|
|
||||||||
Ann Thorac Surg 1996;62:1215-1221
© 1996 The Society of Thoracic Surgeons
Departments of Cardiovascular Surgery and Cardiology, Copenhagen University Hospital, Copenhagen, Denmark
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
In the beginning of lung transplantation, it was clear that bronchial complications were a major cause of death in the early postoperative period [2]. When Veith and associates [3] and Cooper [4] analyzed the early clinical cases of lung transplantation, they found that of the first 38 patients who had undergone lung or lobe transplantation, only 1 survived more than 3 weeks. The single dominating cause of death was dehiscence of the bronchial anastomosis. Already in 1950, Metras [5] had proposed that the bronchial artery circulation could be important in lung transplantation. To pursue this theory, Metras performed experimental single-lung transplantations with direct bronchial revascularization in dogs.
Several anatomic [6] and experimental studies [79] have been done to develop an effective and reliable method for revascularization of the bronchial arteries. Haglin and colleagues [10] performed direct bronchial artery revascularization in a left lung transplant in 1973. In practice, however, bronchial artery revascularization was considered too difficult and unreliable to be used clinically, and other methods of protecting the bronchial anastomosis were developed and applied. With respect to airway healing, these methods have proved very successful.
It was not until 1992 that Couraud and co-workers [11] reported the first clinical series of 8 patients with en bloc double-lung transplantation with successful bronchial artery revascularization. Since then, reports of successful en bloc double-lung transplantation with bronchial revascularization have been published from Harefield [12] and Copenhagen [13, 14]. The Mayo Clinic has recently reported single and bilateral lung transplantation with bronchial artery revascularization [15, 16].
Based on our own experience, we believe that direct bronchial artery revascularization during lung transplantation is possible with a high success rate. Although bronchial healing can be achieved without revascularization, there could be other reasons to believe that a normal arterial blood supply is beneficial to the transplanted lungs. It is important to evaluate available information about the bronchial artery circulation that indicates a possible beneficial effect of direct bronchial artery revascularization in lung transplantation that goes beyond healing of the airways. It has been suggested that ischemia (versus effective bronchial revascularization) could be important in the development of bronchiolitis obliterans syndrome [13, 17].
We reviewed the published information relevant to direct bronchial artery revascularization.
| Anatomy |
|---|
|
|
|---|
The bronchial arteries follow the bronchial tree deeply into the lung parenchyma, forming a peribronchial plexus, and small arterioles penetrate the muscularis to form a submucosal plexus. The bronchial arteries also contribute arterial blood supply to the visceral pleura, the walls of the pulmonary arteries and veins as vasa vasorum, the mediastinal tissue, esophagus, mediastinal and hilar lymph nodes, pericardium, and vagal and sympathetic nerves; even branches to the myocardium can be found in humans [21].
Studies of the microvascular anatomy have found the venous drainage to be partially to the azygos and hemiazygos veins [20] and partially through anastomoses between the bronchial arteries and pulmonary alveolar microvessels, called "bronchopulmonary arteries" [22]. Tobin [23] described these anastomotic vessels as coiled, possibly because of spiral musculature in their walls, and suggested that this structure "might help to regulate the pressure gradient from bronchial to pulmonary vessels." Tobin also described communications between the bronchial arteries and the pulmonary veins.
In 1990, Schreinemakers and colleagues [6] published an anatomic study of the bronchial arteries in 30 autopsy cases. Based on these studies, a technique for procurement of the lungs along with the retroesophageal right intercostobronchial artery was developed. In 17 of 19 dissections, it was possible to prepare a bronchial artery pedicle with a patch of aorta holding the intercostobronchial artery and its origins, which would have been possible to reimplant in the ascending aorta of a recipient after a double lung transplantation.
