Replacement of the thoracoabdominal aorta for extensive

Anatomy of Spinal Cord Blood Supply in the Pig Justus T. Strauch, MD, Alexander Lauten, MD, Ning Zhang, MD, Thorsten Wahlers, MD, and Randall B. Griepp, MD Department of Cardiothoracic Surgery, The Mount Sinai Medical Center/New York University, New York, New York, and Department of Cardiothoracic Surgery, University Hospital of Cologne, Cologne, Germany Background. In a species as popular in research as the pig, little information is available concerning the blood supply and vascular anatomy of the spinal cord and comparability to humans. Methods. To visualize vessels contributing to the blood supply of the spinal cord, x-ray digital subtraction angiography and vascular casting by injection of a polymerizing solution of methylmethacrylate were used. Results. The pig has larger internal thoracic arteries and subscapular arteries than the human, providing extensive collateral flow to the lower body, which offers blood supply to the spinal cord through the collaterals. The pig has a finecaliber vessel plexus providing blood to the neck area, from which flow will reach both the spinal cord and the base of the brain. The segmental thoracic and lumbar arteries are relatively small in pigs, and they almost all originate as a single branch from the aorta and divide after 3 to 4 mm. The segmental vessels show a clear diminution after 2 to 3 cm at the level of the vertebral bodies. Pigs show major differences in the anatomy of the aortic bifurcation compared with humans. The median sacral artery in pigs is a large-caliber vessel, of a size almost comparable to the common iliac artery, with an isolated single dorsal branch leading to the spinal cord. Conclusions. Documenting the anatomic differences in spinal cord blood supply between pigs and humans will aid in the planning of future experimental studies and in determining the clinical relevance of such studies. (Ann Thorac Surg 2007;83:2130 4) 2007 by The Society of Thoracic Surgeons Replacement of the thoracoabdominal aorta for extensive aortic aneurysmal disease often involves the sacrifice of segmental vessels, altering temporarily or permanently the blood supply to the spinal cord. Different techniques to minimize postoperative ischemia of the spinal cord have been reported. Reimplantation of sacrificed segmental arteries to restore blood flow is a common approach. Strategies range from time-consuming reimplantation of all sacrificed segmental vessels, to reimplantation of all those from which backflow is observed, to anastomosis of only those vessels considered indispensable by various criteria [1 3]. Adjunctive procedures to preserve spinal cord integrity during surgery including monitoring of motor evoked potentials, cerebrospinal fluid drainage, distal aortic perfusion, and neuroprotective hypothermia have reduced the incidence of postoperative dysfunction of the cord. However, a small number of patients still suffer from paraplegia or paraparesis postoperatively. At a minimum, such patients have an impaired quality of life and a prolonged hospital stay, and incur significant costs for treatment. More ominously, there is a significant association of postoperative paraplegia with mortality. Factors influencing long-term spinal cord function are more complex and less well understood. In a number of patients, evidence of impaired spinal cord function only becomes manifest days or weeks after the operation despite Accepted for publication Jan 29, 2007. Address correspondence to Dr Strauch, University Hospital of Cologne, Department of Cardiothoracic Surgery, Kerpener Strasse 62, 50924 Cologne, Germany; e-mail: justus.strauch@uk-koeln.de. normal neurologic recovery early postoperatively. In many of these cases, the loss of spinal cord function can be linked to episodes of hypotension, suggesting that inadequate perfusion is the underlying mechanism. Occasionally, spinal cord function can be restored by countermeasures such as spinal cord drainage and hypertensive therapy, but these attempts fail in a significant number of cases. For better understanding of the pathophysiology and the mechanisms responsible for the development of both immediate and delayed paraplegia, extensive research is still necessary to investigate the dynamics of blood supply to the spinal cord and the reaction of the spinal cord to ischemia. Little is known about the existence and possible duration of a period of increased vulnerability that may occur as a consequence of operative ischemia and sudden reduction in blood flow, or about time-related changes in spinal cord blood supply and the development of collaterals. In humans, blood flow to the spinal cord relies not only on branches of the vertebral, deep cervical, intercostal, and lumbar arteries, all of which contribute to the feeding of the anterior spinal artery, but also on more distant arteries such as the hypogastric and subclavian arteries, which feed into the same network. However, in a species as popular and widespread in research as the pig, little information is available concerning the blood supply and vascular anatomy of the spinal cord and its comparability to the human. For the purpose of devising an animal model for investigating delayed paraplegia, we believe this knowledge is essential. The aim of this study was to point out the differences between the anatomic situation in humans and in pigs to be able to better interpret the results of future experiments involving spinal cord blood supply in the pig. 2007 by The Society of Thoracic Surgeons 0003-4975/07/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2007.01.060

