- Email: [email protected]
The use of expandable metallic stents for acute tracheal stenosis in the growing lamb
The Use of Expandable By Dimitris
Metallic Stents for Acute Tracheal in the Growing Lamb
E. Tsakayannis, Aqeel M. Siddiqui, Harry Kozakewich, Kerilyn Juan C. Ibla, Stanton D. Perry, and Craig W. Lillehei Boston, Massachusetts
Purpose: Expandable metallic stents (Palmaz been used in the treatment of tracheobronchial children and adults. The authors investigated the management of acute airway stenosis animal model.
stents) have obstruction in their utility in in a growing
Methods: A model for tracheal stenosis was developed in young lambs (mean age, 4 weeks; mean weight, 8.6 kg). Via an anterior tracheotomy, a circumferential mucosal injury to the trachea was produced with electrocautery in 31 lambs. In the control group (n = IO) no further intervention was used. In the treatment groups, either serial balloon dilatation of the stricture was performed (n = 61, or expandable metallic stents were inserted across the stricture (n = 15). All animals were monitored daily for signs of respiratory distress. Body weights, fluoroscopic airway measurements and rigid bronchoscopy were performed at least weekly. Results: The average weekly rate of airway growth was 8.2% ? 5.5% of the tracheal cross-sectional area (CSA). All animals displayed severe stenosis (mean, 90.2% ? 7.5% of CSA) within 13.1 t- 4 days after the injury. All animals in the control group had severe respiratory distress, weight loss and died within 14.6 Ifr. 2.8 days after injury. Serial balloon
T
RACHEOBRONCHIAL STENOSIS can result from congenital, inflammatory, neoplastic, or traumatic causes.’ Strictures can also be found at bronchial anastomoses such as those after lung transplantation.2 Currently a wide variety of therapeutic options are available for the management of tracheobronchial stenoses.3 These range from surgical resection to bronchoscopic dilatation, laser therapy, or placement of stents. Experience using expandable metallic stents in the tracheobronchial tree primarily has been in adults4-6and only few children.7J Experimental evidence for their use in acute airway obstruction and during airway growth is quite limited. We investigated
From the Departments of Surgery, Pathology, and Cardiology, Children’s Hospital, Boston, MA. Presented at the 1997 Annual Meeting of the Section on Surgeqv of the American Academy of Pediatrics, New Orleans, Louisiana, October 31 -November 2,1997. Address reprint requests to Craig W Lillehei, MD, Department of Surgery, Fegan 3, Children k Hospital, 300 Longwood Ave, Boston, MA 02115. Copyright o I998 by WB. Saunders Company 0022.3468/98/3307-OOI4$03.00/0 1038
Stenosis
K. Nobuhara,
dilatation of the stricture alone failed to relieve symptoms in all six animals in this group, who died within 20 ? 1 days after the injury, despite two to three dilatations each. With placement of expandable metallic stents, only 3 of 15 lambs died (two of pneumonia, one of iatrogenic perforation). The remaining 12 remained symptom-free and gained weight during a 2-month follow-up period. However, fluoroscopic examination showed partial collapse of the stents in all of these animals (mean, 44.7% i 21.6% of CSA) requiring an average of 2 2 0.7 bronchoscopic dilatations. Pathological evaluation showed more pronounced granulation tissue in the stented animals. Conc/usions:The authors conclude that expandable metallic stents provide an effective tool in the management of acute tracheal stenosis. However, airway growth, tissue reaction, and the mechanical properties of the stent require close monitoring and stent adjustment. J Pediatr Surg 33:1038-1042. Copyright o 1998 by W.B. Saunders Company, INDEXWORDS:Tracheal lambs.
stenosis,
expandable
metallicstents,
the utility of metallic endovascular (Palmaz) stents during acute tracheal stenosis using a growing lamb model. MATERIALS
AND
METHODS
Initial Procedure The following protocol was approved by the Animal Care and Use Committee at Boston Children’s Hospital and complies with the “Guidelines for the Care and Use of Laboratory Animals” (NIH publication No. 85-23, revised 1985). Tracheal strictures were created in 31 neonatal lambs (Parsons Farms, Springfield, MA) weighing approximately 10 kg. Animals were sedated using ketamine, 15 mg/kg followed by induction of general anesthesia with halothane, 1% to 2%. After intubation with a 5.5mm uncuffed endotracheal tube, a midline skin incision was performed over the trachea. Through an anterior transverse tracheotomy several tracheal rings below the cricoid cartilage, the inner trachea was exposed. Using electrocautery, a l-cm wide circumferential, mucosal, and perichondreal bum was produced 2 cm distal to the tracheotomy. The tracheal incision was then closed using 5-O polypropylene suture. After recovering, the animals were placed back in their pens and monitored daily for symptoms. All animals received at least weekly bronchoscopy using a 5-mm rigid bronchoscope to evaluate stricture formation. Airway growth was assessed radiographically by calculating the cross-sectional area (CSA) of a normal region of the trachea from a lateral neck film. We used an Journal
of PediatricSurgery,
Vol33,No7
(July),1998:pp
1038-1042
EXPANDABLE
STENTS
FOR TRACHEAL
STENOSIS
1039
average of the diameter of the normal trachea above (A) and below (B) the stricture to calculate the CSA as CSA = T(A + B/4)2. Percent stenosis = (CSA [stricture]/CSA [normal trachea]) X 100. When the lambs became symptomatic, approximately 2 weeks postoperatively, they received either an intervention or served as controls. Intervention consisted of balloon dilatation either with or without stent placement.
