- Email: [email protected]
A new intratracheal stent made from nitinol, an alloy with “shape memory effect”
A new intratracheal stent made from nitinol, an alloy with "shape memory effect" Temporary or permanent tracheal splinting in pediatric patients may be indicated in tracheomalacia or bronchomalacia, repair of congenital tracheal stenosis, and after tracheal resection. This study presents the results of the development of a new intraluminal airway stent made from titanium alloy, a metal with "shape memory effect." At low temperatures (martensitic state) the titanium alloy stent can be fashioned into a specific shape; then when heated to a higher temperature (austenitic state) the stent alters its shape, only to regain its original shape when recooled to the lower temperature. The stent, connected to a small electric power supply, was introduced into 20 young rabbits with the use of a 2.5 em rigid bronchoscope. After implantation in the martensitic state the stent was warmed to 40° C, the austenitic state, by an electric current of 1.5 to 3 ampere for 1 to 2 seconds. After a period of 8 to 10 weeks the stent was removed (in its martensitic state) through the same-sized bronchoscope after being cooled with 3 to 4 ml of 80 % alcohol solution at 6° C. No signs of airway obstruction developed in any of the animals after implantation or extraction of the stent. The biomechanical properties of the trachea, as shown by strain measurements with the use of incremental forces, showed significant differences between the stented and unstented segments (p < 0.005). The titanium alloy intratracheal stent adequately fulfilled the requirements of a temporary intraluminal airway splint, and because of its unique feature of shape memory effect the stent could be inserted, fixed, and removed easily, even in very small airways. (J 'fHORAC CARDIOVASC SURG 1994;107:1255-61)
Itzhak Vinograd, MD,a Baruch Klin, MD,a Tamar Brosh, PhD,b Mark Weinberg, MD, PhD,a Yosef Flomenblit, Phl)," and Zvi Nevo, Phl),? Zerifin and Tel Aviv, Israel
Lacheal operations in infants and children for the treatment of airway stenosis and tracheobronchomalacia frequently require a luminal stent. External tracheal splints made of synthetic materials or autologous rib struts can prevent airway collapse.' However, those grafts are not always successful and their implantation is complicated and permanent. Intratracheal stents made from From the Department of Pediatric Surgery; the Laboratory of Surgical Research. Assaf Harofeh Medical Center, Zerifin; The Maurice and Gabriela Goldschleger School of Dental Medicine"; and the Department of Chemical Pathology," The Sackler Faculty of Medicine, Tel Aviv University, Israel. Supported by a grant from The Sander Research Fund, Tel Aviv University. Received for publication June 28, 1993. Accepted for publication Sept. 21,1993. Address for reprints: 1. Vinograd, MD, Head, Department of Pediatric Surgery, Assaf Harofeh Medical Center, Zerifin 70300, Israel. Copyright © 1994 by Mosby-Year Book, Inc. 0022-5223/94 $3.00
+ 0 12/1/51762
stainless steel or other materials have been used with limited success.v 3 Because in infants and children the narrowest section of the trachea is at the level of the cricoid ring and airway lumina are very small, the development, design, and production of an intraluminal stent is complicated. In a search to define the optimal intraluminal stent, Loeff and associates- proposed the following requirements: (1) simplicity of insertion, fixation, and removal; (2) biocompatibility; (3) no obstruction of airway tributaries; (4) improved clearance of secretions; and (5) accommodation to varying tracheal dimensions. The shape memory phenomenon (Marmen effect) was discovered in nitinol, nickel-titanium alloy, at the U.S. Naval Ordnance Laboratory." The effect occurs when certain metals apparently "plastically" deform at a lower temperature, then revert to their original shape when heated to a somewhat higher temperature. The "shape memory effect" has been studied by a number of workers and is also found in various alloy systems such as Au-Cd, Cu-AI-Ni, and others. All those materials share the common property of exhibiting martensitic transformation, 1 255
1256
The Journal of Thoracic and Cardiovascular Surgery May 1994
Vinograd et al.
A • Austenitic State M • Martensltlc State
Ext. Dlam. (mm) 10
8
A
6
M 2
o
L-
_
6
20
40
Temp. (c)
Fig. 1. Martensitic transformation of nitinol intratracheal stent. Shape memory effect occurred between martensitic state at 6° C and austenitic state at 40° C.
