|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Respiratory Division, Departments of Internal Medicine, Anesthesiology, ENT, and Pathology, University Hospital, Basel, Switzerland. Supported by the Foundation of Pneumology, Basel, Switzerland; and Rüsch AG, Kernen, Germany.
| Abstract |
|---|
|
|
|---|
Setting: Animal laboratory of a university hospital.
Animals and interventions: In a series with 12 mini-pigs, we induced a stable fibromalacic tracheal stenosis that was 50% to 70% of the normal tracheal diameter. After dilation we inserted a 16 x 40-mm self-expandable silicone stent into the stenotic part of the trachea in 10 of the mini-pigs. Five of the stents had a smooth outer surface, and five had additional silicone retaining spikes. Because of a long stenosis in two of the mini-pigs, two overlapping stents (one smooth and one with spikes) were inserted.
Measurements and results: Stent deployment was successful and resulted in the disappearance of the slight to moderate stridor in all of the mini-pigs. Over a mean (± SD) observation period of 24 days (range, 10 to 41 days), all of the mini-pigs redeveloped stridor. Three of them died unexpectedly of suffocation: in all three a smooth stent had migrated, leading to total obstruction of the stenosis. In total, five of the six smooth stents migrated, and only one of the six spiked stents migrated. There was considerable granulation tissue formation at the ends of all of the stents. In the two control mini-pigs, a 12 x 35-mm Dumon stent was inserted. Both Dumon stents migrated, and one of them had considerable granuloma formation at its ends. At the end of the observation period, all stents were removed endoscopically and were found not to have deteriorated over time.
Conclusions: Our model proved to be suitable for the evaluation of the technical aspects of the Polyflex stent. Spikes on the outer stent surface are more effective in preventing migration than smooth-surface stents. Long-term compatibility, however, seems to be difficult to test with our model because both the Polyflex and the Dumon stents had excessive granulation tissue formation at both ends, a factor whichin the case of the Dumon stentdoes not occur to such a degree in benign human airway stenoses. Our results indicate a need for prospective long-term studies in benign human airway stenoses.
Key Words: animal model, central airway obstruction, stents, tracheal stenosis
| Introduction |
|---|
|
|
|---|
Apart from the various silicone stents, expandable metal stents have also been used widely. Although most of them were developed for vascular use, insertion in the tracheobronchial tree proved to be quite successful from a technical point of view.5 ,6 ,7 These stents consist of various forms of more or less densely woven wire struts or meshes, and are usually inserted under fluoroscopic control. Although they provide excellent initial palliation, their long-term results are often disappointing. Theoretically, these stents are epithelialized and therefore should be well tolerated. In reality they often lead to complications. In cases of malignant disease, tumor ingrowth through the wire mesh often occludes the stent as early as 2 to 3 months after placement.8 In benign disease the same problem arises through excessive granulation tissue formation. Furthermore, all of these stents are more difficult to remove than the Dumon stent. Many institutions therefore have abandoned the use of uncovered metal stents. However, metal stents have clear advantages: a very favorable ratio between the inner diameter and wall thickness, excellent adaptability to varying airway diameters, and the near absence of kinking. A current development is a metal stent covered with a thin layer of synthetic material that should prevent stent reocclusion. An example of such a device is the new covered Wallstent (Airway Wallstent; Schneider Europe AG; Bülach, Switzerland), which has recently been commercialized in Europe.9 ,10 The latest development in self-expandable airway stents is the Polyflex (Willy Rüsch AG; Kernen, Germany), which consists of a polyester wire mesh embedded in a silicone covering.
Before using newly developed stents in humans it would be helpful to test them in animal models. Marquette et al11 have recently described a model of tracheal stenosis in pigs that could be used for this purpose. We undertook the current study with a threefold purpose: we first wanted to validate the model of Marquette et al11 for the establishment of a stable tracheal stenosis. We then tested two different prototypes of the Polyflex stent: one has an almost smooth outer surface, and one has additional spikes on the outside. We hypothesized that the version with spikes would be superior to the smooth one in preventing stent migration. Finally, long-term tissue compatibility of the Polyflex stent was analyzed in comparison to the Dumon stent.
| Materials and Methods |
|---|
|
|
|---|
Anesthesia
All surgical and bronchologic procedures were performed with the
test animals under general anesthesia. Premedication was administered
IM with ketamine (15 mg/kg), midazolam (0.3 mg/kg), and atropine (0.03
mg/kg). After insertion of a peripheral IV catheter, anesthesia was
induced with IV propofol (2 to 3 mg/kg) followed by a continuous
infusion of propofol (6 to 10 mg/kg/h) for maintenance. For the
external surgical procedure the mini-pigs were intubated and ventilated
in a spontaneous-assisted mode. For the endoscopic procedures the
mini-pigs were intubated with a rigid bronchoscope and ventilated with
the same mode. Heart rate and oxygen saturation were monitored by pulse
oxymetry throughout the procedure. At the end of the operation,
extubation was performed as soon as spontaneous respiration was
sufficient to maintain oxygen saturation above 90%. During recovery
the mini-pigs were warmed with heating lamps. At the end of the final
procedure, the animals were killed with an IV bolus injection of 100
mg/kg of pentobarbital.
Model of Tracheal Stenosis
The pig model of a tracheal stenosis recently described by
Marquette et al11
consisted of a malacic and a fibrous
component mimicking the more complicated type of benign postintubation
or tracheostomy stenoses in humans. Briefly described, the
trachea was exposed via a midline cervical incision above the thoracic
inlet; and the ventral 50% of the tracheal cartilaginous rings six
through eight were removed extramucosally, which means the trachea was
not opened. Two weeks later, during rigid bronchoscopy, a 2-cm smear of
a highly caustic solution of NaOH (20%) was applied with a cotton swab
at the level of the resected arches. The posterior membrane was
spared this treatment to avoid the development of an esophagotracheal
fistula. The application was stopped at the first signs of a brownish
discoloration of the mucosa. Another 2 weeks later, a second rigid
bronchoscopy was performed to assess the degree of tracheal stenosis
and to insert the prototype stent, or to perform a second application
of NaOH if the degree of the stenosis was < 50% of the normal
diameter.
In a pilot series, two mini-pigs died when we used this approach: one asphyxiated when its trachea was totally obstructed by granulation tissue, and the other had a fatal hemoptysis because of caustic arrosion of an artery. Furthermore, an exact identification of the flaccid part of the trachea was not done as easily as expected, even with additional pressure on the trachea from outside, because inflammatory reparative changes of the mucosa made the recognition of the tracheal rings quite difficult. We subsequently modified the technique of Marquette et al11 : at the end of the resection of the cartilaginous arches, the entire surface was marked with two to three drops of India ink, which indicated the exact extent of the flaccid segment by a bluish halo seen on bronchoscopy 2 weeks later. The application of NaOH was strictly limited to the anterior 50% of the tracheal mucosa, and was stopped when the mucosa turned white before additional brownish discoloration ensued. This led to a less severe reaction resulting in a stable 50% to 70% tracheal stenosis of 2 to 2.5 cm in length, and no respiratory distress in the animals (Fig 1 ). Only when feeding did some of the mini-pigs exhibit a slight stridor without any signs of distress. We applied this modified technique to the series of 12 mini-pigs used in our testing of the new stent.
|
|
|
Follow-up and Outcome Evaluation
All of the mini-pigs were monitored daily for signs of
respiratory distress coughing, sputum production, wheezing, stridor,
and tachypneaas well as changes in weight or feeding habits. The
duration of the experiment was planned to be a maximum of 3 months in
order to evaluate short-term stent performance and long-term tissue
tolerance. Following stent insertion, monthly routine bronchoscopies
were planned, as were additional bronchoscopies if warranted by
clinical signs. If follow-up bronchoscopy revealed that a stent had
migrated, it was exchanged. If excessive granulation tissue formation
occurred on stent ends, the tissue was removed mechanically. If
the situation was judged to be irreparable, the animal was killed at
the end of the procedure with an IV injection of pentobarbital. For the
postmortem examination, the lung, trachea, and esophagus up to the
glottis were removed in a single block and examined histologically.
| Results |
|---|
|
|
|---|
A summary of the most important stent-related results is provided in Table 1 . The loading of the stents into the stent carrier was not accomplished as easily as expected. Sometimes the stents did not elongate in a tubular fashion and folded inward, and their reopening was made more difficult because some of the polyester wires were kinked. This problem was more prevalent in the stents with spikes; there was an increased tendency to fold in through the additional wall thickness at the spike level. However, stent deployment and expansion were achieved in all of the test animals, and the stents fitted the airway walls snugly (Fig 4 ). The overlapping stents placed in the two mini-pigs with longer stenoses fitted perfectly into each other. The overlapping area did not protrude noticeably into the lumen because of the thin wall thickness of the Polyflex stent. None of the stents moved when gently pulled by a forceps, and the procedure could thus be terminated with the stents in place in all of the test animals.
|
|
|
|
| Discussion |
|---|
|
|
|---|
In order to obtain a stable degree of tracheal obstruction without respiratory compromise, the stenosis model was modified slightly: the initial flaccid segment was marked more clearly with India ink, and the caustic agent (NaOH) was applied more carefully and more sparingly. Aside from these changes, the experimental model of Marquette et al11 proved to be excellent in testing the technical aspects of insertion, removal, and early tolerance of stents.
The application device had some shortcomings because some of the stentsespecially the ones with additional spikes on the outsidetended to fold in, instead of elongating in tubular fashion when compressed and loaded into the stent carrier. To avoid this difficulty, a central thin catheter should be added at the level of the wire mesh basket, which should prevent the stent from folding when it is compressed and elongated. The results of the study clearly show that the smooth-surface prototype tends to migrate (five of six stents migrated), and that stents having retaining spikes prevented this complication (one out of six migrated). These findings were the main benefit derived from our study, as it will obviate the need to test smooth Polyflex stents in benign human airway stenoses.
For at least 10 days all of the mini-pigs were asymptomatic, without stridor, coughing, or increased sputum production. This proved that early tolerance and mucus clearance were not problems. Unfortunately, our results were quite disappointing in the long run because all of the mini-pigs redeveloped stridor, a situation that necessitated additional bronchoscopies in 9 of the 12 animals. In the six mini-pigs that had stent migration, stridor was a result of severe obstruction at the level of the original stenosis and was more pronounced than it was before bronchoscopic dilation. In the remaining six animals (four having Polyflex stents, and two having Dumon stents) recurring stridor was clearly a result of the formation of granulation tissue, as it also was with one of the Dumon stents that had partially migrated. Marquette et al11 described similar problems occurring with different stent models in his experiments, with none of the animals surviving with a stent for > 2 weeks. In our study, the extent of granulation tissue formation with both the Polyflex and the Dumon stents was clearly more pronounced than what has been our own experience when using the Dumon stent in humans with similar benign stenoses. This is in accordance with a recent report by Martinez-Ballarin et al,12 who confirmed the good tolerance of the Dumon stent in a series with 63 patients who had benign tracheobronchial stenoses, only 4 of whom (6.3%) developed granulation tissue. We therefore speculate that it is primarily the mini-pig model that leads to more severe long-term reactions than generally occur in humans.
Wassermann et al13 reported the first results of the use of Polyflex stents in humans, and major granulation tissue formation was not observed with 21 stents in 11 patients. In that study, however, 9 of the 11 patients suffered from malignant disease, a factor that precludes direct comparison with benign situations. The question of whether the important granulation tissue formation in the mini-pigs was a result of the Polyflex stent per se, or whether mini-pigs tend to produce granulation tissue more readily than humans cannot be answered unequivocally with the currently available data. Long-term tolerance of stents in benign airway obstruction might be difficult, or even impossible, to test with our current animal model.
The high migration rate of the smooth-surface Polyflex stent seems to confirm our assumption that covered stents do not anchor sufficiently by means of their radial expansion force alone, but need additional mechanical studs, spikes, hooks, or rings to prevent migration. We made similar observations with the Airway Wallstent, a self-expandable covered metal stent. Contrary to the Polyflex stent, the Airway Wallstent has sharp wire ends protruding from the covering at both ends on compression. These wire ends prevent stent migration by lodging in the mucosa.14 Our migration rate with the smooth Polyflex prototype was clearly higher than that reported by Wassermann et al,13 who cited only two dislocations in 21 smooth-stent insertions; but the high percentage of malignant disorders in that series does not allow direct comparison with our study.
We conclude that the modified experimental fibromalacic stricture of Marquette et al11 is a useful model of a tracheal stenosis to test all short-term technical aspects of stents. This obviates the need to test them directly in humans. Of the two Polyflex stent prototypes tested, the one with spikes on the outer surface helped to prevent migration and makes this stent look promising. Long-term compatibility, however, seems to be difficult or even impossible to test with our model, as both the Polyflex and the Dumon stents led to excessive granulation tissue formation at the stent ends, a factor whichin the case of the Dumon stentdoes not occur to such a degree in benign human airway stenoses. Lacking a perfect animal model for testing long-term stent tolerance, clinical trials in humans will have to assess this aspect of the Polyflex stent.
| Footnotes |
|---|
Correspondence to: Chris T Bolliger, MD, Respiratory Division, University Hospital, Petersgraben 4, 4031 Basel, Switzerland; e-mail: bolligerct@ubaclu.unibas.ch
Received for publication February 5, 1998. Accepted for publication July 7, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. R. Gildea, S. C. Murthy, D. Sahoo, D. P. Mason, and A. C. Mehta Performance of a self-expanding silicone stent in palliation of benign airway conditions. Chest, November 1, 2006; 130(5): 1419 - 1423. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. B. Pillai, J. Smith, A. Hasan, and D. Spencer Review of pediatric airway malacia and its management, with emphasis on stenting Eur. J. Cardiothorac. Surg., January 1, 2005; 27(1): 35 - 44. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Ernst, D. Feller-Kopman, H. D. Becker, and A. C. Mehta Central Airway Obstruction Am. J. Respir. Crit. Care Med., June 15, 2004; 169(12): 1278 - 1297. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |