Anatomical Study of Blood Supply to the Spinal Cord

Anatomical Study of Blood Supply to the Spinal Cord Kiyofumi Morishita, MD, PhD, Gen Murakami, MD, PhD, Yasuaki Fujisawa, MD, PhD, Nobuyoshi Kawaharada, MD, PhD, Jhoji Fukada, MD, PhD, Tatsuya Saito, MD, and Tomio Abe, MD, PhD Departments of Thoracic and Cardiovascular Surgery, and Anatomy, Sapporo Medical University School of Medicine, Sapporo, Japan Background. Low incidences of spinal cord ischemia after thoracoabdominal aortic aneurysm repair, despite sacrifice of all segmental arteries, have recently been reported. This, however, cannot be explained by previous anatomical findings, which prompted us to perform an anatomical study of blood supply to the spinal cord. Methods. Fifty-five spinal cords from Japanese formolfixed cadavers (mean age, 79 10 years) were studied. Diameters of the anterior spinal artery (ASA) above and below the junction with the arteria radicularis magna (ARM) and diameters of the ARM were measured using the NIH image program (National Institutes of Health Image 1.58). Results. The degree of narrowing of the ASA, defined as the diameter above the ARM expressed as a percentage of the diameter below the ARM, ranged from 23% to 161% and averaged 66% 30%. The degree of narrowing was plotted against the ARM diameter divided by the ASA diameter above the junction to examine the impact of the degree of narrowing on distal spinal blood flow from the ARM. The degree of narrowing was related to distal spinal blood flow from the ARM (r 0.56, p < 0.0001). Conclusions. The degree of narrowing of the ASA varies considerably. Furthermore, distal spinal blood the narrow point of the ASA becomes narrower. These anatomical findings of spinal blood supply should be useful for elucidating the mechanisms of spinal cord injury after repair of extensive thoracoabdominal aneurysms. (Ann Thorac Surg 2003;76:1967 71) 2003 by The Society of Thoracic Surgeons Spinal cord ischemia still occurs in 5% to15% of patients undergoing extensive thoracoabdominal aortic aneurysm repair [1 3], although many efforts have been made to reduce the rate of paraplegia or paraparesis. Spinal cord injury generally results from temporary or permanent interruption of spinal cord blood supply. The complexity of spinal cord circulation has puzzled surgeons for many years. In addition to considerable variations in normal anatomy, occlusion of segmental arteries due to mural thrombus or atherosclerotic change has made it difficult to elucidate the pathogenesis of spinal cord ischemia. This is why there are conflicting views regarding the appropriate strategy (sacrifice vs aggressive reattachment) for reconstruction of the segmental arteries in thoracoabdominal aortic aneurysm repair. Griepp and colleagues [4] sequentially clamped each pair of intersegmental arteries and subsequently sacrificed them if no change in somatosensory evoked potentials occurred within 8 to 10 minutes after occlusion. The Presented at the Poster Session of the Thirty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 31 Feb 2, 2003. Address reprint requests to Dr Morishita, Department of Thoracic and Cardiovascular Surgery, Sapporo Medical University School of Medicine, South 1 West 16, Chuo-ku, Sapporo 060-8556, Japan; e-mail: kmori@sapmed.ac.jp. rate of paraplegia in their patients was only 2% despite the fact that intersegmental arteries had not been reattached. They speculated that the anterior spinal artery (ASA) was functionally continuous with multiple input arteries throughout its length. Acher and colleagues [5] reported that immediate oversewing of all intercostal arteries resulted in a rate of spinal cord ischemia of only 3.4%. In contrast, Safi and colleagues [6] claimed that reattachment of T9 to T12 significantly prevented neurologic deficit. Svensson and colleagues [7] also concluded that the failure of successful reattachment of critical segmental arteries caused spinal cord ischemia. It is generally believed that the ASA becomes extremely narrow above the junction with the arterial radicularis magna (ARM) and that spinal blood circulation below the junction depends on the ARM [8]. However, the anatomical findings cannot explain the low postoperative rate of spinal cord ischemia despite sacrifice of all intersegmental arteries. This has prompted us to perform an anatomical study of distal spinal blood supply. Material and Methods Fifty-five spinal cords from Japanese formol-fixed cadavers (29 males and 26 females) with a mean age of 79 10 2003 by The Society of Thoracic Surgeons 0003-4975/03/$30.00 Published by Elsevier Inc doi:10.1016/s0003-4975(03)01254-2

