Tracheal Stricture and Fistula: Management With a Barbed Silicone-Covered Retrievable Expandable Nitinol Stent

Transcription

1 Vascular and Interventional Radiology Original Research Kim et al. Tracheal Stent Vascular and Interventional Radiology Original Research Yong Hee Kim 1 Ji Hoon Shin 2 Ho-Young Song 2 Jin Hyoung Kim 2 Kim YH, Shin JH, Song HY, Kim JH Keywords: fistula, obstruction, prosthesis, stenosis, stent, trachea DOI: /AJR Received May 8, 2009; accepted after revision July 15, Department of Thoracic and Cardiovascular Surgery, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea. 2 Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 388-1, Pungnap-2dong, Songpa-gu, Seoul , Korea. Address correspondence to J. H. Shin (jhshin@amc.seoul.kr). WEB This is a Web exclusive article. AJR 2010; 194:W232 W X/10/1942 W232 American Roentgen Ray Society Tracheal Stricture and Fistula: Management With a Barbed Silicone-Covered Retrievable Expandable Nitinol Stent OBJECTIVE. The purpose of this study was to evaluate the safety and effectiveness of a barbed silicone-covered retrievable expandable nitinol stent in preventing stent migration in patients with tracheal strictures or fistulas. SUBJECTS AND METHODS. Under fluoroscopic guidance, barbed silicone-covered retrievable expandable nitinol stents were placed in 15 patients with tracheal strictures, two patients with fistulas, two patients with combined strictures and fistulas, and one patient with variable extrathoracic airway obstruction. The three pairs of barbs were attached to the external stent surface at the middle of the stent at equal intervals. Technical success, improvement in respiratory status, complications, and related interventions were evaluated. RESULTS. The technical success rate was 100%, and respiratory status improved or the fistula closed in all 20 patients. Complications included sputum retention (three patients), tumor overgrowth (three patients), pain (one patient), and granulation tissue formation (one patient). No stent migration occurred, even in the three patients without fixed strictures. Four stents subsequently were removed because of complications, and one stent was removed because the patient s condition improved. Stent removal was not difficult and was uneventful. The silicone membranes and barbs of the removed stents were intact. CONCLUSION. Use of a barbed silicone-covered retrievable expandable nitinol stent relieves dyspnea and facilitates fistula closure in patients with benign or malignant tracheal strictures or fistulas. The barbed design of the stent is important in preventing migration. T racheobronchial obstruction due to benign or malignant disease can cause dyspnea, stridor, and obstructive pneumonia and is occasionally life-threatening because of the possibility of suffocation. Several groups of investigators [1 5] have used covered expandable metallic stents to manage airway obstruction. Covered stents have the advantages of preventing tumor ingrowth and formation of granulation tissue, of being suitable for esophagorespiratory fistulas, and of being relatively easy to retrieve [1 6]. However, in previous studies including patients [3, 4, 7 9], the rates of migration of covered tracheobronchial metallic stents were 5 17%. Stent migration has been reported to be more common for tracheal stents than for bronchial stents [3]. If tracheobronchial stent migration occurs, the stricture is unveiled, the obstructive symptoms recur, and the stent has to be removed because it causes problems such as obstruction of the main and smaller bronchi. To prevent tracheal stent migration, we designed a barbed silicone-covered retrievable expandable nitinol stent. The purpose of this prospective study was to evaluate the safety and effectiveness of this stent in 20 patients with tracheal stricture or fistula. Subjects and Methods Patients Written informed consent was obtained from each patient, and our institutional review board approved this prospective study. From January 2007 to March 2008, 20 consecutively registered patients (12 men, eight women; mean age, 61.2 years; range, years) with tracheal stricture or fistula underwent fluoroscopically guided placement of a barbed silicone-covered retrievable expandable nitinol stent. The causes of tracheal stricture or fistula were benign in eight patients and malignant in 12 patients (Table 1). In benign disease, stent placement was indicated for patients whose condition placed them at high risk during surgery and those who refused surgery. Fistulas were present in four W232 AJR:194, February 2010