and accordingly the shape memory effect has been variously related to the martensitic phase change: changes in the crystalloid structure that are strictly thermodependent. The manner in which the memory effect is related to the martensitic transformation is not clear and numerous divergent views and mechanisms have been proposed. Nickel titanium is considered to be a prototype alloy for the Marmen effect. The high biocompatibility and excellent tissue tolerance of this metal are established facts. 5 The current study presents the results of the development of a new intraluminal airway stent made from titanium alloy. The stent, which has a specially designed coil shape, fulfilled the specific requirements of a temporary intraluminal airway splint. Materials and methods
The stents were constructed in a metallurgic laboratory from nitinol wire. So that integration of the stent into the tracheal mucosa would not occur, the stent was made from a flat sheet of nitinol about 1.8 mm thick and 0.18 mm wide. At room temperature the flat sheet was formed into a spiral shape and "Onnected by its two edges to 0.3 mm electric wires. At 4° , ) 6° C (the martensitic state), the stent was compacted t ) an external diameter of 2.2 mm and a length of 4.2 c n. At 40° C (the austenitic
state), the stent was transformed to an external diameter of 7.2 mm and a length of 3.3 em (Fig. 1). The length of the stent and its diameter could be altered to fit various tracheal dimensions. The open spiral design permitted endobronchial or tracheal placement with minimal impingement and pressure on the respiratory epithelium and without obstruction of main or lobar bronchi. The experimental protocols were reviewed and approved by the Animal Experimentation Committee of the Laboratory of Surgical Research, The Assaf Harofeh Medical Center. The intraluminal tracheal and bronchial stents were studied in 20 young rabbits (1.8 to 2.1 kg). The rabbit model was chosen because its tracheal size approximates that of an infant weighing 3 to 5 kg. Each of the 20 rabbits was premedicated with atropine sulfate (0.03 mg/kg), and cefazolin (50 mg/kg) was given intramuscularly 1/2 hour before the procedure. The animal was anesthetized with 5 rug/kg xylazine hydrochloride (Rompun, Bayer, Leverkusen, Germany) and 0.35 rug/kg ketamin intramuscularly. Implantation of the intraluminal stent was accomplished through a pediatric bronchoscope (Karl Storz GmbH & Co., Tuttlingen, Germany) with an external diameter of 2.5 em and a length of 20 em. The stent was implanted with a special stent introducer that was guided into the trachea through
The Journal of Thoracic and Cardiovascular Surgery Volume 107, Number 5
Vinograd et al.
I 257
~
+-
A
c
B
o
Fig. 2. Schematic illustration of techniqueof implantation (A and B) and removal (C and D) of nitinolintratracheal stent. A, Stent is introduced into trachea in martensitic state. B, with electriccurrent of 3 to 5 volts, stent is converted to austenitic state. C and D, By injection of 3 to 5 ml of cooled 80% alcohol solution, stent is extracted in its martensitic state with small grasper. a, Trachea; b, bronchoscope; c, nitinol intratracheal stent.
the rigid bronchoscope. The introducer was connected to a small portable electric power supply, with 0.3 mm diameter wires attached to the stent. With the use of an electric current of 3 to 5 volts, 1.5 to 3 amperes, for I to 2 seconds, the nitinol stent was warmed to 40 0 C, thereby converting it to its austenitic state wherein it was fully expanded (Fig. 2). At that point, the introducer was disconnected from the stent by a gentle twisting movement and removed. An external energy source was required so as to achieve instant expansion of the stent and thereby prevent its migration from the designated point of implantation in the airway lumen. Bronchoscopy was again done to verify the position and full expansion of the stent. In all animals the bronchoscope was extracted on stent implantation and before recovery from anesthesia. Stents were inserted into 20 animals: in 16 the stent was placed in the trachea (in 4 at the level of the carina) and in 4 animals it was placed in a main bronchus. Subsequently, the intraluminal stent was extracted with the use of the same-sized bronchoscope. A small grasper was applied to the proximal edge of the stent. At the level of the stent 2 to 3 ml of 80% alcohol solution at 6 0 C was injected through a thin plastic tube that had
been previously cooled to 4 0 to 6 0 C by nitrous oxide vapor. The stent was transformed to its martensitic state and removed from the trachea through the bronchoscope as an almost flat sheet. The animals were monitored daily for signs of respiratory distress or failure to thrive. They were weighed and evaluated with chest radiographs and bronchoscopy, then killed after 8 to 10 weeks by means of a barbiturate overdose. Postmortem examination of the upper airway, tracheobronchial tree, and lungs was done in all animals. The animals were divided into two groups: In group A (10 animals) the stent was removed in vivo and the animal subsequently killed; in group B (8 animals) the animals were killed with the intraluminal stent in place. In both groups the entire trachea was extracted and then divided into two segments: (I) normal trachea or bronchus and (2) poststented (group A) or with stent in situ (group B). In group A, eight tracheal and two bronchial segments were fixed in formalin and transverse histologic sections prepared from both segments. The sections were stained with hematoxylin-eosin for light microscopy. The histologic slides of the trachea were also used for morphometric measurement of the tracheal diameters with the IBAS
12 58
The Journal of Thoracic and Cardiovascular Surgery May 1994
Vinograd et al.
A
B
Fig. 3. Plain radiograph showing expanded nitinol intratracheal stent in trachea 6 weeksafter bronchoscopicimplantation.
image analyzer (Karl Zeiss, Inc., Thornwood, N.Y.). The differences between internal cross-sectional areas of the normal and the splinted trachea were evaluated by a paired t test. In group B (8 animals) two tracheal segments of 3.5 mm in length were evaluated for biomechanical properties. The trachea was placed on a warm plate at 37° C and tracheal deformation was measured with a dial gauge (Mitutowo, Tokyo, Japan). The deformation was evaluated as the difference between Do (external diameter of the segment before deformation) and D] (external diameter after deformation) after the application of a force (in 10 gm increments) perpendicularly to the long axis of the trachea. Tracheal strain was determined as follows: .
Do-D 1
Stram=~x
100
and expressed as percent strain. A paired t test was done to assess significant differences. Tracheal segment strains with the same loads were done after in vitro removal of the stents in five animals of group B.
