Ann Thorac Surg 1997;64:349-354
© 1997 The Society of Thoracic Surgeons
Original Articles: General Thoracic
Experimental Reconstruction of the Mediastinal Trachea With a Wing-Shaped Reversed Esophageal Flap
Masanobu Kiriyama, MD,
Akira Masaoka, MD,
Yosuke Yamakawa, MD,
Hiroshi Niwa, MD,
Yoshitaka Fujii, MD
Department of Surgery II, Nagoya City University Medical School, Nagoya, Japan
Accepted for publication February 11, 1997.
 |
Abstract
|
|---|
Background. In cases of extensive tracheal resection in which direct end-to-end anastomosis is impossible there is a need for reconstruction. Nevertheless, with the present lack of reliable artificial trachea, no reconstruction method is available to assure safe replacement of the mediastinal trachea.
Methods. After tubular resection of the mediastinal trachea in mongrel dogs, the trachea was reconstructed using a wing-shaped reversed esophageal flap. A silicone tube was used as an internal stent.
Results. In group I (16 animals), three tracheal rings were resected; in group II (4 animals), six tracheal rings; in group III (6 animals), eight tracheal rings; and in group IV (5 animals), eight tracheal rings and the lining of the greater omentum. Safe reconstruction was accomplished in all cases in groups I and II, 2 of 6 cases in group III, and 2 of 5 cases in group IV. The omentopexy failed to prevent incomplete closure, yet served to minimize inflammation in the mediastinum.
Conclusions. A reversed esophageal autograft can be considered as a tracheal replacement.
|
Introduction
|
|---|
Although aggressive surgical intervention has become the rule for tracheal lesions, reconstruction is necessary after extensive tracheal resection in cases in which direct end-to-end anastomosis is impossible. A number of case have been reported in which, after subtotal tracheal removal, a permanent tracheostomy has been left on the caudal trachea without performing reconstruction [1, 2]. Nevertheless, with the present lack of an artificial trachea, no reconstruction method is available to assure safe replacement of the mediastinal trachea. Thus, we used mongrel dogs in an attempt to reconstruct the mediastinal trachea with a wing-shaped reversed esophageal flap.
|
Material and Methods
|
|---|
The experimental animals used for the study of mediastinal trachea reconstruction were 31 mongrel dogs ranging in weight from 8 to 17 kg (mean weight, 11 kg). The animals were divided into four groups in terms of the number of resected tracheal rings and the presence or absence of greater omentum lining: group I (16 animals), resection of three tracheal rings; group II (4 animals), resection of six tracheal rings; group III (6 animals), resection of eight tracheal rings; and group IV (5 animals), resection of eight tracheal rings and the lining of the greater omentum.
Anesthesia was induced by 3 mL of ketamine hydrochloride injected intramuscularly. After intravenous administration of sodium thiopental and succinyl choline chloride, an endotracheal tube was inserted and controlled respiration was obtained with a mixture of air and oxygen (oxygen 50%). With the animal in the left lateral position, a right thoracotomy was then done at the third or fourth intercostal space to expose the trachea and esophagus. The trachea was then isolated from the subclavian artery, and the esophagus was prepared for the proposed esophageal flap. The mediastinal trachea was dissected and excised from the carina (Fig 1A
). After the tracheal sectioning, the operative field intubation was performed, and the esophagal flap was made. The esophagus was cut at the position near the middle of the tracheal stumps. A transverse cut was made 3 to 4 cm cephalad to the first incision. One fourth of the circumference of the anterior wall was left and served as the pedicle. The width of the pedicled portion was approximately 1 cm; the flap length averaged around 3 cm in groups I, II, and III, and about 4 cm in group IV. Next, an incision was made lengthwise in the dorsal surface of the esophageal flap to create a wing-shaped flap (Fig 1B). This wing-shaped esophageal flap was then reversed so that the outer surface of the esophageal flap would face the luminal side, with the esophageal mucous membrane exposed to the outside when creating a cylindric tracheal replacement (Fig 1C). End-to-end anastomosis of the trachea and the esophageal flap was performed using continuous sutures with 3-0 Vicryl (Ethicon, Somerville, NJ) stitches. A silicone tube (11 mm outer diameter, 8 mm inner diameter, 6 to 7 cm in length) as an interior stent was placed and the tube was fixed to a silicone pledget (approximately 1 x 1 cm) on the anterior wall of the trachea with 3-0 nylon stitches. Finally, the anterior surface was sutured, and the tracheal reconstruction was completed. The esophagus was reconstructed by end-to-end anastomosis (Fig 1D, Fig 2).