1968 MORISHITA ET AL Ann Thorac Surg BLOOD SUPPLY TO THE SPINAL CORD 2003;76:1967 71 Fig 2. Distribution of narrowing in the ASA. For representation of the degrees of narrowing in the ASA, its diameter above the junction was expressed as a percentage of the diameter below the junction. (ASA anterior spinal artery.) Fig 1. Measurement of the ASA and the ARM. Diameters of the ASA 1 cm above (A) and below (B) the junction with the ARM were measured. The diameter of the ARM 1 cm away from the junction was also measured (C). (ARM arterial radicularis magna; ASA anterior spinal artery.) years (range, 49 to 97 years) were studied. The causes of death did not include any significant aortic disease. Dissection of the spinal cord has been described in detail elsewhere [9]. Briefly, with each cadaver placed in the prone position, laminectomy of the vertebrae was performed. The spinal cord was removed from the fifth thoracic vertebral level to the second lumbar vertebral level. The left T12 intercostal artery was marked to investigate the vertebral level and laterality of the ARM. Dye was not used in this study, based on our previous work that there is no difference between diameters of the ARMs that are injected with dye and those that are not in formol-fixed cadavers [9]. The ASA and ARM were dissected along their course. After checking the continuity of the ASA, its diameters 1 cm above and below the junction with the ARM were measured. The diameter of the ARM 1-cm away from the junction was also measured (Fig 1). The NIH image program was used for the measurements [10]. The ARM was defined as the largest of the anterior radicular arteries joining the ASA with a hairpin turn. Statistical Analysis All data are expressed as means standard deviations. Differences in measurements were analyzed using a one-way analysis of variance (ANOVA), followed by Fisher s PLSD test if permitted by the F value. Relations among variables were assessed using linear regression analysis. This analysis was performed using StatView J-5.0 software (SAS Institute, Cary, NC). A value of p less than 0.05 was considered statistically significant. Results The ASA was continuous from the fifth thoracic vertebral level to the cauda equina in all cases. The ARM arose from T7 T8 segment in 10%, T9 T11 in 68%, T12 L1 in 16%, and L2 L4 in 6%. The origin of the ARM was from the left side in 43 cases (78%). Diameters of the ASA 1 cm above the junction with the ARM ranged from 0.096 mm to 0.720 mm (mean diameter, 0.337 0.122 mm), diameters of the ASA 1 cm below the junction with the ARM ranged from 0.266 mm to 0.839 mm (mean diameter, 0.545 0.141 mm), and diameters of the ARM 1 cm away from the junction ranged from 0.257 mm to 1.066 mm (mean diameter, 0.608 0.165 mm). The diameters were significantly different (ASA above junction vs ASA below junction, p 0.001; ASA above junction vs ARM, p 0.001; ASA below junction vs ARM, p 0.0371). For comparison of the degrees of narrowing in the ASA, the diameter above the junction was expressed as a percentage of the diameter below the junction. The percentage ranged from 23% to 161% and averaged 66% 30% (Fig 2). The degree of narrowing of the ASA varied from patient to patient. There were three cadavers with a percentage of more than 120%, indicating that the ASA diameter above the junction is much wider than that below the junction. Each had the ARM and the lower lumbar artery supplying the lower lumbar spinal cord. The lower lumbar artery also showed a hairpin bend. As blood in the distal spinal cord circulation is mainly supplied by the ASA and the ARM, the ARM diameter

Ann Thorac Surg MORISHITA ET AL 2003;76:1967 71 BLOOD SUPPLY TO THE SPINAL CORD 1969 Fig 3. Correlation between narrowing of the ASA and impact of blood flow on distal spinal cord circulation. To show which artery impacts distal spinal blood flow, the ARM diameter was divided by the ASA diameter above the junction. (ARM arterial radicularis magna; ASA anterior spinal artery.) was divided by the ASA diameter above the junction to show which artery impacts distal spinal blood flow. We plotted the above-described ratios against ASA diameters above the junction expressed as percentages of the ASA diameters below the junction. The relationships between the ratios and the percentages were linear and had coefficients of correlation (r 0.56, p 0.0001; Fig 3). This indicated that distal spinal blood supply becomes progressively dependent on the ARM as the narrow point of the ASA becomes narrower. Comment Previous anatomical studies revealed that the ASA becomes extremely narrow above the junction with the ARM and that the distal spinal cord is mainly perfused through the ARM [8 11]. Based on these anatomical findings, most surgeons have performed reattachment of lower thoracic and lumbar segmental arteries from which the ARM is thought to originate. In contrast to the strategy of reattachment of segmental arteries, some authors have recently contended that sacrifice of segmental arteries can be performed safely [4, 5]. Surprisingly, only a few of the patients operated on by these authors suffered from postoperative spinal cord injury. This, however, cannot be explained by the previously reported anatomy of blood supply to the spinal cord. Theoretically, sacrifice of segmental arteries would cause distal cord spinal ischemia. Biglioli and colleagues have suggested that such ligation of the segmental arteries can be justified due to anatomic continuity of the ASA [12]. However, as pointed out by Svensson [13], the narrow point of the ASA reduces blood flow of the distal spinal cord. He used the law of Hagen-Poiseuille to show that blood cannot sufficiently flow down the ASA through the narrow point. According to his theory, if all segmental arteries are sacrificed, the distal flow through the narrow point will be so small that ischemia cannot be prevented from occurring. In such a situation, instead of segmental arteries, a collateral pathway can supply blood to the distal spinal cord. It is expected that most patients have collateral circulation for supplying blood to the distal spinal cord based on the fact that few patients have suffered from spinal cord ischemia despite the sacrifice of all segmental arteries. However, in an angiographic study by Kieffer and colleagues, the ARM was visualized via anastomotic circulation in a fourth of almost 400 patients with thoracic or thoracoabdominal aortic aneurysms [14]. In other words, collateral circulation takes the place of segmental arteries to supply blood to the distal spinal cord in only one fourth of patients. Another anatomical explanation is required for surgeons to understand the mechanism of postoperative spinal cord ischemia. We hypothesized that the degree of narrowing of the ASA varied from patient to patient. However, we have little information on the degree of ASA narrowing. The present study focused on degrees of narrowing of the ASA. The degree of narrowing, defined as diameter of the ASA above the junction expressed as a percentage of its diameter below the junction, ranged from 23% to 161% in the cadavers we examined. The presence of a slight narrowing may account for low incidences of spinal cord ischemia despite sacrifice of segmental arteries. In addition to the large range of degrees of narrowing of the ASA, our study showed that distal spinal blood the narrow point of the ASA becomes narrower. For example, in patients with an extremely narrow point, the ARM provides the distal spinal cord with 202 times greater blood flow than that through the upper ASA [11]. When the size of the ASA below the junction is equal to that above the junction, blood flow through the ARM is reduced to about double that from the upper ASA. When the diameter of the ASA below the junction is narrower than that above the junction, blood flow from the ASA is almost the same as that from the ARM. Interestingly, the lower lumbar arteries with a hairpin bend inevitably supplied the lumbar spinal cord in the latter cases. The anatomical findings in the present study suggest that if the narrow point of the ASA is extremely narrow, reattachment of the intersegmental arteries may be required; if the ASA has no narrow point, or only a slightly narrow point, sacrifice of the intersegmental arteries can be justified, and if, conversely, the diameter of the ASA above the junction is larger than that below the junction, the lower lumbar artery may play an important role in lumbar spinal cord circulation [3, 15]. There are many ways (such as administration of neuroprotective agents) other than the anatomical maintenance of spinal cord blood supply to prevent spinal cord injury. However, the most important problem that surgeons face intraoperatively is which segmental arteries should be reattached. Aggressive reattachment increases aortic cross-clamping time, though some of the reattached arteries may not need to be reattached [13]. On the other hand, some

1970 MORISHITA ET AL Ann Thorac Surg BLOOD SUPPLY TO THE SPINAL CORD 2003;76:1967 71 authors [4, 5] have reported that a low rate of paraplegia was achieved without reattachment of a single intersegmental artery. However, their strategy is not effective for extensive aortic dissection. Most surgeons agree that reattachment of only the arteries that need to be reattached should be performed. To achieve such reattachment, an accurate diagnostic tool for preoperatively identifying the anatomy of spinal cord circulation, including the ASA, is required. Spinal arteriography has been performed for preoperative localization of the segmental arteries supplying the spinal cord, but this technique has not demonstrated the ASA in detail [14]. Magnetic resonance imaging (MRI) angiography has emerged as a new noninvasive method for detection of the ARM [16, 17]. Based on the anatomical findings in the present study, we have recently started examining the narrow point of the ASA as well as localization of the ARM. Although currently available MRI technology does not enable precise measurement of the diameters of the ASA, a narrow point can be identified. When the ASA does not have a narrow point, the segmental arteries can be sacrificed. If the ASA has a narrow point, the