Fig. 4. Histopathologic view of trachea after extraction of stent. A, Trachea is patent with no evidence of foreign body inflammatory reaction. B, Submucosal mucous gland layer was significantly thinner in 4 of 10 animals.
Results
The animals were ambulatory and eating within 2 to 4 hours of the procedure. No signs of airway obstruction developed in any of the animals at any stage after implantation or extraction of the stent. Two rabbits died unexpectedly 2 and 4 weeks after the procedure. In both animals the trachea was patent with no evidence of obstruction or pulmonary infection and both animals were excluded from the study. Stents were well-tolerated in the animals as evidenced by (1) continuous mean weekly weight gain of 180 gm (range 160 to 210 gm/week), (2) minimal signs or symptoms of respiratory distress, and (3) no deaths. There was no migration of the stent within the trachea or bronchus, which was confirmed by repeated bronchoscopy and chest radiographs in the second, fourth, and sixth postoperative weeks (Fig. 3). Bronchoscopic examinations showed increased secretions in most animals, particularly in the first 2 weeks
The Journal of Thoracic and Cardiovascular Surgery Volume 107, Number 5
after implantation, but the secretions diminished significantly in the ensuing weeks. No evidence was found of gross inflammation or granulation formation of the mucosa. In group A, histologic studies of eight tracheal segments and 2 main bronchi after extraction of the stent and of another 10 segments of normal airway showed no inflammatory response of the mucosa. The cartilage rings showed no evidence of change or damage. In four stented segments, the submucosal mucous gland layer was found to be significantly thinner than in the normal segments of the same trachea (Fig. 4). Measurement of the tracheal cross-sectional area showed no significant difference: in the stented segment it was 0.197 ± 0.021 cm 2 as compared with 0.188 ± 0.027 crrr' in the normal trachea. In group B, study of the biomechanical properties of 12 tracheal segments and two segments from the main bronchi, with and without stents in eight animals, showed that with the same applied force (100 gm) the mean deformation of the trachea with and without the stent was 17% ± 9% and 37% ± 13%, respectively; p < 0.005 (Fig. 5). After the in vitro extraction of the stent from the trachea, the deformation after the application of an applied force of 100 gm was 37% in the normal trachea and 35% in the segment from which the stent had been removed (Fig. 6): this difference was not statistically significant.
Discussion Some of the most common indications for temporary or permanent tracheal splinting in pediatric patients include tracheomalacia or bronchomalacia, repair of congenital tracheal stenosis, and repair of the defect left by tracheal resection. Tracheobronchomalacia results in intrathoracic airway collapse with obstruction of the lumen during expiration. A flaccid tracheobronchial tree in children with this condition is incompatible with normal respiratory function, resulting in recurrent pneumonia, hypoxic "dying spells," and inability to be extubated," Symptoms depend on the location and length of the abnormal airway segment and the severity of the structural abnormality. Severe cases of airway collapse unresponsive to aggressive medical management require surgical intervention, which may include aortopexy/' tracheopexy,? external splinting/' reconstruction, or even pneumonectomy." The various surgical procedures may prevent tracheal collapse; however, the bronchi are difficult to treat surgically. The surgical techniques always require thoracotomy and in the more severe cases implantation of grafts and foreign materials, which carry with them the risk of rejection, infection, and high morbidity. After tracheal resection and end-to-end anastomosis
Vinograd et al.
12 5 9
(particularly in tracheas with small lumina) and repair of congenital stenosis with pericardial or periosteal grafts, different intrinsic mural or external supportive temporary stents have been used to maintain the patency of the reconstructed airway.l'vII Recently, several innovative implantable devices have been described to prevent collapse or stenosis of the trachea and other organs.'? Wallace and associates'< adapted an intravascular stent made of expandable wire for use within the tracheobronchial tree. Loeff and associates- used stainless steel wire coils coated with a thin silicone rubber layer. The stent was implanted with the use of either an open surgical technique or a balloon-type introducer mounted with a spring for endoscopic placement. In experimental and clinical experience with the use of various intratracheal stents, difficulties were encountered with endoscopic placement and removal, there were problems with tracheal secretions, and unexpected histologic changes of squamous metaplasia and inflammation in the respiratory epithelium and the submucosa occurred.b 13 Interest in the use of titanium for intratracheal stents was aroused because of two basic qualities of the alloy: (1) its proven highly noncorrosive nature, biocompatibility, and good tissue tolerance and (2) its unique phenomenon of shape memory effect that provides an unprecedented possibility for endoscopic placement and removal, even through the lumen of a very small airway. One of the main disadvantages of the intratracheal stents used hitherto was their lack of biocompatibility.v 3 Studies indicated that stainless steel may be highly corrosive in the presence of chloride ions in the body fluids. When the metal implants release corrosive metal ions into the surrounding tissues, the body may respond with a number of adverse reactions, such as chronic inflammation, pain, discolora. tion, hypersensitivity, fibrosis, necrosis, and in some instances tumors.lv 15 Although the reactions are generally localized and in the majority not severe, these possible effects suggested that other materials should be considered for implantation. Experimental and clinical use of titanium urethral stents showed no incorporation of the stent into the urethral wall, and the inflammatory reaction was minimal. 16 In the present study the inflammatory reaction to nitinol intratracheal stents was minimal. The slight shrinking of the mucous glands in the submucosa layer observed in 4 of 10 rabbits could be attributed to overexpansion of the stent. In the human being this problem can be solved by more precise determination of the tracheal internal lumen by computed tomographic and bronchoscopic examinations. The nitinol intratracheal stents can be produced with external dimensions that are tailored to fit an individual trachea or bronchus without any difficulty, by lim-
1260
The Journal of Thoracic and Cardiovascular Surgery May 1994
Vinograd et al.
