Noninvasive Correction of a Fractured Endoluminal Nitinol Tracheal Stent in a Dog

Introduction

Tracheal collapse is a frequently encountered, progressive, chronic respiratory disease of middle-aged, small and toy breeds of dogs.¹ Initially, medical therapy is recommended and results in long-term resolution of clinical signs in >70% of affected dogs.¹ When appropriate medical management does not produce satisfactory results, surgical correction may be considered.¹ Surgical options include dorsal tracheal membrane plication, placement of extraluminal prosthetic support, or insertion of an endoluminal prosthesis.^1–3 Polypropylene C-shaped and spiral extraluminal stents have had good results when used for extrathoracic tracheal collapse, but their use has been unrewarding for intrathoracic collapse.^4–6 Complications reported with extraluminal prostheses include persistent coughing, infection, disruption of tracheal blood flow, laryngeal paralysis, and death.^1,4,6–8

The use of endoluminal stents has been described for the treatment of tracheal collapse in the dog.^9–15 A variety of endoluminal stent devices are available, including commercial and custom-made polyvinyl prostheses.^a,b,c,d Endoluminal stents have many advantages, such as potential use in extra- and intrathoracic tracheal collapse, noninvasive placement, ease and rapidity of insertion, and better adaptability to various airway diameters (e.g., trachea, bronchus).¹⁶ Reported complications from endoluminal stents in dogs include fracture and migration of the prosthesis, deformation and collapse of the stent, pneumonia, chronic cough, formation of granulomas at the extremities of the stent, erosion of the tracheal epithelium, and acute pulmonary edema.^10,11,13,17 Mucosal hyperplasia and formation of granulation tissue at the proximal and distal ends of the prosthesis have been previously reported in dogs and also occur in 10% of humans.^12,18 The purpose of this article is to describe the successful correction of a fractured endoluminal nitinol tracheal stent in a dog.

Case Report

An 11-year-old, 5.9-kg, castrated male Pomeranian was presented to the internal medicine service of the University of Montreal for persistent coughing and intermittent dyspnea. The animal had a long medical history of severe extrathoracic tracheal collapse. Five years earlier, multiple extraluminal C-shaped polypropylene (custom-made) prostheses were inserted surgically over the affected trachea. The dog did well for several years, but within the 4 months prior to presentation, the cough returned and was treated with cough suppressants, anti-inflammatory medications, and a bronchodilator. The dog was otherwise in good health.

On initial examination, the dog was anxious, panting, and had increased upper respiratory sounds. Frequent episodes of coughing and dyspnea with mild cyanosis were noted. A dry cough was easily induced on tracheal palpation. The rest of the physical examination was unremarkable, except for a mildly increased body condition score of 6/9.

A complete blood count and serum biochemical profile were normal. Thoracic radiographs revealed a tracheal collapse at the thoracic inlet and the cranial portion of the thoracic trachea. Fluoroscopy revealed tracheal collapse involving a large portion of the cranial thoracic trachea. The C-shaped extraluminal rings were visible, and the cervical trachea was unremarkable.

Because this dog did not tolerate any physical activity and his quality of life was considered unacceptable despite medical therapy, placement of either intrathoracic external ring prostheses or an endoluminal expandable stent was considered. After discussion with the owners, the decision was made to place an endoluminal stent prosthesis.

Anesthesia was induced with propofol^e (4 mg/kg intravenously [IV]) after premedication with butorphanol^f (0.1 mg/kg intramuscularly [IM]). Tracheobronchoscopy with a flexible, 5.0-mm bronchoscope^g revealed tracheal collapse at the thoracic inlet distal to the previously placed external tracheal prosthesis. The lumen of the trachea and mainstem bronchi was otherwise normal. The endotracheal tube was withdrawn just distal to the larynx, and positive ventilatory pressure was applied at 15 to 20 mm Hg while lateral and ventrodorsal radiographic views were taken. Maximal tracheal diameter was measured, and an appropriately sized self-expanding nitinol endoluminal tracheal stent^c was selected. The stent was chosen so that the expanded diameter would be 1.3 times the maximal measurement of the tracheal diameter to ensure stent stability via sufficient radial forces.¹⁰ A 14-mm diameter and 60-mm long nitinol stent^c was placed under fluoroscopic guidance using a previously described technique.^16,19 Two radiographic views confirmed correct placement and expansion of the stent. The stent extended to a position that was 8 mm proximal to the carina and covered the entire length of the tracheal collapse [Figure 1].

