- Email: [email protected]
Technical Pearls for Swine Lung Transplantation
Journal of Surgical Research 171, e107–e111 (2011) doi:10.1016/j.jss.2011.05.067
Technical Pearls for Swine Lung Transplantation Ashkan Karimi, M.D.,* Jessica A. Cobb, Ph.D.,* Edward D. Staples, M.D.,* Maher A. Baz, M.D.,† and Thomas M. Beaver, M.D., M.P.H.*,1 *The Division of Thoracic and Cardiovascular Surgery; and †The Division of Pulmonary, Critical Care, and Sleep Medicine, University of Florida, Gainesville, Florida Submitted for publication April 21, 2011
Background. Since the advent of ex vivo lung perfusion (EVLP), there has been increased focus on swine models of lung transplantation; however, the anatomic differences between human and swine lungs and the technical challenges in performing porcine lung transplantation are not well described in the surgical literature. Methods. Surgically important anatomic variations are described, and the technical measures taken to address them during harvest and transplantation are introduced. Results. There are three surgically important anatomic variations in pigs. First, the right cranial lobe bronchus arises directly from the trachea, which makes right lung transplantation technically challenging if not prohibitive. Second, the left hemi-azygos vein is fully developed and courses upward through the posterior mediastinum, where it crosses the left pulmonary hilum and drains directly into the coronary sinus. During transplantation, this vein is ligated and dissected away to expose the underlying left pulmonary hilar structures. Third, the right inferior pulmonary vein crosses the midline to drain into the left atrium immediately adjacent to the left inferior pulmonary vein. During donor lung preparation, the right inferior pulmonary vein is ligated distally from the left atrium, which leaves an adequate atrial cuff around the left sided pulmonary veins for later anastomosis. Conclusion. Experimental porcine lung transplantation is technically demanding. We have found recognition of the above described anatomical differences and technical nuances facilitate transplantation and provide reproducible results. Ó 2011 Elsevier Inc. All rights
Key Words: animal model; lung transplantation; ex vivo lung perfusion.
INTRODUCTION
Lung transplantation remains the only viable option for end stage pulmonary disease; however, because of a shortage of donors, many patients are deprived of this life saving treatment. Since the introduction of ex vivo lung perfusion (EVLP) for rehabilitation of marginal donor lungs, many centers have employed swine models to gain experience and to initiate experimental protocols [1, 2]. EVLP has also been used for drug and gene delivery to the donor lungs [3], and its usage is most likely to expand in laboratory and clinical settings in near future. Because of the similarity to humans, the porcine model of lung transplantation has been mainly used in such experiments; however, the technique of swine lung transplantation is not clearly described in the surgical literature. The purpose of this article is to highlight the important anatomical differences between human and swine anatomy and to share technical nuances in performing swine transplantation. METHODS We discovered anatomical differences between swine and human lungs that made performing experimental lung transplantation challenging. Herein, we describe these differences and introduce the technical measures we took to address them during harvest and transplantation.
reserved.
RESULTS 1 To whom correspondence and reprint requests should be addressed at Division of Thoracic and Cardiovascular Surgery, University of Florida, Room NG 32, 1600 Archer Road, Gainesville, FL 32610-0129. E-mail: [email protected].
Surgically important aspects of swine lung anatomy are described, followed by a detailed explanation of our harvest and left lung transplantation techniques.
e107
0022-4804/$36.00 Ó 2011 Elsevier Inc. All rights reserved.