Ann Thorac Surg STRAUCH ET AL 2007;83:2130 4 ANATOMY OF PORCINE SPINAL CORD BLOOD SUPPLY 2131 Material and Methods Animals The vascular architecture of the aortic branches supplying the spinal cord was studied in juvenile Yorkshire pigs, a popular strain (Thomas D. Morris, Inc, Reisterstown, MD). The pigs were 3 months of age and weighed 20 to 22 kg. All animals received humane care in accordance with the Principles of Laboratory Animal Care prepared by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals formulated by the Institute of Laboratory Animal Resources and published by the National Institutes of Health. Anesthesia Anesthesia was induced using ketamine hydrochloride (10 mg/kg intramuscularly) and sodium thiopental (20 mg/kg). After endotracheal intubation, the pigs were ventilated mechanically with an inspired oxygen fraction of 0.5 and isoflurane of 2% to maintain anesthesia. Pancuronium (0.1 mg/kg) was administered intravenously to achieve muscular paralysis and heparin (300 IU/kg) to avoid the formation of blood clots that could occlude peripheral arterial vessels. During angiography, the ventilator rate and tidal volume were adjusted to maintain arterial carbon dioxide tension at approximately 35 to 40 mm Hg. End-expiratory carbon dioxide was continuously monitored (model 2010 200 R; PPG Biomedical Systems, Lenexa, KS). In pigs undergoing vascular casting, animals were sacrificed shortly after aortic cannulation. Illustration of the Arterial Vascular System To visualize vessels contributing to the blood supply of the spinal cord, x-ray digital subtraction angiography and vascular casting by injection of a polymerizing solution of methylmethacrylate were used. Thus, two-dimensional angiographic images and a three-dimensional cast of each animal s arterial vascular system were obtained, illustrating vessels and collaterals by visualizing even arteries of very small caliber. Two-Dimensional Visualization of the Arterial System: X-Ray Digital Subtraction Angiographic Imaging With the animal in a right-sided position, the chest was entered through the fourth intercostal space and the pericardium opened. To avoid accidental dislocation of the catheter, the anticipated site for catheterization on the ascending aorta was prepared using a 4-0 pursestring suture. Catheterization was carried out using the Seldinger technique with a regular 5F sheath inserted into the aorta through a conventional 18G needle. Flush arteriograms (ascending, descending, and abdominal aorta, iliac and median sacral arteries) and selective angiograms (truncus bicaroticus and left subclavian arteries) were obtained during injection of 60 ml of undiluted Conray contrasting agent (Iothalamate Meglumine Injection; Mallinckrodt Inc, St. Louis, MO) into the ascending aorta. For image recording, a C-arched angiography system (BV 29; Phillips Medical Systems North America, Denver, CO) was used. Three-Dimensional Visualization of the Descending Aorta and Branches: Vascular Casting A left anterior thoracotomy was carried out in the fourth intercostal space to allow exposure of the heart and the great vessels. The ascending aorta was identified, and a 4-0 pursestring put in place for cannulation. Vessel loops were placed around the ascending and descending aorta to allow occlusion by placement of vessel clamps. Using a retroperitoneal approach, the renal arteries, the celiac trunk, the superior and inferior mesenteric arteries, and the femoral arteries were identified and ligated to avoid loss and diversion of the polymerizing solution of methylmethacrylate to the kidneys and to organs of the peritoneal cavity. The ascending aorta was cannulated with a 16F aortic cannula normally used for extracorporeal circulation. By incising the superior and inferior venae cavae, the animal was exsanguinated. The arterial system was flushed through the aortic cannula using 4 L of 4 C cold saline solution until the return from the venous system was clear and free of blood. Vessel clamps were positioned on the descending aorta and on the ascending aorta proximal to the cannulation site. The injection of 1.5 L of Batson s No. 17 solution (Polysciences, Warrington, PA) was started into the ascending aorta at a constant rate of 150 ml/min. On appearance of the solution in the superior vena cava, the superior vena cava was occluded to allow complete filling of the vessels of the upper body. After the first 500 ml was injected, the distal clamp was opened and the Batson s No. 17 solution was allowed to perfuse the lower body. By clamping the inferior vena cava on appearance of the solution, the entire vascular system was allowed to fill. After complete injection and decannulation, the animal was transferred into a water Fig 1. Angiogram showing large subscapular arteries and internal thoracic arteries feeding collateral pathways in pigs (lateral view).

2132 STRAUCH ET AL Ann Thorac Surg ANATOMY OF PORCINE SPINAL CORD BLOOD SUPPLY 2007;83:2130 4 Fig 2. Angiogram showing segmental arteries in the lumbar area, here L3, L4, and L5 (lateral view). bath of 4 C for 6 hours to allow the methylmethacrylate to cure. Baths of sodium hydroxide were used to dissolve the tissue surrounding the cast. With proper safety measures, a 5N solution of the base was prepared, and the solution in the bath was changed regularly. Results The pig has larger internal thoracic arteries and subscapular arteries than the human, providing extensive collateral Fig 4. Supraaortic cast shows the importance of the large vertebral arteries, internal thoracic arteries, and large arteriovenous shunt networks laterally in the neck. There is a large first vertebral artery branch on each side. flow to the lower body, which offers blood supply to the spinal cord through the collaterals. The pig has a finecaliber vessel plexus providing blood to the neck area, from which flow will reach both the spinal cord and the base of the brain. The segmental thoracic and lumbar arteries are relatively small in pigs, and they almost all originate as a single branch from the aorta and divide after 3 to 4 mm. The segmental vessels show a clear diminution after 2 to 3 cm at the level of the vertebral bodies. Pigs show major differences in the anatomy of the aortic bifurcation compared with humans. The median sacral artery in pigs is a largecaliber vessel, of a size almost comparable to the common iliac artery, with an isolated single dorsal branch leading to the spinal cord. Angiography Angiography reveals large subscapular arteries and internal thoracic arteries feeding collateral pathways in pigs (Fig 1; lateral view). Also present in angiographic views are segmental arteries in the lumbar area, here L3, L4, and L5 (Fig 2; lateral view), and the median sacral artery and its Fig 3. Angiogram showing median sacral artery and its major side branches (lateral and anteroposterior view). Fig 5. Portion of the thoracoabdominal aorta with its segmental arteries, accompanying veins, and intercostal arteries.