_
• p
c
Normal trachea
N·a
Eill Trach. with N·a
NITI-ITS
0.008
10
o
eo
100
200
180
ApplIeCl
'oroe (or)
Fig. 5. Study of biomechanical propertiesof trachea with and without intratracheal stent (NiTi-ITS).
_ 8tr.1nA""(:.... ..:....'
Normal trachea
N·S
I:23J AI. remov. NITI-ITS N·S
------
50 40 30 20 10
o
eo
100
150 Applied 'oree (lIr'
Fig. 6. Study of mechanical propertiesof trachea after in vitro removal of stent (NiTi-ITS).
iting its expansion in the austenitic state at 40 0 C when constructed in the laboratory. Because of the shape memory effect, once converted into the martensitic state in the trachea, the stent will retain the same external diameter. At that point, additional expansion is very small and exclusively thermic dependent; overexpansion is unlikely inasmuch as body temperature does not normally rise higher than 41 0 C. This characteristic is very important because after expansion the nitinol stent does not increase its pressure on the tracheal wall, in contrast to the expandable metallic stent with which there is constant elastic pressure on the trachea. One of the main functions of the tracheal stent is to
increase tracheal wall rigidity. During normal respiration, the intrathoracic pressure ranges from -5 em H 20 on inspiration to + 30 em H 20 and much higher during coughing. Collapse is prevented by the rigidity of the airway wall. In tracheomalacia, however, the tracheal wall is less rigid than normal and tends to collapse, producing airway obstruction. Our current preliminary experience showed that the nitinol stent significantly increased the rigidity of the tracheal wall. The resistance to external pressure was significantly greater for the stented trachea, as compared with that for a normal segment (p < 0.005). An applied force of 50 gm to the segment with the nitinol stent created a deformation of only 50%. After removal
The Journal of Thoracic and Cardiovascular Surgery Volume 107, Number 5
of the stent the mechanical properties of the trachea were not affected. The stented and normal tracheal segments showed equal strain. This study showed that the nitinol intratracheal stent fulfilled the five basic requirements of an intratracheal stent proposed by Loeff and associates.': (I) by taking advantage of its unique feature of shape memory effect, the stent can be inserted, fixed, and removed easily, even in a very small airway; (2) the stent proved its biocompatibility, and inflammatory and adverse reactions in the implanted trachea were minimal: (3 and 4) because of its coil-shape construction and because it is made from a very thin flat wire sheet, obstruction of air tributaries was minimal and pressure on a large surface area of the respiratory epithelium was avoided; and (5) the length of the stent could be adjusted and the diameter varied to fit any tracheal or bronchial dimensions. In infants and children with very narrow airways, the nitinol intratracheal stent may be used as a supplementary method of achieving temporary airway splinting. REFERENCES I. Filler RM, Buck JR, Bahoric A, et a1. Treatment of
segmental tracheornalacia and bronchomalacia by implantation of an airway splint. J Pediatr Surg 1982;17:597-603. 2. Johnston MR, Lober N, Hillyer P, et a1. External stent for repair of secondary bronchornalacia. Ann Thorac Surg 1980;30:291-6. 3. Loeff DS, Filler RM, Gorenstein A, et a1. A new tracheobronchial reconstruction: experimental and clinical studies. J Pediatr Surg 1988;23:1113-7. 4. Wayman CM, Shimizu K. The shape memory ("Marmen") effect in alloys. Metal Sci J 1972;6:175-83.
Vinograd et al.
I 26 I
5. Williams DF. Titanium as a metal for implantation. Part 2. Biological properties and clinical applications. Biomed Eng 1977;1:266-9. 6. Blair GK, Cohen R, Filler RM. Treatment of tracheomalacia: eight years' experience. J Pediatr Surg 1986;21: 781-5. 7. Cohen D. Tracheopexy: aorta-tracheal suspension for severe tracheomalacia. Aust Pediatr J 1981;17:117-21. 8. Vinograd I, Filler RM, Bahoric A. Long term functional results of prosthetic airway splinting in tracheomalacia and bronchomalacia. J Pediatr Surg 1987;22:38-41. 9. Gupta TGCM, Goldberg SJ, Lewis E, Fonkalsrud EW. Congenital bronchomalacia. Am J Dis Child 1968;115:8890. 10. Loeff DS, Filler RM, Vinograd I, et a1. Congenital tracheal stenosis: a review of 22 patients from 1965 to 1987. J Pediatr Surg 1988;23:744-8. II. Idriss FS, Serafin YD, Ilbawi MN, et a1. Tracheoplasty with pericardial patch for extensive tracheal stenosis in implants and children. J THORAC CARDIOVASC SURG 1984;88:527-36. 12. Mair EA, Parsons DS, Lolly KP. Treatment of severe bronchomalacia with expanding endobronchial stents. Arch Otolaryngol Head Neck Surg 1990;116:1087-90. 13. Wallace JM, Charnsangavej L, Ogawa K, et a1. Tracheobronchial tree: expandable metallic stents used in experimental and clinical applications. Radiology 1986;158:30912. 14. Harris B. Corrosion of stainless stet" surgical implants. J Med Eng TechnoI1979;3:117-22. 15. Sunderman FW Jr. Carcinoger .city of metal alloys in orthopedic prostheses: clinical snd experimental studies. Fundam Appl Toxicol 1989;13:..05-11. 16. Parra RO. Treatment of posteri, r urethral strictures with titanium urethral stent. J Urol I: 91;146:997-1000.
