Endoscopic removal of a self-expanding metallic airway stent: A case report

Abstract

Self-expanding metallic stents are sometimes placed for the management of obstructing airway lesions or conditions such as airway wall malacia or tracheal stenosis. However, endoscopic removal of these devices from the airway can pose extreme challenges for both clinical airway management as well as for the administration of general anesthesia. We report on a 61-year-old man with a complex cardiac history presenting for endoscopic stent removal necessitated by the formation of extensive granulation tissue. Comorbidities included a history of myocardial infarction, an ischemic cardiomyopathy with severe left heart failure (ejection fraction of 25%), mild right heart failure, 2+ tricuspid regurgitation status post tricuspid valve repair, and atrial fibrillation. An automatic external (wearable) cardiac defibrillator (Zoll Life Vest) was also in place. Induction of anesthesia was carried out using etomidate, with maintenance of anesthesia carried out with a propofol infusion (total intravenous anesthesia). Rocuronium was used for neuromuscular blockade. A size 4 iGel supraglottic airway and, later, rigid bronchoscopy formed the basis for airway management. Stable conditions were met through the 2-h procedure, and the patient recovered uneventfully. Our successful experience in this case leads us to propose further use of a supraglottic airway in conjunction with total intravenous anesthesia for these procedures.

Key Words: Airway management, Flexible bronchoscope, Rigid bronchoscopy, Self-expanding metallic stents, Supraglottic airway, Total intravenous anesthesia

Core tip: Endoscopic removal of self-expanding metallic airway stents may be necessitated by the formation of extensive granulation tissue, but can pose difficult challenges to both the proceduralist and the anesthesiologist. Total intravenous anesthesia utilizing a propofol infusion and rocuronium for neuromuscular blockade can be useful in such cases. Induction of general anesthesia with etomidate can be useful in patients with poor ventricular function. Airway management can be achieved with an iGel supraglottic airway and later, rigid bronchoscopy.

INTRODUCTION

The incidence of central airway obstruction is on the rise, paralleled with the increasing incidence of lung cancer[1-4]. The advent of endoscopic methods has allowed for less invasive management of these patients, with airway stenting as one of several options available to patients with central airway obstruction[3,5-9]. Silicone and self-expandable metallic airway stents (SEMS) are of the two main options in managing the airway in such cases. While silicone stents are often the preferred modality, SEMS are frequently used in cases of airway wall malacia or stenosis, especially in the context of malignancy.

SEMS have the advantage over silicone stents of a lower migration rate, thinner wall construction (allowing for greater cross-sectional airway diameter), and conformation to irregular airways[6]. However, they are prone to excessive granulation and stent fracture[6]. Indications for SEMS removal include excessive or recurrent granulation, stent failure or fracture, infection, mucous plug formation, stent migration, or achievement of successful treatment[10-12]. While SEMS placement is usually achievable with relative ease, endoscopic removal of these stents can be very difficult[10,11,13]. Complications during stent removal may include pneumothorax; tracheal, bronchial, laryngeal or pulmonary artery damage, and airway obstruction during or following stent removal. Furthermore, when a rigid bronchoscope is used, leaks around the bronchoscope can pose a problem, especially when using potent inhalational agents such as sevoflurane[14]. These issues can pose a great challenge to all participants[14].

CASE REPORT

The patient was a 61-year-old man who presented for SEMS removal from the lower trachea via rigid bronchoscopy. The indications for the stent removal were recurrent bacterial infections, and granulation tissue/stricture formation. His cardiac history included a history of myocardial infarction, ischemic cardiomyopathy with severe left heart failure (ejection fraction of 25%), mild right heart failure, 2+ tricuspid regurgitation status post tricuspid valve repair, and atrial fibrillation. An automatic external (wearable) cardiac defibrillator (AECD) (Zoll Life Vest)[15] was in place. Other past medical history included hypertension, hyperlipidemia, and chronic obstructive pulmonary disease. In 1998, following a severe myocardial infarction he underwent endotracheal intubation, tracheostomy and mechanical ventilation for 3 mo, and subsequently developed tracheal stenosis. His tracheostomy remained for one year before being reversed. Following reversal he continued to have recurrent tracheal stenosis, and underwent a tracheal resection, re-anastomosis and placement of two SEMS, which had remained in place for 13 years. In an attempt to minimize recurrent granulation tissue after his re-anastomosis he also received multiple radiation treatments to his trachea.