Couraud and associates [11] were the first to publish a series of patients having en bloc double lung transplantation with bronchial artery revascularization. The clinical series was preceded by an anatomic study of the tracheobronchial blood supply in 20 human cadavers and by angiographic studies in 50 patients having thoracic operations. A right intercostobronchial artery was found in 95% of the cases. In 50% of the cases, orifices of additional bronchial arteries were found 0.5 to 1.5 cm from the intercostobronchial artery orifice. A common bronchial artery trunk giving branches to both the right and left sides was found in about half of the cases.
| Physiology and Pathophysiology |
|---|
|
|
|---|
The flow of the bronchial arteries has been measured in humans during cardiopulmonary bypass conditions. Baile and colleagues [26] found the flow of the bronchial arteries to be 3.23% ± 4% of cardiac output. Deal and co-workers [27] found it to be 3.8% ± 2.7% of cardiac output during the same conditions. In dogs, the bronchial artery flow has been found to be temperature dependent, with a maximum flow at 36°C [28]. Also in dogs, Deffebach and associates [29] demonstrated that the bronchial artery flow is capable of a sevenfold increase as a response to lung injury by intravascular administration of
-naphthylthiourea or by airway instillation of glucose with glucose oxidase.
The function of the bronchial arteries during pulmonary artery occlusion and reperfusion has been studied in rabbits by Kowalski and associates [30]. An inverse correlation was found between bronchial artery flow during ischemia and the lung wet-to-dry ratio after reperfusion. The authors speculated whether a critical level of bronchial artery flow is necessary during pulmonary artery occlusion and reperfusion to prevent lung injury. Their findings have been supported by Pearse and Wagner [31], who suggested that the bronchial artery flow decreases edema formation after ischemia and reperfusion.
Charan and colleagues [32] studied the role and response of the bronchial arteries during severe lung infection in sheep suffering from multiple lung abscesses. It was found that the walls of the abscess cavities were surrounded by a dense vascular plexus that was being supplied exclusively by the bronchial arteries. Cudkowicz [33] studied the vasculature of tuberculous lungs in five postmortem examinations and found that the tuberculous cavities had a rich blood supply. In lungs with bronchiectasis, Cudkowicz found evidence of proliferation and enlargement of the bronchial arteries along with large bronchopulmonary anastomoses.
Development of bronchiolitis obliterans syndrome as a consequence of airway ischemia has been suggested by Yousem and co-workers [17]. Closset and associates [34] investigated this possible correlation in 15 dogs undergoing severing of the left bronchial arteries by hilar stripping and bronchotomy followed by immediate reanastomosis. With a follow-up time of up to 3.5 months, no sign of obliterative bronchiolitis was found. Closset and associates concluded that ischemia is not an important factor in obliterative bronchiolitis.
| Experimental Animal Studies of Bronchial Artery Revascularization in Lung Transplantation |
|---|
|
|
|---|
Already in 1950, Metras [5] performed single-lung transplantation with direct revascularization of the bronchial arteries in dogs. In 1964, Nettleblad and colleagues [35] reported successful revascularization of a bronchial artery in canine left lower lobe transplantation. A donor aortic pouch including the bronchial arteries was anastomosed to the recipient's descending aorta in a side-to-side fashion.
In 1970, Mills and co-workers [36] published a controlled study of the effects of reestablishment of the bronchial artery blood supply on outcome after left lung homotransplantation in 21 mongrel dogs. A preliminary anatomic study had been performed to determine the bronchial artery supply of the left lung. In group 1 (10 dogs), the left bronchial artery with a patch of aortic wall was anastomosed to the recipient's descending aorta. In group 2 (11 dogs), the bronchial artery supply was not reestablished and the left bronchial artery was ligated. In group 1, two instances of bronchial complications were found. One dog died of bronchial disruption on the seventh postoperative day, and the bronchial artery was found to be occluded. The other died from pneumonia 23 days postoperatively, and an ulcerated area was found in the donor bronchus. The bronchial artery was patent at autopsy. In group 2, nine bronchial complications occurred, including ulceration in 5 dogs, bronchial disruption in 3, and development of bronchial stenosis in 1.
