Bioabsorbable Self-reinforced Poly-l -Lactide, Metallic, and Silicone Stents in the Management of Experimental Tracheal Stenosis

laboratory and animal investigations Bioabsorbable Self-reinforced Poly-LLactide, Metallic, and Silicone Stents in the Management of Experimental Tracheal Stenosis* Antti Korpela, MD; Pertti Aarnio, MD; Hannu Sariola, MD; Pertti To¨rma¨la¨, PhD; and Ari Harjula, MD

The aim of the present study was to compare, in rabbits, the biocompatibility and suitability of a bioabsorbable spiral stent made of self-reinforced poly-L-lactide (SR-PLLA) in the management of experimental tracheal stenosis with stents made of metal and silicone. Tracheobronchial stenosis, and its management, is still problematic because stenoses are not always amenable to surgical resection and reconstruction, especially concerning anastomotic problems and stenosis after lung transplantation. Stenosis can be handled with stenting, although the ideal stent has yet to be developed; all the stents available have their disadvantages. Because stenting of the airways can be only temporary, stents made of bioabsorbable materials, theoretically, offer benefits. Tracheal stenosis was created in rabbits by the extramucosal resection of cartilaginous arches of the cervical trachea. After a few weeks, the animals were operated on again, and those stenoses that had developed were dilated with a balloon. Stents then were implanted in the area of stenosis to keep the dilated trachea open. All the animals in the group with silicone stents had to be killed because of respiratory difficulties: their stents had a tendency to occlude because of internal encrustation, and they developed a hyperplastic polyp at the ends of the stents. The SR-PLLA and metallic stents were tolerated well, and after follow-up ended the animals were put to death. This experimental study showed that silicone stents had a tendency to occlude and that stents made of metal and of SR-PLLA were well tolerated and can be used in the management of airway stenosis. (CHEST 1999; 115:490 – 495) Key words: airway stenosis; airway stent; bioabsorbable material; poly-l-lactide Abbreviations: SEM 5 scanning electron microscope; SR-PLLA 5 self-reinforced poly-l-lactide

stenosis can be caused by the postsurgical A irway state, by malacia, by obstructing tumors, or by extrinsic compression, any of which may result in For related material see pages 496 and 532

*From the Department of Thoracic and Cardiovascular Surgery (Drs. Korpela and Harjula), Helsinki University Central Hospital, Helsinki, Finland; Institute of Biotechnology (Dr. Sariola), Helsinki; Satakunta Central Hospital (Dr. Aarnio), Pori, Finland; and Institute of Biomaterials (Dr. To¨rma¨la¨), Tampere University of Technology, Tampere, Finland. Manuscript received May 21, 1997; revision accepted July 7, 1998. Correspondence to: Antti Korpela, MD, Department of Surgery, Pa¨ija¨t-Ha¨me Central Hospital, Keskussairaalankatu 7, 15850 Lahti, Finland 490

progressive dyspnea and hypoxemia. The most common indication for tracheal resection and reconstruction is still postintubation tracheal stenosis.1 Tracheobronchial obstruction caused by malignant diseases can cause severe symptoms such as dyspnea, cough, and suffocation in the patient. Bronchial anastomotic complications, like stenoses, also have been very common in lung transplant patients.2 Surgical resection and end-to-end anastomosis is the method of choice to handle airway stenosis, but this is not always possible due to the nature and cause of the stenosis and the general state of the patient. Tracheobronchial malacia is also difficult to correct surgically. Experiments have been performed in which tracheal prostheses or homografts have been substituted for the trachea. EndobronLaboratory and Animal Investigations