e108
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
Swine Lung Anatomy
The trachea has a length of 15–20 cm, contains 32–45 rings and divides into right and left main bronchus at the level of fifth thoracic vertebrae [4]. The lobes of the right lung are designated as the cranial, middle, caudal, and accessory and the lobes of the left lung as the middle and the caudal lobe [5]. In pigs, unlike humans, the right cranial lobe bronchus, also called the tracheal bronchus, arises directly from the trachea before its bifurcation (Fig. 1) [5]. This anatomic variation makes right lung transplantation technically difficult and is avoided. On the right side, the cranial and the middle lobes are separated by a deep cardiac notch and an incomplete fissure separates the middle lobe from the caudal lobe. The right accessory lobe is situated between the base of the heart and the diaphragm and encircles the terminal intrathoracic portion of the inferior vena cava [4]. On the left side, there are two
lobes, which are the middle and the caudal lobe and an incomplete fissure separates them. The left lung has no cranial lobe and no tracheal bronchus [4]. According to the Nakakuki classification of the bronchial tree, the pig lung has the dorsal (D), lateral (L), ventral (V), and medial (M) bronchiole systems that are named according to their orientation as they come off the main bronchus [5]. The right middle lobe is ventilated by the first bronchiole of the lateral group (L1), the accessory lobe by the first bronchiole of the ventral group (V1) and the remaining bronchioles of the dorsal, lateral and ventral bronchiole systems ventilate the right caudal lobe together with the bronchioles from the medial system. On the left side, the middle lobe receives the first branch from the lateral system (L1), and the caudal lobe basically receives the rest of the branches that come off from the left main bronchus (Fig. 1). The pulmonary artery divides into right and left branches that run along the dorsolateral side of the corresponding main bronchus and divide in concordance with the rest of the bronchial tree. The pulmonary veins are located ventromedial to the bronchial structures. Pulmonary veins from right cranial and middle lobes converge to form the right superior pulmonary vein and the pulmonary vein from right caudal lobe forms the right inferior pulmonary vein (RIPV), which also receives venous drainage from the right accessory lobe. On the left side, the pulmonary vein from the middle lobe forms the left superior pulmonary vein (LSPV), and the vein from the caudal lobe forms the left inferior pulmonary vein (LIPV). Importantly, the RIPV enters the left atrium in the midline adjacent to the LIPV. The left middle lobe is less developed than caudal lobe and correspondingly the LSPV is much smaller than LIPV [6]. The systemic vascular system is also different in pigs. The left hemi-azygos vein courses next to the descending aorta in the posterior mediastinum and then crosses the left pulmonary hilum to drain directly into the coronary sinus and right atrium. The importance of these anatomic variations is further discussed below in the surgical technique section of this article. Donor Lung Procurement
FIG. 1. Schematic view of the bronchial tree: D ¼ dorsal; L ¼ lateral; V ¼ ventral; M ¼ medial bronchial system; subscripts shows the order they come off from the main bronchus. III ¼ tracheal bronchus (right cranial lobe bronchus); a ¼ apical; b ¼ basal segment. L1 ¼ first branch from lateral system (right/left middle lobe bronchus); V1 ¼ first branch from ventral system (right accessory lobe bronchus). The rest of the bronchioles from dorsal, lateral and ventral systems plus the bronchioles from the medial system ventilate the caudal lobes. RPA ¼ right pulmonary artery; LPA ¼ left pulmonary artery. Source: Courtesy of Nakakuki S., Department of Veterinary Anatomy, Tokyo University of Agriculture and Technology, Japan.
Under an approved Institutional Animal Care and Use Committee (IACUC) protocol and standard general anesthesia procurement is performed through an almost complete sternotomy. The Manubrium is relatively thick in pigs and difficult to cut with vascular structures directly below it; however, it does not need to be divided entirely for adequate exposure. The pleural cavities are entered on both sides and the inferior pulmonary ligaments are released with careful attention to avoid damaging lung parenchyma. Silk sutures are placed around the superior and inferior vena cava
KARIMI ET AL.: SWINE LUNG TRANSPLANTATION TECHNIQUE
and the left hemi-azygos vein is ligated as it drains into the coronary sinus. Ligation of the left hemi-azygos vein allows a bloodless field during administration of preservation solution, but this is not absolutely necessary. Next a 4-0 Prolene (Ethicon Inc., New Brunswick, NJ) purse-string suture is placed in the main pulmonary artery. A bolus of 15,000 units of IV heparin (300 units per kilogram) is given; and the pulmonary artery is cannulated with a 21F arterial cannula and hooked to the preservation solution tubing, which has been flushed of air. We employ Perfadex (Vitrolife, G€ oteborg, Sweden) w50 mL/kg and use a 3.3 mmol/mL THAM solution (Vitrolife, G€ oteborg, Sweden) to adjust the pH to 7.4. A bolus of 500 mcg of prostaglandin E1 (Pharmacia & Upjohn, Bridgewater, NJ) is injected into the main pulmonary artery and then the superior and inferior vena cava are ligated as far away from the heart as feasible. After Perfadex administration is complete, the ascending aorta is cross-clamped, and the trachea is stapled under positive airway pressure of 20–25 mmHg. The rest of the posterior mediastinal attachments are released and the heart-lung bloc is explanted. On the back table, the heart is excised with careful attention to leave behind an adequate margin of left atrium. The lung bloc is then stored on ice at 4 C with a small amount of Perfadex solution in the bag. Left Lung Transplantation
The recipient pig is anesthetized and placed in the right lateral decubitus position. The pig’s left front leg is fixed in a fully extended position and a standard left anterolateral thoracotomy incision is performed one finger breadth below the tip of scapula and extended ventrocaudally along the rib. In our experience, a wide incision gives adequate exposure with no need to excise a rib; however, caution should be exercised in opening the retractor too far as bleeding can ensue from collateral branches that feed the hemi-azygos vein. Upon entrance into the chest cavity, the inferior pulmonary ligament is released and the left lung is moved into the field, allowing dissection of the left hilum. The left hemi-azygos vein that crosses ventral to pulmonary vessels is ligated. After adequate careful exposure of the left hilar structures, an umbilical tape can be placed around the right pulmonary artery for later clamping or ligation as some studies desire to assess recipient pig’s performance solely on the left transplanted lung. The donor lung is prepared on the back table. In pigs, unlike humans, the RIPV and LIPV orifices are located next to each other as they enter the left atrium (Fig. 2). For the donor venous pedicle preparation, the RIPV is ligated distally from the left atrium that leaves an adequate atrial cuff around the left sided pulmonary veins.