The dog was discharged from the hospital 48 hours following the procedure and was treated with amoxicillin-clavulanic acid^h (15 mg/kg per os [PO] q 12 hours for 10 days), sustained-release theophyllineⁱ (10 mg/kg PO q 12 hours), and codeine^j (0.2 mg/kg PO q 8 hours). Owners were instructed to confine the dog and restrict it from strenuous exercise.

During the following weeks, the owners noted marked improvement in the dog’s breathing, although intermittent coughing episodes followed by expectoration continued. Two months following placement of the stent, the dog was presented for recheck examination. A cough was easily induced on palpation of the cervical trachea. Thoracic radiographs confirmed the stent was in place, and a mild generalized bronchointerstitial pattern in the caudal lung lobes was noted. Tracheobronchoscopy showed moderate tracheal hyperemia, abundant mucus, a small amount of proliferative tissue at the edges of the stent, and mild collapse of the right mainstem bronchus. A tracheal lavage was performed and showed suppurative septic inflammation. Culture and sensitivity testing identified Bordetella bronchiseptica that was resistant to amoxicillin-clavulanic acid. Enrofloxacin^k (5 mg/kg PO q 24 hours for 14 days) and prednisone^l (0.3 mg/kg PO q 24 hours for 7 days) were added to the treatment regimen.

The dog’s clinical signs improved with antibiotic therapy. Three months later, however, the dog became anxious and developed frequent episodes of productive coughing, gagging, and dyspnea. Physical examination revealed moderate inspiratory and expiratory dyspnea, with a respiratory rate of 40 breaths per minute and a heart rate of 200 beats per minute. The dog became cyanotic during the physical examination and required oxygen therapy, dexamethasone^m (0.1 mg/kg IV), and butorphanol^f (0.1 mg/kg IM). Thoracic radiographs revealed a kink in the distal portion of the stent, suggesting failure of the implant [Figure 2].

Tracheobronchoscopy was performed, and the proximal portion of the implant looked normal, with partial epithelialization and formation of small nodules around the proximal edge of the stent. The lumen of the stent was markedly narrowed distally, and multiple ruptured fragments of nitinol wire were protruding into the tracheal lumen [Figure 3]. Fracture of the caudal portion of the stent, with subsequent collapse of the trachea, was confirmed. Tracheal lavage samples were submitted for cytology and culture. The caudal portion of the trachea and bronchi were not entered with the scope for fear of causing further damage to the stent.

Treatment options that were considered included endoscopic removal of the fractured stent and placement of a new nitinol prosthesis; surgical resection and anastomosis of the fractured portion of the stent and associated trachea; or placement of a second endoluminal stent inside the fractured portion of the prosthesis. Removal was considered extremely risky because of the partial integration of the stent into the tracheal mucosa. Surgical resection and anastomosis were declined by the owner. The placement of a new stent was done the following day.

Anesthesia was induced, and placement of a second endoluminal nitinol stent was performed as described for the first procedure. Special care was taken during the insertion of the second stent to avoid friction between the delivery system and the fractured stent. The diameter of the prosthesis chosen was 14 mm, but the stent was 20 mm longer to ensure maximal support of the trachea. Expansion of the stent was monitored under fluoroscopy and was considered complete within a few minutes.

Tracheobronchoscopy, performed immediately following stent placement, revealed the fractured stent was completely covered by the newly expanded one. Nitinol filaments were no longer protruding into the tracheal lumen and were flattened against the tracheal wall by the new stent [Figure 4]. Mild hemorrhage was noted in several areas of the mucosa, and a considerable amount of mucus was present. Moderate collapse of the right mainstem bronchus remained.

The dog recovered uneventfully from anesthesia. Postoperative medications included one dose of dexamethasone^m (0.1 mg/kg IM) administered at extubation, butorphanol given as a constant-rate infusion (0.05 mg/kg per hour IV), ampicillinⁿ (20 mg/kg IV q 8 hours), and codeine^g (0.2 mg/kg PO q 8 hours). Cytological examination of the tracheal wash revealed severe suppurative inflammation, epithelial dysplasia, and metaplasia. Aerobic bacterial cultures were negative. The dog was discharged 3 days later on amoxicillin-clavulanic acid (15 mg/kg PO q 12 hours for 5 days) and codeine (0.2 mg/kg PO q 8 hours for 14 days). The animal showed a significant improvement in breathing but coughed occasionally, despite medical therapy.

Within 4 weeks of the second procedure, the dog was breathing well, coughing occasionally, and was able to tolerate moderate physical activity. Twelve weeks after placement of the second stent, increased coughing and exercise intolerance were noted. The owners were unable to bring the dog to the hospital. The dog experienced a severe, acute episode of dyspnea and cyanosis and acutely died.