chial methods for handling airway stenoses have been developed, including laser surgery, cryotherapy, endobronchial resection, brachytherapy (in malignant obstructions), and dilatation and stenting of the stenotic area of the airway.3 The most common types of stents in use are made of silicone4 or metallic wire.5 Silicone stents are relatively thick, and these stents cause mucociliary function to be lost in the stented area and have a tendency for secretions to accumulate in their lumen, which can cause obstruction. Metallic stents, once they have been covered by the epithelium, can be removed by rigid bronchoscopy, but open surgery may be required. The ideal stent for tracheobronchial stenosis has yet to be designed, but Colt and Dumon6 have listed the requirements for such a stent. Bioabsorbable airway stents could offer benefits that the stents now available for clinical use do not offer. Extraction of the stent would not be necessary, and resorption of the device would not interfere with the normal function of the airway. The time of absorption of the stent can be modified, depending on the choice of the basic molecule with which it is manufactured, the shape, the degree of polymerization, the internal arrangement of the material components, and the site of implantation. Bioabsorbable materials degrade and are metabolized to water and carbon dioxide by hydroxylation inside normal tissue. A very strong but elastic bioabsorbable stent can be designed; self-expanding models can be made. The biocompatibility of bioabsorbable materials has been good in other organs. Rods of bioabsorbable polyglycolic acid and polylactic acid are in clinical use for the fixation of bone fractures, with favorable results.7 SR-PLLA spiral stents have been shown to have good biocompatibility with the rabbit urethra.8 Spiral stents made of self-reinforced polyglycolic acid were used in the human urethra to prevent postoperative urinary retention after visual laser ablation, with good results.9 In this study, we examined the treatment of tracheal stenosis with three types of intratracheal stents in an animal model: the silicone tube, metallic tubular mesh, and an SR-PLLA spiral, which is a new material for stents. Poly-l-lactide was chosen for an absorbable bronchial stent because it has the longest biodegradation time of the basic molecules available. Materials and Methods The silicone stents were 1.0-mm thick tubes with a 5 or 6 mm outer diameter and a length of 12 mm. The SR-PLLA stents were made of 0.7-mm thick wire, and were 5.5 mm in outer diameter and 12 mm in length. The length of the metallic Schneider stent

(Wallstent; Pfizer; Bern, Switzerland) was 12 mm, and the nominal outer diameter was 6 mm. Altogether, 32 rabbits underwent the first operation in which the arch of tracheal cartilage was partly excised in order to create a tracheal stenosis. One animal died before insertion of the stent, and another died during that procedure. Group A, using silicone stents, comprised 10 rabbits with a mean weight of 2.5 kg (range, 2.1 to 3.0 kg); group B, using SR-PLLA stents, comprised 11 rabbits with a mean weight of 2.7 kg (range, 2.2 to 3.5 kg); and group C, using metallic stents, comprised 9 rabbits with a mean weight of 2.7 kg (range, 2.1 to 3.0 kg). The rabbits were anesthetized with atropine (0.75 mg/kg of body weight), ketamine (20 mg/kg of body weight), medetomide (300 mg/kg of body weight), and diazepam (1.0 mg/kg of body weight). Procaine penicillin (150,000 IU) was used as an antibiotic prophylaxis. All these drugs were administered subcutaneously. The animals maintained spontaneous breathing during the operation. The cervical trachea was prepared by sight through the midline incision, and three to five cartilage arches spanning an area half the circumference of the trachea were excised submucously; an area of about 5 3 8 mm was then without the support of the cartilage. The wound was closed in layers. After a follow-up period of 4 to 8 weeks, the rabbits were operated on again using the same anesthetic procedure. The area previously operated on was prepared by sight, and a transverse incision was made in the trachea that was about two thirds of its circumference and that was between cartilage distal to the stenosed area where the cartilage previously had been partly excised. A 6-mm thick angioplasty balloon catheter was inserted through the bronchotomy at the stenosed area, which was then dilatated by filling the balloon. The pressure of the balloon was regulated manually by visual inspection in order to avoid overdistension and extra damage to the bronchial wall. Then the stent was placed in the dilatated area, the bronchotomy was closed with an uninterrupted 5– 0 polypropylene suture, and the wound was closed in layers. If the rabbits had stridor or difficulties with breathing during follow-up, they were killed. If they were doing well, they were put to death 3, 6, or 9 months postoperatively. The stented area in the cervical trachea was excised for further investigation. The specimen was divided longitudinally into two pieces; one half was fixed in 4% formalin solution for histologic examination, and the other half was fixed in 2% buffered glutaraldehyde solution for scanning electron microscope (SEM) studies. From paraffin-embedded tissue specimens, 4-mm thick longitudinal sections were cut and stained with hematoxylin and eosin. The magnitude of the inflammation, the fibrosis, and the changes in the epithelium were assessed semiquantitatively as follows: 0 5 normal; 1 5 mild changes; 2 5 moderate changes; and 3 5 severe changes. Tissue samples for SEM studies were dehydrated in ethanol, were critical-point dried (Balzers CPD 020; Balzers Union Ltd; Vaduz, Liechtenstein), were mounted on aluminum studs, and were coated with gold with a sputtering device (model JFC-1100; Jeol Ltd; Tokyo, Japan). A digital SEM (model DSM 962; Carl Zeiss; Oberkochen, Germany) was used at a 10-kV acceleration voltage in SEM studies at the Electron Microscopy unit of the Institute of Biotechnology at the University of Helsinki. Changes in the epithelial cell layer and in ciliated cells were evaluated. Due to the nature of findings in histologic and SEM studies, quantitative analysis was not performed. All animals received humane care in compliance with the “European Convention for Protection of Vertebrate Animals Used For Experimental and Other Scientific Purposes,” ratified in Strasbourg in 1986 by the Council of Europe. The study protocol was approved by the institutional committee for test animal research. CHEST / 115 / 2 / FEBRUARY, 1999