View larger version (28K):
[in this window]
[in a new window]
|
Fig 1. . Operative procedures of wing-shaped reversed esophageal flap for tracheal replacement (for details see Material and Methods section).
|
|
View larger version (135K):
[in this window]
[in a new window]
|
Fig 2. . Operative findings ( black arrow = wing-shaped reversed esophageal flap; white arrow = esophageal anastomosis).
|
|
In the greater omental lining group (group IV), a laparotomy was performed by transverse incision of the upper abdomen. The right gastroomental artery was preserved and a pedicled omental flap (approximately 20 cm in length) was made. This flap was then raised to the lumen through an opening in the anterior diaphragm and sutured to cover the tracheal reconstruction region and esophageal anastomosis area.
The animals fasted for 6 days postoperatively; intravenous drip infusions were given. On day 7 the animals were given water and allowed to return to a fixed diet. By day 10 they were eating normally. The reconstructed trachea was observed through a fiberoptic bronchoscope and chest Xp after the dogs had been anesthetized by an intramuscular injection of ketamine hydrochloride. The interposed esophageal segment was macroscopically examined when the animals died.
Autopsy specimens were fixed in formalin, and specimens stained with hematoxylin and eosin were made from both ends of the esophageal graft and the esophageal end-to-end anastomosis. Histologic examination was done with a light microscopy. All animals received humane care throughout the study, as recommended by the National Institutes of Health.
|
Results
|
|---|
The operation time was 2 hours 30 minutes on average in groups I, II, and III, and 3 hours 30 minutes in group IV with greater omental lining. In group I (16 animals), the resection length averaged 13 ± 1.0 mm. All the dogs (dogs 1–16) showed a favorable postoperative course without complications (Table 1). They were sacrificed at various postoperative stages from 1 week to 4 months postoperatively.
In group II (4 animals), the resection length averaged 29 ± 2.5 mm for the six tracheal rings, and all 4 dogs (dogs 17–20) displayed a favorable, uncomplicated course. They were capable of long-term survival and were sacrificed from 1 to 4 months postoperatively.
In group III (6 animals), the length of the eight resected tracheal rings averaged 38 ± 3.0 mm. The course was uncomplicated in 2 of the dogs (dogs 25 and 26) until they were sacrificed at 1 month and 4 months postoperatively. In the other dogs (dogs 21–24) tracheoesophageal fistula developed due to incomplete closure, and those dogs died within 7 days postoperatively.
In group IV (5 animals), the resection length averaged 40 ± 0.9 mm. The postoperative course was good in 2 dogs (dogs 30 and 31) until they were sacrificed at 1 month and 3 months postoperatively. In 1 of these dogs (dog 31) a diaphragmatic hernia developed. The other 3 dogs (dogs 27–29) died within 8 days postoperatively of a tracheoesophageal fistula due to faulty closure.
Bronchoscopic Findings
Bronchoscopic observation of long-term survivors showed that the reconstructed trachea attached to the inner stent during both inspiration and expiration. The bronchoscope could be easily slipped in between the stent and the reconstructed trachea, and observation of the lumen side of the reconstructed trachea was possible throughout the entire course. The trachea and the reconstructed trachea appeared reddish at 1 to 2 months postoperatively. However, at 4 months and thereafter, the lumen of the reconstructed trachea was smooth and covered with epithelium. No inflammation was observed thereafter in the vicinity of the anastomosis. Also, no secretion was seen in the inside of the stent (Fig 3). At 4 months and thereafter, there was no difference among groups I, II, III, and IV.
View larger version (80K):
[in this window]
[in a new window]
|
Fig 3. . Bronchoscopic findings at 3 months postoperatively. No secretion was seen in the inside of the stent. No inflammation was observed in the tracheal and bronchial lumen.
|
|
Plain Roentgenographic Findings
No abnormal shadows were noted in the lung field as shown by simple roentgenograms at 3 months postoperatively. The lumen of the reconstructed trachea was supported by the internal stent, and no stenosis was observed (Fig 4).
