- Email: [email protected]
A New Method of Tracheal Reconstruction
A New Method of Tracheal Reconstruction B. F. Akl, M.D., J. Mittelman, M.D., D. E. Smith, M.D., and C. Butler, M.D. ABSTRACT When resection of more than 50% of the trachea is necessary, tracheal reconstruction becomes a very difficult problem. We tested a new method of tracheal reconstruction using the left main bronchus. The procedure was performed in 6 adult mongrel dogs. Through a midline sternotomy, a left pneumonectomy is performed, preserving as much of the left main bronchus as possible. The carina is preserved and mobilized, and the left main bronchus is passed under the aortic arch and reversed. The distal end of the bronchus is anastomosed to the proximal end of the resected trachea. The distal end of the resected trachea is closed by stapling. Four animals survived the procedure without apparent functional difficulty and remained healthy until they were killed five to ten months postoperatively. The anastomosis was well healed without stenosis. Anatomical measurements in 10 human cadavers revealed that the length of the left main bronchus to the level of the upper lobe takeoff is approximately 50% that of the trachea. The diameter of the left main bronchus is approximately 75% that of the trachea. We believe that the technique described can extend the limits of tracheal resection in a selected group of patients for whom there is currently no good alternative.
When limited tracheal resection for a benign or a malignant process is performed, reconstruction of the airway by end-to-end anastomosis has become the standard and preferred method. Adjunctive procedures such as laryngeal release and mobilization of the right hilum have allowed reconstruction of the trachea after resection of up to 50% of its length. However, when
From the Departments of Surgery and Pathology, The University of New Mexico School of Medicine, and The Veterans Administration Medical Center, Albuquerque, NM. Presented at the Twenty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Hilton Head Island, SC, Nov 4-6, 1982. Address reprint requests to Dr. Akl, Division of Cardiovascular and Thoracic Surgery, 2211 Lomas Blvd NE, Albuquerque, NM 87131.
265
resection of more than 50% of the trachea is necessary, airway reconstruction becomes a very difficult and challenging problem. Currently available prostheses have major limitations: the potential for infection; the lack of mucosal lining; the potential for migration or erosion of surrounding structures or both; and the nonapplicability of the prostheses to small children. One of us (B. F. A , ) recently was faced with such a clinical problem. A neonate was seen with a pulmonary artery "sling." This was treated successfully; however, the patient was found to have a congenital tracheal stenosis with complete rings involving the entire length of the trachea but sparing the carina. The effort to manage this patient, whose case has been reported in detail elsewhere,+led to this experimental study.
Material and Method Six adult mongrel dogs weighing 15 to 20 kg were used in the experiment. Anesthesia was induced with Pentothal (thiopental sodium). An endotracheal tube was placed, positive-pressure ventilation was initiated with a pressure-cycled Harvard ventilator, and halothane was used for the maintenance of anesthesia. A midline sternotomy was performed, and the left pleura was opened. The left hilum was dissected and a left pneumonectomy done in the standard fashion except that most of the left side of the bronchus was preserved by dividing it past the takeoff of the left upper lobe bronchus. The trachea then was dissected free between the course of the carotid vessels. The right pleural space was opened and the azygos vein doubly ligated and divided. By blunt dissection, a tunnel around the left main bronchus was developed as it coursed under the aortic arch. The trachea was transected, and a segment about 4 to 5 cm in length
*Akl BF, Yabek S, Berman W: Total tracheal reconstruction in a three-month-old infant. J Thorac Cardiovasc Surg (in press, 1983)
266 The Annals of Thoracic Surgery Vol 36 No 3 September 1983
Fig 2. Healed anastomosis demonstrating the benefit of the 90-degree axial rotation of the bronchus. Note the match of membrane to cartilage.
A
B
Fig 1. (A) The left main bronchus is reversed and ( B ) used to replace the resected trachea.