70 60 1
Group 1: Balloon Dilatation With Stent Placement These animals were intubated with a 5-mm rigid bronchoscope equipped with a side port for delivery of the anesthetic. Under fluoroscopic guidance, a balloon (4-cm length, 14-mm diameter) catheter fitted with a metallic expandable endovascular stent (Palmaz, P308, Johnson & Johnson, Warren. NJ) was inserted through the bronchoscope to the level of the stricture. The balloon was fully dilated within the stricture to allow complete expansion of the stent. The catheter was then withdrawn and the stent visualized with a fiberoptic lens through the bronchoscope. Animals were killed 2 months after the initial procedure.
Group 2: Balloon Dilatation Without Stent Placement Animals in this group were intubated with a rigid 5-mm bronchoscope. Anesthesia was delivered via a side port on the bronchoscope. Under fluoroscopic guidance, a 4-cm long balloon catheter was placed through the bronchoscope to the level of the stricture. The balloon was fully expanded to 12 mm in diameter within the stricture for 15 seconds and released. After the catheter was removed, a fiberoptic lense was inserted into the bronchoscope to visualize the expanded stricture. After recovery, animals were returned to their pens and monitored daily. With the recurrence of symptoms, animals were redilated until balloon dilatation failed to relieve symptoms, at which time they were killed.
Group 3: Control Group After the tracheal stricture was produced, monitored with no further intervention.
animals
in this group
were
Specimens After death of the animals, the laryngotracheal complexes were removed and fixed in 10% formalin for several days. Specimens subsequently were x-rayed in both antero-posterior and lateral views and sectioned either in the coronal or saggital plane for H&E staining.
RESULTS
Thirty-one lambs were used with a mean age of 4 weeks and an average weight of 8.6 + 1.6 kg. Airway growth of the stented animals is shown in Fig 1. The average weekly rate of airway growth was 8.2% k 5.5% of the tracheal cross sectional area (CSA). The experimental model produced severe tracheal stenosis with a mean narrowing of 90.2% f 7.5% of the tracheal CSA. The tracheal stenosis was easily reproducible and evident within 13.1 2 4 days after the injury. The animal data are shown in Table 1. Animals in the control group had severe respiratory distress and died within 14.6 t 2.8 days after the injury. Respiratory distress was evident in the lambs who received serial balloon dilatations without stent placement despite having an average of two to three dilatations each. All died within 20 + 1 days after the injury. In the group in which expandable metallic stents were placed, 12 of 15 animals
1
2
3
/
r-
4
5
/
6
7
8
9
Time (Weeks) Fig 1. Weekly percentage airway normal trachea compared with initial
growth CSA.
measured
in CSA
of
survived, remaining symptom-free and gaining weight over the 2-month follow-up period. Fluoroscopic and endoscopic examination in the stented animals showed partial collapse of all the stents by 44.1% k 21.6% of their initial CSA. Over the course of the study, an average number of 2 + 0.7 bronchoscopic dilatations were required to expand the collapsed stent to its previous size. Despite partial collapse of the stents, these animals showed no clinical symptoms of respiratory distress. Pathology The tracheas from the control and balloon dilated groups appeared similar and were characterized grossly by a circumferential stricture with mucopurulent exudate. Histologically, they showed mucosal, muscle, and cartilage necrosis with inward buckling of the tracheal wall that produced the narrowing. Mucosal ulcers were incompletely healed with superficial bacterial invasion, particularly of the exposed necrotic cartilage, accompanied by inflammatory exudate, granulation tissue, and fibrocartilage repair. The stented tracheas had either a widely patent lumen or were narrowed from partial collapse of the stent in the anterior-posterior direction. One- to three-millimeter polyps through the fenestrations of the stent often interfered Table
1. Characteristics
of Experimental
Parameter
Dilation
Number Degree of stricture Respiratory distress
6 >90%
Stricture recurrence Complications Survival
Yes 6 (100%) Pneumonia, 0 (0%)
1
and Control
Groups
stem
Controls
15
10
>90% No 0 (0%)
>90% Yes 10 (100%)
Perforation, Pneumonia, 12 (80%)
1 2 0 (0%)
TSAKAYANNIS
1040
ET AL
with easy stent removal (Fig 2). Tracheas examined histologically after stent removal showed patchy mucosal ulcerations with acute inflammatory exudate and bacterial colonization or superficial invasion. A marked mucosal chronic inflammatory infiltrate was also present. The mucosal polyps were characterized by inflamed mucosa or granulation tissue. Transmural fibrosis also was present (Fig 3). DISCUSSION
Stents have been used as a temporary maneuver for both benign and malignant tracheobronchial obstruction. Expandable intraluminal stenting was developed for the nonoperative management of stenotic vascular disease.9 However, because these stents are increasingly used in extravascular sites, a formal assessment of their utility in such locations is warranted. The use of stents in the management of acute airway stenosis is of particular interest.‘O Different classes of stents have been developed for airway stenosis. Silicon rubber stents have been shown to interfere with mucociliary clearance and have a high rate of migration. Self-expanding metallic stents (Gianturco, Cook Group Europe, Bjaeverskoo, Denmark; Wallstent, Medinvent SA, Lausanne, Switzerland) expand to a fixed diameter.7,12J3 Ulceration and erosion has therefore resulted from excessive intraluminal pressure. Balloonexpandable metallic stents (Palmaz, Strecker; Boston Scientific, Watertown, MA) have been used successfully for acute airway stenoses.7J0J1 These stents are well tolerated in the tracheobronchial tree, allowing for better mucociliary clearance. Migration is minimized by tissue
Fig 2. Gross pathology comparing a control trachea (A) to stented trachea (B). Of note is the inflammatory reaction with granulation tissue growing through the fenestrations of the stent.