Ann Thorac Surg STRAUCH ET AL 2007;83:2130 4 ANATOMY OF PORCINE SPINAL CORD BLOOD SUPPLY 2133 Fig 8. Segmental arteries here the thoracic arteries which have a common origin from the aorta. Fig 6. Portion showing two segmental arteries here the smallcaliber lumbar arteries which often have a common origin from the aorta. No major connection to the anterior spinal artery is seen. major side branches (Fig 3; lateral and anteroposterior view). Fig 7. The median sacral artery in pigs is large and bifurcates almost immediately after its origin from the descending aorta. It has major branches. Three-Dimensional Vascular Cast The supraaortic cast shows the importance of the large vertebral arteries, internal thoracic arteries, and large arteriovenous shunt networks laterally in the neck. There is a large first vertebral artery branch on each side (Fig 4). A portion of the thoracoabdominal aorta with its segmental arteries, accompanying veins, and intercostal arteries is shown in Figure 5. Figure 6 shows two segmental arteries here the small-caliber lumbar arteries which often have a common origin from the aorta. No major connection to the anterior spinal artery is seen. The median sacral artery in pigs is large and bifurcates almost immediately after its origin from the descending aorta (Fig 7). It has major branches. Postmortem Specimen Segmental arteries have a common origin from the aorta; the thoracic arteries are shown in Figure 8. Comment This study paves the way for use of the pig model for research to investigate the dynamics of blood supply to the spinal cord and of the reactions of the spinal cord to transient and permanent ischemia [5, 6]. Studies in animals are essential to understand physiologic principles and to provide a suitable environment to safely verify and refine protective measures subsequently to be used clinically. The pig has become a popular and widely accepted model for investigating different strategies for preventing neurologic dysfunction during operations on the thoracoabdominal aorta [7 9]. In humans, blood flow to the spinal cord depends on branches of the vertebral, deep cervical, intercostal, and lumbar arteries that contribute to the feeding of the anterior spinal artery. Since the advent of routine replacement of the thoracoabdominal aorta for extended aneurysms, a number of imaging methods, such as roentgenography, contrast angiography, computerized tomography, and magnetic resonance imaging, have succeeded to a greater or lesser

2134 STRAUCH ET AL Ann Thorac Surg ANATOMY OF PORCINE SPINAL CORD BLOOD SUPPLY 2007;83:2130 4 extent in visualizing the blood supply to the spinal cord in humans [10]. This knowledge has been supplemented by observations at the time of surgery. Both techniques used in this study to visualize the anatomy of the vascular supply to the spinal cord in pigs revealed interesting differences from the human situation. These differences should be noted by any researcher considering establishing a spinal cord ischemia model. In correlation with its weight and body surface area, the pig has much larger internal thoracic arteries and subscapular arteries than the human, providing extensive collateral flow to the lower body, which offers additional blood supply to the spinal cord through the chest and abdominal walls. The pig also has a large fine-caliber vessel plexus providing blood to the neck area, from which we believe flow will reach both the spinal cord and the base of the brain. Furthermore, large bilateral vertebral arteries feed the circle of Willis in pigs, with abundant small branches. The first two branches on each side are major tributaries, which arise in a right angle fashion from the vertebral artery, and head toward the spinal cord in the cervical area: it seems evident that they are of major importance for the spinal cord above the segmental arteries. We also found large arteriovenous shunts between the subclavian artery branches in the deep neck muscles and the muscles of mastication (Figs 1 and 4): we are not sure whether they have a beneficial effect on spinal cord blood supply or whether they provide an opportunity for steal. It is of course true that the anatomy of the neck and muscles of mastication are very different in the pig and the human, and the vessels reflect these differences. The arteriovenous shunt connections may be important to take into consideration not only for studies of spinal cord blood supply but also for studies dealing with cerebral blood flow and cerebral protection models. In the pig, these vessels are supplying a much larger volume of tissue in the head and neck than in humans. The blood distribution in the pig s head is different from that in humans: a high percentage of flow goes into a large and densely vascularized nose plexus. Against the background of body weight, the segmental thoracic and lumbar arteries are relatively small in pigs, and they almost all originate as a single branch from the aorta and divide after 3 to 4 mm. The segmental vessels show a clear diminution after 2 to 3 cm at the level of the vertebral bodies. In the experimental model, one might want to dissect the thoracic and abdominal aorta circumferentially in each case, so as not to miss the occasional segmental artery that arises bilaterally, most frequently in the high lumbar area. With the methods we used, we were not able to visualize and describe direct connections between the intersegmental arteries and the anterior spinal artery (Figs 2, 5 and 8) [11, 12]. There are a total of 16 to 17 segmental arteries in pigs: usually 9 to 11 thoracic, and 6 lumbar arteries. Pigs show major differences in the anatomy of the aortic bifurcation compared with humans. The median sacral artery in pigs (roughly equivalent to the hypogastric arteries in humans) is a large-caliber vessel, of a size almost comparable to the common iliac artery, with an isolated single dorsal branch leading to the spinal cord area. One centimeter after its origin from the aortic bifurcation, the median sacral artery also splits, again with big branches going dorsally and dorsocaudally (Figs 3 and 7) to supply blood to the spinal cord and to the muscles of the buttocks [13 15]. We believe that there is a large quantity of flow going from these pelvic vessels to the lower spinal cord, and that they have to be taken into account in spinal cord studies. These major collateral vessels need to be identified and isolated in each experimental setup to properly interpret results involving intersegmental artery sacrifice and reconstitution. Documenting the anatomic differences in spinal cord blood supply between the pig and human will aid in the planning of future experimental studies and in determining the clinical relevance of such studies. References 1. Griepp RB, Ergin MA, Galla JD, et al. Looking for the artery of Adamkiewicz: a quest to minimize paraplegia after operations for aneurysms of the descending thoracic and thoracoabdominal aorta. J Thorac Cardiovasc Surg 1996;112:1202 15. 2. Kouchoukkos NT, Rokkas CK. Hypothermic cardiopulmonary bypass for spinal cord protection: rationale and clinical results. Ann Thorac Surg 1999;67:1940 2. 3. Coselli JS, LeMaire SA, Miller CC. Mortality and paraplegia after thoracoabdominal aortic aneurysm repair: a risk factor analysis. Ann Thorac Surg 2000;69:409 14. 4. Christiansson L, Ulus AT, Hellberg A, Bergqvist D, Wiklund L, Karacagil S. Aspects of the spinal cord circulation as assessed by intrathecal oxygen tension monitoring during various arterial interruptions in the pig. J Thorac Cardiovasc Surg 2001;121:762 72. 5. Qayumi AK, Janusz MT, Lyster DM, Gillespie KD. Animal model for investigation of spinal cord injury by aortic crossclamping. J Invest Surg 1997;10:47 52. 6. Hellberg A, Christiansson L, Ulus AT, Bergqvist D, Wiklund L, Karacagli S. A prolonged spinal cord ischemia model in pigs. Passive shunting offers stable central hemodynamics during aortic occlusion. Eur J Vasc Endovasc Surg 2000;19:318 23. 7. dehaan P, Kalkman CJ, Meylaerts SAG, Lips J, Jacobs MJ. Development of spinal cord ischemia after clamping of noncritical segmental arteries in the pig. Ann Thorac Surg 1999;68:1278 84. 8. Svensson LG, Patel V, Coselli JS, et al. Preliminary report of localization of spinal cord blood supply by hydrogen during aortic operations. Ann Thorac Surg 1990;49:528 35. 9. Svensson LG, Patel V, Robinson MF, et al. Influence of preservation or perfusion of intraoperatively identified spinal cord blood supply on spinal motor evoked potentials and paraplegia after aortic surgery. J Vasc Surg 1991;13:355 65. 10. Kieffer E, Fukui S, Chiras J, et al. Spinal cord arteriography: a safe adjunct before descending thoracic or thoracoabdominal aortic aneurysmectomy. J Vasc Surg 2002;35:262 8. 11. Alleyne CH Jr, Cawley CM, Shengelaia GG, Barrow DL. Microsurgical anatomy of the artery of Adamkiewicz and its segmental artery. J Neurosurg 1998;89:791 5. 12. Domisse GF. The blood supply of the spinal cord. J Bone Joint Surg B 1974l56:225 35. 13. Wissdorf H. The blood supply of the spinal column and the spinal cord of the pig. Tieraerztliche Hochschule Hannover, Med Vet Habilschrift. Berlin: Parez-Verlag, 1972:1 72. 14. Hannon JP, Bossone CA, Wade CE. Normal physiological values for conscious pigs used in biomedical research. Lab Anim Sci 1990;40:293 8. 15. D Ambra MN, Dewhirst W, Jacobs MJ, et al. Cross-clamping the thoracic aorta: effect on intracranial pressure. Circulation 1988;78:198 202.