View larger version (117K):
[in this window]
[in a new window]
|
Fig 4. . Plain roentgenogram of a dog of group III at 3 months postoperatively. The lumen of the reconstructed trachea was supported by the internal stent, and no stenosis was observed.
|
|
Autopsy Findings
In the vicinity of the reconstructed trachea in group I, there was edema at 1 week postoperatively (dogs 1 and 2), little connective tissue, and almost no adherence. But at 2 weeks (dogs 3 and 4), some adhesion to the surrounding tissue due to connective tissue growth was observed. At 3 weeks and thereafter a thick connective tissue covering was observed, and there was strong adhesion with the surrounding tissue. In all animals of group I, sacrificed from 1 week to 4 months postoperatively, the wing-shaped reversed esophageal flap adhesion was good, no inflamed abscess was noted in the reconstructed area, and there was no separation or faulty closure between the cephalad and caudal trachea with the esophageal flap. In the tracheal lumen of all 16 dogs the wing-shaped reversed esophageal flap made contact with the internal stent; no granulation tissue formation was noted. The lumen surface was reddish at the first and second week, but from the third week on the epithelium covering provided a smooth surface.
In all animals (dogs 17–20) of group II and in 2 animals (dogs 25 and 26) of group III at 1 month and 4 months, just as with group I at the same period, there was a thick covering with connective tissue and strong adhesion with the surrounding tissues; there was no inflammation or granulation, and smooth epithelium covered the luminal surface (Fig 5).
View larger version (123K):
[in this window]
[in a new window]
|
Fig 5. . Autopsy findings at 4 months postoperatively (autopsy specimen from animal 26 of group III). The suture lines healed well. The lumen was smoothly covered by epithelium, and there was no formation of granulation.
|
|
In 4 animals (dogs 21–24) of group III, incomplete closure was found between the area near the pedicled part of the end-to-end anastomosis. Mediastinitis developed, and tracheoesophageal fistula resulted. In 2 animals (dogs 30 and 31) in group IV, the reconstructed trachea was covered with the greater omentum and there was virtually no adhesion with the surrounding tissue. Connective tissue in this region was extremely thin. The take was good, no inflammation or separation of the anastomosis region was noted, the lumen was smoothly covered by epithelium, and there was no formation of granulation. In dog 31, a diaphragmatic hernia resulted from the use of pedicled greater omentum flap. Three of the animals (dogs 27–29) were found to have faulty closure at the same location as the 4 animals (dogs 21–24) in group III, and tracheoesophageal fistulas formed. However, the mediastinitis remained local.
In all four groups, when the inner stent was removed, the wall of the reconstructed trachea was as rigid as that of the esophagus.
Histologic Findings
Long-term survivors were sacrificed as indicated in Table 1
, and the recovery of the reconstructed trachea was examined histologically (Table 2). Squamous epithelium appeared at 3 weeks postoperatively at the cephalad and caudal end of the grafted esophagus, and the entire lumen of the reconstructed trachea was covered by the squamous epithelium at 1 month and thereafter (Fig 6). Moreover, ciliated epithelium was observed on either end of the graft at 3 weeks postoperatively, but even after 4 months the ciliated epithelial cells reached no further than the vicinity of the anastomosis (Fig 7). No differences in healing conditions were recognized among the four groups at each time point. In the reconstructed trachea at 3 weeks postoperatively the luminal side (esophageal external membrane) showed cell infiltration, mostly by neutrophils. However, neutrophil infiltration disappeared almost entirely after a month. Microabscesses were found along the suture in the anastomotic region at 3 weeks postoperatively. Sutures were absorbed at 3 months, and by this time abscesses had healed. The muscle of the esophageal flap showed no remarkable change throughout the course. The esophageal gland at 1 month postoperatively showed accumulation of mucus within the glandular cells, which eventually filled the cytoplasm. Reduction in cytoplasm was seen at 3 months postoperatively and thereafter, which is probably due to disuse atrophy.