A postmortem examination was performed 24 hours after death. Both endoluminal nitinol prostheses were still in place, and there was obvious damage to the distal portion of the first stent. Extratracheal polypropylene rings, applied in the cervical portion of the trachea when the dog was 5 years old, were found and were covered by fibrous tissue. The second stent was intact. No evidence of tracheal perforation or hemorrhage was present. A single nodule (1 to 2 mm) was observed at the distal end of the first endoluminal stent and was microscopically an inflammatory granuloma containing numerous plasma cells and fibrous tissue. Numerous struts of the endoprosthesis in the tracheal mucosa were surrounded by concentrically arranged fibrous tissue. The distal portion of the trachea (beyond the prostheses) and extrapulmonary portions of the main bronchi were visibly collapsed and flaccid. Microscopic examination showed mild mucopurulent inflammation of large intrapulmonary bronchi, with no evidence of microorganisms. The rest of the necropsy examination was within normal limits. A diagnosis of tracheobronchial collapse at the carina and mucopurulent bronchitis was established. It was thought that severe collapse at the carina led to tracheal obstruction, respiratory distress, severe hypoxemia, and death.

Discussion

The stents used in this report were self-expanding, proximal-release, uncovered nitinol tracheobronchial stents.^c Nitinol prostheses have been widely used in people for treatment of airway stenosis secondary to tracheobronchial neoplasia.^19–22 The biomechanical properties of nitinol make it a very attractive material in the fabrication of endoprostheses. Nitinol is a flexible metal with good biocompatibility and physical properties that resemble those of tracheal cartilage.²³ Its elasticity and flexibility allow precise placement of a prosthesis, thus minimizing the risk of stent migration. Nitinol stents have been used previously with good results in dogs with tracheal collapse.^10,14 The dog of this report initially had a good response to the stent, but clinical deterioration occurred several months later. This clinical course was similar to another reported case of a fractured endotracheal stent in a dog.¹⁶ Failure of the implant in the current dog was suspected based on thoracic radiographs and was obvious on tracheal endoscopy. Retrospectively, images from the 2-month tracheobronchoscopy were reviewed, and a few nitinol filaments were ruptured at that time. These abnormalities were difficult to visualize because of the presence of abundant mucus in the area and lack of experience in the interpretation of such lesions.

This is the second report of a ruptured endoluminal nitinol tracheal stent in a dog. In the previous report, a tracheotomy was performed to remove the fractured stent along with a portion of the trachea.¹⁶ In the case reported here, noninvasive correction was successfully achieved by the placement of a second stent. The dog recovered rapidly and uneventfully, which represents an advantage over the surgical approach.

Fractured stents have been occasionally reported in dogs and humans in various anatomical locations, including the digestive tract, respiratory tract, and vascular system.^16,24–27 Placement of stents in areas of frequent movement and use of longer prostheses are known to increase the risk of failure, since fracture rates as high as 37.2% were reported in vascular femoropopliteal nitinol stenting in people.²⁴ Cyclic movements and high-stress forces applied to the stents may induce fatigue and eventually cause fractures. Forces applied to the tracheal wall of dogs during coughing are undefined and may be significant. The fatigability of various materials used in stenting devices in the airways of dogs must be carefully studied to determine the true risks associated with these prostheses.

Based on necropsy examination of the dog reported here, the cause of death appeared to be secondary to acute respiratory obstruction secondary to tracheal collapse at the level of the carina. Collapse at the carina may be explained by progressive chondromalacia of the tracheal and bronchial rings.¹ Collapse at the carina may have resulted from a distal shift of the equal pressure point. Any obstruction to airflow results in a shift of the equal pressure point toward the alveoli.²⁸ In this dog, collapse of the right mainstem bronchus may have led to reduced airflow in smaller airways, resulting in airway pressures exceeding pleural pressure. Proximal to the bronchial obstruction, airway pressure would be lower than pleural pressure, thus leading to collapse at the carina.²⁹ Perhaps insertion of a prosthesis into the right mainstem bronchus would have prevented the distal shift of the equal pressure point and thus prevented collapse at the carina.

Conclusion

An 11-year-old Pomeranian developed a fracture of a tracheobronchial endoluminal nitinol stent that was successfully treated with the placement of a second endoluminal prosthesis, superimposed over the first one. This case represents an interesting, noninvasive alternative to surgical correction of a fractured stent. Further investigation of the forces applied to tracheal stents in dogs and long-term clinical studies evaluating outcomes after stent placement are needed.

Acknowledgments

The authors thank Dr. Richard Drolet and Mr. Marco Langlois for their technical assistance.