e109
FIG. 2. Donor lung bloc on the back table with the heart excised except for the left atrial cuff: long arrow ¼ right inferior pulmonary vein crossing the midline to drain into left atrium immediately adjacent to left inferior pulmonary vein; short arrow ¼ left atrial cuff.
Subsequently, the left pulmonary artery and bronchus are divided away from the hilum and the donor lung is brought into the field for transplantation. Because there is no double lumen endotracheal tube for single lung ventilation and because there is limited space we cross-clamp the left hilum all at once with a Satinsky side biting clamp. Caution should be used to avoid the RIPV to allow it to continue to drain into the left atrium. Thereafter, the left pulmonary artery, left bronchus, and LIPV are transected distally and the recipient’s lung is excised. Due to the size of the hilar vessels, magnifying loupes are employed. The left atrium, bronchus, and artery are trimmed appropriately. The left pulmonary veins near the atrium are thin-walled and friable and need to be treated carefully. Furthermore, the LSPV is diminutive and fragile in pigs, and, therefore, we ligate this vein immediately before or just after applying the hilar cross-clamp, not earlier to avoid left middle lobe congestion. We have found it easier to use the LIPV with extension into the adjacent left atrium as the recipient pulmonary venous pedicle (Fig. 3) [6]. Importantly, in our technique the order of anastomosis is different from humans. We perform the venous anastomosis first as it has a very short stump that would make it inaccessible if the other anastomoses were performed first; it is completed with a running 4-0 Prolene RB-1 needle (Ethicon Inc.) imbricating the recipient and donor left atria. Again it is friable and should be treated carefully. The bronchial anastomosis is completed next with a running 3-0 or 4-0 Prolene SH needle (Ethicon Inc.). Lastly, the pulmonary arterial anastomosis is completed with a running 4-0 Prolene RB-1 needle (Ethicon Inc.). Prior to reperfusion, a small vascular clamp is placed on the recipient side of the pulmonary arterial anastomosis and the hilar cross-clamp is released de-airing the donor lung in a retrograde
e110
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
FIG. 3. This image shows the recipient left hilum before transplantation (see text for technical details): long arrow ¼ right inferior pulmonary vein; short arrow ¼ left inferior pulmonary vein; arrow head ¼ left hemi-azygos vein divided as it enters into the coronary sinus; asterisk ¼ divided left superior pulmonary vein; LPA ¼ left pulmonary artery; LB ¼ left bronchus.