In 1990, Nazari and associates [7] published a study of left lung transplantation in 33 pigs. Bronchial artery revascularization was performed in 24 pigs, in which an aortic patch containing the bronchial artery orifice was anastomosed to the recipient's descending aorta. The perioperative mortality rate was high, with only 8 of 28 pigs alive on the 11th postoperative day. Little can be concluded from this study regarding the effect of bronchial artery revascularization except that the anatomy of the pig allows easy identification and preservation of the left bronchial arteries.
In 1991, Laks and colleagues [9] published a method for bronchial artery revascularization tested in 5 baboons. Three had double-lung transplantation; 1 had a single left lung and one had a single right lung transplanted. A pouch of descending aorta from the proximal to the left subclavian artery to the level of the pulmonary hilum (holding the bronchial artery branches) was prepared and anastomosed to the recipient's left subclavian artery or ascending aorta. Angiograms of the 3 surviving baboons (1 double, 1 single right, 1 single left lung) performed after 14 days and postmortem examinations after 22, 30, and 30 days demonstrated patent anastomoses, no pouch thrombi, and normal healing of the tracheal and bronchial anastomoses.
The effect of direct bronchial artery revascularization on the airway mucosal blood flow was investigated by Aoki and colleagues [37] in 1991. In a model of modified unilateral lung transplantation in pigs, the mucosal blood flow was measured at the donor main carina and at the upper lobe carina using laser Doppler velocimetry and radioisotope studies with radiolabeled erythrocytes. The mucosal blood flow was significantly higher in animals with than in those without bronchial artery revascularization.
Fujino and co-workers [8] published a similar study in 1992. In 23 mongrel dogs, radical hilar stripping and transection of the right main bronchus and the right bronchial artery were performed. The right main bronchus was immediately reanastomosed. Group A had the bronchial artery ligated. In group B, the bronchial artery was anastomosed to the right internal thoracic artery using microsurgical technique. The effect of revascularization was estimated by laser Doppler velocimetry. The peripheral bronchial mucosal blood flow immediately after operation had diminished to 59% of the original in group A and to 78% in group B. A statistically significant difference in mucosal blood flow between the groups continued on days 1, 3, 5, and 7. From day 10 on, there was no difference. Two dogs in group B were found to have occlusion of the anastomosis when sacrified after 14 days. No airway complications were observed in either group.
| Clinical Experience |
|---|
|
|
|---|
In 1970, Mills and colleagues [36] polled the 18 investigators responsible for the first 25 human lung allotransplantations registered in the International Transplant Registry in Chicago. Seventeen investigators returned the questionnaire and reported 13 instances of bronchial complications, including ulceration, necrosis, and disruption.
In 1973, Haglin and associates [10] reported a case of direct left bronchial artery revascularization of a left lung using a method almost the same as described earlier by Mills and colleagues [36]. In addition, the right lung was transplanted without bronchial revascularization after an ischemic time of 11 hours. The patient died of septicemia and pulmonary insufficiency on the 11th postoperative day. At autopsy, the left (revascularized) bronchial wall remained viable, whereas the right (nonrevascularized) bronchial wall showed extensive necrosis.
As mentioned earlier, Couraud and associates [11] were the first to publish a series of patients having en bloc double-lung transplantation with bronchial artery revascularization. Organ procurement was done according to the method described by Schreinemakers and colleagues [6]. Couraud and associates included a tracheal anastomosis. The bronchial artery revascularization was done with a saphenous vein graft from the recipient's ascending aorta to an aortic patch including the origin of the right intercostobronchial artery and, in addition in a few cases, one or more left bronchial arteries. The distal anastomosis of the saphenous vein graft was done on a side table before introducing the donor lungs into the recipient. The proximal saphenous vein to recipient aorta anastomosis was done at the end of the operation. It was stressed that dissection around the carina and individual mobilization of the bronchial arteries should be avoided so as not to interfere with the bronchial blood supply.