2 Tracheal Stent TABLE 1: Barbed Silicone-Covered Retrievable Expandable Nitinol Stent Placement (n = 20) Patient No. Age (y) Sex Diagnosis Site of Lesion in Thorax Stricture Diameter (mm) Hugh-Jones Grade Complications Stent Removal Before After Type Treatment Indication 1 67 M Postintubation stenosis Upper Granulation tissue Stent placement after stent removal Granulation tissue Interval After Insertion (d) Follow- Up Period (d) Survival Status Alive 2 78 M Postintubation stenosis Middle None None NA 20 Dead Alive 3 34 M In-stent restenosis Middle, lower 4 70 F Variable extrathoracic airway obstruction Sputum retention Stent removal Sputum retention Upper Sputum retention Stent removal Sputum retention Alive 5 62 M Postintubation stenosis Lower None None NA 89 Alive 6 80 F Tracheal stenosis and Middle None None NA 74 Alive esophagotracheal fistula due to esophageal stenting 7 68 M Esophagotracheal fistula due Middle NA NA NA None None NA 30 Alive to tracheostomy 8 75 F Endobronchial tuberculosis Lower None None NA 30 Alive 9 72 M Thyroid carcinoma Upper None None NA 45 Dead F Thyroid carcinoma Middle None None NA 51 Dead F Lung carcinoma Lower Pain Stent removal Pain Alive M Lung carcinoma Middle None None NA 88 Alive M Lung carcinoma Lower None None NA 52 Dead F Adenoid cystic carcinoma Middle None None Elective Alive removal NA 121 Dead M Esophageal carcinoma Middle, lower Tumor overgrowth (proximal) Carinal type stent placement M Esophageal carcinoma Lower None None NA 112 Alive NA 52 Dead M Esophageal carcinoma Middle, lower Sputum retention, tumor overgrowth (distal) Carinal type stent placement M Tracheomediastinal fistula, esophageal carcinoma F Esophagotracheal fistula, tracheal compression by esophageal stent, breast carcinoma F Metastasis of rectal carcinoma Lower NA NA NA None None NA 150 Dead Lower None None NA 35 Dead Middle Tumor overgrowth (proximal and distal) Conservative NA 272 Dead Note NA = not available. AJR:194, February 2010 W233

3 Kim et al. Fig year-old woman with tracheal adenoid cystic carcinoma (patient 14). A and B, Axial (A) and 3D reconstructed (B) CT images show marked narrowing (arrows) of middle portion of trachea. C, Photograph shows mm barbed retrievable tracheal stent with three pairs of barbs (arrowheads). D, Radiograph obtained immediately after stent placement shows partial waist formation at middle of stent. E, Radiograph obtained during stent removal shows retrieval hook (arrow) grasping drawstrings attached to proximal part of stent. Stent was removed because of follow-up CT findings (not shown) after completion of chemoradiation therapy. F, Follow-up CT scan obtained 9 months after stent removal shows marked resolution of stricture. patients (patients 6, 7, 18, and 19). Two of these patients (patients 7 and 18) had no tracheal strictures. Patient 4, who had variable extrathoracic airway obstruction (dynamic narrowing of the extrathoracic airway lumen with more limited inspiratory flow than corresponding expiratory flow and inspiratory flattening on the flow volume loop [10]), had the smallest airway diameter (3 mm) during inspiration, and her condition was categorized as tracheal stricture, although no fixed stricture was present. D A E B The diagnoses were made with bronchoscopy, CT, and the relevant clinical histories. The diagnosis of malignant disease was established with bronchoscopic biopsy. The lesion sites included the upper trachea (three patients), middle trachea (seven patients), lower trachea (seven patients), and middle to lower trachea (three patients). One patient (patient 3) with in-stent restenosis had a history of polyurethane stent placement for postintubation stenosis 4 years previously. Abundant granulation tissue formation into the stent lumen developed because of degradation of the polyurethane covering material. In two patients (patients 18 and 19) with a fistula, radiation therapy and chemotherapy were performed for underlying cancer treatment. In three patients with benign causes of strictures, tracheal stent placement was chosen because of the absence of clinical response to previous balloon dilation (patients 1 and 2) or for bronchoscopic removal of granulation tissue (patient 3). Tracheal stent placement was the primary treatment of the other 17 patients. C F W234 AJR:194, February 2010