manner, and employing a 21-guage needle to further de-air the pulmonary artery as necessary. Following reperfusion atelectasis resolves slowly over time. Arterial blood gases can be obtained directly from the pulmonary vein of the transplanted lung, and tissue samples are obtained from the lung parenchyma according to protocol. DISCUSSION
Swine is the most commonly used large animal model of lung transplantation; often in studies of drug effects on ischemia/reperfusion injury [3, 7–10], EVLP [1, 3] and in xenotransplantation [11, 12]. An adequate understanding of pig lung anatomy is important for those beginning experimental protocols. In our review of the literature, we could not find an article dedicated to the technique of swine lung transplantation; nevertheless, several articles briefly describe their donor and recipient preparation [1, 7–10, 12]. In our described model, a single stage cross-clamp is applied to the left pulmonary hilum and obviates the need for sequential clamping. We also differ in the order of performing the anastomoses as venous is performed first, followed by bronchial and arterial. Others have suggested ligation of the left pulmonary veins, followed by clamping of the left bronchus and pulmonary artery and cutting them distally for left pneumonectomy; with subsequent performance of the bronchial and arterial anastomoses in sequence and then clamping the recipient’s left atrium that is sewed to the donor’s left atrial cuff [8, 10]. Functional assessment of the transplanted lung is an important part of experimental protocols. Oxygenation
capacity (PaO2/FiO2) with arterial blood gases, pulmonary artery pressures and vascular resistance, and dynamic and static lung compliance are used to monitor the transplanted lung performance during reperfusion [1, 7–12]. There are three approaches to monitor the transplanted lung. Some authors do not exclude the recipient’s native right lung from the circulation or ventilation during assessments [7]. Others prefer to ligate the right pulmonary artery shortly after reperfusion so that the recipient pig is solely dependent on the transplanted lung during assessments [9, 10, 12]; however, it is unlikely that pigs will tolerate reperfusion times in excess of 1-2 h following right pulmonary artery ligation as there is then massive shunting of the entire cardiac output to the donor lung, which has already sustained endothelial damage during cold ischemia, with subsequent lung edema, hypoxemia, and death [9]. Some xenotransplantation models employing right pulmonary artery ligation have 100% mortality in the control group within 20 s of occlusion [12]; however, in an allotransplantation model, Pizanis et al. have reported 100% (n ¼ 7) survival in the control group during 6 h of reperfusion with the right pulmonary artery and bronchus ligated 10–20 min after reperfusion [12]. An alternative approach is continue ventilating and perfusing both lungs with sampling blood gases only from the left pulmonary veins to assess transplanted lung oxygenation capacity during reperfusion (usually 4–6 h) [1, 8]. The right pulmonary artery can then be ligated 15–30 min before a final blood gas is obtained to assess the recipient’s dependence solely on the transplanted lung [1, 8]. We would recommend the third approach as it is likely to produce a more predictable model and allows for functional assessment of the transplanted lung over longer reperfusion times. Of note, as described earlier, the left and right inferior pulmonary veins drain next to each other into the left atrium. Accordingly, samples from the left pulmonary vein should be obtained as close as possible to the lung parenchyma to avoid mixing of its blood with venous return from the recipient’s native right lung. CONCLUSION
Due to the size of the porcine hilar vessels, lung transplantation can be challenging, and both magnifying loupes and surgical experience are required. We have found the above described technical nuances facilitate transplantation and provide reproducible results. REFERENCES 1. Cypel M, Yeung JC, Hirayama S, et al. Technique for prolonged normothermic ex vivo lung perfusion. J Heart Lung Transplant 2008;27:1319.
KARIMI ET AL.: SWINE LUNG TRANSPLANTATION TECHNIQUE 2. Steen S, Liao Q, Wierup PN, et al. Transplantation of lungs from non-heart-beating donors after functional assessment ex vivo. Ann Thorac Surg 2003;76:244. 3. Cypel M, Liu M, Rubacha M, et al. Functional repair of human donor lungs by IL-10 gene therapy. Sci Transl Med 2009; 1:4ra9. 4. Dondelinger RF, Ghysels MP, Brisbois D, et al. Relevant radiological anatomy of the pig as a training model in interventional radiology. Eur Radiol 1998;8:1254. 5. Nakakuki S. Bronchial tree, lobular division and blood vessels of the pig lung. J Vet Med Sci 1994;56:685. 6. Swindle M. Swine in the Laboratory. 2nd ed. Boca Raton, FL: Taylor and Francis Group, 2007:213–214. 7. Gomez CB, del Valle HF, Bertolotti A, et al. Effects of shortterm inhaled nitric oxide on interleukin-8 release after single-lung transplantation in pigs. J Heart Lung Transplant 2005;24:714.
e111
8. Reece TB, Ellman PI, Maxey TS, et al. Adenosine A2A receptor activation reduces inflammation and preserves pulmonary function in an in vivo model of lung transplantation. J Thorac Cardiovasc Surg 2005;129:1137. 9. Korom S, Hillinger S, Cardell M, et al. Sildenafil extends survival and graft function in a large animal lung transplantation model. Eur J Cardiothorac Surg 2006;29:288. 10. Pizanis N, Milekhin V, Tsagakis K, et al. PDE-5 inhibitor donor intravenous preconditioning is superior to supplementation in standard preservation solution in experimental lung transplantation. Eur J Cardiothorac Surg 2007;32:42. 11. Gaca JG, Lesher A, Aksoy O. Disseminated intravascular coagulation in association with pig-to-primate pulmonary xenotransplantation. Transplantation 2002;73:1717. 12. Daggett CW, Yeatman M, Lodge AJ, et al. Total respiratory support from swine lungs in primate recipients. J Thorac Cardiovasc Surg 1998;115:19.
Technical Pearls for Swine Lung Transplantation Ashkan Karimi, M.D.,* Jessica A. Cobb, Ph.D.,* Edward D. Staples, M.D.,* Maher A. Baz, M.D.,† and Thomas M. Beaver, M.D., M.P.H.*,1 *The Division of Thoracic and Cardiovascular Surgery; and †The Division of Pulmonary, Critical Care, and Sleep Medicine, University of Florida, Gainesville, Florida Submitted for publication April 21, 2011
Background. Since the advent of ex vivo lung perfusion (EVLP), there has been increased focus on swine models of lung transplantation; however, the anatomic differences between human and swine lungs and the technical challenges in performing porcine lung transplantation are not well described in the surgical literature. Methods. Surgically important anatomic variations are described, and the technical measures taken to address them during harvest and transplantation are introduced. Results. There are three surgically important anatomic variations in pigs. First, the right cranial lobe bronchus arises directly from the trachea, which makes right lung transplantation technically challenging if not prohibitive. Second, the left hemi-azygos vein is fully developed and courses upward through the posterior mediastinum, where it crosses the left pulmonary hilum and drains directly into the coronary sinus. During transplantation, this vein is ligated and dissected away to expose the underlying left pulmonary hilar structures. Third, the right inferior pulmonary vein crosses the midline to drain into the left atrium immediately adjacent to the left inferior pulmonary vein. During donor lung preparation, the right inferior pulmonary vein is ligated distally from the left atrium, which leaves an adequate atrial cuff around the left sided pulmonary veins for later anastomosis. Conclusion. Experimental porcine lung transplantation is technically demanding. We have found recognition of the above described anatomical differences and technical nuances facilitate transplantation and provide reproducible results. Ó 2011 Elsevier Inc. All rights
Key Words: animal model; lung transplantation; ex vivo lung perfusion.
INTRODUCTION
Lung transplantation remains the only viable option for end stage pulmonary disease; however, because of a shortage of donors, many patients are deprived of this life saving treatment. Since the introduction of ex vivo lung perfusion (EVLP) for rehabilitation of marginal donor lungs, many centers have employed swine models to gain experience and to initiate experimental protocols [1, 2]. EVLP has also been used for drug and gene delivery to the donor lungs [3], and its usage is most likely to expand in laboratory and clinical settings in near future. Because of the similarity to humans, the porcine model of lung transplantation has been mainly used in such experiments; however, the technique of swine lung transplantation is not clearly described in the surgical literature. The purpose of this article is to highlight the important anatomical differences between human and swine anatomy and to share technical nuances in performing swine transplantation. METHODS We discovered anatomical differences between swine and human lungs that made performing experimental lung transplantation challenging. Herein, we describe these differences and introduce the technical measures we took to address them during harvest and transplantation.
reserved.
RESULTS 1 To whom correspondence and reprint requests should be addressed at Division of Thoracic and Cardiovascular Surgery, University of Florida, Room NG 32, 1600 Archer Road, Gainesville, FL 32610-0129. E-mail: [email protected].
Surgically important aspects of swine lung anatomy are described, followed by a detailed explanation of our harvest and left lung transplantation techniques.
e107
0022-4804/$36.00 Ó 2011 Elsevier Inc. All rights reserved.