Fig 3. Histology of the same specimens shown in Fig 2. Control trachea (A) shows marked narrowing compared with the stented trachea (B). The stent has been removed for processing of the specimen.
ingrowth through the wire mesh. However, these stents may collapse and also initiate an intense inflammatory reaction.14 Experience using expandable metallic stents in children is limited. Filler et al7 reported a series of seven children in whom expandable metallic stents were placed for recurrent stricture after tracheoplasty for congenital tracheal stenosis. Of these patients, three had recurrent airway obstruction after 1 month requiring redilatation. One patient died after recurrence of symptoms and removal of stent. This preliminary series emphasized the utility of expandable metallic stents but also raised questions regarding the long-term behavior and tolerance of the stent during growth. By creating a model of tracheal stenosis in a growing animal, we hoped to better define the usefulness of the Palmaz stent for acute airway obstruction in the pediatric population. The animal model of tracheal stenosis used in this study is a modification of that used by Duff et all5 and Wenig et alI6 in beagles. In our study, the Palmaz stent clearly had a beneficial effect in the immediate symptomatic period. Without intervention, animals became severely stridorous and either died acutely or had to be killed because of respiratory distress. Simple balloon dilatation of the stricture proved of only marginal benefit, providing temporary symptomatic relief and very limited prolongation of survival. All surviving animals in the stented group were symptom-free for the remainder of the study. However, during routine fluoroscopic examinations, par-
EXPANDABLE
STENTS
FOR TRACHEAL
1041
STENOSIS
tial stem collapse was seen, requiring repeated dilatations. Based on our findings, stent collapse refers to the narrowing of the stent without airway compromise. Stricture recurrence refers to narrowing of the airway caused by transmural fibrosis. Thus, collapse may mean that the stents are too pliable to withstand the external forces of neck movement and, to a lesser degree, scar contracture. Furthermore, dilatations are required to accommodate tracheal cross-sectional growth. These stents also pose risks. For example, one animal died of an iatrogenic perforation, stressing the importance of selecting the correct balloon size for dilation. Two animals died of pneumonia despite appropriate treatment with antibiotics, possibly from hindered ciliary clearance of bronchial secretions. Ideally, a stent for the management of acute tracheo-
bronchial stenosis should have several characteristics. It must be safe and easy to place and, if necessary, remove yet avoid unwanted migration. It should be well tolerated by the tissue with minimal inflammatory reaction or interference with mucociliary clearance. It needs to be constructed of such a material to prevent collapse, but not produce excessive pressure on the underlying mucosa. This study describes an easily reproducible model of tracheal stenosis in a growing animal, which can be used to test newer airway stents and perfect their properites before further clinical application.
ACKNOWLEDGMENTS
The authors thank Chrissy Rader, BA, and Michelle Ferretti, BA, for their technical assistance.