491

Results The rabbits tolerated both operations well, except for one animal that died before insertion of the stent 5 weeks after the first operation, and another animal that died during the second operation. One rabbit in group B had to be killed 7 weeks postimplantation because of severe difficulties in breathing. The small trachea of this rabbit had caused difficulties during the insertion of the stent in the second operation. At autopsy, severe stenosis at the bronchotomy site was found, but the stented area itself was fully open; this rabbit was excluded from the group. All rabbits developed a clinically significant tracheal stenosis within 8 weeks. In the second operation before balloon dilatation, the stenosis was observed to be . 50% of the luminal area. The animals had stridor in excitement, both in inspirium and expirium, as the clinical symptom. The follow-up time in group A ranged from 6 to 39 weeks (mean, 22 weeks); all animals in this group, except one, were put to death because of stridor and difficulties in breathing. Viewed macroscopically at autopsy, a light brown obstructive material was apparent inside the lumen of the stent, and mucosal hyperplasia had developed at the ends of the stents, which caused stenosis in the tracheal lumen. In group B, the rabbits were followed for 13 to 39 weeks (mean, 24 weeks); one rabbit died of uncertain causes 24 weeks postimplantation. The stent had disappeared, but the lumen was well open; otherwise, the findings were unexceptional. Another rabbit died 31 weeks after surgery. In the lumen of the trachea, parts of the spirals of the stent were noticed; this disintegration of the stent may have caused the death of the rabbit. One rabbit was killed 26 weeks postoperatively because of stridor caused by parts of spirals of the stent occluding the lumen of the trachea. Findings at autopsy for the other seven rabbits were acceptable: two stents had disappeared, but each lumen was still open. In five animals, the stents were in place, the lumen was fully open, and stenosis had not developed again. In group C, rabbits were followed for 13 to 38 weeks (mean, 25 weeks), and all were killed at the end of the follow-up period. At autopsy, all the stents were satisfactorily in place, partly covered by the growing epithelium, and the lumen was open with no stenosis having recurred. For light microscopic studies, the stents had to be removed before the cutting of the paraffin-embedded tissue samples. Epithelial ulceration was noticed under the silicone stents as well as under the spirals of the SR-PLLA stents and under the filaments of the metallic stents (Fig 1). Masses of polypoid granulation tissue and moderate eosinophilia were 492

Figure 1. Tracheal epithelium of a rabbit with a metallic stent, 6 months postimplantation. The groove was caused by the wire of the stent (center) (hematoxylin-eosin, original 3 200).

noticed at the ends of the silicone stents, and moderate chronic lymphocytic inflammation also was noticed under the stents (Fig 2). In group B, reservecell hyperplasia was detected between the stent spirals; some degree of eosinophilia and chronic lymphocytic inflammation was apparent in the same areas. Between the metallic stent filaments there was reserve-cell hyperplasia, eosinophilia, and some degree of chronic lymphocytic inflammation. No foreign-body reaction occurred in any group; all stents were histologically well tolerated, although SR-PLLA and metallic stents seemed to cause somewhat less eosinophilia and chronic lymphocytic inflammation than stents made of silicone. SEM studies showed a well-preserved epithelial cell layer and an uninterrupted carpet of ciliated cells between the spirals of the SR-PLLA stents (Fig 3). In grooves where the spirals had rested, the ciliated cells had disappeared. The findings in the

Figure 2. Masses of polypoid granulation tissue (left) at the end of the silicone stent, 6 months postimplantation (hematoxylineosin, original 3 20. Laboratory and Animal Investigations

Figure 3. Scanning electron micrograph of the epithelium between stent spirals, showing a continuous ciliated epithelial cell layer and solitary nonciliated cells, comparable to normal tracheal epithelium in rabbits. The image is from a rabbit with an SR-PLLA stent, 7 months postoperative. (original 31,600; bar 5 20 mm).

ciliated cell area were comparable to normal tracheal epithelium observed in specimens of normal rabbit trachea. The epithelial cell layer had disappeared under the silicone stents, yielding an uncovered basal membrane with cell debris (Fig 4), but in some areas the epithelial cell layer had been preserved with

Figure 5. Scanning electron micrograph of a specimen from a rabbit with a metallic stent, 13 weeks postoperative. Filaments of the stent, partly covered by the epithelium, are visible (original 375; bar 5 500 mm).

solitary ciliated cells. In group C, the metallic net of the stents was visible, with the filaments of the stents, partly covered by the epithelium, growing between the filaments (Fig 5). In those areas where filaments had been covered by the epithelium and in areas between filaments, an almost normal ciliated cell layer was detectable (Fig 6). In some areas, the ciliated cells were solitary and the surface of the trachea was covered by nonciliated epithelial cells. The morphology of the cilia was found to be abnormally flat in three animals from the group using the SR-PLLA stents, in four animals from the group using the metallic stents, and in one animal from the group using the silicone stents. There was no obvious cause for these findings. Discussion Many nonsurgical palliative methods have been developed for patients who are not amenable to the

Figure 4. Scanning electron micrograph showing uncovered basal membrane, red blood cells, and cell debris; the trachea specimen was taken from under a silicone stent, 18 weeks postoperative (original 31,800; bar 5 20 mm).

Figure 6. Scanning electron micrograph of a specimen from a rabbit with a metallic stent, 14 weeks postoperative, showing the epithelium with ciliated cells and some nonciliated cells between filaments of the stent. (original 31,550; bar 5 20 mm). CHEST / 115 / 2 / FEBRUARY, 1999

493

surgical correction of airway stenosis. These methods include: resection of the stenosis with diathermy, laser, cryosurgery, or forceps; dilatation of the stenosis; brachytherapy (with malignant stenoses); and stenting the stenotic zone of the tracheobronchial tree. In addition to the silicone Montgomery Ttube,10,11 one of the most widely used types of silicone stents was designed by Dumon (Novatech; Aubagne, France).4 Dumon silicone stents were used by Martinez-Ballarin et al12 in the management of benign tracheobronchial stenosis in 63 patients. Temporary stents were removed from 21 patients after 18 months, and, although stenosis recurred in 4 patients, the therapy proved successful in 17. The complication rate involving stents was 31.7%, with complications including migration of the stent in 11 patients, granuloma formation in 4 patients, and airway obstruction due to heavy secretion in 4 patients. One of their patients died from hypersecretion and obstruction of the airway. Because of the poor prognosis of patients with malignant airway obstruction, palliative methods are needed to maintain airway patency. Bolliger et al13 used 38 Dumon stents in 31 patients for endoscopic palliation of malignant airway stenosis. Their patients underwent laser resection and insertion of the stent, followed by percutaneous or endobronchial radiotherapy. Complications were noticed with 10 stents, with migration being the most common. Both Dumon silicone stents and metallic stents, with or without a silicone covering, were used to treat patients with severe dyspnea for airway stenoses caused by malignant tumors. Bolliger et al13 concluded that stents play an important role in the improvement of respiratory symptoms and in the quality of life of patients with malignant tracheobronchial stenosis that was impossible to cure by surgery. Bare metallic stents are effective in extrinsic compression, but Dumon stents or covered metallic stents are preferable for use in intraluminal stenosis.14 Rousseau et al5 used 39 Wallstent and 35 Gianturco (Cook Europe; Bjaerverskov, Denmark) stents in 55 patients with 33 tracheal and 29 bronchial noninflammatory (except for one) lesions. The group using the Wallstent showed a total of six complications, with one tumoral proliferation successfully treated endoscopically with a laser. Gianturco stents were used exclusively for tracheobronchomalacia. In this group, Rousseau et al5 noticed a high rate (31%) of complications such as braces in the filament branches with or without migration of the stent that potentially can lead to obstruction or wall perforation. They summarized that Wallstent insertion is a safe procedure and a good alternative to silicone stent insertion for the treatment of post-transplanta494

tion bronchial stenoses without inflammatory lesions. Both silicone15 and self-expandable metallic stents5,16 thus have performed satisfactorily in the treatment of post-transplantation airway stenoses. Extramucosal resection of the cartilaginous arches of the trachea in pigs has been used to create an experimental malacia.17 When Marquette et al18 resected three tracheal rings and, in addition, used caustic agent intrabronchially during rigid bronchoscopy to create a tracheal stenosis, the clinically significant tracheal stenosis developed in their adult mini-pigs in 8 weeks. We used almost the same method, but without any caustic agent to create an experimental cervical tracheal stenosis, in our rabbits in order to evaluate various possibilities for treating the stenosis with dilatation and different types of intratracheal stents. In our study, the material of the stents was well tolerated, and no foreign-body reaction was noticed on histologic evaluation. The findings in histologic and SEM studies in the groups using metallic and SR-PLLA stents were so similar and the differences were so minor that no significant difference between these groups could be found. The changes in the group using silicone stents were more obvious. Clinically, silicone stents performed the worst. All rabbits in that group had to be killed due to respiratory symptoms. In the group using SR-PLLA stents, parts of the disintegrating spiral could have caused obstruction of the airways; one rabbit died spontaneously, and the other was killed because of respiratory symptoms. In the group using the Wallstent, no complications developed, and the animals could be followed to the end of the follow-up period. In SEM studies, a well-preserved carpet of ciliated cells was noticed in the stent material in the groups using SR-PLLA and metallic stents. Eight animals showed structural abnormality of the cilia, which were flattened. However, this can be an artifact, due to the fixation procedures for SEM studies. In a prospective study of intubated patients, Konrad et al19 found that impaired mucociliary transport in intubated patients is associated with the loss of cilia rather than with ultrastructural abnormalities of the cilia. We had no method to measure the mucociliary transport across the stents. The finding in SEM studies of well-preserved ciliary cells between the filaments of the stents in the groups using metallic and biosabsorbable stents suggests that mucociliary transport can work after the ciliated epithelium has covered the stents. Conclusion An ideal stent for the treatment of tracheobronchial stenosis has not yet been developed. In the Laboratory and Animal Investigations

present experimental study, we compared three types of stents. These stents, made of silicone, SR-PLLA, and metallic fibers, were used for the management of surgically induced tracheal stenosis in rabbits. The metallic stents, overall, were tolerated the best; no complications developed, and all the animals could be followed to the end of the follow-up period. Silicone stents had a tendency to occlude, and all animals with those stents had respiratory symptoms when they were killed. Both silicone and metallic stents are already in clinical use, but all stents that are now available have their disadvantages. The SR-PLLA stents were well tolerated, and mucosa with ciliated cells was preserved between the stent spirals, allowing bronchial mucus transport to continue. Thus, SR-PLLA seems to be a promising material for use in airway stents. References 1 Grillo HC, Donahue DM, Mathisen DJ, et al. Postintubation tracheal stenosis. J Thorac Cardiovasc Surg 1995; 109:486 – 493 2 Shennib H, Massard G. Airway complications in lung transplantation. Ann Thorac Surg 1994; 57:506 –511 3 Hetzel MR, Smith SGT. Endoscopic palliation of tracheobronchial malignancies. Thorax 1991; 46:325–333 4 Dumon J-F. A dedicated tracheobronchial stent. Chest 1990; 97:328 –332 5 Rousseau H, Dahan M, Lauque D, et al. Self-expandable prostheses in the tracheobronchial tree. Radiology 1993; 188:199 –203 6 Colt HG, Dumon J-F. Tracheobronchial stents: indications and applications. Lung Cancer 1993; 9:301–306

7 Rokkanen P, Bo¨stman O, Vainionpa¨a¨ S, et al. Biodegradable implants in fracture fixation: early results of treatment of fractures of the ankle. Lancet 1985; 1:1422– 424 8 Kemppainen E, Talja M, Riihela¨ M, et al. A biosorbable urethral stent. Urol Res 1993; 21:235–238 9 Talja M, Tammela T, Petas A, et al. Biodegradable selfreinforced polyglycolic acid spiral stent in prevention of postoperative urinary retention after visual laser ablation of the prostate-laser prostatectomy. J Urol 1995; 154:2089 –2092 10 Montgomery WW. T-tube tracheal stent. Arch Otolaryngol 1965; 82:320 –321 11 Gaissert HA, Grillo HC, Mathisen DJ, et al. Temporary and permanent restoration of airway continuity with tracheal T-tube. J Thorac Cardiovasc Surg 1994; 107:600 – 606 12 Martinez-Ballarin JI, Diaz-Jimenez JP, Castro MJ, et al. Silicone stents in the management of benign tracheobronchial stenoses. Chest 1996; 109:626 – 629 13 Bolliger CT, Probst R, Tschopp K, et al. Silicone stents in the management of inoperable tracheobronchial stenoses. Chest 1993; 104:1653–1659 14 Tojo T, Iioka S, Kitamura S, et al. Management of malignant tracheobronchial stenosis with metal stents and Dumon stents. Ann Thorac Surg 1996; 61:1074 –1078 15 Sonett JR, Keenan RJ, Ferson PF, et al. Endobronchial management of benign, malignant, and lung transplantation airway stenoses. Ann Thorac Surg 1995; 59:1417–1422 16 Brichon PY, Blanc-Jouvan F, Rousseau H, et al. Endovascular stents for bronchial stenosis after lung transplantation. Transplant Proc 1992; 24:2656 –2659 17 Johnston MR, Loeber N, Hillyer P, et al. External stent for repair of secondary tracheomalacia. Ann Thorac Surg 1980; 30:291–296 18 Marquette CH, Mensier E, Copin M-C, et al. Experimental models of tracheobronchial stenoses: a useful tool for evaluating airway stents. Ann Thorac Surg 1995; 60:651– 656 19 Konrad F, Schiener R, Marx T, et al. Ultrastructure and mucociliary transport of bronchial respiratory epithelium in intubated patients. Intensive Care Med 1995; 21:482– 489

CHEST / 115 / 2 / FEBRUARY, 1999

495