The additional ischemic time needed for performing the bronchial artery revascularization was estimated as 1 hour. The bronchial blood supply was evaluated by repeated endoscopic studies and in 7 patients by arteriography 2 weeks postoperatively. The arteriography results showed an open vein graft with visualization of bronchial arteries in 5 patients, whereas 2 showed thrombotic occlusion of the vein graft. Normal tracheal healing was observed in all patients. One patient died of gastrointestinal hemorrhage on day 30, and 1 died of a fulminant Pseudomonas aeruginosa pneumonia on day 35.
Later in 1992, Couraud and associates [39] enlarged their clinical experience by 3 patients, 2 having en bloc double and 1 having right single lung transplantation. No patients had anastomosis healing problems or stenosis. In 2 of the additional patients, arteriography results showed a functional saphenous vein graft and visualization of the bronchial arteries. One patient died of hepatorenal failure on day 16.
In 1993, Daly and colleagues [12] published the Harefield series of 8 patients who had nine en bloc double-lung transplantations with bronchial artery revascularization. The method of organ procurement was modified from those described previously [6, 9]. The bronchial arteries were identified and confirmed by probing. The largest vessel going in the direction of the carina was chosen for revascularization with the left internal mammary artery. The mammary to bronchial artery anastomosis was done at the end of the operation. The orifice of the bronchial artery was brought into view by pulling on the proximal end of the donor aorta. Orifices in the donor descending aorta producing back-bleeding were oversewn. The total additional time needed to perform the bronchial artery revascularization was estimated to be 30 to 60 minutes, and the donor organ ischemic time was prolonged by 20 to 22 minutes.
The bronchial artery blood supply was evaluated by repeated bronchoscopic examinations, and arteriography was performed 12 days to 2 months postoperatively. In 1 patient with angiographic evidence of bronchial artery occlusion, a large ulcer developed in the tracheal anastomosis. In all other patients, the internal mammary artery graft was patent, with good bronchial artery perfusion and normal healing, including 1 patient who had retransplantation after 1 month because of pulmonary emboli. The internal mammary artery graft was chosen instead of a saphenous vein graft because of expected better long-term patency.
In 1994, Daly and colleagues [15] published the results of direct bronchial artery revascularization in 10 single-lung transplantations from the Mayo Clinic. The additional ischemic time for the revascularization procedure was estimated as 15 to 20 minutes. One patient died perioperatively of donor organ dysfunction. In the surviving 9 patients, angiography results demonstrated excellent perfusion of the bronchial arteries in 7 and no perfusion in 2. Bronchial healing was excellent in all. These investigators demonstrated that it is possible to perform bronchial artery revascularization of two single lungs from the same donor by dividing the donor aorta and the mediastinal tissue between two bronchial arteries.
McGregor and co-workers [16], also in 1994, documented 4 patients who had bronchial artery revascularization among the first 15 lung transplant recipients at the Mayo Clinic having different procedures for bronchial anastomosis protection. We presume that these 4 patients represent part of the group of 10 patients published earlier [15].
In 1994, we published the results of our first 14 en bloc double lung transplantations with bronchial artery revascularization [13] using the left internal thoracic artery. The method used was a modification of that described by Daly and colleagues [12].
Our own experience with bronchial artery revascularization now includes 63 patients: 49 double lung, 5 single lung, and 9 heart lung transplantations. Preparation of the donor organs for revascularization adds about 15 to 20 minutes of ischemic time. On one occasion, we performed bronchial artery revascularization of two single lungs from the same donor by dividing the block so that the bronchial arteries were present for both sides. Our ambition is to achieve complete bronchial artery revascularization, ie, anastomoses to all identified bronchial arteries. When more than one artery is revascularized, one common or sequential mammary anastomoses are performed. Each anastomosis takes about 10 minutes or less. After the block has been introduced, the bronchial artery revascularization is performed first. In en bloc double lung transplantation, the left lung is hung backwards over the heart, which gives excellent exposure of the bronchial artery openings in the donor descending aorta. Immediate bronchial reperfusion is begun and maintained during the rest of the procedure. We have encountered anastomotic bleeding on three occasions. In one instance, the extra hemostatic sutures resulted in occlusion of the anastomosis. The problem was diagnosed immediately by arteriography. The patient underwent reoperation, and the bronchial artery revascularization was successfully reestablished.