REFERENCES 1. Brichon for bronchial
PY, Blanc-Jouvan stenosis after
F, Pison C, et al: Endovascular lung transplantation. Transplant
stents Proc
24:2656-2659,1992 2. Sonett JR, Keenan
RJ, Ferson PF, et al: Endobronchial management of benign, malignant, and lung transplantation airway stenoses. Ann Thorac Surg 59:1417-1422, 1995 3. Tsang V, Williams A, Goldstraw P: Sequential silastic and expandable metal stenting for tracheobronchial strictures. Ann Thorac Surg 53:856-8560, 1992 4. George PJ: Irving JD, Khaghani A, et al: Role of the Gianturco expandable metal stent in the management of tracheobronchial obstmction. Cardiovasc Intervent Radio1 15:375-381, 1992 5. Higgins R, Mcneil K, Dennis C, et al: Airway stenoses after lung transplantation: Management with expandable metal stems. J Heart Lung Transplant 13:774-778. 1994 6. Shah R, Sabanathan S, Mearns AJ, et al: Self-expanding metallic stents in the management of major airway problems. J Cardiovasc Surg 36:343-348,1995 7. Filler RM, Forte V, Fraga JC, et al: The use of expandable metallic airway stents for tracheobronchial obstruction in children. J Pediatr Surg 30:1050-1056, 1995 8. Bugmann P. Rouge JC, Bemer M, et al: Use of Gianturco 2 stents in the treatment of vascular compression of the tracheobronchial tree in childhood, a feasible solution when surgery fails. Chest 106: 1580- 1582, 1994
9. Vorwerk D, Gunther RW: Stent placement in iliac arterial lesions: Three years of clinical experience with the wallstent. Cardiovasc Intervent Radio1 15:285-290, 1992 10. Zollikofer CL, Antonucci F, Stuckmann G: Historical overview on the development and characteristics of stents and future outlooks. Cardiovasc Intervent Radio1 15:272-278, 1992 11. Palmaz JC: Intravascular stenting: From basic research to clinical application. Cardiovasc Intervent Radio1 15:279-284, 1992 12. Mair EA, Parsons DS, Lally Kp, et al: Comparison of expandable endotracheal stents in the treatment of surgically induced piglet tracheomalacia. Laryngoscope lOl:lOOl-1007, 1991 13. Gaer JAR, Tsang V, Khaghani A, et al: Use of endotracheal silicone stents for relief of tracheobroncial obstruction. Ann Thorac Surg 54:512-516, 1992 14. Marquette CH, Mensier E, Copin M-C, et al: Experimental models of tracheobronchial stenoses: A useful tool for evaluating airway stents. Ann Thorac Surg 60:651-656, 1995 15. Duff BE, Wenig BL, Applebaum EL, et al: Tracheal reconstruction using an epithelial equivalent. Laryngoscope 104:409-414, 1994 16. Wenig BL, Reuter VC, Steinberg BM, et al: Tracheal reconstmction: In vitro and in vivo animal pilot study. Laryngoscope 97:959-965, 1987
Discussion R.G. Azizkhan (Buffalo, NY): This is a very interesting and exciting paper. Your lamb model recapitulates the findings that we have seen in human patients with tracheal stenosis that have undergone tracheal stenting. We have also found, in infants, that in most cases the stent does partially collapse during the recovery period, and these patients require repeated endoscopy to remove granulation tissue, which does go through the interstices as you have very nicely demonstrated in your animal model. Also, the increasing fibrosis does seem to have an impact on collapsing the stent.
Do you feel that the collapse of the stents in the animal model are related to the unique neck motions of the animal in that these stents are extremely fragile, it doesn’t take very much to collapse them? Or is it related primarily to the fibrosis that you have seenin your model? Have you looked at the impact of the stents in the tracheas that have not had any type of injury? Do you still see the same intense inflammatory reaction? If not, what is the impact of this in terms of removing the stents after the placement? Because we have had trouble removing the stents in some of the patients that have had tracheal
1042
stenosis and placement of the stents for a prolonged period. The danger is tearing the trachea in the process of removing the stents. So with these questions in mind, this is a very good model to look at other types of stent material such as absorbable stents, which may have a much more longterm efficacy in this type of difficult problem. Thank you for the opportunity to comment. D.E. Tsakayannis (response): Thank you very much for your kind remarks. On the first question, I think that part of the collapse was caused by the fibrous tissue, but mainly the mechanical properties of the stent. The growth of about 8% a week is also significant. We may need to look at newer types of stents such as those made of nitinol. This nickel-titanium compound is more rigid, and may thereby improve outcome. Regarding your second question, we did not actually have a group with normal tracheas. I think this is an excellent remark, and we should be looking at that. In regard to your third question, I think that with this model we can definitely use a newer type of stent, or even modify a particular stent for the trachea. Since these stents were made for blood vessels, it may be that for the trachea different tensile strengths and mechanical properties will be required. In the meantime, we can use those things that are available but are looking at different stents for the future. A. Kosloske (Lubbock, TX): I rise with a note of caution. About 2 years ago a pediatric surgical colleague of mine inserted a Palmaz stent into the left main stem bronchus of a retarded boy who had severe bronchial malacia, successfully reopening the airway. This patient experienced a fatal complication of the stent. Approximately 2 months later he had a massive episode of hemoptysis and at autopsy he had formed a bronchial aortic fistula with a stent eroded through the wall of the bronchus.
TSAKAYANNIS
ET AL
So I must caution the audience against the clinical use of these stents in any part of the airway that pulsates immediately adjacent to a pulsatile vessel. The stents may be safe in a relatively quiet part of the airway, but much more experimental work needs to be done. D.E. Tsakayannis (response): I agree 100% with your statement. I think it is very interesting to note that stents were first used clinically, but now we are looking back to the experimental evidence. I think you make a very solid point on how careful we should be in the future with stents. E. Doolin (Camden, NJ): The Palmaz stent is convenient and easy to use. That is why it is popular, especially for tracheal and bronchial malacia. However, what makes it expand can make it collapse. And you have proven that. Do you think you have found in the inflammatory stenosis a subset that deserve a more resilient stent like a Dumont stent? In your model, which doesn’t exist much in nature, and the subglottic equivalent studied last year, the burn of the airway creates enough regulation of the contracting type III collagen and the down-regulation of the mature type I collagen. Have you studied the biochemical components of your scar to see if they are dissimilar to what we see in nature and therefore make them more at risk for restriction? D.E. Tsakayannis (response): Thank you for your questions. In regard to your first question, I have not used the Dumont stent because, as you said, the PaImaz is much more easily accessible. Given that Dumont is a silicon stent, it may have a higher migration rate. With regard to your second question, we have not identified the specific collagen type, but I think that, as you pointed out, it may not just represent re-stricture, but may have an element of tracheomalacia as well.