was removed. A sterile endotracheal tube was inserted in the distal end of the trachea and guided into the right main bronchus, and ventilation was switched to that system. The left bronchus was passed under the aortic arch, with care taken not to avulse the bronchial vessels, although no special effort was made to visualize these structures. The bronchus was transected at the level of the takeoff of the left upper lobe bronchus and rotated approximately 90 degrees along its axis and anastomosed to the proximal end of the trachea with interrupted 3-0 Vicryl sutures (Fig 1).The rotation of the bronchus allowed the membranous portion of the left main bronchus to be supported by the cartilaginous portion of the trachea and vice versa. This appeared to open up the anastomosis, since the cartilages of the left main bronchus were a little softer than the tracheal cartilages (Fig 2). Once the suture line was completed, the second endotracheal tube was withdrawn, the dis-
tal trachea was stapled just above its bifurcation, and ventilation was reestablished through the original endotracheal tube. Additional sutures were placed as necessary to stop the air leak. No other coverage of the suture line was provided. The pleural pericardial reflection on the left side was sutured to the back of the sternum in an attempt to stabilize the mediastinal structures. A chest tube was placed on the right side through an intercostal space and connected to a Heimlich flutter valve. The left pleural space was not drained. The sternlotomy incision was closed in routine fashion. As soon as adequate spontaneous respirations resumed, all tubes but the chest tube were removed. That tube was removed when no air leak was apparent. Antibiotics were used for three days in the perioperative period. A normal diet was resumed the day after operation. Results Canine Study Two dogs died, 1 of pleural and mediastinal infection and 1 of tension pneumothorax. The 4 that survived underwent bronchoscopy at weekly intervals for three weeks and then at monthly intervals until they were killed. The healing of the suture line was smooth. No granulation tissue or suture material had to be
267 Akl et al: Tracheal Reconstruction
Fig 3 . Healed anastomosis in the dog with the shortest interval between the procedure and death (five months). There is complete healing, and no suture material is visible.
removed. No stenosis developed, and no dilations were required. The dogs had survived for five to ten months, and appeared healthy when they were killed and the tracheobronchial trees examined. All suture lines were completely and smoothly healed (Fig 3 ) . There was no evidence of pneumonia or bronchiectasis. There was no stenosis at the anastomotic site in any of the dogs. Microscopic examination showed smooth lining by normal-looking mucosa. The cilia appeared normal. In the dog with the shortest interval between the procedure and death, there was some granulation tissue around the site of the sutures. All the suture material, however, had disappeared. Human Anatomical Study Anatomical measurements were made on 10 specimens of the tracheobronchial tree removed from cadavers of adults dying of a variety of unrelated causes (Fig 4). The average length of the trachea from the cricoid cartilage to the carina was 11 cm (10.0 to 12.5 cm). The average tracheal diameter measured in the coronal plane in the middle portion of the trachea was 2 cm (1.5 to 2.5 cm). The length of the left main bronchus to the level of takeoff of the left upper lobe
Fig 4. Anatomical measurements in 10 human cadavers. (A = 10.0-12.5 cm; B = 5.0-6.5 cm; C = 6.5-8.5 cm; D = 1.5-2.5cm:d = 1.25-1.75cm.)
bronchus averaged 5.5 cm (5.0 to 6.5 cm), and its diameter averaged 1.5 cm (1.25 to 1.75 cm).
Comment In 1951, Belsey [l]wrote the following about tracheal resection and reconstruction: ”The intrathoracic portion of the trachea is the last unpaired organ in the body to fall to the surgeon, and the successful solution of the problem of its reconstruction may mark the end of the ‘expansionist’ epoch in the development of surgery.” Since then, tracheal resection and reanastomosis for a variety of benign and malignant processes has become a standard accepted procedure. It gained even more support after Grillo 12, 31 demonstrated in his extensive clinical experience that up to one-half of the tracheal length can be removed and continuity of the airway reestablished by end-to-end anastomosis, particularly if adjunctive procedures such as laryngeal release procedures are utilized. Refinement of anesthetic techniques and the availability of cardiopulmonary bypass to use in some special circumstances were essential to these advances.
268 The Annals of Thoracic Surgery Vol 36 No 3 September 1983
There remain, however, a number of patients who require an even longer resection and in whom tracheal reconstruction would not be possible with the current methods. A variety of materials and methods have been used for airway reconstruction, all with limited success [I, 4-71. The Neville prosthesis [S] may be more promising, although not ideal. The potential for infection and the risk of migration and erosion into surrounding structure are major hazards. In the 1951 report already quoted from, Belsey [ 11 outlined the essential characteristics of a reconstructed airway: lateral rigidity; longitudinal elasticity and flexibility; an adequate and airtight lumen; and an uninterrupted lining of ciliated columnar epithelium. Our proposed procedure has a good potential for overcoming most of the problems presented here and comes close to fulfilling most of the requirements outlined by Belsey: 1. It uses autologous tissue of similar structure, thereby providing the elasticity, the semirigidity, and the mucosal lining of the normal trachea. 2. Although the length of the left main bronchus is not equal to that of the trachea, it is possible to reestablish continuity of the airway, even after resection of most of the tracheal length, if adjunctive procedures such as laryngeal release and mobilization of the right hilum and flexion of the neck are used. 3. It may provide a good potential for growth when it is used in infants. End-to-end anastomoses with interrupted sutures have shown such potential [9-111.