The arterial blood supply of the bronchial tree is evaluated by repeated endoscopic and arteriographic studies. In the first double lung series, 10 of 12 patients showed successful revascularization at angiography, and excellent tracheal healing was observed. In 2 patients, the internal thoracic artery was open but there was no communication to the bronchial arteries. Both patients experienced submucosal necrosis in the central bronchi and subsequent stenosis of the left main bronchus; eventually a left-side pneumonectomy had to be performed. Early survival was 100%.
Our results show that of the 63 patients in whom bronchial artery revascularization was attempted, 53 had angiography performed 1 day to 5 months postoperatively, showing successful revascularization in 50 (in 1 after reoperation for anastomotic occlusion) and failed revascularization in 3 (including the 2 patients in the early series). Nine of 10 patients not examined by arteriography for various reasons had normal bronchial healing, whereas 1 had poor bronchial healing, suggesting failed revascularization. Sixteen patients have been reexamined after 2 years, showing 100% long-term patency of the bronchial artery revascularization. Using the Kaplan-Meier method (and the International Society for Heart and Lung Transplantation working formulation [40]), the proportion of patients having double lung transplantation with bronchial artery revascularization who were free from bronchiolitis obliterans syndrome after 1, 2, and 3 years was, respectively, 85%, 82%, and 67%. One of 6 patients having single lung transplantation with bronchial artery revascularization had bronchiolitis obliterans syndrome after 12 months. None of our 9 patients having heart lung transplantation with bronchial artery revascularization have so far developed bronchiolitis obliterans syndrome.
Table 1
summarizes the published results of lung transplantation with bronchial artery revascularization.
|
| Comment |
|---|
|
|
|---|
Our interpretation of available studies is that it seems unlikely that the bronchial arteries can be neglected during transplantation without adverse consequences, ie, the recipient will have lungs with a severed physiologic response to pathologic conditions as a result of the absence of the bronchial artery blood supply.
In the early era of lung transplantation, healing of the airway anastomosis was a major concern. The method used for anastomotic protection and airway revascularization in the first successful human lung transplantations, single as well as en bloc double, was to wrap the airway anastomosis with omentum [41]. Wrapping the anastomosis with omentum has been shown to enhance early ingrowth of vessels around the airway anastomosis [42]. For single-lung transplantation, this method produced good early results, whereas double lung transplantation performed en bloc with a tracheal anastomosis was still associated with a high incidence of tracheal anastomosis and bronchial problems [43]. Based on these observationssuggesting that the intrapulmonary part of the bronchial tree is less dependent on the arterial blood supply than are the main bronchi and tracheathe technique of performing double-lung transplantation as sequential bilateral lung transplantation was introduced [44]. This method has been associated with a low rate of bronchial complications and good early results [45]. With improved technique, the incidence of airway problems today is very low, even without wrapping the bronchial anastomoses. This method is, however, associated with a prolonged ischemic time, especially for the second lung.
Although the logical method to prevent bronchial and lung ischemiadirect bronchial artery revascularizationwas introduced experimentally by Metras already in 1950 [5] and clinically by Couraud and associates [11, 39] in 1992, the Harefield group in 1993 [12], the Mayo Clinic in 1994 [15], and the Copenhagen group in 1994 [13, 14], most lung transplant surgeons have not yet decided to use the method. The main arguments for not performing bronchial artery revascularization have been as follows: (1) good early results without bronchial artery revascularization, (2) lack of evidence of the effectiveness of the method, (3) prolonged operation time and prolonged ischemic time, and (4) increased risk of bleeding.
Until this publication, 33 en bloc double-lung and 11 single-lung transplantations with bronchial artery revascularization have been published. Concerning early results and effectiveness of the method, 28 of 33 reported patients with bronchial artery revascularization during en bloc double lung transplantations were studied by angiography. Twenty-three showed good revascularization of the bronchial arteries, and none of these patients had airway healing problems. In 5, the bronchial artery revascularization had failed; 3 of these had airway healing problems including necrosis, and 2 had subsequent development of severe bronchial stenosis. As mentioned earlier, we can add 31 en bloc double lung transplantation patients, in whom angiography verified successful revascularization in 30 and failure in 1. This failure was in a retransplantation patient who died of pneumonia 30 days postoperatively.