segmental artery that gives rise to the ARM is reattached. We are investigating whether performing reattachment of the segmental arteries based on such an assessment has an effect on postoperative paraplegia. There are several limitations in the present study. First, we did not investigate the effect of arteriosclerosis on our data. We did actually examine the presence of arteriosclerosis in spinal cord circulation. However, the results were not reported in this paper because description of findings of arteriosclerosis in our cadavers will be published in another article focusing on vascular arteriosclerosis of the spinal cord. Briefly, there was no arteriosclerosis in the intercostal arteries, anterior radicular arteries, or anterior spinal artery, though some of them had the orifices of their segmental arteries occluded by arteriosclerosis. Jacobs and colleagues [15], based on their experience, speculated that only the orifices of the segmental arteries are occluded with aortic plaques and that their lumen can be still patent. Previous anatomical studies have shown that arteriosclerosis seldom occurs in the spinal artery [18, 19]. It therefore seems that the presence of arteriosclerosis does not greatly affect the anatomy of spinal cord circulation. Second, only Japanese cadavers were used in this study. It is unknown whether the anatomical findings in this study are applied to Western people. The present study demonstrated that the degree of narrowing of the ASA varies from patient to patient and that distal spinal blood supply becomes progressively dependent on the ARM as the narrow point of the ASA becomes narrower. However, there have been no reported findings regarding these issues for other races. A similar study in a Western country is needed to determine whether there is ethnic variability in this anatomic factor. Third, we did not distend the vessels by injecting dye. Unlike fresh cadavers, the tissues of formol-fixed cadavers are so hard that the arteries cannot be distended by even high-pressured injection of dye. Since this was confirmed in a previous study [9], we did not use dye. Consequently, the ASA and ARM were smaller than previously reported diameters. However, the main focus of this study was not to measure the ASA and ARM diameters but to investigate degrees of narrowing of the ASA. The investigation was free from bias by expressing the ASA diameter above the junction as a percentage of its diameter below the junction. In conclusion, the degree of narrowing of the ASA varies considerably. Furthermore, distal spinal blood the narrow point of the ASA becomes narrower. These anatomical findings of spinal blood supply should be useful for elucidating the mechanisms of spinal cord injury after repair of extensive thoracoabdominal aneurysms. References 1. Cambria RP, Clouse WD, Davison JK, Dunn PF, Corey M, Dorer D. Thoracoabdominal aneurysm repair: results with 337 operations performed over a 15-year interval. Ann Surg 2002;236:471 9. 2. Coselli JS, LeMaire SA, Conklin LD, Kok soy C, Schmittling ZC. Morbidity and mortality after extent II thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 2002;73:1107 16. 3. Estrera AL, Miller CC III, Huynh TTT, Porat E, Safi HJ. Neurologic outcome after thoracic and thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 2001;72:1225 31. 4. 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. 5. Acher CW, Wynn MM, Hoch JR, Kranner PW. Cardiac function is a risk factor for paralysis in thoracoabdominal aortic replacement. J Vasc Surg 1998;27:821 30. 6. Safi HJ, Miller CC III, Carr C, Iliopoulos DC, Dorsay DA, Baldwin JC. Importance of intercostal artery reattachment during thoracoabdominal aortic aneurysm repair. J Vasc Surg 1998;27:58 68. 7. Svensson LG, Crawford ES, Hess KR, Coselli JS, Safi HJ. Experience with 1509 patients undergoing thoracoabdominal aortic operations. J Vasc Surg 1993;17:357 70. 8. Dommisse GF. The blood supply of the spinal cord. A critical vascular zone in spinal surgery. J Bone Joint Surg Br 1974; 56:225 35. 9. Koshino T, Murakami G, Morishita K, Mawatari T, Abe T. Does the Adamkiewicz artery originate from the larger segmental arteries? J Thorac Cardiovasc Surg 1999;117:898 903. 10. Sidney LS, Edward DS, Ralph SQ. Digital photography for the light microscope: results with a gated, video-rate CCD camera and NIH image software. Bio Techniques 1995;19: 946 55. 11. Svensson LG, Klepp P, Hinder RA. Spinal cord anatomy of the baboon: comparison with man and implications on spinal cord blood flow during thoracic aortic cross-clamping. S Afr J Surg 1986;24:32 4. 12. Biglioli P, Spirito R, Roberto M, et al. The anterior spinal artery: the main arterial supply of the human spinal cord a preliminary anatomic study. J Thorac Cardiovasc Surg 2000;119:376 9. 13. Svensson L. Commentary to the paper entitled Does the Adamkiewicz artery originate from the larger segmental arteries? J Thorac Cardiovasc Surg 1999;117:903 5.