At the time of initial presentation to Cleveland Clinic, he had 50%-74% narrowing of the upper and lower third of the trachea. The two stents were found embedded throughout the whole trachea. He previously had a total of 50 bronchoscopic procedures to remove granulation tissue and scarring. He initially presented to our facility for attempted removal of an infected pacemaker/international classification of diseases lead resulting in recalcitrant bacteremia. Prior to undergoing this procedure, given his history of tracheal stenosis, our interventional pulmonary team was consulted to maximize his airway for endotracheal tube placement. During his initial bronchoscopy it was felt that an attempt at removal was reasonable given the tremendously negative effect on his quality of life from the stents. The SEMS in the upper trachea was removed successfully using a combination of the electrocautery blade and rigid debulking. At this time, we focused on removing his lower SEMS, which remained a challenge given its position in the lower portion of the airway.

The patient had a Mallampati class III airway[16,17] with no other significant airway abnormalities. His height was 177.8 cm, and weight was 106.8 kg. Physical examination was within normal limits with the exception of a productive cough, expiratory wheeze and occasional stridor. Given his cardiac history, he was judged to be American Heart Association IV. Blood pressure on day of the procedure was 109/67 mmHg, with a pulse rate of 61/min. Hemoglobin was 9.7 g/dL. Blood biochemistry was within normal limits. His electrocardiogram showed normal sinus rhythm, and echocardiography revealed a left ventricular ejection fraction of 25%. His chest X-ray showed bilateral scattered densities, but otherwise free of focal consolidations or masses. On the morning of the procedure, he had received the following oral medications: potassium 10mEq, primidone 50 mg, gabapentin 300 mg, and metoprolol 25 mg. He had been off clopidogrel for 6 d[18,19].

We proceeded with general anesthesia, opting for total intravenous anesthesia (TIVA). His wearable AECD was removed after induction due to the likelihood of using cautery. The patient was induced with the following intravenous drugs: etomidate 14 lidocaine 50 mg, midazolam 2 mg, of rocuronium 50 mg, and a 120 mcg/kg per minute propofol infusion. Following induction, he was intubated with a size 4 iGel supraglottic airway (SGA)[20-22] without difficulty. After SGA placement, the patient’s blood pressure dropped slightly to 100/65. Subsequently, the propofol infusion was decreased to 90 mcg/kg per minute; a phenylephrine infusion of 50 mcg/min was started for a total of 15 min. Blood pressure responded adequately. After the patient was fully anesthetized, a flexible bronchoscope was introduced through the iGel SGA and advanced to the tracheobronchial tree. A full airway inspection was performed with the flexible bronchoscope. The iGel SGA was then removed and an orange-stripe (13.2 mm) bronchial rigid scope was introduced (Bryan-Dumon). Ventilation was achieved through the side arm of the rigid bronchoscope.

Several precautionary measures were taken prior to the removal phase in the event that the stent could not be removed or unforeseen airway obstruction occurred. Ventilation with 100% oxygenation was achieved prior to the periods of apnea required during cautery. To prepare for possible SEMS fracture during the removal phase, balloon dilators were ready so as to facilitate in dissecting the SEMS from the airway wall, expanding the airway in case of collapse and tamponading bleeding. An electrocautery knife was then used to make long cuts through the overgrown granulation tissue and scar in the lower trachea. This was done in order to facilitate debulking of this tissue with the rigid beveled tip. Once the tissue was debulked, sections of the SEMS were able to be separated from the airway wall. As more of the stent was liberated, it was felt appropriate to attempt removal. Approximately 90%-95% of the lower tracheal stent was removed via forceps. The patient was maintained with two more 10 mg boluses of rocuronium through the case. Pressure controlled ventilation with 15 lpm O₂ delivery was used with appropriate pauses during cauterization. He was maintained on an average pressure of 27 cm H₂O, positive end expiratory pressure of 3 cm H₂O, minute ventilation of 3.7 L/min, inspiratory to expiratory ratio of 1:2, and a respiratory rate of 12/min. Through the 2 h case, the patient’s blood pressure ranged from 100/65 to 145/100 mmHg, and heart rate ranged from 50 to 65/min. Oxygen saturation from pulse oximetry remained between 90%-100%. The patient received 700 mL of normal saline solution, and had an estimated blood loss of 5 mL. Towards the end of the case, the patient received 50 mcg of fentanyl, 20 mg of lidocaine, and 4 mg of ondansetron. For reversal of the rocuronium, he received 2.5 mg of neostigmine and 0.5 mg of glycopyrrolate. After confirming spontaneous breathing, strong hand grip, and sustained head lift for 5 s, the patient was extubated without difficulty. He was able to communicate shortly after extubation. He recovered in the post-anesthesia care unit without any post-operative nausea or vomiting or respiratory complications. End-of-operation blood pressure was 130/70 mmHg, heart rate was 51/min, respiratory rate was 16/min, and oxygen saturation was 97% on 3 lpm O₂ delivered via nasal cannula.