There are two major drawbacks, however. The need to sacrifice the left lung limits the applicability of the procedure in patients with a severe decrease of lung function. Also, it is not applicable if the pathological process involves the carina. The concern about the blood supply to the left main bronchus after such mobilization is valid. However, it did not seem to be a problem in our experimental model. There was no necrosis of the suture line, and there was good pulsatile bleeding from the cut edge of the bronchus at
the time of its anastomosis. As mentioned earlier, this procedure has been tried in l patient, a neonate. Even though it was necessary to use the entire length of the left bronchus down to the level of the lower lobe bronchus division, the blood supply appeared excellent and no necrosis was apparent at the suture line when the patient died forty-eight hours later. In anatomical dissections of 150 cadavers, Cauldwell and colleagues [121 demonstrated great variability in the number and the site of origin of the bronchial arteries and classified them into nine different types. In every instance, there was at least one bronchial artery to each lung. In the majority of specimens there was more than one artery, at least on one side. These were all vessels arising from the descending aorta. Detailed study of the gross and microscopic blood supply of the trachea was performed by Salassa and co-workers [ 131. Upper tracheal blood supply was found to derive from the inferior thyroid, subclavian, supreme intercostal, internal thoracic, and innominate arteries. Two important longitudinal anastomoses were found between these vessels andl the bronchial arteries. Transverse interconnecting vessels ran in the soft tissues between the cartilages. A rich submucosal capillary network was demonstrated also. This rich blood supply of the tracheobronchial tree originating from multiple sources is probably the reason why no major problem with blood supply to the anastomosis was encountered in our study, even after extensive mobilization of the left main bronchus. The ciliary function in the reversed bronchial segment is not certain. The cilia appeared normal by light microscopy. Whether or not they continued to function normally after mobilization and at least partial denervation of the bronchus remains speculative. In fact, mucus may be transported in the wrong direction by the cilia in the reversed bronchus. However, this did not seem to interfere with the mechanism of clearing of normal secretions in our dogs, nor did it lead to any pulmonary infections or suppurations. Even though this method of tracheal reconstruction may not be completely physiological, it appears to offer a very strong potential for
269 Akl et al: Tracheal Reconstruction
patients for whom no other method of reconstruction is currently available. The number of patients that will benefit from such a procedure will be very small a n d will consist mostly of infants a n d children who need total tracheal reconstruction for extensive congenital or acquired tracheal pathological conditions. In some adult patients with a n extensive pathological process of the trachea, it may also be worth consideration. There are some patients who have a less than optimal resection of a malignant process because of fear of the difficulty of reconstruction. In these patients, a more radical resection m a y become feasible.
References Belsey R: Resection and reconstruction of the intrathoracic trachea. Br J Surg 38:200, 1951 Grillo HC: Surgery of the trachea. In Keen G (ed): Operative Surgery and Management. Bristol, Eng, Wright, 1981, pp 651-660 Grillo HC: Surgical treatment of postintubation tracheal injuries. J Thorac Cardiovasc Surg 78:860, 1979 Fonkalsrud EW, Plested WG, Mulder DG: Tracheobronchial reconstruction with autologous
periosteum. J Thorac Cardiovasc Surg 52:666, 1966 5. Narodick BG, Pemberton AH, Worman LW: Tracheoplasty by means of a periosteal pedicle graft. J Thorac Cardiovasc Surg 47:572, 1964 6. Ohlsen L, Nordin U: Tracheal reconstruction with perichondrial grafts. Scand J Plast Reconstr Surg 10:135, 1976 7. Sat0 R, Hasegawa I, Nakagawa J: Experimental study of tracheal reconstruction. J Thorac Surg 34:526, 1957 8. Neville WE, Bolanowski JP, Soltanzadeh H: Prosthetic reconstruction of the trachea and carina. J Thorac Cardiovasc Surg 72:525, 1976 9. Harrison MR, Heldt GP, Brasch RC, et al: Resection of distal tracheal stenosis in a baby with agenesis of the lung. J Pediatr Surg 15:938, 1980 10. Maeda M, Grillo HC: Tracheal growth following anastomosis in puppies. J Thorac Cardiovasc Surg 64:304, 1972 11. Mansfield PB: Tracheal resection in infancy. J Pediatr Surg 15:79, 1980 12. Cauldwell EW, Siekert RG, Lininger RE, et al: The bronchial arteries. Surg Gynecol Obstet 86:395, 1948 13. Salassa JR, Pearson BW, Payne WS: Gross and microscopical blood supply of the trachea. Ann Thorac Surg " 24:lOO. 1977
When limited tracheal resection for a benign or a malignant process is performed, reconstruction of the airway by end-to-end anastomosis has become the standard and preferred method. Adjunctive procedures such as laryngeal release and mobilization of the right hilum have allowed reconstruction of the trachea after resection of up to 50% of its length. However, when
From the Departments of Surgery and Pathology, The University of New Mexico School of Medicine, and The Veterans Administration Medical Center, Albuquerque, NM. Presented at the Twenty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Hilton Head Island, SC, Nov 4-6, 1982. Address reprint requests to Dr. Akl, Division of Cardiovascular and Thoracic Surgery, 2211 Lomas Blvd NE, Albuquerque, NM 87131.