e108
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
Swine Lung Anatomy
The trachea has a length of 15–20 cm, contains 32–45 rings and divides into right and left main bronchus at the level of fifth thoracic vertebrae [4]. The lobes of the right lung are designated as the cranial, middle, caudal, and accessory and the lobes of the left lung as the middle and the caudal lobe [5]. In pigs, unlike humans, the right cranial lobe bronchus, also called the tracheal bronchus, arises directly from the trachea before its bifurcation (Fig. 1) [5]. This anatomic variation makes right lung transplantation technically difficult and is avoided. On the right side, the cranial and the middle lobes are separated by a deep cardiac notch and an incomplete fissure separates the middle lobe from the caudal lobe. The right accessory lobe is situated between the base of the heart and the diaphragm and encircles the terminal intrathoracic portion of the inferior vena cava [4]. On the left side, there are two
lobes, which are the middle and the caudal lobe and an incomplete fissure separates them. The left lung has no cranial lobe and no tracheal bronchus [4]. According to the Nakakuki classification of the bronchial tree, the pig lung has the dorsal (D), lateral (L), ventral (V), and medial (M) bronchiole systems that are named according to their orientation as they come off the main bronchus [5]. The right middle lobe is ventilated by the first bronchiole of the lateral group (L1), the accessory lobe by the first bronchiole of the ventral group (V1) and the remaining bronchioles of the dorsal, lateral and ventral bronchiole systems ventilate the right caudal lobe together with the bronchioles from the medial system. On the left side, the middle lobe receives the first branch from the lateral system (L1), and the caudal lobe basically receives the rest of the branches that come off from the left main bronchus (Fig. 1). The pulmonary artery divides into right and left branches that run along the dorsolateral side of the corresponding main bronchus and divide in concordance with the rest of the bronchial tree. The pulmonary veins are located ventromedial to the bronchial structures. Pulmonary veins from right cranial and middle lobes converge to form the right superior pulmonary vein and the pulmonary vein from right caudal lobe forms the right inferior pulmonary vein (RIPV), which also receives venous drainage from the right accessory lobe. On the left side, the pulmonary vein from the middle lobe forms the left superior pulmonary vein (LSPV), and the vein from the caudal lobe forms the left inferior pulmonary vein (LIPV). Importantly, the RIPV enters the left atrium in the midline adjacent to the LIPV. The left middle lobe is less developed than caudal lobe and correspondingly the LSPV is much smaller than LIPV [6]. The systemic vascular system is also different in pigs. The left hemi-azygos vein courses next to the descending aorta in the posterior mediastinum and then crosses the left pulmonary hilum to drain directly into the coronary sinus and right atrium. The importance of these anatomic variations is further discussed below in the surgical technique section of this article. Donor Lung Procurement
FIG. 1. Schematic view of the bronchial tree: D ¼ dorsal; L ¼ lateral; V ¼ ventral; M ¼ medial bronchial system; subscripts shows the order they come off from the main bronchus. III ¼ tracheal bronchus (right cranial lobe bronchus); a ¼ apical; b ¼ basal segment. L1 ¼ first branch from lateral system (right/left middle lobe bronchus); V1 ¼ first branch from ventral system (right accessory lobe bronchus). The rest of the bronchioles from dorsal, lateral and ventral systems plus the bronchioles from the medial system ventilate the caudal lobes. RPA ¼ right pulmonary artery; LPA ¼ left pulmonary artery. Source: Courtesy of Nakakuki S., Department of Veterinary Anatomy, Tokyo University of Agriculture and Technology, Japan.
Under an approved Institutional Animal Care and Use Committee (IACUC) protocol and standard general anesthesia procurement is performed through an almost complete sternotomy. The Manubrium is relatively thick in pigs and difficult to cut with vascular structures directly below it; however, it does not need to be divided entirely for adequate exposure. The pleural cavities are entered on both sides and the inferior pulmonary ligaments are released with careful attention to avoid damaging lung parenchyma. Silk sutures are placed around the superior and inferior vena cava
KARIMI ET AL.: SWINE LUNG TRANSPLANTATION TECHNIQUE
and the left hemi-azygos vein is ligated as it drains into the coronary sinus. Ligation of the left hemi-azygos vein allows a bloodless field during administration of preservation solution, but this is not absolutely necessary. Next a 4-0 Prolene (Ethicon Inc., New Brunswick, NJ) purse-string suture is placed in the main pulmonary artery. A bolus of 15,000 units of IV heparin (300 units per kilogram) is given; and the pulmonary artery is cannulated with a 21F arterial cannula and hooked to the preservation solution tubing, which has been flushed of air. We employ Perfadex (Vitrolife, G€ oteborg, Sweden) w50 mL/kg and use a 3.3 mmol/mL THAM solution (Vitrolife, G€ oteborg, Sweden) to adjust the pH to 7.4. A bolus of 500 mcg of prostaglandin E1 (Pharmacia & Upjohn, Bridgewater, NJ) is injected into the main pulmonary artery and then the superior and inferior vena cava are ligated as far away from the heart as feasible. After Perfadex administration is complete, the ascending aorta is cross-clamped, and the trachea is stapled under positive airway pressure of 20–25 mmHg. The rest of the posterior mediastinal attachments are released and the heart-lung bloc is explanted. On the back table, the heart is excised with careful attention to leave behind an adequate margin of left atrium. The lung bloc is then stored on ice at 4 C with a small amount of Perfadex solution in the bag. Left Lung Transplantation
The recipient pig is anesthetized and placed in the right lateral decubitus position. The pig’s left front leg is fixed in a fully extended position and a standard left anterolateral thoracotomy incision is performed one finger breadth below the tip of scapula and extended ventrocaudally along the rib. In our experience, a wide incision gives adequate exposure with no need to excise a rib; however, caution should be exercised in opening the retractor too far as bleeding can ensue from collateral branches that feed the hemi-azygos vein. Upon entrance into the chest cavity, the inferior pulmonary ligament is released and the left lung is moved into the field, allowing dissection of the left hilum. The left hemi-azygos vein that crosses ventral to pulmonary vessels is ligated. After adequate careful exposure of the left hilar structures, an umbilical tape can be placed around the right pulmonary artery for later clamping or ligation as some studies desire to assess recipient pig’s performance solely on the left transplanted lung. The donor lung is prepared on the back table. In pigs, unlike humans, the RIPV and LIPV orifices are located next to each other as they enter the left atrium (Fig. 2). For the donor venous pedicle preparation, the RIPV is ligated distally from the left atrium that leaves an adequate atrial cuff around the left sided pulmonary veins.
