexible bronchoscopies are reportedly necessary to remove secretions in up to 40% of patients with indwelling stents.⁸ Additional rigid bronchoscopic procedures are seldom required in patients with malignant CAO, in part because of their limited survival.^10,28

In this patient’s case, pulmonary function testing might reveal reduced expiratory flows, but findings are nonspecific and would not preclude a diagnostic bronchoscopy. Computed tomography might identify stent migration, tumor shrinkage, or obstruction by mucus, but would not be therapeutic. Consulting palliative care medicine could be premature because a reversible stent-related complication is most likely.

In fact, outpatient flexible bronchoscopy showed that the stent had migrated distally to the main carina (Figure 7). Rigid bronchoscopy with stent revision was promptly scheduled for the same morning. The patient had begun to cough violently. By the time of rigid bronchoscopy, the stent had partially migrated into the right main bronchus, crossing over the carina and nearly occluding the entrance of the left main bronchus. The stent was removed with rigid forceps. The airway was patent, and no evidence of recurrent endoluminal disease was found (see Figure 7). One year later, the patient continued to do well, had not required stent reinsertion or repeat bronchoscopic interventions, and had no respiratory complaints. Chest CT showed only chronic radiation pneumonitis changes with fibrosis (see Figure 7).

Figure 7 A, Stent dislodged from the lower trachea 1 month after insertion, signaling successful concurrent chemoradiotherapy. B, The stent had migrated 1 cm distally and was sitting on the main carina, partially occluding the entrance to the right and left mainstem bronchi. C, Rigid bronchoscopic view of the stent migrated farther down into the right mainstem bronchus and causing near complete occlusion of the left mainstem bronchus. D, Rigid bronchoscopic view after stent removal shows patent airway and no evidence of obstruction. E, Flexible bronchoscopic view 1 year after the initial procedure and after completion of systemic therapy shows maintained airway patency and no evidence of tumor. F, Computed tomography 1 year after the procedure shows a patent trachea, no mediastinal tumor, and mild radiation fibrosis within the field of prior external beam radiation.