265
resection of more than 50% of the trachea is necessary, airway reconstruction becomes a very difficult and challenging problem. Currently available prostheses have major limitations: the potential for infection; the lack of mucosal lining; the potential for migration or erosion of surrounding structures or both; and the nonapplicability of the prostheses to small children. One of us (B. F. A , ) recently was faced with such a clinical problem. A neonate was seen with a pulmonary artery "sling." This was treated successfully; however, the patient was found to have a congenital tracheal stenosis with complete rings involving the entire length of the trachea but sparing the carina. The effort to manage this patient, whose case has been reported in detail elsewhere,+led to this experimental study.
Material and Method Six adult mongrel dogs weighing 15 to 20 kg were used in the experiment. Anesthesia was induced with Pentothal (thiopental sodium). An endotracheal tube was placed, positive-pressure ventilation was initiated with a pressure-cycled Harvard ventilator, and halothane was used for the maintenance of anesthesia. A midline sternotomy was performed, and the left pleura was opened. The left hilum was dissected and a left pneumonectomy done in the standard fashion except that most of the left side of the bronchus was preserved by dividing it past the takeoff of the left upper lobe bronchus. The trachea then was dissected free between the course of the carotid vessels. The right pleural space was opened and the azygos vein doubly ligated and divided. By blunt dissection, a tunnel around the left main bronchus was developed as it coursed under the aortic arch. The trachea was transected, and a segment about 4 to 5 cm in length
*Akl BF, Yabek S, Berman W: Total tracheal reconstruction in a three-month-old infant. J Thorac Cardiovasc Surg (in press, 1983)
266 The Annals of Thoracic Surgery Vol 36 No 3 September 1983
Fig 2. Healed anastomosis demonstrating the benefit of the 90-degree axial rotation of the bronchus. Note the match of membrane to cartilage.
A
B
Fig 1. (A) The left main bronchus is reversed and ( B ) used to replace the resected trachea.
was removed. A sterile endotracheal tube was inserted in the distal end of the trachea and guided into the right main bronchus, and ventilation was switched to that system. The left bronchus was passed under the aortic arch, with care taken not to avulse the bronchial vessels, although no special effort was made to visualize these structures. The bronchus was transected at the level of the takeoff of the left upper lobe bronchus and rotated approximately 90 degrees along its axis and anastomosed to the proximal end of the trachea with interrupted 3-0 Vicryl sutures (Fig 1).The rotation of the bronchus allowed the membranous portion of the left main bronchus to be supported by the cartilaginous portion of the trachea and vice versa. This appeared to open up the anastomosis, since the cartilages of the left main bronchus were a little softer than the tracheal cartilages (Fig 2). Once the suture line was completed, the second endotracheal tube was withdrawn, the dis-
tal trachea was stapled just above its bifurcation, and ventilation was reestablished through the original endotracheal tube. Additional sutures were placed as necessary to stop the air leak. No other coverage of the suture line was provided. The pleural pericardial reflection on the left side was sutured to the back of the sternum in an attempt to stabilize the mediastinal structures. A chest tube was placed on the right side through an intercostal space and connected to a Heimlich flutter valve. The left pleural space was not drained. The sternlotomy incision was closed in routine fashion. As soon as adequate spontaneous respirations resumed, all tubes but the chest tube were removed. That tube was removed when no air leak was apparent. Antibiotics were used for three days in the perioperative period. A normal diet was resumed the day after operation. Results Canine Study Two dogs died, 1 of pleural and mediastinal infection and 1 of tension pneumothorax. The 4 that survived underwent bronchoscopy at weekly intervals for three weeks and then at monthly intervals until they were killed. The healing of the suture line was smooth. No granulation tissue or suture material had to be
267 Akl et al: Tracheal Reconstruction
Fig 3 . Healed anastomosis in the dog with the shortest interval between the procedure and death (five months). There is complete healing, and no suture material is visible.