Itzhak Vinograd, MD,a Baruch Klin, MD,a Tamar Brosh, PhD,b Mark Weinberg, MD, PhD,a Yosef Flomenblit, Phl)," and Zvi Nevo, Phl),? Zerifin and Tel Aviv, Israel
Lacheal operations in infants and children for the treatment of airway stenosis and tracheobronchomalacia frequently require a luminal stent. External tracheal splints made of synthetic materials or autologous rib struts can prevent airway collapse.' However, those grafts are not always successful and their implantation is complicated and permanent. Intratracheal stents made from From the Department of Pediatric Surgery; the Laboratory of Surgical Research. Assaf Harofeh Medical Center, Zerifin; The Maurice and Gabriela Goldschleger School of Dental Medicine"; and the Department of Chemical Pathology," The Sackler Faculty of Medicine, Tel Aviv University, Israel. Supported by a grant from The Sander Research Fund, Tel Aviv University. Received for publication June 28, 1993. Accepted for publication Sept. 21,1993. Address for reprints: 1. Vinograd, MD, Head, Department of Pediatric Surgery, Assaf Harofeh Medical Center, Zerifin 70300, Israel. Copyright © 1994 by Mosby-Year Book, Inc. 0022-5223/94 $3.00
+ 0 12/1/51762
stainless steel or other materials have been used with limited success.v 3 Because in infants and children the narrowest section of the trachea is at the level of the cricoid ring and airway lumina are very small, the development, design, and production of an intraluminal stent is complicated. In a search to define the optimal intraluminal stent, Loeff and associates- proposed the following requirements: (1) simplicity of insertion, fixation, and removal; (2) biocompatibility; (3) no obstruction of airway tributaries; (4) improved clearance of secretions; and (5) accommodation to varying tracheal dimensions. The shape memory phenomenon (Marmen effect) was discovered in nitinol, nickel-titanium alloy, at the U.S. Naval Ordnance Laboratory." The effect occurs when certain metals apparently "plastically" deform at a lower temperature, then revert to their original shape when heated to a somewhat higher temperature. The "shape memory effect" has been studied by a number of workers and is also found in various alloy systems such as Au-Cd, Cu-AI-Ni, and others. All those materials share the common property of exhibiting martensitic transformation, 1 255
1256
The Journal of Thoracic and Cardiovascular Surgery May 1994
Vinograd et al.
A • Austenitic State M • Martensltlc State
Ext. Dlam. (mm) 10
8
A
6
M 2
o
L-
_
6
20
40
Temp. (c)
Fig. 1. Martensitic transformation of nitinol intratracheal stent. Shape memory effect occurred between martensitic state at 6° C and austenitic state at 40° C.
and accordingly the shape memory effect has been variously related to the martensitic phase change: changes in the crystalloid structure that are strictly thermodependent. The manner in which the memory effect is related to the martensitic transformation is not clear and numerous divergent views and mechanisms have been proposed. Nickel titanium is considered to be a prototype alloy for the Marmen effect. The high biocompatibility and excellent tissue tolerance of this metal are established facts. 5 The current study presents the results of the development of a new intraluminal airway stent made from titanium alloy. The stent, which has a specially designed coil shape, fulfilled the specific requirements of a temporary intraluminal airway splint. Materials and methods
The stents were constructed in a metallurgic laboratory from nitinol wire. So that integration of the stent into the tracheal mucosa would not occur, the stent was made from a flat sheet of nitinol about 1.8 mm thick and 0.18 mm wide. At room temperature the flat sheet was formed into a spiral shape and "Onnected by its two edges to 0.3 mm electric wires. At 4° , ) 6° C (the martensitic state), the stent was compacted t ) an external diameter of 2.2 mm and a length of 4.2 c n. At 40° C (the austenitic
state), the stent was transformed to an external diameter of 7.2 mm and a length of 3.3 em (Fig. 1). The length of the stent and its diameter could be altered to fit various tracheal dimensions. The open spiral design permitted endobronchial or tracheal placement with minimal impingement and pressure on the respiratory epithelium and without obstruction of main or lobar bronchi. The experimental protocols were reviewed and approved by the Animal Experimentation Committee of the Laboratory of Surgical Research, The Assaf Harofeh Medical Center. The intraluminal tracheal and bronchial stents were studied in 20 young rabbits (1.8 to 2.1 kg). The rabbit model was chosen because its tracheal size approximates that of an infant weighing 3 to 5 kg. Each of the 20 rabbits was premedicated with atropine sulfate (0.03 mg/kg), and cefazolin (50 mg/kg) was given intramuscularly 1/2 hour before the procedure. The animal was anesthetized with 5 rug/kg xylazine hydrochloride (Rompun, Bayer, Leverkusen, Germany) and 0.35 rug/kg ketamin intramuscularly. Implantation of the intraluminal stent was accomplished through a pediatric bronchoscope (Karl Storz GmbH & Co., Tuttlingen, Germany) with an external diameter of 2.5 em and a length of 20 em. The stent was implanted with a special stent introducer that was guided into the trachea through
The Journal of Thoracic and Cardiovascular Surgery Volume 107, Number 5
Vinograd et al.
I 257
~
+-
A
c
B
o
Fig. 2. Schematic illustration of techniqueof implantation (A and B) and removal (C and D) of nitinolintratracheal stent. A, Stent is introduced into trachea in martensitic state. B, with electriccurrent of 3 to 5 volts, stent is converted to austenitic state. C and D, By injection of 3 to 5 ml of cooled 80% alcohol solution, stent is extracted in its martensitic state with small grasper. a, Trachea; b, bronchoscope; c, nitinol intratracheal stent.
the rigid bronchoscope. The introducer was connected to a small portable electric power supply, with 0.3 mm diameter wires attached to the stent. With the use of an electric current of 3 to 5 volts, 1.5 to 3 amperes, for I to 2 seconds, the nitinol stent was warmed to 40 0 C, thereby converting it to its austenitic state wherein it was fully expanded (Fig. 2). At that point, the introducer was disconnected from the stent by a gentle twisting movement and removed. An external energy source was required so as to achieve instant expansion of the stent and thereby prevent its migration from the designated point of implantation in the airway lumen. Bronchoscopy was again done to verify the position and full expansion of the stent. In all animals the bronchoscope was extracted on stent implantation and before recovery from anesthesia. Stents were inserted into 20 animals: in 16 the stent was placed in the trachea (in 4 at the level of the carina) and in 4 animals it was placed in a main bronchus. Subsequently, the intraluminal stent was extracted with the use of the same-sized bronchoscope. A small grasper was applied to the proximal edge of the stent. At the level of the stent 2 to 3 ml of 80% alcohol solution at 6 0 C was injected through a thin plastic tube that had
been previously cooled to 4 0 to 6 0 C by nitrous oxide vapor. The stent was transformed to its martensitic state and removed from the trachea through the bronchoscope as an almost flat sheet. The animals were monitored daily for signs of respiratory distress or failure to thrive. They were weighed and evaluated with chest radiographs and bronchoscopy, then killed after 8 to 10 weeks by means of a barbiturate overdose. Postmortem examination of the upper airway, tracheobronchial tree, and lungs was done in all animals. The animals were divided into two groups: In group A (10 animals) the stent was removed in vivo and the animal subsequently killed; in group B (8 animals) the animals were killed with the intraluminal stent in place. In both groups the entire trachea was extracted and then divided into two segments: (I) normal trachea or bronchus and (2) poststented (group A) or with stent in situ (group B). In group A, eight tracheal and two bronchial segments were fixed in formalin and transverse histologic sections prepared from both segments. The sections were stained with hematoxylin-eosin for light microscopy. The histologic slides of the trachea were also used for morphometric measurement of the tracheal diameters with the IBAS
12 58
The Journal of Thoracic and Cardiovascular Surgery May 1994
Vinograd et al.
A
B
Fig. 3. Plain radiograph showing expanded nitinol intratracheal stent in trachea 6 weeksafter bronchoscopicimplantation.
image analyzer (Karl Zeiss, Inc., Thornwood, N.Y.). The differences between internal cross-sectional areas of the normal and the splinted trachea were evaluated by a paired t test. In group B (8 animals) two tracheal segments of 3.5 mm in length were evaluated for biomechanical properties. The trachea was placed on a warm plate at 37° C and tracheal deformation was measured with a dial gauge (Mitutowo, Tokyo, Japan). The deformation was evaluated as the difference between Do (external diameter of the segment before deformation) and D] (external diameter after deformation) after the application of a force (in 10 gm increments) perpendicularly to the long axis of the trachea. Tracheal strain was determined as follows: .
Do-D 1
Stram=~x
100
and expressed as percent strain. A paired t test was done to assess significant differences. Tracheal segment strains with the same loads were done after in vitro removal of the stents in five animals of group B.
Fig. 4. Histopathologic view of trachea after extraction of stent. A, Trachea is patent with no evidence of foreign body inflammatory reaction. B, Submucosal mucous gland layer was significantly thinner in 4 of 10 animals.
Results
The animals were ambulatory and eating within 2 to 4 hours of the procedure. No signs of airway obstruction developed in any of the animals at any stage after implantation or extraction of the stent. Two rabbits died unexpectedly 2 and 4 weeks after the procedure. In both animals the trachea was patent with no evidence of obstruction or pulmonary infection and both animals were excluded from the study. Stents were well-tolerated in the animals as evidenced by (1) continuous mean weekly weight gain of 180 gm (range 160 to 210 gm/week), (2) minimal signs or symptoms of respiratory distress, and (3) no deaths. There was no migration of the stent within the trachea or bronchus, which was confirmed by repeated bronchoscopy and chest radiographs in the second, fourth, and sixth postoperative weeks (Fig. 3). Bronchoscopic examinations showed increased secretions in most animals, particularly in the first 2 weeks
The Journal of Thoracic and Cardiovascular Surgery Volume 107, Number 5
after implantation, but the secretions diminished significantly in the ensuing weeks. No evidence was found of gross inflammation or granulation formation of the mucosa. In group A, histologic studies of eight tracheal segments and 2 main bronchi after extraction of the stent and of another 10 segments of normal airway showed no inflammatory response of the mucosa. The cartilage rings showed no evidence of change or damage. In four stented segments, the submucosal mucous gland layer was found to be significantly thinner than in the normal segments of the same trachea (Fig. 4). Measurement of the tracheal cross-sectional area showed no significant difference: in the stented segment it was 0.197 ± 0.021 cm 2 as compared with 0.188 ± 0.027 crrr' in the normal trachea. In group B, study of the biomechanical properties of 12 tracheal segments and two segments from the main bronchi, with and without stents in eight animals, showed that with the same applied force (100 gm) the mean deformation of the trachea with and without the stent was 17% ± 9% and 37% ± 13%, respectively; p < 0.005 (Fig. 5). After the in vitro extraction of the stent from the trachea, the deformation after the application of an applied force of 100 gm was 37% in the normal trachea and 35% in the segment from which the stent had been removed (Fig. 6): this difference was not statistically significant.
Discussion Some of the most common indications for temporary or permanent tracheal splinting in pediatric patients include tracheomalacia or bronchomalacia, repair of congenital tracheal stenosis, and repair of the defect left by tracheal resection. Tracheobronchomalacia results in intrathoracic airway collapse with obstruction of the lumen during expiration. A flaccid tracheobronchial tree in children with this condition is incompatible with normal respiratory function, resulting in recurrent pneumonia, hypoxic "dying spells," and inability to be extubated," Symptoms depend on the location and length of the abnormal airway segment and the severity of the structural abnormality. Severe cases of airway collapse unresponsive to aggressive medical management require surgical intervention, which may include aortopexy/' tracheopexy,? external splinting/' reconstruction, or even pneumonectomy." The various surgical procedures may prevent tracheal collapse; however, the bronchi are difficult to treat surgically. The surgical techniques always require thoracotomy and in the more severe cases implantation of grafts and foreign materials, which carry with them the risk of rejection, infection, and high morbidity. After tracheal resection and end-to-end anastomosis
Vinograd et al.
12 5 9
(particularly in tracheas with small lumina) and repair of congenital stenosis with pericardial or periosteal grafts, different intrinsic mural or external supportive temporary stents have been used to maintain the patency of the reconstructed airway.l'vII Recently, several innovative implantable devices have been described to prevent collapse or stenosis of the trachea and other organs.'? Wallace and associates'< adapted an intravascular stent made of expandable wire for use within the tracheobronchial tree. Loeff and associates- used stainless steel wire coils coated with a thin silicone rubber layer. The stent was implanted with the use of either an open surgical technique or a balloon-type introducer mounted with a spring for endoscopic placement. In experimental and clinical experience with the use of various intratracheal stents, difficulties were encountered with endoscopic placement and removal, there were problems with tracheal secretions, and unexpected histologic changes of squamous metaplasia and inflammation in the respiratory epithelium and the submucosa occurred.b 13 Interest in the use of titanium for intratracheal stents was aroused because of two basic qualities of the alloy: (1) its proven highly noncorrosive nature, biocompatibility, and good tissue tolerance and (2) its unique phenomenon of shape memory effect that provides an unprecedented possibility for endoscopic placement and removal, even through the lumen of a very small airway. One of the main disadvantages of the intratracheal stents used hitherto was their lack of biocompatibility.v 3 Studies indicated that stainless steel may be highly corrosive in the presence of chloride ions in the body fluids. When the metal implants release corrosive metal ions into the surrounding tissues, the body may respond with a number of adverse reactions, such as chronic inflammation, pain, discolora. tion, hypersensitivity, fibrosis, necrosis, and in some instances tumors.lv 15 Although the reactions are generally localized and in the majority not severe, these possible effects suggested that other materials should be considered for implantation. Experimental and clinical use of titanium urethral stents showed no incorporation of the stent into the urethral wall, and the inflammatory reaction was minimal. 16 In the present study the inflammatory reaction to nitinol intratracheal stents was minimal. The slight shrinking of the mucous glands in the submucosa layer observed in 4 of 10 rabbits could be attributed to overexpansion of the stent. In the human being this problem can be solved by more precise determination of the tracheal internal lumen by computed tomographic and bronchoscopic examinations. The nitinol intratracheal stents can be produced with external dimensions that are tailored to fit an individual trachea or bronchus without any difficulty, by lim-
1260
The Journal of Thoracic and Cardiovascular Surgery May 1994
Vinograd et al.
_
• p
c
Normal trachea
N·a
Eill Trach. with N·a
NITI-ITS
0.008
10
o
eo
100
200
180
ApplIeCl
'oroe (or)
Fig. 5. Study of biomechanical propertiesof trachea with and without intratracheal stent (NiTi-ITS).