4 Tracheal Stent Stent Design Stents were woven from a single thread of 0.2-mm-diameter nitinol wire in a tubular configuration and were covered with a silicone membrane (NuSil, NuSil Technology) made with the dipping method. The stent and introducer set were constructed by a local manufacturer (S&G Biotech). The stent measured 18 or 20 mm in diameter and mm in length when fully expanded (Fig. 1). There was no flaring at either end of the stent. To prevent stent migration, three pairs of barbs were attached to the external surface at the middle of the stent at the same intervals. The barbs were 3 mm long and angulated 60 toward the distal end of the stent. To make the stent removable, a nylon loop 2 mm in diameter was hooked inside each bend of the upper end of the stent, and two nylon threads were passed through each of these nylon loops to form a larger loop (drawstring) that filled the inside circumference of the proximal end of the stent. The stent retrieval set consisted of a 13-French sheath, a 10-French dilator, a hookwire, and a inch exchange guidewire. The end of the retrieval hookwire was shaped like a question mark to hook the stent drawstring, as previously described [5, 11]. The stent introducer set and stent retrieval set were similar to those described in previous reports [5, 9]. Stent Placement and Removal Technique Before the procedure, topical anesthesia of the pharynx and larynx was routinely achieved with an aerosol spray. Sedation drugs were routinely used. Under bronchoscopic guidance, a inch exchange guidewire was inserted through the mouth across the stricture and into the distal portion of the trachea. The bronchoscope was removed while the guidewire was in place. Under fluoroscopic guidance, a straight, 5-French, graduated-size catheter was passed over the guidewire into the distal part of the stricture to measure the length of the stricture. Nonionic contrast medium was injected through the graduated-size catheter to opacify the stricture. The location of the stricture was marked on the patient s skin with radiopaque markers under fluoroscopic guidance. A superstiff J-tipped guidewire was inserted through the graduated-size catheter, and stent placement was performed over the stiff wire. The middle of the stent was located at the middle of the stricture or fistula. The indications for stent removal included elective purposes when the stent was no longer needed and stent-related complications, such as stent migration and marked granulation tissue formation above or below the stent. The stents were removed with a retrieval hook under fluoroscopic guidance [11]. When the hook grasped and pulled the drawstring of the proximal end of the stent into a sheath, the proximal end collapsed, and the stent was removed with a standard technique. Follow-Up and Analysis All patients underwent clinical and fluoroscopic examinations 1 3 days, 1 month, and every 3 or 6 months after stent placement for verification of stent location and patency. In cases of stent removal, clinical and fluoroscopic examinations were performed immediately, 1 month, and every 3 or 6 months after removal. Technical success was defined as successful insertion of a stent in the proper position. The Hugh- Jones classification was used to evaluate improvement in respiratory function before and 1 3 days after stent placement in all patients except the two with fistulas only (patients 7 and 18). This classification is a five-grade system used to assess breathlessness on the basis of daily activities, grade 5 indicating the most severe form of dyspnea [9]. Complications and related interventions were evaluated. Removed stents were examined grossly for intactness of the barbs and membrane degradation. In terms of complications, sputum retention was considered present when it was symptomatic and had increased since stent placement. Tumor overgrowth, tumor ingrowth, and the presence of granulation tissue were confirmed with bronchoscopy and biopsy. Stent migration was defined as a difference of more than 1 cm in stent location during follow-up. Survival data, including cause of death and additional treatments, were evaluated. Results Technical and Clinical Outcome A total of 21 barbed silicone-covered retrievable expandable nitinol stents were successfully placed in the proper position in all 20 patients. In one patient, a second barbed tracheal stent was placed after removal of the initial stent. Because no complications occurred immediately after the procedure, the technical success rate was 100%. After the procedure, all 18 patients with tracheal strictures had one or more grades of improvement on the Hugh-Jones scale (Table 1). Esophagograms of the four patients with tracheal fistulas with or without tracheal strictures showed complete sealing-off of the tracheal fistulas after tracheal stent placement (Fig. 2). Five stents were removed during the indwelling period of days (mean, 48.5 days). Two stents were removed 7 and 15 days after placement because of aggravating sputum retention (patients 3 and 4). One stent was removed 107 days after placement because of granulation tissue formation at the upper part of the stent (patient 1). One stent was removed 28 days after placement because of pain refractory to analgesics (patient 11). The fifth stent was electively removed 84 days after placement from a patient with adenoid cystic carcinoma (patient 14) (Fig. 1) because follow-up CT 75 days after stent placement showed a smaller mass encircling the trachea after chemoradiation therapy. All five stents removed during follow-up were removed with a standard technique, and no difficulties were encountered (Fig. 1). Balloon dilation before removal was necessary for the patient in whom granulation tissue had formed at the upper part of the stent. No clinically significant bleeding occurred after stent removal. The silicone membranes and barbs of the removed stents remained intact. Complications Eight instances of complications of stent placement occurred in seven patients. The complications included sputum retention in three patients (15%), tumor overgrowth at either end of the stents in three patients (15%), pain in one patient (5%), and granulation tissue formation at the upper part of the stent in one patient (5%). There were no cases of stent migration. Because of the complications, the stents were removed from two of the three patients with sputum retention and from the patient with pain. Patient 1 underwent placement of the same type of barbed tracheal stent after the initial stent was removed because of formation of granulation tissue. The second stent was placed to cover the granulation tissue completely. In two patients with tumor overgrowth in the proximal or distal part of the stent, a second, carinal type of tracheal stent was inserted. The two patients with sputum retention or tumor overgrowth at either end of the stent underwent conservative treatment such as mucolytic therapy or postural drainage. Follow-Up Patient 1, who had postintubation stenosis, underwent tracheal resection because of respiratory difficulty even after the second stent placement. Patient 4, who had variable extrathoracic airway obstruction of the trachea, underwent tracheostomy because of aggravation of dyspnea after stent removal. Patient 19, who had breast cancer metastasis involving the trachea and esophagobronchial fistula, underwent insertion of a left bronchial stent to control the fistula. During follow-up AJR:194, February 2010 W235

5 Kim et al. (mean, days; range, days), one patient with benign and eight patients with malignant causes of strictures died a mean of 88.7 days (range, days) after stent placement. Six patients died of disease progression, two patients died of pneumonia, and one patient died of cardiac arrest. A Fig year-old woman with underlying corrosive esophageal stricture and concomitant esophageal carcinoma (patient 6). A, Esophagogram shows tracheal stricture (arrow) and esophagotracheal fistula caused by esophageal stent that had been placed for esophageal carcinoma. B, Bronchoscopic image shows marked tracheal narrowing by esophageal stent (arrows). C, Esophagogram obtained immediately after placement of mm barbed retrievable tracheal stent shows sealing off of fistula and widening of tracheal lumen. Discussion Migration is one of the most common complications associated with the use of stents. Migrated stents can irritate the adjacent tissue and block the adjacent lumen, which should remain patent. Migration of gastrointestinal and biliary stents can cause bowel perforation, fistulas with the adjacent bowel, and bowel obstruction [12, 13]. In airways, migrated stents can interfere with air flow by exposing the stricture and the migrated stent itself. Migrated airway stents also can obstruct adjacent bronchi, especially in patients with covered stents. When airway stents migrate, patients can cough them up [9, 14]. The expectorated stents reported in the literature all have been bronchial stents. In cases of stricture recurrence and obstructive symptoms, stent repositioning, stent removal, or replacement with another stent is required [3 5, 7, 8]. Because of the not rare occurrence