View larger version (121K):
[in this window]
[in a new window]
|
Fig 6. . Microscopic findings of the specimen obtained from animal 7 at 1 month postoperatively. The external esophageal membrane that forms the surface of the tracheal lumen was covered by squamous epithelium. (Hematoxylin and eosin; x10 before 23% reduction.)
|
|
View larger version (135K):
[in this window]
[in a new window]
|
Fig 7. . Microscopic findings of the specimen obtained from animal 11 at 1 month postoperatively. Ciliated epithelium was observed in the anastomotic region. (Hematoxylin and eosin; x10 before 23% reduction.)
|
|
|
Comment
|
|---|
The surgical method of choice for obstruction of the trachea is excision of the trachea and direct end-to-end anastomosis. The length of the trachea that allows safe resection and direct end-to-end anastomosis differs according to the patient's age, physical condition, and history of previous treatment. In the adult, about a half of the trachea with eight to ten rings can be safely resected [3, 4]. A replacement trachea is needed when a longer portion needs to be resected. To date many attempts have been made using various artificial and biological materials. Essential requirements for the tracheal replacement according to Belsey [5] are (1) lateral rigidity, (2) longitudinal elasticity and flexibility, (3) an adequate and airtight lumen, and (4) an uninterrupted lining of ciliated columnar epithelium. Neville [6] stated that the replacement must be (1) airtight, (2) of adequate consistency, (3) well accepted by the host, (4) having minimal inflammatory reaction, (5) impervious to fibroblastic and bacterial invasion of the lumen, and (6) showing ingrowth of respiratory epithelium along the lumen.
Among the replacement tracheas used to date are those made with artificial materials. Since Neville and associates [7] used silicone rubber in 1972 for an artificial trachea, many authors have reported its use even today (ie, Neville prosthesis). Between 1970 and 1988, Neville and associates [8] used straight grafts for 48 cases of either benign or malignant disease and bifurcated grafts in 14 cases, but the results have not been satisfactory.
Examples of biological materials are tracheal homograft and autograft. The preservation methods used in tracheal transplant experiments have been various [9–12]. However, necrosis of the trachea transplant has occurred due to infection and rejection at 1 to 3 weeks postoperatively, and there were no long-term survivors. In all of these cases involving biological materials the cause of failure was considered to be rejection.
Experimental reconstruction of the trachea with vascularized autologous organs including small bowel or pedicled esophagus has also been reported [13]. In reconstruction using free small bowel [14], vascular anastomosis is required, the mucous membrane must be carefully treated to control digestive juice secretion, and other steps should be taken to support the lumen.
There have been reports on the use of esophagus for tracheal reconstruction. Clinically, the esophageal anterior wall has been used as material to supplement the membrane portion for complete congenital tracheal stenosis [15, 16] and for membranous laceration caused by insertion of a Dumon stent [17]. For congenital tracheal aplasia the esophagus itself has been used for tracheal replacement after severance from the stomach [18]. Kato and colleagues [19] resected seven tracheal rings from the cervical trachea in 6 mongrel dogs and used a pedicled esophageal interposition with feeding vessels and then employed a silicone T tube as the internal stent to achieve long-term survival in 3 animals.
The esophagus is near the dorsal side of the trachea and has muscle lining with ample blood supply; it is airtight and elastic. Other than rigidity it fulfills all of Belsey's conditions. We reversed the esophageal flap to avoid the problem of secretion into the trachea.
Although there was concern about possible secretion to the mediastinum, at 3 months postoperatively the esophageal glands showed atrophy from disuse and no effect of secretion was found. After 3 weeks postoperatively squamous epithelium appeared on the graft and ciliary epithelium was seen near the anastomosis. At 1 month postoperatively and thereafter, squamous epithelium covered the entire reconstructed trachea, and no granulation was observed.
We used a silicone tube for lumen support. Although there was connective tissue growth in the esophageal graft, it was not rigid even at 1 to 4 months. Clinically speaking, long-term use of a silicone T tube is generally feasible without complication [8, 20].
A 3-cm-long wing-shaped reversed esophageal flap could safely reconstruct resection of up to six tracheal rings as seen in groups I and II. However, tracheoesophageal fistulas developed in 4 of the 6 group III cases with resection of eight tracheal rings. In group IV, where eight tracheal rings were resected and the esophageal flap was grafted with omental coverage, tracheobronchial fistulas occurred in 3 of 5 animals. The cause of this failure is unknown but the slight retraction of the trachea might have contributed. Although the greater omental lining was not sufficiently effective to prevent failed closure, it partially limited contamination from the opening and decreased the local inflammation.