Metallic Stents for Acute Tracheal in the Growing Lamb
E. Tsakayannis, Aqeel M. Siddiqui, Harry Kozakewich, Kerilyn Juan C. Ibla, Stanton D. Perry, and Craig W. Lillehei Boston, Massachusetts
Purpose: Expandable metallic stents (Palmaz been used in the treatment of tracheobronchial children and adults. The authors investigated the management of acute airway stenosis animal model.
stents) have obstruction in their utility in in a growing
Methods: A model for tracheal stenosis was developed in young lambs (mean age, 4 weeks; mean weight, 8.6 kg). Via an anterior tracheotomy, a circumferential mucosal injury to the trachea was produced with electrocautery in 31 lambs. In the control group (n = IO) no further intervention was used. In the treatment groups, either serial balloon dilatation of the stricture was performed (n = 61, or expandable metallic stents were inserted across the stricture (n = 15). All animals were monitored daily for signs of respiratory distress. Body weights, fluoroscopic airway measurements and rigid bronchoscopy were performed at least weekly. Results: The average weekly rate of airway growth was 8.2% ? 5.5% of the tracheal cross-sectional area (CSA). All animals displayed severe stenosis (mean, 90.2% ? 7.5% of CSA) within 13.1 t- 4 days after the injury. All animals in the control group had severe respiratory distress, weight loss and died within 14.6 Ifr. 2.8 days after injury. Serial balloon
T
RACHEOBRONCHIAL STENOSIS can result from congenital, inflammatory, neoplastic, or traumatic causes.’ Strictures can also be found at bronchial anastomoses such as those after lung transplantation.2 Currently a wide variety of therapeutic options are available for the management of tracheobronchial stenoses.3 These range from surgical resection to bronchoscopic dilatation, laser therapy, or placement of stents. Experience using expandable metallic stents in the tracheobronchial tree primarily has been in adults4-6and only few children.7J Experimental evidence for their use in acute airway obstruction and during airway growth is quite limited. We investigated
From the Departments of Surgery, Pathology, and Cardiology, Children’s Hospital, Boston, MA. Presented at the 1997 Annual Meeting of the Section on Surgeqv of the American Academy of Pediatrics, New Orleans, Louisiana, October 31 -November 2,1997. Address reprint requests to Craig W Lillehei, MD, Department of Surgery, Fegan 3, Children k Hospital, 300 Longwood Ave, Boston, MA 02115. Copyright o I998 by WB. Saunders Company 0022.3468/98/3307-OOI4$03.00/0 1038
Stenosis
K. Nobuhara,
dilatation of the stricture alone failed to relieve symptoms in all six animals in this group, who died within 20 ? 1 days after the injury, despite two to three dilatations each. With placement of expandable metallic stents, only 3 of 15 lambs died (two of pneumonia, one of iatrogenic perforation). The remaining 12 remained symptom-free and gained weight during a 2-month follow-up period. However, fluoroscopic examination showed partial collapse of the stents in all of these animals (mean, 44.7% i 21.6% of CSA) requiring an average of 2 2 0.7 bronchoscopic dilatations. Pathological evaluation showed more pronounced granulation tissue in the stented animals. Conc/usions:The authors conclude that expandable metallic stents provide an effective tool in the management of acute tracheal stenosis. However, airway growth, tissue reaction, and the mechanical properties of the stent require close monitoring and stent adjustment. J Pediatr Surg 33:1038-1042. Copyright o 1998 by W.B. Saunders Company, INDEXWORDS:Tracheal lambs.
stenosis,
expandable
metallicstents,
the utility of metallic endovascular (Palmaz) stents during acute tracheal stenosis using a growing lamb model. MATERIALS
AND
METHODS
Initial Procedure The following protocol was approved by the Animal Care and Use Committee at Boston Children’s Hospital and complies with the “Guidelines for the Care and Use of Laboratory Animals” (NIH publication No. 85-23, revised 1985). Tracheal strictures were created in 31 neonatal lambs (Parsons Farms, Springfield, MA) weighing approximately 10 kg. Animals were sedated using ketamine, 15 mg/kg followed by induction of general anesthesia with halothane, 1% to 2%. After intubation with a 5.5mm uncuffed endotracheal tube, a midline skin incision was performed over the trachea. Through an anterior transverse tracheotomy several tracheal rings below the cricoid cartilage, the inner trachea was exposed. Using electrocautery, a l-cm wide circumferential, mucosal, and perichondreal bum was produced 2 cm distal to the tracheotomy. The tracheal incision was then closed using 5-O polypropylene suture. After recovering, the animals were placed back in their pens and monitored daily for symptoms. All animals received at least weekly bronchoscopy using a 5-mm rigid bronchoscope to evaluate stricture formation. Airway growth was assessed radiographically by calculating the cross-sectional area (CSA) of a normal region of the trachea from a lateral neck film. We used an Journal
of PediatricSurgery,
Vol33,No7
(July),1998:pp
1038-1042
EXPANDABLE
STENTS
FOR TRACHEAL
STENOSIS
1039
average of the diameter of the normal trachea above (A) and below (B) the stricture to calculate the CSA as CSA = T(A + B/4)2. Percent stenosis = (CSA [stricture]/CSA [normal trachea]) X 100. When the lambs became symptomatic, approximately 2 weeks postoperatively, they received either an intervention or served as controls. Intervention consisted of balloon dilatation either with or without stent placement.
70 60 1
Group 1: Balloon Dilatation With Stent Placement These animals were intubated with a 5-mm rigid bronchoscope equipped with a side port for delivery of the anesthetic. Under fluoroscopic guidance, a balloon (4-cm length, 14-mm diameter) catheter fitted with a metallic expandable endovascular stent (Palmaz, P308, Johnson & Johnson, Warren. NJ) was inserted through the bronchoscope to the level of the stricture. The balloon was fully dilated within the stricture to allow complete expansion of the stent. The catheter was then withdrawn and the stent visualized with a fiberoptic lens through the bronchoscope. Animals were killed 2 months after the initial procedure.
Group 2: Balloon Dilatation Without Stent Placement Animals in this group were intubated with a rigid 5-mm bronchoscope. Anesthesia was delivered via a side port on the bronchoscope. Under fluoroscopic guidance, a 4-cm long balloon catheter was placed through the bronchoscope to the level of the stricture. The balloon was fully expanded to 12 mm in diameter within the stricture for 15 seconds and released. After the catheter was removed, a fiberoptic lense was inserted into the bronchoscope to visualize the expanded stricture. After recovery, animals were returned to their pens and monitored daily. With the recurrence of symptoms, animals were redilated until balloon dilatation failed to relieve symptoms, at which time they were killed.
Group 3: Control Group After the tracheal stricture was produced, monitored with no further intervention.
animals
in this group
were
Specimens After death of the animals, the laryngotracheal complexes were removed and fixed in 10% formalin for several days. Specimens subsequently were x-rayed in both antero-posterior and lateral views and sectioned either in the coronal or saggital plane for H&E staining.
RESULTS
Thirty-one lambs were used with a mean age of 4 weeks and an average weight of 8.6 + 1.6 kg. Airway growth of the stented animals is shown in Fig 1. The average weekly rate of airway growth was 8.2% k 5.5% of the tracheal cross sectional area (CSA). The experimental model produced severe tracheal stenosis with a mean narrowing of 90.2% f 7.5% of the tracheal CSA. The tracheal stenosis was easily reproducible and evident within 13.1 2 4 days after the injury. The animal data are shown in Table 1. Animals in the control group had severe respiratory distress and died within 14.6 t 2.8 days after the injury. Respiratory distress was evident in the lambs who received serial balloon dilatations without stent placement despite having an average of two to three dilatations each. All died within 20 + 1 days after the injury. In the group in which expandable metallic stents were placed, 12 of 15 animals
1
2
3
/
r-
4
5
/
6
7
8
9
Time (Weeks) Fig 1. Weekly percentage airway normal trachea compared with initial
growth CSA.
measured
in CSA
of
survived, remaining symptom-free and gaining weight over the 2-month follow-up period. Fluoroscopic and endoscopic examination in the stented animals showed partial collapse of all the stents by 44.1% k 21.6% of their initial CSA. Over the course of the study, an average number of 2 + 0.7 bronchoscopic dilatations were required to expand the collapsed stent to its previous size. Despite partial collapse of the stents, these animals showed no clinical symptoms of respiratory distress. Pathology The tracheas from the control and balloon dilated groups appeared similar and were characterized grossly by a circumferential stricture with mucopurulent exudate. Histologically, they showed mucosal, muscle, and cartilage necrosis with inward buckling of the tracheal wall that produced the narrowing. Mucosal ulcers were incompletely healed with superficial bacterial invasion, particularly of the exposed necrotic cartilage, accompanied by inflammatory exudate, granulation tissue, and fibrocartilage repair. The stented tracheas had either a widely patent lumen or were narrowed from partial collapse of the stent in the anterior-posterior direction. One- to three-millimeter polyps through the fenestrations of the stent often interfered Table
1. Characteristics
of Experimental
Parameter
Dilation
Number Degree of stricture Respiratory distress
6 >90%
Stricture recurrence Complications Survival
Yes 6 (100%) Pneumonia, 0 (0%)
1
and Control
Groups
stem
Controls
15
10
>90% No 0 (0%)
>90% Yes 10 (100%)
Perforation, Pneumonia, 12 (80%)
1 2 0 (0%)
TSAKAYANNIS
1040
ET AL
with easy stent removal (Fig 2). Tracheas examined histologically after stent removal showed patchy mucosal ulcerations with acute inflammatory exudate and bacterial colonization or superficial invasion. A marked mucosal chronic inflammatory infiltrate was also present. The mucosal polyps were characterized by inflamed mucosa or granulation tissue. Transmural fibrosis also was present (Fig 3). DISCUSSION
Stents have been used as a temporary maneuver for both benign and malignant tracheobronchial obstruction. Expandable intraluminal stenting was developed for the nonoperative management of stenotic vascular disease.9 However, because these stents are increasingly used in extravascular sites, a formal assessment of their utility in such locations is warranted. The use of stents in the management of acute airway stenosis is of particular interest.‘O Different classes of stents have been developed for airway stenosis. Silicon rubber stents have been shown to interfere with mucociliary clearance and have a high rate of migration. Self-expanding metallic stents (Gianturco, Cook Group Europe, Bjaeverskoo, Denmark; Wallstent, Medinvent SA, Lausanne, Switzerland) expand to a fixed diameter.7,12J3 Ulceration and erosion has therefore resulted from excessive intraluminal pressure. Balloonexpandable metallic stents (Palmaz, Strecker; Boston Scientific, Watertown, MA) have been used successfully for acute airway stenoses.7J0J1 These stents are well tolerated in the tracheobronchial tree, allowing for better mucociliary clearance. Migration is minimized by tissue
Fig 2. Gross pathology comparing a control trachea (A) to stented trachea (B). Of note is the inflammatory reaction with granulation tissue growing through the fenestrations of the stent.