removed. No stenosis developed, and no dilations were required. The dogs had survived for five to ten months, and appeared healthy when they were killed and the tracheobronchial trees examined. All suture lines were completely and smoothly healed (Fig 3 ) . There was no evidence of pneumonia or bronchiectasis. There was no stenosis at the anastomotic site in any of the dogs. Microscopic examination showed smooth lining by normal-looking mucosa. The cilia appeared normal. In the dog with the shortest interval between the procedure and death, there was some granulation tissue around the site of the sutures. All the suture material, however, had disappeared. Human Anatomical Study Anatomical measurements were made on 10 specimens of the tracheobronchial tree removed from cadavers of adults dying of a variety of unrelated causes (Fig 4). The average length of the trachea from the cricoid cartilage to the carina was 11 cm (10.0 to 12.5 cm). The average tracheal diameter measured in the coronal plane in the middle portion of the trachea was 2 cm (1.5 to 2.5 cm). The length of the left main bronchus to the level of takeoff of the left upper lobe
Fig 4. Anatomical measurements in 10 human cadavers. (A = 10.0-12.5 cm; B = 5.0-6.5 cm; C = 6.5-8.5 cm; D = 1.5-2.5cm:d = 1.25-1.75cm.)
bronchus averaged 5.5 cm (5.0 to 6.5 cm), and its diameter averaged 1.5 cm (1.25 to 1.75 cm).
Comment In 1951, Belsey [l]wrote the following about tracheal resection and reconstruction: ”The intrathoracic portion of the trachea is the last unpaired organ in the body to fall to the surgeon, and the successful solution of the problem of its reconstruction may mark the end of the ‘expansionist’ epoch in the development of surgery.” Since then, tracheal resection and reanastomosis for a variety of benign and malignant processes has become a standard accepted procedure. It gained even more support after Grillo 12, 31 demonstrated in his extensive clinical experience that up to one-half of the tracheal length can be removed and continuity of the airway reestablished by end-to-end anastomosis, particularly if adjunctive procedures such as laryngeal release procedures are utilized. Refinement of anesthetic techniques and the availability of cardiopulmonary bypass to use in some special circumstances were essential to these advances.
268 The Annals of Thoracic Surgery Vol 36 No 3 September 1983
There remain, however, a number of patients who require an even longer resection and in whom tracheal reconstruction would not be possible with the current methods. A variety of materials and methods have been used for airway reconstruction, all with limited success [I, 4-71. The Neville prosthesis [S] may be more promising, although not ideal. The potential for infection and the risk of migration and erosion into surrounding structure are major hazards. In the 1951 report already quoted from, Belsey [ 11 outlined the essential characteristics of a reconstructed airway: lateral rigidity; longitudinal elasticity and flexibility; an adequate and airtight lumen; and an uninterrupted lining of ciliated columnar epithelium. Our proposed procedure has a good potential for overcoming most of the problems presented here and comes close to fulfilling most of the requirements outlined by Belsey: 1. It uses autologous tissue of similar structure, thereby providing the elasticity, the semirigidity, and the mucosal lining of the normal trachea. 2. Although the length of the left main bronchus is not equal to that of the trachea, it is possible to reestablish continuity of the airway, even after resection of most of the tracheal length, if adjunctive procedures such as laryngeal release and mobilization of the right hilum and flexion of the neck are used. 3. It may provide a good potential for growth when it is used in infants. End-to-end anastomoses with interrupted sutures have shown such potential [9-111.