Ten of the 11 reported patients with single lung transplantations were checked by arteriography. Two revascularization procedures had failed. All patients healed their bronchial anastomoses without complications. We can add five single-lung procedures, with successful revascularization verified by angiography in four. In the last patient, angiography was not performed. All patients healed without airway complications.
We also have added nine bronchial artery revascularizations in heart lung transplantations. Successful revascularization was verified by arteriography in six. Arteriography was not performed in the remaining three. Tracheal anastomosis healing was normal in all.
Prolongation of the ischemic time required for performing the bronchial artery revascularization has been 20 to 60 minutes, depending on the method used. With our method, the ischemic time is shortened rather than prolonged because early bronchial artery reperfusion is allowed.
When calculating the ischemic time involved in performing bronchial artery revascularization, it is necessary to compare the time used for this procedure with the increased surgical trauma and the potential risk of abdominal complications involved with wrapping the anastomosis with omentum. Bilateral sequential lung transplantation using a clamshell incision takes longer than en bloc double-lung transplantation including bronchial artery revascularization.
Postoperative bleeding as a consequence of bronchial artery revascularization has been rare, and in all instances has been controllable.
| Conclusion |
|---|
|
|
|---|
However, airway healing is not the only issue. Long-term results after lung transplantation are still not convincing, and the early results after attempted direct bronchial artery revascularization are definitely good enough to justify its use in large series to allow studies of long-term results.
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G Meachery, A De Soyza, A Nicholson, G Parry, A Hasan, K Tocewicz, T Pillay, S Clark, J L Lordan, S Schueler, et al. Outcomes of lung transplantation for cystic fibrosis in a large UK cohort Thorax, August 1, 2008; 63(8): 725 - 731. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Macedo, E. Fadel, G.-M. Mazmanian, V. de Montpreville, M. German-Fattal, S. Mussot, A. Chapelier, and P. G. Dartevelle Heterotopic en bloc tracheobronchial transplantation with direct revascularization in pigs J. Thorac. Cardiovasc. Surg., June 1, 2004; 127(6): 1593 - 1601. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Dodd-o, L. E. Welsh, J. D. Salazar, P. L. Walinsky, E. A. Peck, J. G. Shake, D. J. Caparrelli, B. T. Bethea, S. M. Cattaneo, W. A. Baumgartner, et al. Effect of bronchial artery blood flow on cardiopulmonary bypass-induced lung injury Am J Physiol Heart Circ Physiol, February 1, 2004; 286(2): H693 - H700. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Schroder, F. Scholl, E. Daon, A. Goodwin, W. H. Frist, J. R. Roberts, K. G. Christian, M. Ninan, A. P. Milstone, J. E. Loyd, et al. A modified bronchial anastomosis technique for lung transplantation Ann. Thorac. Surg., June 1, 2003; 75(6): 1697 - 1704. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-P. Brouland, T. Egan, J. Roussi, M. Bonneau, G. Pignaud, C. Bal, M. Vaiman, P. Andre, P. Herve, G. M. Mazmanian, et al. In Vivo Regulation of von Willebrand Factor Synthesis : von Willebrand Factor Production in Endothelial Cells After Lung Transplantation Between Normal Pigs and von Willebrand Factor-Deficient Pigs Arterioscler Thromb Vasc Biol, December 1, 1999; 19(12): 3055 - 3062. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. Herold, H. Jakob, M. Kamler, R. Thiele, U. Tochtermann, J. Weinmann, J. Motsch, M. M. Gebhard, and S. Hagl Interruption of bronchial circulation leads to a severe decrease in peribronchial oxygen tension in standard lung transplantation technique Eur J Cardiothorac Surg, February 1, 1998; 13(2): 176 - 183. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |