Montgomery T-tube tracheal stent implantation with high-flow nasal oxygen under conscious sedation: A case report

Introduction

Subglottic stenosis following prolonged endotracheal intubation or tracheostomy has become an increasingly recognized clinical problem. Surgical resection with tracheal reconstruction can provide definitive treatment; however, it is invasive and not suitable for all patients. The Montgomery T-tube offers a less invasive alternative that restores airway patency while preserving phonation. Placement usually requires rigid bronchoscopy under general anesthesia, which necessitates cervical hyperextension and is therefore inappropriate in patients with cervical spine instability.^1,2

Our patient developed subglottic stenosis after tracheostomy for spinal cord injury with incomplete paraplegia and could be weaned from the tracheostomy tube, achieving adequate ventilation and phonation with a T-tube. However, rigid bronchoscopy was challenging because of cervical instability, and general anesthesia carried an increased risk of respiratory depression and airway collapse. We therefore elected to perform implantation using fiberoptic bronchoscopy under intravenous conscious sedation.

Intravenous conscious sedation allows a mild and reversible depression of consciousness while preserving the patient’s ability to respond to verbal or tactile stimuli. This technique improves patient comfort and procedural tolerance while avoiding some of the physiological disadvantages associated with general anesthesia. ³ Dexmedetomidine provides arousable sedation with minimal respiratory depression compared with propofol or midazolam, and its sympatholytic and analgesic-sparing properties make it particularly well-suited for airway procedures.^4,5 To achieve balanced sedation and analgesia, dexmedetomidine was combined with esketamine and sufentanil.^6,7 This multimodal regimen provides cooperative sedation, effective analgesia, and airway reflex suppression while maintaining stable hemodynamics and spontaneous respiration.

Conscious sedation is increasingly used for bronchoscopy; however, it carries an inherent risk of hypoventilation. ⁸ High-flow nasal cannula (HFNC) oxygen therapy delivers heated and humidified gas flows of 8–60 L/min at a stable fraction of inspired oxygen (21%–100%). Compared with conventional oxygen therapy, HFNC improves patient comfort, reduces mucosal dryness, lowers respiratory rate, and decreases the work of breathing. It is therefore gaining wider application in perioperative and anesthetic practice.^9,10

Because our patient could not tolerate cervical hyperextension and was at increased risk from general anesthesia, we adopted a novel strategy involving T-tube implantation using fiberoptic bronchoscopy under intravenous conscious sedation, supported by HFNC. To the best of our knowledge, this combination of techniques has rarely been reported in patients with such complex conditions. Here, we describe the patient’s clinical course, sedation protocol, and procedural details, with the aim of providing a reference for clinicians managing similar cases.

Case report

A man in his early 20 s presented to The Fourth Affiliated Hospital of Soochow University with difficulty in extubation after 4 months of tracheotomy. In early 2021, the patient underwent surgery under general anesthesia for os odontoideum complicated by C1–C2 instability and spinal cord injury with incomplete paraplegia. Postoperatively, he underwent tracheal intubation followed by tracheostomy and was maintained on invasive positive pressure ventilation (IPPV) in the intensive care unit. Despite successful weaning from mechanical ventilation, he developed dyspnea upon tube occlusion, resulting in prolonged dependence on the tracheal cannula. In mid-2021, 4 months after tracheostomy, he was admitted to our hospital due to recurrent chest tightness and hemoptysis. Physical examination revealed dysphonia, left-sided weakness with hypertonia, and positive Babinski and Hoffmann signs. Arterial blood gas analysis at an FiO₂ of 33% showed a pH of 7.38, PaCO₂ of 46.5 mmHg, and PaO₂ of 111 mmHg, indicating adequate oxygenation with mild hypercapnia. Neck computed tomography (CT) showed postoperative changes in the trachea and cervical spine, including cervical spine recurvature and upper trachea stenosis (Figure 1(a) and (b)). Chest CT scan revealed bilateral lower lobe pulmonary infection with partial atelectasis (Figure 1(c) and (d)). Empirical antibiotic therapy with ceftazidime (2 g twice daily) and etimicin (0.3 g once daily) was initiated and continued for approximately 2 weeks. Bronchoscopy performed 2 days after admission revealed upper tracheal stenosis caused by granulation tissue hyperplasia and mucosal hypertrophy above the tracheostomy tube (Figure 2(a) and (b)). The degree of stenosis was classified as Grade II according to the Myer–Cotton classification, with approximately 60% luminal narrowing. Additionally, the proximal margin of the stenosis was located more than 1 cm below the vocal cords. To facilitate subsequent decannulation, removal of the tracheal cannula and placement of a T-tube were indicated. Given the cervical spine instability and the risks associated with general anesthesia and neck hyperextension, we planned T-tube insertion using a flexible fiberoptic bronchoscopic technique under intravenous conscious sedation combined with HFNC oxygen therapy.

OPEN IN VIEWER

Figure 1.

Neck and chest computed tomography. (a) Postoperative changes and cervical spine recurve. (b) Upper tracheal narrowing. Green arrows indicate tracheal stenosis, and red arrows indicate tracheostomy stoma and (c, d) bilateral lower lobe pulmonary infection with partial atelectasis.

OPEN IN VIEWER

Figure 2.

Bronchoscopic findings and stent. (a, b) Granulation tissue and mucosal hypertrophy above tracheostomy tube causing ∼60% stenosis (Myer–Cotton Grade II). (c) Montgomery T-tube (model and dimensions). (d, e) Process of T-tube insertion using fiberoptic bronchoscopy and guidewire and (f–h) bronchoscopic views showing final stent position (upper branch, lumen, and lower branch).

Outcome and follow-up

During the period in which the T-tube remained in place, the patient regained normal speech function but continued to experience intermittent coughing, sputum production, and occasional episodes of chest tightness. Nebulized therapy was routinely administered via the upper limb of the T-tube. Surveillance bronchoscopy performed approximately 15 months after T-tube placement demonstrated that the stent remained in situ approximately 1 cm below the glottis, with marked improvement in glottic edema and no evidence of migration or granulation tissue formation (Figure 3(a)). The upper end, inner lumen, and lower end of the T-tube were clearly visualized, and the bilateral segmental bronchi appeared patent (Figure 3(b) to (d)). In late 2022, the T-tube was successfully removed and immediately replaced with a metal cannula (Figure 3(e) and (f)). No residual edema was noted at the vocal cords, and the trachea above the metal cannula remained unobstructed. The metal cannula was later removed, and the tracheostomy was successfully closed.

OPEN IN VIEWER

Figure 3.

Bronchoscopic follow-up images at 15 months postplacement. (a) Stent in situ 1 cm below the glottis. (b–d) Upper end, inner lumen and lower end of the T-tube, respectively and (e, f) Trachea and vocal cords after T-tube removal and metal cannula insertion.

Discussion

With advances in critical care medicine, subglottic stenosis after tracheal intubation or tracheostomy is being recognized with increasing frequency. ¹² Surgical resection with tracheal reconstruction remains the definitive treatment in selected cases; however, its application is limited by surgical trauma, perioperative complications, and poor baseline status in many patients. ² The Montgomery T-tube provides an effective and less invasive alternative for benign subglottic stenosis and is usually inserted under general anesthesia using rigid bronchoscopy. ¹³ However, rigid bronchoscopy requires cervical hyperextension and may be challenging or even life-threatening in patients with limited cervical mobility. ¹⁴ In our patient, who had previously undergone cervical spine surgery, rigid bronchoscopy was not recommended.

Dexmedetomidine was selected as the sedative agent for conscious sedation in this case, rather than the more commonly used agents propofol and midazolam.^3,15 Total intravenous anesthesia with propofol is widely used but presents challenges in maintaining spontaneous respiration, preserving airway protective reflexes, and preventing laryngospasm. ¹⁶ Similarly, benzodiazepine-based sedation with midazolam may result in respiratory depression and delayed recovery, making dexmedetomidine a safer alternative for maintaining spontaneous breathing and patient cooperation.^17,18 Dexmedetomidine exerts its sedative and hypnotic effects via activation of α2-adrenoceptors in the locus coeruleus, mimicking physiological sleep pathways. Its key advantages include arousable sedation without significant respiratory depression, analgesic and sympatholytic properties that stabilize hemodynamics, and suppression of salivary secretions, making it particularly suitable for bronchoscopic procedures.^19,20 Jeyabalan and colleagues reported that intravenous conscious sedation not only maintained the diagnostic yield of endoscopic ultrasound–guided transbronchial needle aspiration but also improved patient satisfaction with the procedure. ³

To further optimize sedation and analgesia while maintaining spontaneous ventilation, esketamine and sufentanil were coadministered alongside dexmedetomidine. Esketamine, an N-methyl-D-aspartate receptor antagonist, enhances analgesia and produces mild dissociative sedation while preserving spontaneous ventilation. ²¹ Sufentanil, a μ-opioid receptor agonist, provides potent analgesia and suppresses airway reflexes. ²² Their synergistic interaction enables adequate procedural sedation and analgesia at lower individual doses, reducing the risk of hemodynamic instability and respiratory compromise.^6,23 This multimodal regimen is therefore well-suited for airway interventions performed under cooperative sedation while preserving spontaneous breathing.