Fig 3. Histology of the same specimens shown in Fig 2. Control trachea (A) shows marked narrowing compared with the stented trachea (B). The stent has been removed for processing of the specimen.
ingrowth through the wire mesh. However, these stents may collapse and also initiate an intense inflammatory reaction.14 Experience using expandable metallic stents in children is limited. Filler et al7 reported a series of seven children in whom expandable metallic stents were placed for recurrent stricture after tracheoplasty for congenital tracheal stenosis. Of these patients, three had recurrent airway obstruction after 1 month requiring redilatation. One patient died after recurrence of symptoms and removal of stent. This preliminary series emphasized the utility of expandable metallic stents but also raised questions regarding the long-term behavior and tolerance of the stent during growth. By creating a model of tracheal stenosis in a growing animal, we hoped to better define the usefulness of the Palmaz stent for acute airway obstruction in the pediatric population. The animal model of tracheal stenosis used in this study is a modification of that used by Duff et all5 and Wenig et alI6 in beagles. In our study, the Palmaz stent clearly had a beneficial effect in the immediate symptomatic period. Without intervention, animals became severely stridorous and either died acutely or had to be killed because of respiratory distress. Simple balloon dilatation of the stricture proved of only marginal benefit, providing temporary symptomatic relief and very limited prolongation of survival. All surviving animals in the stented group were symptom-free for the remainder of the study. However, during routine fluoroscopic examinations, par-
EXPANDABLE
STENTS
FOR TRACHEAL
1041
STENOSIS
tial stem collapse was seen, requiring repeated dilatations. Based on our findings, stent collapse refers to the narrowing of the stent without airway compromise. Stricture recurrence refers to narrowing of the airway caused by transmural fibrosis. Thus, collapse may mean that the stents are too pliable to withstand the external forces of neck movement and, to a lesser degree, scar contracture. Furthermore, dilatations are required to accommodate tracheal cross-sectional growth. These stents also pose risks. For example, one animal died of an iatrogenic perforation, stressing the importance of selecting the correct balloon size for dilation. Two animals died of pneumonia despite appropriate treatment with antibiotics, possibly from hindered ciliary clearance of bronchial secretions. Ideally, a stent for the management of acute tracheo-
bronchial stenosis should have several characteristics. It must be safe and easy to place and, if necessary, remove yet avoid unwanted migration. It should be well tolerated by the tissue with minimal inflammatory reaction or interference with mucociliary clearance. It needs to be constructed of such a material to prevent collapse, but not produce excessive pressure on the underlying mucosa. This study describes an easily reproducible model of tracheal stenosis in a growing animal, which can be used to test newer airway stents and perfect their properites before further clinical application.
ACKNOWLEDGMENTS
The authors thank Chrissy Rader, BA, and Michelle Ferretti, BA, for their technical assistance.
REFERENCES 1. Brichon for bronchial
PY, Blanc-Jouvan stenosis after
F, Pison C, et al: Endovascular lung transplantation. Transplant
stents Proc
24:2656-2659,1992 2. Sonett JR, Keenan
RJ, Ferson PF, et al: Endobronchial management of benign, malignant, and lung transplantation airway stenoses. Ann Thorac Surg 59:1417-1422, 1995 3. Tsang V, Williams A, Goldstraw P: Sequential silastic and expandable metal stenting for tracheobronchial strictures. Ann Thorac Surg 53:856-8560, 1992 4. George PJ: Irving JD, Khaghani A, et al: Role of the Gianturco expandable metal stent in the management of tracheobronchial obstmction. Cardiovasc Intervent Radio1 15:375-381, 1992 5. Higgins R, Mcneil K, Dennis C, et al: Airway stenoses after lung transplantation: Management with expandable metal stems. J Heart Lung Transplant 13:774-778. 1994 6. Shah R, Sabanathan S, Mearns AJ, et al: Self-expanding metallic stents in the management of major airway problems. J Cardiovasc Surg 36:343-348,1995 7. Filler RM, Forte V, Fraga JC, et al: The use of expandable metallic airway stents for tracheobronchial obstruction in children. J Pediatr Surg 30:1050-1056, 1995 8. Bugmann P. Rouge JC, Bemer M, et al: Use of Gianturco 2 stents in the treatment of vascular compression of the tracheobronchial tree in childhood, a feasible solution when surgery fails. Chest 106: 1580- 1582, 1994
9. Vorwerk D, Gunther RW: Stent placement in iliac arterial lesions: Three years of clinical experience with the wallstent. Cardiovasc Intervent Radio1 15:285-290, 1992 10. Zollikofer CL, Antonucci F, Stuckmann G: Historical overview on the development and characteristics of stents and future outlooks. Cardiovasc Intervent Radio1 15:272-278, 1992 11. Palmaz JC: Intravascular stenting: From basic research to clinical application. Cardiovasc Intervent Radio1 15:279-284, 1992 12. Mair EA, Parsons DS, Lally Kp, et al: Comparison of expandable endotracheal stents in the treatment of surgically induced piglet tracheomalacia. Laryngoscope lOl:lOOl-1007, 1991 13. Gaer JAR, Tsang V, Khaghani A, et al: Use of endotracheal silicone stents for relief of tracheobroncial obstruction. Ann Thorac Surg 54:512-516, 1992 14. Marquette CH, Mensier E, Copin M-C, et al: Experimental models of tracheobronchial stenoses: A useful tool for evaluating airway stents. Ann Thorac Surg 60:651-656, 1995 15. Duff BE, Wenig BL, Applebaum EL, et al: Tracheal reconstruction using an epithelial equivalent. Laryngoscope 104:409-414, 1994 16. Wenig BL, Reuter VC, Steinberg BM, et al: Tracheal reconstmction: In vitro and in vivo animal pilot study. Laryngoscope 97:959-965, 1987