e109
FIG. 2. Donor lung bloc on the back table with the heart excised except for the left atrial cuff: long arrow ¼ right inferior pulmonary vein crossing the midline to drain into left atrium immediately adjacent to left inferior pulmonary vein; short arrow ¼ left atrial cuff.
Subsequently, the left pulmonary artery and bronchus are divided away from the hilum and the donor lung is brought into the field for transplantation. Because there is no double lumen endotracheal tube for single lung ventilation and because there is limited space we cross-clamp the left hilum all at once with a Satinsky side biting clamp. Caution should be used to avoid the RIPV to allow it to continue to drain into the left atrium. Thereafter, the left pulmonary artery, left bronchus, and LIPV are transected distally and the recipient’s lung is excised. Due to the size of the hilar vessels, magnifying loupes are employed. The left atrium, bronchus, and artery are trimmed appropriately. The left pulmonary veins near the atrium are thin-walled and friable and need to be treated carefully. Furthermore, the LSPV is diminutive and fragile in pigs, and, therefore, we ligate this vein immediately before or just after applying the hilar cross-clamp, not earlier to avoid left middle lobe congestion. We have found it easier to use the LIPV with extension into the adjacent left atrium as the recipient pulmonary venous pedicle (Fig. 3) [6]. Importantly, in our technique the order of anastomosis is different from humans. We perform the venous anastomosis first as it has a very short stump that would make it inaccessible if the other anastomoses were performed first; it is completed with a running 4-0 Prolene RB-1 needle (Ethicon Inc.) imbricating the recipient and donor left atria. Again it is friable and should be treated carefully. The bronchial anastomosis is completed next with a running 3-0 or 4-0 Prolene SH needle (Ethicon Inc.). Lastly, the pulmonary arterial anastomosis is completed with a running 4-0 Prolene RB-1 needle (Ethicon Inc.). Prior to reperfusion, a small vascular clamp is placed on the recipient side of the pulmonary arterial anastomosis and the hilar cross-clamp is released de-airing the donor lung in a retrograde
e110
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
FIG. 3. This image shows the recipient left hilum before transplantation (see text for technical details): long arrow ¼ right inferior pulmonary vein; short arrow ¼ left inferior pulmonary vein; arrow head ¼ left hemi-azygos vein divided as it enters into the coronary sinus; asterisk ¼ divided left superior pulmonary vein; LPA ¼ left pulmonary artery; LB ¼ left bronchus.