of stent migration and the need for urgent management of migrated stents, prevention of stent migration probably is the best policy. There have been several reports of efforts to prevent stent migration in the esophagus and urethra [15 19]. Fixation of stents by endoscopic clips, polypectomy snares, and silk thread have been reported [17 19]. Fixation, however, requires additional endoscopic procedures and can cause cosmetic problems because the snare or silk thread is attached to the patient s ear. Stent design modification such as enlarging the stent end flanges or making a double-layered stent decreases the migration rate but does not completely prevent the problem [15, 16]. Barbs have been used for prevention of esophageal stent migration [20, 21], but there have been only limited reports of their clinical efficacy. Reports regarding the prevention of migration of a covered stent in the airway are relatively limited. Partially covered stents have been used with both ends bare, but stent removal has been difficult or impossible [2, 4, 6, 22]. One or both ends of tracheal or bronchial stents are usually flared to decrease migration [3, 7, 9]. Some investigators [23, 24] have reported the use of an external fixation apparatus to prevent stent migration in the subglottic trachea, but the apparatus is available only for use in the subglottic area. Literature reports on the use of barbs in airway stents are limited. B We found that barbed tracheal stents were safe and effective in preventing stent migration. In this study we encountered neither complications related to the barbs nor stent migration. The barbs were 3 mm long and projected 60 toward the distal end of the stent. Therefore, the perpendicular distance of the barbs from the stent surface was less than 3 mm. We conjecture that this distance does not seem to be injurious to the surrounding vascular structures. To prevent migration, angulation of the barbs toward the distal end of the stent would seem to be an inferior design in comparison with perpendicular barbs. However, we encountered no stent migration, undoubtedly because of the large number of barbs placed at equal intervals. In addition, the design can decrease the risk of perforation and the difficulty of stent removal. In a report of fatal hemorrhage in a patient with a barbed esophageal stent [21], the barbs were 10 mm long and projected almost straight out laterally. The cause of death determined at autopsy was perforation of the aortic arch by the barbs. It is remarkable that stent migration did not occur in the three patients in our study who did not have fixed tracheal strictures. We sometimes encounter clinical situations in which stent placement is necessary for a stricture-free airway. One of the three patients C W236 AJR:194, February 2010

6 Tracheal Stent had variable extrathoracic airway obstruction of the trachea, and the other two had a fistula between the trachea and the esophagus or mediastinum. Although they are advantageous for preventing migration, stents that are long enough or have sufficient conformability sometimes cannot overcome the dynamic size of the trachea or bronchus, especially in patients with tracheobronchomalacia or severe coughing. To prevent stent migration, we advocate the use of barbs that anchor the mucosa in three directions. Stent removal was safe and performed without difficulty in our patients who underwent removal. The indwelling period was days. The direction of the barbs toward the distal end of the stents probably made the removal easy and caused no clinically significant mucosal bleeding during or after the removal procedures. We used our standard technique [11] to remove all five stents: The drawstrings were grasped by a retrieval hook, and each stent was removed with its proximal portion collapsed. Our success suggests that anchoring with the barbs is sufficient to prevent stent migration but not so strong as to impede stent removal. Although barbed tracheal stents were safe in our study, the possibility of perforation of adjacent tissue and resulting hemorrhage should be emphasized. Unlike the proximity of the aorta and the esophagus, the distance between the trachea and the aorta and the major neck arteries is sufficient to reduce the risk of perforation. In addition, because the tracheal wall contains cartilage and is much thicker than the esophageal wall, the risk of injury to the trachea by the barbs would seem to be much less than the risk of injury to the esophagus. Further study with a