Discussion R.G. Azizkhan (Buffalo, NY): This is a very interesting and exciting paper. Your lamb model recapitulates the findings that we have seen in human patients with tracheal stenosis that have undergone tracheal stenting. We have also found, in infants, that in most cases the stent does partially collapse during the recovery period, and these patients require repeated endoscopy to remove granulation tissue, which does go through the interstices as you have very nicely demonstrated in your animal model. Also, the increasing fibrosis does seem to have an impact on collapsing the stent.
Do you feel that the collapse of the stents in the animal model are related to the unique neck motions of the animal in that these stents are extremely fragile, it doesn’t take very much to collapse them? Or is it related primarily to the fibrosis that you have seenin your model? Have you looked at the impact of the stents in the tracheas that have not had any type of injury? Do you still see the same intense inflammatory reaction? If not, what is the impact of this in terms of removing the stents after the placement? Because we have had trouble removing the stents in some of the patients that have had tracheal
1042
stenosis and placement of the stents for a prolonged period. The danger is tearing the trachea in the process of removing the stents. So with these questions in mind, this is a very good model to look at other types of stent material such as absorbable stents, which may have a much more longterm efficacy in this type of difficult problem. Thank you for the opportunity to comment. D.E. Tsakayannis (response): Thank you very much for your kind remarks. On the first question, I think that part of the collapse was caused by the fibrous tissue, but mainly the mechanical properties of the stent. The growth of about 8% a week is also significant. We may need to look at newer types of stents such as those made of nitinol. This nickel-titanium compound is more rigid, and may thereby improve outcome. Regarding your second question, we did not actually have a group with normal tracheas. I think this is an excellent remark, and we should be looking at that. In regard to your third question, I think that with this model we can definitely use a newer type of stent, or even modify a particular stent for the trachea. Since these stents were made for blood vessels, it may be that for the trachea different tensile strengths and mechanical properties will be required. In the meantime, we can use those things that are available but are looking at different stents for the future. A. Kosloske (Lubbock, TX): I rise with a note of caution. About 2 years ago a pediatric surgical colleague of mine inserted a Palmaz stent into the left main stem bronchus of a retarded boy who had severe bronchial malacia, successfully reopening the airway. This patient experienced a fatal complication of the stent. Approximately 2 months later he had a massive episode of hemoptysis and at autopsy he had formed a bronchial aortic fistula with a stent eroded through the wall of the bronchus.
TSAKAYANNIS
ET AL
So I must caution the audience against the clinical use of these stents in any part of the airway that pulsates immediately adjacent to a pulsatile vessel. The stents may be safe in a relatively quiet part of the airway, but much more experimental work needs to be done. D.E. Tsakayannis (response): I agree 100% with your statement. I think it is very interesting to note that stents were first used clinically, but now we are looking back to the experimental evidence. I think you make a very solid point on how careful we should be in the future with stents. E. Doolin (Camden, NJ): The Palmaz stent is convenient and easy to use. That is why it is popular, especially for tracheal and bronchial malacia. However, what makes it expand can make it collapse. And you have proven that. Do you think you have found in the inflammatory stenosis a subset that deserve a more resilient stent like a Dumont stent? In your model, which doesn’t exist much in nature, and the subglottic equivalent studied last year, the burn of the airway creates enough regulation of the contracting type III collagen and the down-regulation of the mature type I collagen. Have you studied the biochemical components of your scar to see if they are dissimilar to what we see in nature and therefore make them more at risk for restriction? D.E. Tsakayannis (response): Thank you for your questions. In regard to your first question, I have not used the Dumont stent because, as you said, the PaImaz is much more easily accessible. Given that Dumont is a silicon stent, it may have a higher migration rate. With regard to your second question, we have not identified the specific collagen type, but I think that, as you pointed out, it may not just represent re-stricture, but may have an element of tracheomalacia as well.