Oxygen therapy is a critical component of both bronchoscopy and anesthesia. HFNC delivers warmed, humidified gas flows up to 60 L/min with a stable fraction of inspired oxygen (21%–100%). Compared with conventional oxygen delivery devices, such as nasal cannula or non-IPPV (NIPPV), HFNC provides superior patient comfort. ²⁴ In recent years, HFNC has been increasingly adopted in the management of hypoxemic respiratory failure, including severe pneumonia, acute respiratory distress syndrome, ventilator weaning, and postoperative respiratory support. ²⁵ In a multicenter open-label trial, Frat and colleagues demonstrated that HFNC significantly reduced mortality compared with mask oxygen therapy or NIPPV in patients with acute hypoxemic respiratory failure. ²⁶ Depending on clinical needs, oxygen therapy can be delivered by nasal cannula, face mask, NIPPV, or invasive ventilation. HFNC provides an additional option for bronchoscopy. ²⁷ Compared with conventional methods, HFNC not only supplies high flows of humidified gas at stable oxygen concentrations but also improves oxygenation in acute respiratory failure by flushing anatomical dead space, facilitating carbon dioxide clearance, reducing expiratory resistance, increasing lung volume, recruiting collapsed alveoli, and preserving mucociliary function. ²⁸ Conclusively, these physiological effects provide a strong rationale for the use of HFNC during bronchoscopy. In a retrospective study of 10 patients undergoing bronchoscopy or interventional therapy with HFNC, all procedures were completed successfully without episodes of hypoxemia. ²⁷

In our case, the combination of fiberoptic bronchoscopy, intravenous conscious sedation, and HFNC enabled the successful implantation of a Montgomery T-tube in a patient with subglottic stenosis and cervical instability. This experience suggests that this approach may be safely and effectively applied in carefully selected patients when spontaneous breathing is preserved and airway visualization is adequate. This strategy may also be applicable to a wider patient population. In obese individuals and those with chronic obstructive pulmonary disease, limited respiratory reserve increases the risk of hypoventilation under general anesthesia. Conscious sedation allows preservation of spontaneous ventilation and may therefore enhance procedural safety in these patients. In addition, elderly individuals and patients with cardiovascular comorbidities may experience less anesthetic stress and more stable hemodynamics, making them suitable candidates for this approach.

Despite these advantages, several limitations should be acknowledged. The technique may be unsuitable for extremely severe or circumferential tracheal stenosis, in which airway obstruction or hypoxia cannot be safely managed without invasive ventilation. It also requires close coordination between the bronchoscopist and anesthesia team, meticulous titration of sedative agents, and continuous monitoring to maintain airway patency and adequate oxygenation. Future studies involving larger patient cohorts are warranted to validate the safety and efficacy of this multimodal approach, refine patient selection criteria, and establish standardized sedation protocols for interventional airway procedures.

Although pretreatment of granulation tissue is generally recommended, direct T-tube placement was chosen in this patient to minimize the risk of mucosal injury, demonstrating a potential approach for patients with mixed stenosis or fragile mucosa. Although conscious sedation carries inherent risks of coughing or movement, careful titration of sedatives and local anesthesia allowed the procedure to be performed safely in this patient. Therefore, the present findings should be interpreted as applicable primarily to carefully selected patients and may not be generalizable to all cases of subglottic stenosis.

In summary, benign airway stenosis after intubation or tracheostomy can be managed effectively with T-tube placement, where anesthetic strategy and oxygen delivery are being key determinants of procedural success. In this case, fiberoptic bronchoscopy with intravenous conscious sedation supported by HFNC enabled safe T-tube implantation with good tolerance and comfort. This experience suggests that conscious sedation combined with HFNC may offer a safe and well-tolerated alternative to general anesthesia with rigid bronchoscopy in selected patients. Further clinical studies are required to confirm the safety and broader applicability of this approach in patients with moderate-to-severe subglottic stenosis.

Supplemental Material