DISCUSSION

Two of the main concerns in removal of SEMS are the risk of airway obstruction from stent fracture or bleeding during stent removal. Additionally, the use of electrocautery devices and balloon dilatation necessitates full communication between the surgeon and the anesthesiologist. We initially elected to use an iGel, a supraglottic airway, given the patient’s previous upper tracheal stent, which might have led to difficulties with attempted use of an endotracheal tube (ETT). A previous case series comparing the i-Gel to ETTs found comparable ventilation overall[23]. Further, iGels are superior to ETTs regarding ease of insertion; compared to laryngeal mask airways, iGels also guarantee proper positioning for supraglottic ventilation. Given the need to switch between ventilation with various devices, the iGel allowed for a rapid and easy transition between ventilation through the bronchoscope and through the iGel. While the iGel does not guarantee a definite airway compared to the ETT[14], we shortly replaced the iGel with a rigid bronchoscope that allowed for a more defined airway. During the removal phase, a clear communication line was made between the interventional pulmonologist and the anesthesiologist[14]. The patient was oxygenated with 100% oxygen, which helped maintain adequate oxygenation throughout periods of apnea during cauterization.

Overall, our patient did well with the supraglottic airway management device followed by ventilation via rigid bronchoscope, requiring only one trial of placement without trauma, and minimal use of anesthetic agents through the two-hour case. He recovered without difficulty, with only a brief period of coughing, and recovered well the post-anesthetic care unit to go home that same day.

Current practice has leaned towards using balanced anesthesia with both intravenous and inhalational agents, although the debate on this topic remains in evolution[24-27]. As discussed below, in bronchoscopic cases we tend to avoid the use of potent inhalational agents both because of concerns for operating room pollution and the fact that delivery of the agent to the patient can at times be unreliable[14]. One of the concerns with the use of TIVA is inadvertent consciousness during the procedure[28,29]; in our patient, we were able to monitor depth of anesthesia with Bispectral Index monitoring, maintaining a level of less than 50 through the case.

This concern notwithstanding, however, there are several advantages of using TIVA in bronchoscopy procedures. As noted above, even with procedures not involving the airway, there are concerns over contamination of the work-space environment by inhalant anesthetics. In removal of the SEMS, the airway circuit had to be periodically disrupted and disconnected to allow for introduction of different types of bronchoscopes, introducing the possibility of leakage of various gas into the operating room; the use of TIVA allows for avoidance of contamination. Studies comparing TIVA and inhalational anesthetics found better surgical field view in endoscopic sinus surgery with decreased operative time and reduced intraoperative bleeding[30,31]. During the surgery, there was little bleeding in the trachea, and the surgical field view was clear. Other benefits of TIVA include less post-operative nausea and vomiting[32-35], resulting in better post-operative recovery. In our case, the patient reported no post-operative nausea or vomiting, and was able to be discharged the same day of surgery.

Anesthetic agents can have an adverse effect on patients with heart failure, causing diminished cardiac reserve by depressing the sympathetic system and myocardium, as well as modifying cardiovascular control mechanisms[36]. Etomidate is a carboxylated imidazole derivative that enhances GABA activity, and is commonly used as an induction agent in patients with a history of heart failure as it causes minimal cardiac depression compared to induction agents such as propofol. The standard induction dose of etomidate is 0.3 mg/kg[37]. Only in supratherapeutic doses does etomidate have a negative inotropic effect[37]. Etomidate infusions have been shown to inhibit 11-beta-hydroxylase activity, leading to suppression of adrenocorticoid synthesis[38]; it is debatable whether this happens with single boluses of etomidate[39-44]. Another advantage of using etomidate for induction includes its rapid onset effect with short period of action; etomidate has also been shown to maintain hemodynamic stability in patients with hypovolemia or shock[45]. We used a 14 mg induction bolus for our patient, and he was able to maintain adequate hemodynamic stability through the case, only requiring 15 min of phenylephrine infusion of 50 mcg/min.

Cases similar to ours have been reported. Chawla et al[11] described a case of a patient who developed cough and dyspnea one and a half years after SEMS placement. The authors identified “stenosis above the level of stent with granulation tissue inside the stent, stent fracture in lower part and stent migration to right main bronchus” , necessitating stent removal under general anesthesia with the assistance of a rigid bronchoscope. Fruchter et al[13] conducted a retrospective review of patients requiring tracheobronchial stent placement after lung transplantation in which the SEMS had to be removed. In six of the 24 patients receiving a SEMS, the stent “had to be removed due to excessive granulation tissue formation and stent obstruction” with the average time from placement to removal being 30 mo. Of these six patients, the stent was able to be completely removed in all but one case, and no major complications were encountered. This series differed from our case in that the procedures were carried out under conscious sedation rather than general anesthesia, and utilized flexible bronchoscopy rather than rigid bronchoscopy.

In conclusion, a patient with multiple severe comorbidities did well with our anesthetic technique of using a supraglottic airway management device, TIVA and etomidate for removal of a SEMS. Future SEMS removal may proceed with such a similar technique.