_ 8tr.1nA""(:.... ..:....'
Normal trachea
N·S
I:23J AI. remov. NITI-ITS N·S
------
50 40 30 20 10
o
eo
100
150 Applied 'oree (lIr'
Fig. 6. Study of mechanical propertiesof trachea after in vitro removal of stent (NiTi-ITS).
iting its expansion in the austenitic state at 40 0 C when constructed in the laboratory. Because of the shape memory effect, once converted into the martensitic state in the trachea, the stent will retain the same external diameter. At that point, additional expansion is very small and exclusively thermic dependent; overexpansion is unlikely inasmuch as body temperature does not normally rise higher than 41 0 C. This characteristic is very important because after expansion the nitinol stent does not increase its pressure on the tracheal wall, in contrast to the expandable metallic stent with which there is constant elastic pressure on the trachea. One of the main functions of the tracheal stent is to
increase tracheal wall rigidity. During normal respiration, the intrathoracic pressure ranges from -5 em H 20 on inspiration to + 30 em H 20 and much higher during coughing. Collapse is prevented by the rigidity of the airway wall. In tracheomalacia, however, the tracheal wall is less rigid than normal and tends to collapse, producing airway obstruction. Our current preliminary experience showed that the nitinol stent significantly increased the rigidity of the tracheal wall. The resistance to external pressure was significantly greater for the stented trachea, as compared with that for a normal segment (p < 0.005). An applied force of 50 gm to the segment with the nitinol stent created a deformation of only 50%. After removal
The Journal of Thoracic and Cardiovascular Surgery Volume 107, Number 5
of the stent the mechanical properties of the trachea were not affected. The stented and normal tracheal segments showed equal strain. This study showed that the nitinol intratracheal stent fulfilled the five basic requirements of an intratracheal stent proposed by Loeff and associates.': (I) by taking advantage of its unique feature of shape memory effect, the stent can be inserted, fixed, and removed easily, even in a very small airway; (2) the stent proved its biocompatibility, and inflammatory and adverse reactions in the implanted trachea were minimal: (3 and 4) because of its coil-shape construction and because it is made from a very thin flat wire sheet, obstruction of air tributaries was minimal and pressure on a large surface area of the respiratory epithelium was avoided; and (5) the length of the stent could be adjusted and the diameter varied to fit any tracheal or bronchial dimensions. In infants and children with very narrow airways, the nitinol intratracheal stent may be used as a supplementary method of achieving temporary airway splinting. REFERENCES I. Filler RM, Buck JR, Bahoric A, et a1. Treatment of
segmental tracheornalacia and bronchomalacia by implantation of an airway splint. J Pediatr Surg 1982;17:597-603. 2. Johnston MR, Lober N, Hillyer P, et a1. External stent for repair of secondary bronchornalacia. Ann Thorac Surg 1980;30:291-6. 3. Loeff DS, Filler RM, Gorenstein A, et a1. A new tracheobronchial reconstruction: experimental and clinical studies. J Pediatr Surg 1988;23:1113-7. 4. Wayman CM, Shimizu K. The shape memory ("Marmen") effect in alloys. Metal Sci J 1972;6:175-83.
Vinograd et al.
I 26 I
5. Williams DF. Titanium as a metal for implantation. Part 2. Biological properties and clinical applications. Biomed Eng 1977;1:266-9. 6. Blair GK, Cohen R, Filler RM. Treatment of tracheomalacia: eight years' experience. J Pediatr Surg 1986;21: 781-5. 7. Cohen D. Tracheopexy: aorta-tracheal suspension for severe tracheomalacia. Aust Pediatr J 1981;17:117-21. 8. Vinograd I, Filler RM, Bahoric A. Long term functional results of prosthetic airway splinting in tracheomalacia and bronchomalacia. J Pediatr Surg 1987;22:38-41. 9. Gupta TGCM, Goldberg SJ, Lewis E, Fonkalsrud EW. Congenital bronchomalacia. Am J Dis Child 1968;115:8890. 10. Loeff DS, Filler RM, Vinograd I, et a1. Congenital tracheal stenosis: a review of 22 patients from 1965 to 1987. J Pediatr Surg 1988;23:744-8. II. Idriss FS, Serafin YD, Ilbawi MN, et a1. Tracheoplasty with pericardial patch for extensive tracheal stenosis in implants and children. J THORAC CARDIOVASC SURG 1984;88:527-36. 12. Mair EA, Parsons DS, Lolly KP. Treatment of severe bronchomalacia with expanding endobronchial stents. Arch Otolaryngol Head Neck Surg 1990;116:1087-90. 13. Wallace JM, Charnsangavej L, Ogawa K, et a1. Tracheobronchial tree: expandable metallic stents used in experimental and clinical applications. Radiology 1986;158:30912. 14. Harris B. Corrosion of stainless stet" surgical implants. J Med Eng TechnoI1979;3:117-22. 15. Sunderman FW Jr. Carcinoger .city of metal alloys in orthopedic prostheses: clinical snd experimental studies. Fundam Appl Toxicol 1989;13:..05-11. 16. Parra RO. Treatment of posteri, r urethral strictures with titanium urethral stent. J Urol I: 91;146:997-1000.