greater number of patients and a comparative study design is necessary for resolving the safety issues. This study had the limitations of small sample size, lack of comparative control, and lack of bronchoscopic follow-up to determine airway patency. Despite the limitations, we conclude that a barbed siliconecovered retrievable expandable nitinol stent can be used to relieve dyspnea and close fistulas in patients with benign or malignant tracheal strictures or fistulas and that the barbed design is important for preventing stent migration. References 1. Noppen M, Stratakos G, D Haese J, Meysman M, Vinken W. Removal of covered self-expandable metallic airway stents in benign disorders: indications, technique, and outcomes. Chest 2005; 127: Madden BP, Park JE, Sheth A. Medium-term follow-up after deployment of Ultraflex expandable metallic stents to manage endobronchial pathology. Ann Thorac Surg 2004; 78: Kim JH, Shin JH, Song HY, Shim TS, Yoon CJ, Ko GY. Benign tracheobronchial strictures: longterm results and factors affecting airway patency after temporary stent placement. AJR 2007; 188: Breitenbucher A, Chhajed PN, Brutsche MH, Mordasini C, Schilter D, Tamm M. Long-term follow-up and survival after Ultraflex stent insertion in the management of complex malignant airway stenoses. Respiration 2008; 75: Kim JH, Shin JH, Shim TS, et al. Results of temporary placement of covered retrievable expandable nitinol stents for tuberculous bronchial strictures. J Vasc Interv Radiol 2004; 15: Shin JH, Song HY, Shim TS. Management of tracheobronchial strictures. Cardiovasc Intervent Radiol 2004; 27: Kim JH, Shin JH, Song HY, Lee SC, Kim KR, Park JH. Use of a retrievable metallic stent internally coated with silicone to treat airway obstruction. J Vasc Interv Radiol 2008; 19: Monnier P, Mudry A, Stanzel F, et al. The use of the covered Wallstent for the palliative treatment of inoperable tracheobronchial cancers: a prospective, multicenter study. Chest 1996; 110: Shin JH, Kim SW, Shim TS, et al. Malignant tracheobronchial strictures: palliation with covered retrievable expandable nitinol stent. J Vasc Interv Radiol 2003; 14: Das AK, Davanzo LD, Poiani GJ, et al. Variable extrathoracic airflow obstruction and chronic laryngotracheitis in Gulf War veterans. Chest 1999; 115: Kim JH, Shin JH, Shim TS, et al. Efficacy and safety of a retrieval hook for removal of retrievable expandable tracheobronchial stents. J Vasc Interv Radiol 2004; 15: Alvi A, Pitt JP. A case of entero-colic fistula formation resulting from stent migration inserted for obstructing left colonic tumour. Int J Colorectal Dis 2009; 24: Saranga Bharathi R, Rao P, Ghosh K. Iatrogenic duodenal perforations caused by endoscopic biliary stenting and stent migration: an update. Endoscopy 2006; 38: Zakaluzny SA, Lane JD, Mair EA. Complications of tracheobronchial airway stents. Otolaryngol Head Neck Surg 2003; 128: Woo CW, Song HY, Yoon CJ, et al. Is a flared stent effective for decreasing stent migration in canine prostatic urethra? Acta Radiol 2008; 49: Verschuur EM, Homs MY, Steyerberg EW, et al. A new esophageal stent design (Niti-S stent) for the prevention of migration: a prospective study in 42 patients. Gastrointest Endosc 2006; 63: Kato H, Fukuchi M, Miyazaki T, et al. Endoscopic clips prevent self-expandable metallic stent migration. Hepatogastroenterology 2007; 54: Manes G, Corsi F, Pallotta S, Massari A, Foschi D, Trabucchi E. Fixation of a covered self-expandable metal stent by means of a polypectomy snare: an easy method to prevent stent migration. Dig Liver Dis 2008; 40: Shim CS, Cho YD, Moon JH, et al. Fixation of a modified covered esophageal stent: its clinical usefulness for preventing stent migration. Endoscopy 2001; 33: Song HY, Choi KC, Cho BH, Ahn DS, Kim KS. Esophagogastric neoplasms: palliation with a modified Gianturco stent. Radiology 1991; 180: Dirks K, Schulz T, Schellmann B, Stolte M, Lutz H. Fatal hemorrhage following perforation of the aorta by a barb of the Gianturco-Rosch esophageal stent. Z Gastroenterol 2002; 40: Madden BP, Datta S, Charokopos N. Experience with Ultraflex expandable metallic stents in the management of endobronchial pathology. Ann Thorac Surg 2002; 73: Miwa K, Takamori S, Hayashi A, Fukunaga M, Shirouzu K. Fixation of silicone stents in the subglottic trachea: preventing stent migration using a fixation apparatus. Ann Thorac Surg 2004; 78: Colt HG, Harrell J, Neuman TR, Robbins T. External fixation of subglottic tracheal stents. Chest 1994; 105: AJR:194, February 2010 W237