There are two major drawbacks, however. The need to sacrifice the left lung limits the applicability of the procedure in patients with a severe decrease of lung function. Also, it is not applicable if the pathological process involves the carina. The concern about the blood supply to the left main bronchus after such mobilization is valid. However, it did not seem to be a problem in our experimental model. There was no necrosis of the suture line, and there was good pulsatile bleeding from the cut edge of the bronchus at
the time of its anastomosis. As mentioned earlier, this procedure has been tried in l patient, a neonate. Even though it was necessary to use the entire length of the left bronchus down to the level of the lower lobe bronchus division, the blood supply appeared excellent and no necrosis was apparent at the suture line when the patient died forty-eight hours later. In anatomical dissections of 150 cadavers, Cauldwell and colleagues [121 demonstrated great variability in the number and the site of origin of the bronchial arteries and classified them into nine different types. In every instance, there was at least one bronchial artery to each lung. In the majority of specimens there was more than one artery, at least on one side. These were all vessels arising from the descending aorta. Detailed study of the gross and microscopic blood supply of the trachea was performed by Salassa and co-workers [ 131. Upper tracheal blood supply was found to derive from the inferior thyroid, subclavian, supreme intercostal, internal thoracic, and innominate arteries. Two important longitudinal anastomoses were found between these vessels andl the bronchial arteries. Transverse interconnecting vessels ran in the soft tissues between the cartilages. A rich submucosal capillary network was demonstrated also. This rich blood supply of the tracheobronchial tree originating from multiple sources is probably the reason why no major problem with blood supply to the anastomosis was encountered in our study, even after extensive mobilization of the left main bronchus. The ciliary function in the reversed bronchial segment is not certain. The cilia appeared normal by light microscopy. Whether or not they continued to function normally after mobilization and at least partial denervation of the bronchus remains speculative. In fact, mucus may be transported in the wrong direction by the cilia in the reversed bronchus. However, this did not seem to interfere with the mechanism of clearing of normal secretions in our dogs, nor did it lead to any pulmonary infections or suppurations. Even though this method of tracheal reconstruction may not be completely physiological, it appears to offer a very strong potential for
269 Akl et al: Tracheal Reconstruction
patients for whom no other method of reconstruction is currently available. The number of patients that will benefit from such a procedure will be very small a n d will consist mostly of infants a n d children who need total tracheal reconstruction for extensive congenital or acquired tracheal pathological conditions. In some adult patients with a n extensive pathological process of the trachea, it may also be worth consideration. There are some patients who have a less than optimal resection of a malignant process because of fear of the difficulty of reconstruction. In these patients, a more radical resection m a y become feasible.
References Belsey R: Resection and reconstruction of the intrathoracic trachea. Br J Surg 38:200, 1951 Grillo HC: Surgery of the trachea. In Keen G (ed): Operative Surgery and Management. Bristol, Eng, Wright, 1981, pp 651-660 Grillo HC: Surgical treatment of postintubation tracheal injuries. J Thorac Cardiovasc Surg 78:860, 1979 Fonkalsrud EW, Plested WG, Mulder DG: Tracheobronchial reconstruction with autologous
periosteum. J Thorac Cardiovasc Surg 52:666, 1966 5. Narodick BG, Pemberton AH, Worman LW: Tracheoplasty by means of a periosteal pedicle graft. J Thorac Cardiovasc Surg 47:572, 1964 6. Ohlsen L, Nordin U: Tracheal reconstruction with perichondrial grafts. Scand J Plast Reconstr Surg 10:135, 1976 7. Sat0 R, Hasegawa I, Nakagawa J: Experimental study of tracheal reconstruction. J Thorac Surg 34:526, 1957 8. Neville WE, Bolanowski JP, Soltanzadeh H: Prosthetic reconstruction of the trachea and carina. J Thorac Cardiovasc Surg 72:525, 1976 9. Harrison MR, Heldt GP, Brasch RC, et al: Resection of distal tracheal stenosis in a baby with agenesis of the lung. J Pediatr Surg 15:938, 1980 10. Maeda M, Grillo HC: Tracheal growth following anastomosis in puppies. J Thorac Cardiovasc Surg 64:304, 1972 11. Mansfield PB: Tracheal resection in infancy. J Pediatr Surg 15:79, 1980 12. Cauldwell EW, Siekert RG, Lininger RE, et al: The bronchial arteries. Surg Gynecol Obstet 86:395, 1948 13. Salassa JR, Pearson BW, Payne WS: Gross and microscopical blood supply of the trachea. Ann Thorac Surg " 24:lOO. 1977