The airtightness, elasticity, and good epithelial formation of the wing-shaped reversed esophageal flap are definitely advantages. The issue of somewhat weak lumen support can be resolved by using an internal stent. If the reconstruction could be done with less tension on the trachea and the esophagus, the reconstruction would become safer, surer, and feasible over a wider area, paving the way to possible clinical application.
|
Footnotes
|
|---|
Address reprint requests to Dr Kiriyama, Department of Surgery II, Nagoya City University Medical School, Mizuho-ku, Nagoya 467, Japan.
|
References
|
|---|
- Grillo HC. Terminal or mural tracheostomy in the anterior mediastinum. J Thorac Cardiovasc Surg 1966;51:422–7.[Medline]
- Orringer MB, Sloan H. Anterior mediastinal tracheostomy: indications, techniques, and clinical experience. J Thorac Cardiovasc Surg 1976;78:850–9.
- Grillo HC, Dignan EF, Miura T. Extensive resection and reconstruction of mediastinal trachea without prosthesis or graft: an anatomical study in man. J Thorac Cardiovasc Surg 1964;48:741–9.[Medline]
- Grillo HC. Tracheal replacement. Ann Thorac Surg 1990;49:864–5.[Medline]
- Belsey R. Resection and reconstruction of the intrathoracic trachea. Br J Surg 1950;38:200–5.[Medline]
- Neville WE. Reconstruction of the trachea and stem bronchi with Neville prosthesis. Int Surg 1982;67:229–34.[Medline]
- Neville WE, Hamouda F, Anderson J, Divan FM. Replacement of the intrathoracic trachea and both stem bronchi with a molded Silastic prosthesis. J Thorac Cardiovasc Surg 1972;63:569–76.[Medline]
- Neville WE, Bolanowski PJP, Kotia GG. Clinical experience with the silicone tracheal prosthesis. J Thorac Cardiovasc Surg 1990;99:604–13.[Abstract]
- Jackson TL, O'Brien EJ, Tuttle W, Meyer J. The experimental use of homogenous tracheal transplants in the restoration of continuity of the tracheobronchial tree. J Thorac Surg 1950;20:598–612.
- Carter MG, Strieder JW. Resection of the trachea and bronchi: an experimental study. J Thorac Surg 1950;20:613–27.
- Davies OG, Edmiston JM, McCorkle HJ. The repair of experimental tracheal defects with fresh and preserved homologous tracheal grafts. J Thorac Surg 1952;23:367–76.
- Keshishian JM, Blades B, Beattie EJ Jr. Tracheal reconstruction. J Thorac Surg 1956;32:707–27.
- Papp C, McCraw JB, Arnold PG. Experimental reconstruction of the trachea with autogenous materials. J Thorac Cardiovasc Surg 1985;90:13–20.[Abstract]
- Letang E, Llolet JS, Gimferrer JM, Ramirez J, Vicens A. Experimental reconstruction of the canine trachea with a free revascularized small bowel graft. Ann Thorac Surg 1990;49:955–8.[Abstract]
- Ein SH, Friedberg J, Williams WG, Rearon B, Barker GA, Mancer K. Tracheoplasty—a new operation for complete congenital tracheal stenosis. J Pediatr Surg 1982;17:872–8.[Medline]
- Fonkalsrud EW, Martelle RR, Maloney JV, et al. Surgical treatment of tracheal agenesis. J Thorac Cardiovasc Surg 1963;45:520–5.[Medline]
- Niwa H, Masaoka A, Yamakawa Y, Fukai I, Kiriyama M, Shindou J. Esophageal tracheobronchoplasty for membranous laceration caused by insertion of a Dumon stent—maintenance of oxygenation by percutaneous cardiopulmonary support. Eur J Cardiothorac Surg 1995;9:213–5.[Abstract]
- Fonkalsrud EW, Sumida S. Tracheal replacement with autologous esophagus for tracheal stricture. Arch Surg 1971;102:139–42.[Abstract/Free Full Text]
- Kato R, Onuki S, Watanabe M, et al. Tracheal reconstruction by esophageal interposition: an experimental study. Ann Thorac Surg 1990;49:951–4.[Abstract]
- Cooper JD, Pearson FG, Patterson GA, et al. Use of silicone stents in the management of airway problems. Ann Thorac Surg 1989;47:371–8.[Abstract]
Top
Abstract
Introduction