manner, and employing a 21-guage needle to further de-air the pulmonary artery as necessary. Following reperfusion atelectasis resolves slowly over time. Arterial blood gases can be obtained directly from the pulmonary vein of the transplanted lung, and tissue samples are obtained from the lung parenchyma according to protocol. DISCUSSION
Swine is the most commonly used large animal model of lung transplantation; often in studies of drug effects on ischemia/reperfusion injury [3, 7–10], EVLP [1, 3] and in xenotransplantation [11, 12]. An adequate understanding of pig lung anatomy is important for those beginning experimental protocols. In our review of the literature, we could not find an article dedicated to the technique of swine lung transplantation; nevertheless, several articles briefly describe their donor and recipient preparation [1, 7–10, 12]. In our described model, a single stage cross-clamp is applied to the left pulmonary hilum and obviates the need for sequential clamping. We also differ in the order of performing the anastomoses as venous is performed first, followed by bronchial and arterial. Others have suggested ligation of the left pulmonary veins, followed by clamping of the left bronchus and pulmonary artery and cutting them distally for left pneumonectomy; with subsequent performance of the bronchial and arterial anastomoses in sequence and then clamping the recipient’s left atrium that is sewed to the donor’s left atrial cuff [8, 10]. Functional assessment of the transplanted lung is an important part of experimental protocols. Oxygenation
capacity (PaO2/FiO2) with arterial blood gases, pulmonary artery pressures and vascular resistance, and dynamic and static lung compliance are used to monitor the transplanted lung performance during reperfusion [1, 7–12]. There are three approaches to monitor the transplanted lung. Some authors do not exclude the recipient’s native right lung from the circulation or ventilation during assessments [7]. Others prefer to ligate the right pulmonary artery shortly after reperfusion so that the recipient pig is solely dependent on the transplanted lung during assessments [9, 10, 12]; however, it is unlikely that pigs will tolerate reperfusion times in excess of 1-2 h following right pulmonary artery ligation as there is then massive shunting of the entire cardiac output to the donor lung, which has already sustained endothelial damage during cold ischemia, with subsequent lung edema, hypoxemia, and death [9]. Some xenotransplantation models employing right pulmonary artery ligation have 100% mortality in the control group within 20 s of occlusion [12]; however, in an allotransplantation model, Pizanis et al. have reported 100% (n ¼ 7) survival in the control group during 6 h of reperfusion with the right pulmonary artery and bronchus ligated 10–20 min after reperfusion [12]. An alternative approach is continue ventilating and perfusing both lungs with sampling blood gases only from the left pulmonary veins to assess transplanted lung oxygenation capacity during reperfusion (usually 4–6 h) [1, 8]. The right pulmonary artery can then be ligated 15–30 min before a final blood gas is obtained to assess the recipient’s dependence solely on the transplanted lung [1, 8]. We would recommend the third approach as it is likely to produce a more predictable model and allows for functional assessment of the transplanted lung over longer reperfusion times. Of note, as described earlier, the left and right inferior pulmonary veins drain next to each other into the left atrium. Accordingly, samples from the left pulmonary vein should be obtained as close as possible to the lung parenchyma to avoid mixing of its blood with venous return from the recipient’s native right lung. CONCLUSION
Due to the size of the porcine hilar vessels, lung transplantation can be challenging, and both magnifying loupes and surgical experience are required. We have found the above described technical nuances facilitate transplantation and provide reproducible results. REFERENCES 1. Cypel M, Yeung JC, Hirayama S, et al. Technique for prolonged normothermic ex vivo lung perfusion. J Heart Lung Transplant 2008;27:1319.
KARIMI ET AL.: SWINE LUNG TRANSPLANTATION TECHNIQUE 2. Steen S, Liao Q, Wierup PN, et al. Transplantation of lungs from non-heart-beating donors after functional assessment ex vivo. Ann Thorac Surg 2003;76:244. 3. Cypel M, Liu M, Rubacha M, et al. Functional repair of human donor lungs by IL-10 gene therapy. Sci Transl Med 2009; 1:4ra9. 4. Dondelinger RF, Ghysels MP, Brisbois D, et al. Relevant radiological anatomy of the pig as a training model in interventional radiology. Eur Radiol 1998;8:1254. 5. Nakakuki S. Bronchial tree, lobular division and blood vessels of the pig lung. J Vet Med Sci 1994;56:685. 6. Swindle M. Swine in the Laboratory. 2nd ed. Boca Raton, FL: Taylor and Francis Group, 2007:213–214. 7. Gomez CB, del Valle HF, Bertolotti A, et al. Effects of shortterm inhaled nitric oxide on interleukin-8 release after single-lung transplantation in pigs. J Heart Lung Transplant 2005;24:714.
e111
8. Reece TB, Ellman PI, Maxey TS, et al. Adenosine A2A receptor activation reduces inflammation and preserves pulmonary function in an in vivo model of lung transplantation. J Thorac Cardiovasc Surg 2005;129:1137. 9. Korom S, Hillinger S, Cardell M, et al. Sildenafil extends survival and graft function in a large animal lung transplantation model. Eur J Cardiothorac Surg 2006;29:288. 10. Pizanis N, Milekhin V, Tsagakis K, et al. PDE-5 inhibitor donor intravenous preconditioning is superior to supplementation in standard preservation solution in experimental lung transplantation. Eur J Cardiothorac Surg 2007;32:42. 11. Gaca JG, Lesher A, Aksoy O. Disseminated intravascular coagulation in association with pig-to-primate pulmonary xenotransplantation. Transplantation 2002;73:1717. 12. Daggett CW, Yeatman M, Lodge AJ, et al. Total respiratory support from swine lungs in primate recipients. J Thorac Cardiovasc Surg 1998;115:19.