Ann Thorac Surg MORISHITA ET AL 2003;76:1967 71 BLOOD SUPPLY TO THE SPINAL CORD 1971 14. Kieffer E, Fukui S, Chiras J, Koskas F, Bahnini A, Cormier E. Spinal cord arteriography: a safe adjunct before descending thoracic or thoracoabdominal aortic aneurysmectomy. J Vasc Surg 2002;35:262 8. 15. Jacobs MJ, de Mol BA, Elenbaas T, et al. Spinal cord blood supply in patients with thoracoabdominal aortic aneurysms. J Vasc Surg 2002;35:30 7. 16. Yamada N, Okita Y, Minatoya K, et al. Preoperative demonstration of the Adamkiewicz artery by magnetic resonance angiography in patients with descending or thoracoabdominal aortic aneurysms. Eur J Cardiothorac Surg 2000;18:104 11. 17. Kawaharada N, Morishita K, Fukada J, et al. Thoracoabdominal or descending aortic aneurysm repair after preoperative demonstration of the Adamkiewicz artery by magnetic resonance angiography. Eur J Cardiothorac Surg 2002;21:970 4. 18. Blackwood W. Discussion on vascular disease of the spinal cord. Proc Roy Soc Med 1958;51:543 7. 19. Bailey AA. Changes with age in the spinal cord. Arch Neurol Psychiat 1953;70:299 309. The Thoracic Surgery Foundation for Research and Education The Thoracic Surgery Foundation for Research and Education (TSFRE) extends best wishes to all its colleagues and friends for a joyous and peaceful holiday season and a prosperous and happy New Year. The generosity and commitment of those who have given to TSFRE have made possible our support of exciting and innovative research projects. The end of one year and start of another is a time of reflection, introspection, and consideration of the legacy we shall be leaving behind. TSFRE would ask, to acknowledge what cardiothoracic surgery has done for you, that you make an annual donation to the Foundation. Tax-deductible gifts and pledges that sum at $10,000 donated over a period of years will make you a Life Member of TSFRE. Annual gifts can include securities that have appreciated in value, thereby providing additional tax relief. You might also consider the making of a planned gift to TSFRE. The most common planned gift is a charitable bequest to TSFRE through your will after you have properly ensured that your family and other loved ones will be financially secure. Under certain circumstances, a donation from your estate might give TSFRE $10,000 while representing only about $2,500 that would go to your family should the payment of estate, income and capital gains taxes be required. Real Estate as a planned gift can be a very practical mechanism, even if the real estate property has decreased in value. A donor may even give real estate and yet retain the use of the property. Sometimes a real estate gift can provide an ongoing source of income to the donor or to another designated individual. You might also consider establishing a charitable trust in the name of your family thereby providing funds to TSFRE with a provision that the funds from which the gift derives would eventually return to you or your family or other loved ones. This announcement refers to some of the many options available to you and we urge you to consult with your financial advisor once you have made the basic decision that you would like to support research and education in cardiothoracic surgery. TSFRE would also be happy to discuss these and other planned gift options with you. Please feel free to contact Joe Webber, the Director of Development at (978) 927-8330 or e-mail him at jwebber@prri.com. Again, we thank you for your support to TSFRE. We also ask that you acknowledge what cardiothoracic surgery has done for you by supporting the research and education of our successors so that they will continue to serve patients as well as possible. 2003 by The Society of Thoracic Surgeons Ann Thorac Surg 2003;76:1971 0003-4975/03/$30.00 Published by Elsevier Inc