Potential Use of the Intercostal Artery as an In Situ Graft: A Cadaveric Study

Transcription

1 Potential Use of the Intercostal Artery as an In Situ Graft: A Cadaveric Study Lindsay C. H. John, FRCS, Christopher L. H. Chan, MB, BS, and David R. Anderson, FRCS Department of Cardiothoracic Surgery, Guy's Hospital, London, United Kingdom The third to eighth intercostal arteries (ICAs) were bilaterally dissected in 10 cadavers to assess their length and possible routes to coronary arteries if used as in situ grafts. The mean lengths for the intercostal arteries harvested were 27.0 ± 2.9 cm on the left and 27.4 ± 3.2 cm on the right. The shortest anatomic route to the coronary arteries of the in situ ICAs harvested was medial to the lung and either superior to or inferior to the hilum. By using either the superior or inferior routes in situ ICAs were long enough to reach the major coronary artery territories in all cadavers. The most suitable ICAs for grafting the coronary arteries and the shortest routes were as follows: left anterior descending--left fifth ICA by inferior route; circumflex coronary artery-left fifth ICA by inferior route; and right coronary artery-right seventh ICA by inferior route. We conclude that it is anatomically feasible to use the intercostal artery as an in situ graft in coronary artery operation. (Ann Thorac Surg 1995;59:190-5) S uperior long-term patency of internal mammary artery (IMA) grafts clearly has been demonstrated previously [1-3]. Partly as a consequence of these observations, there has been a tendency toward increased use of arterial conduits. Strategies to increase the use of arterial conduit include the use of both IMAs [3], the use of sequential IMAs [4] and the use of other arterial conduits apart from the IMA. Alternative arterial conduits that have been used include the gastroepiploic artery [5], inferior epigastric artery [6], splenic artery [7], and the radial artery [8]. It is at present unknown whether the long-term patency of some of these alternative conduits will match that of the IMA. One of the factors that has been implicated in the excellent longterm patency of the IMA is that in most cases it is used as an in situ graft. The intercostal artery (ICA) has recently been proposed as a new alternative arterial conduit [91 in a study that demonstrated its favorable histologic characteristics. It was reported that the ICA could only be used as a free graft. However, the intrathoracic position of the ICA suggests its potential as an in situ graft. To investigate this possibility a cadaveric study was undertaken to assess the potential routes and feasibility of using the ICA as an in situ coronary bypass graft. Material and Methods For this study 10 cadavers were examined (4 male and 6 female). The age range was 65 to 91 years (mean age, 76 years), and none had died of cardiac disease. In each cadaver the third to eighth ICAs were dissected out bilaterally (Fig 1). The extent of the dissection was from Accepted for publication July 27, Address reprint requests to Dr John, Department of Cardiothoracic Surgery, Guy's Hospital St. Thomas St, London SE1 9RT, UK. approximately 2 cm lateral to the vertebral bodies to the anterior end of the ICA (internal mammary or musculophrenic arteries}, where they were transected. To demonstrate the anatomy, the intercostal arteries were skeletonized. The length of each intercostal artery dissected out was measured. The routes for an in situ intercostal artery graft to the major coronary artery territories were assessed. There were two principal routes, the superior and the inferior. For the superior route the intercostal artery passed from its proximal dissection point, superior to the pulmonary hilum, through the pericardium, anterior to the proximal descending aorta and main pulmonary artery on the left side and the superior vena cava and ascending aorta on the right side and then inferiorly to reach the relevant coronary artery. For the inferior route the ICA passed from its proximal dissection point, inferior to the pulmonary vein (ie, after division of the inferior pulmonary ligament), through the pericardium and directly to the relevant coronary artery. The three major coronary artery territories assessed were (1) the left anterior descending (LAD) coronary artery, defined as approximately the junction between the proximal two thirds and distal third of the artery; (2) the lateral circumflex (LCx) coronary artery, defined as approximately the midpoint of the major branch of the circumflex artery on the lateral surface of the heart; and (3) the right coronary artery (RCA), defined as approximately the site of the RCA just distal to the acute margin of the heart. The distances to the major coronary arteries were measured for the third to the fifth ICAs through the superior route and the fifth to eighth ICAs through the inferior route. This was done bilaterally for the LAD, on the left side for the LCx and on the right for the RCA. All measurements were made for each cadaver by The Society of Thoracic Surgeons / oo (94)oo715-j

2 Ann Thorac Surg JOHN ET AL ;59:190-5 IN SITU INTERCOSTAL ARTERY GRAFTS Fig 1. Third to eighth intercostal arteries dissected out bilaterally in a cadaver. To assess the potential for each ICA as an in situ coronary artery graft, the differences between the length of each ICA and the distance of the routes to the major coronary artery territories were calculated. The optimum in situ ICA graft to each coronary artery was defined as that where there was a maximum excess in length of the dissected ICA compared with the distance of the relevant route. Examples of potential ICA grafts to the major coronary arteries are illustrated in Figs 2 through 5. Fig 2. Left fifth intercostal artery passing through the superior route to the left anterior descending coronary artery. (see Table 2). Any of the right third to eighth ICAs can reach the RCA through the superior or inferior route but the optimum appears to be the right seventh ICA through the inferior route (see Table 2). Results Data are expressed as mean _+ standard deviation. Table 1 illustrates the mean lengths of the third to eighth ICAs dissected out in the 10 cadavers on both sides. The overall mean length on the right side was 27.4 _+ 3.2 cm and cm on the left. Table 2 illustrates the mean excess of length for each ICA dissected out when routed as for an in situ graft to the major coronary arteries. The greater the excess length, the more proximally the ICA can be divided to anastomose to the relevant coronary artery. The more proximal the ICA can be divided, the greater its diameter and therefore also the greater its suitability for use as a coronary graft. On this basis, although any of the right or left third to eighth ICAs can reach the LAD by either the superior or inferior routes (apart from the right eighth ICA by the inferior route) the optimum graft and route would appear to be the left fifth ICA through the inferior route (see Table 2). Similarly, although any of the left third to eighth ICAs can reach the LCx by superior or inferior routes, the optimum graft and route would again appear to be the left fifth ICA through the inferior route Fig 3. Left fifth intercostal artery passing through the superior route to the left anterior descending coronary artery with the lung in situ.

3 192 JOHN ET AL Ann Thorac Surg IN S1TU INTERCOSTAL ARTERY GRAFTS 1995;59:190-5 Table 1. Mean Length (crn) of the Dissected Cadaveric Intercostal Arteries Intercostal Artery Side Right (cm) +_ 2.4 _ _+ 3.6 _+ 3.2 _+ 2.4 Left (crn) _ Fig 4. The left seventh intercostal artery passing through the inferior route to the lateral circumflex coronary artery. Comment Arterial conduits are useful in coronary artery bypass operation for two general reasons. First, superior long- Fig 5. Right eighth intercostal artery passing through the inferior route to the right coronary artery. term patency has been demonstrated with the use of the IMA [1-3]. There has been an expectation that alternative arterial conduits may also demonstrate such a benefit. Second, arterial conduits provide a source of graft material where there is no or poor alternative venous conduit, such as in the presence of varicose disease or in redo procedures. Although the use of the IMA can be extended by the use of bilateral IMAs or by sequential IMA grafting, complete revascularization can often not be achieved by the use of IMAs alone. Alternative arterial conduits apart from the IMA have included the gastroepiploic (GEA), inferior epigastric, radial, and splenic arteries [5-8]. Although short-term patency (up to 2 years) of gastroepiploic artery grafts appears to be comparable with that of IMA grafts [5, 10], long-term patency is not known. The use of the inferior epigastric artery as a graft is a relatively recent innovation [6], and again long-term patency is unknown. The radial artery was first used as a coronary graft in 1971 [11]. Although there has been recent renewed interest in its use [8], the initially reported patency was poor. There have been infrequent reports of the use of the splenic artery as a conduit, either in situ [12] or as a free graft [7]. However, difficulties with its use include the frequency of atherosclerosis and calcification, difficulty in harvesting, the need for splenectomy, and the discrepancy in size compared with the coronary artery. Although long-term patency figures are not available for arterial conduits apart from the IMA, one method by which their potential patency may be assessed is by determining which features of the IMA are responsible for its excellent long-term patency and assessing whether these features are present in the alternatives. Two characteristics of the IMA have been implicated. First, the IMA exhibits favorable histology. Its resistance to atherosclerosis has been related to the perfect continuity of its internal elastic membrane, preventing migration of smooth muscle cells from the media into the intima [13]. The histology of the GEA has been reported to be similar to that of the IMA [10], the implication being that they may therefore have similar patencies. This remains speculative and in addition it has been observed that the GEA is rich with smooth muscle cells in the media whereas the IMA has a large number of elastic fibers in its main trunk [14], which may explain the greater tendency for spasm in the GEA. The second characteristic that might explain the durability of IMA grafts is its release of endotheliumderived relaxing factor and prostacyclin [15-17[. In addi-

4 Ann Thorac Surg JOHN ET AL ;59:190-5 IN SITU INTERCOSTAL ARTERY GRAFTS Table 2. Mean Excess Length (cm) of Each Intercostal Artery Passing to the Major Coronary Arteries Superior Route Inferior Destination/Artery ICA3 ICA4 ICA5 ICA5 ICA6 ICA7 ICA8 To left anterior descending coronary artery Right side (cm) ± _ ± 6.6 Left side (cm) ~ _ _ ± _ 3.8 To lateral circumflex coronary artery Left side (cm) 7 +_ ~ ± To right coronary artery Right side (crn) 1.4 _ z _ _ _ _+ 3.3 ICA = intercostal arteries (3-8 represent the third to eighth ICAs). tion, the response of smooth muscle to endotheliumderived relaxing factor appears to vary between different arteries [18] as well as being greater in the IMA compared with the saphenous vein [15]. It has been reported [19] that the GEA has a strong capacity to secrete such vasodilators and inhibitors of platelet function which may influence long-term patency. Although the GEA has had good short-term patency reported and there are theoretical reasons discussed above why it also may have a good long-term patency, there are potential complications with its use. These include possible gastric devascularization and difficulties if future laparotomy should be required. However, reports to date have not found such potential problems significant [201. An unanswered question of relevance to the use of potential alternative arterial conduits is the significance of in situ versus free grafts. It has been reported that the patency rate of free IMA grafts approaches that of in situ grafts [21]. In addition, it has been speculated that as the IMA may be nourished entirely through its lumen, there may be minimal risk of ischemic injury to the arterial wall in a free graft [22]. However, there is some evidence that free IMA grafts may not be so beneficial. In a reported comparison of patency at a mean of 9.2 months postoperatively between radial artery, in situ IMA, and free IMA grafts, the respective patencies were 93.5%, 100%, and 69.3%, respectively [8]. The GEA has been used as both a free graft and as an in situ graft. Again, there is some evidence that there is improved patency when used as an in situ graft; in one series the 2-month patency of free grafts was 75% compared with 95% for in situ grafts [5]. Although the GEA is used as an in situ graft, this does limit the coronary artery territories that can be grafted. As of necessity radial artery grafts are free grafts, the poor patency of the early radial artery grafts [23] may in part have been due to ischemia of the thick media and the absence of medial vasa vasorum [14]. An ideal arterial conduit would appear to be one that has a similar histology to the IMA, can be used as an in situ graft, and is of sufficient quantity to allow multiple grafts. It recently has been reported [9] that the intercostal artery (ICA) has similar histologic features to the IMA. Specifically, it appears to have multiple elastic lamellae in its media, which is thin, and it also has a thin intima. These features would appear to favor a good long-term patency if used as a graft. The presence of multiple, bilateral ICAs offers the potential for multiple grafts. However, the ICA has to date only been considered suitable as a free graft. The present study has demonstrated in cadavers that it is anatomically feasable to perform in situ ICA grafts to any of the major coronary artery territories. Specifically, the optimum graft to the LAD appears to be the left fifth ICA through the inferior route. The optimum graft to the LCx would also appear to be the left fifth ICA through the inferior route, although for both the LAD and LCx the left sixth ICA through the inferior route appears to be almost as suitable. The right seventh ICA through the inferior route appears to be the optimum graft to the RCA. For the purpose of demonstrating the anatomy, the ICAs were skeletonized in this study. However, it is envisaged that in clinical use they would be dissected as a pedicle. An observation of possible clinical relevance was that despite the cadaver population being relatively elderly, there was no obvious atheromatous disease in any of the intercostal arteries dissected. It remains to be established whether the diameter of the ICA at the anastomotic site and its flow would be suitable for coronary artery revascularization. The free flow of the transected ICA at the midaxillary level has been measured as 80 to 100 ml]min [9]. However, free flow is not necessarily relevant and has not been shown to correlate with measurement of flow after grafting with the IMA, nor with clinical outcome I24]. Of more relevance may be the diameter of the vessel. From the application of the Poiseuille-Hagen formula the smaller the diameter of a tube for a given flow, the greater the pressure drop across that tube. The mean luminal diameter of the fifth ICA at autopsy has been reported previously [9] as being 1.4 +_ 0.3 cm at its origin and

5 194 JOHN ET AL Ann Thorac Surg IN SITU INTERCOSTAL ARTERY GRAFTS 1995;59: cm at its distal end. However, it has been estimated that there would have been a 30%-40% reduction in diameter in this situation due to the flaccid state and rigor mortis. Therefore, the probable diameter of the ICA at the site of anastomosis would be similar to that of a "small" IMA or a GEA in which the mean anastomotic diameter has been estimated as being between 1.25 to 1.5 mm [10]. In addition, the size of a graft is not necessarily static; it has been observed that the diameters of inferior epigastric arteries, when grafted to a large viable myocardial area, were larger 10 days after grafting than when measured before grafting [6]. One potential problem is the risk of medullar ischemia, particularly when harvesting intercostal arteries beyond the fifth pair. We believe that limiting the proximal dissection to approximately 2 cm lateral to the vertebral bodies will avoid dissection of any spinal tributaries and reduce the likelihood of this complication occurring in clinical practice. A potential problem with the use of the ICA as a coronary graft is access for harvesting. This could be overcome by the following methods. First, the ICA could be harvested through a median sternotomy with the pleura widely open while on cardiopulmonary bypass with the lungs collapsed. Alternatively, both the harvesting of the ICA and exposure of the heart could be performed through an anterolateral or posterolateral thoracotomy incision [25, 26]. However, although exposure to the LCx territory would be good, that to the LAD and RCA would be poor unless a bilateral thoracotomy is considered [27]. Finally, there remains the possibility of harvesting ICAs by minimally invasive means using videoscopic thoracoscopy. Such a method has been used for experimental harvesting of IMAs from pigs [28]. In conclusion, this study has demonstrated that in situ ICA artery grafts to the major coronary artery territories is anatomically feasible and offers the possibility of an alternative arterial conduit that has the histologic potential for long-term patency together with the suitability for use in multiple coronary grafting. We are grateful to the Anatomy Department of Guy's Hospital Medical School and to Kevin Fitzpatrick, FRPS, for the photography. References 1. Grondin CM, Campeau L, Lesperance J, Enjalbert M, Bourassa MG. Comparison of late changes in internal mammary artery and saphenous vein grafts in two consecutive patients 10 years after operation. Circulation 1984;70(Suppl 1): Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal mammary artery graft on 10 year survival and after cardiac events. N Engl J Med 1986;314: Spencer FC. The internal mammary artery: the ideal coronary bypass graft? N Engl J Med 1986;314: Kamath M, Matysik L, Schmidt D, Smith L. Sequential internal mammary artery grafts. Expanded utilization of an ideal conduit. J Thorac Cardiovasc Surg 1985;89: Suma H, Wanibuchi Y, Terada T, et al. The right gastroepiploic artery graft. Clinical and angiographic midterm results in 200 patients. J Thorac Cardiovasc Surg 1993;105: Buche M, Schoevaerdts J-C, Louagie Y, et al. Use of the inferior epigastric artery for coronary bypass. J Thorac Cardiovasc Surg 1992;103: Mueller PK, Blakeman BP, Pickelman J. Free splenic artery used in aortocoronary bypass. Ann Thorac Surg 1993;55: Acar C, Jebara VA, Portoghese M, et al. Revival of the radial artery for coronary artery bypass grafting. Ann Thorac Surg 1992;54: Van Son JAM, Smedts F, Korving J, Guyt A, de Kok LB. Intercostal artery: histomorphometric study to assess its suitability as a coronary bypass graft. Ann Thorac Surg 1993;56: Lytle BW, Cosgrove DM, Ratliff NB, Loop FD. Coronary artery bypass grafting with the right gastroepiploic artery. J Thorac Cardiovasc Surg 1989;97: Carpentier A, Guermonprez JL, Deloche A, Frechette C, Dubost C. The aorta-to-coronary radial artery bypass graft: a technique avoiding pathological changes in grafts. Ann Thorac Surg 1973;16: Edwards WS, Lewis CE, Blakeley WR, Napolitano L. Coronary artery bypass with internal mammary and splenic artery grafts. Ann Thorac Surg 1973;15: Sims FH. The internal mammary artery as a bypass graft. Ann Thorac Surg 1987;44: Van Son JAM, Smedts F, Vincent JG, Van Lier HJ, Kubat K. Comparative anatomic studies of various arterial conduits for myocardial revascularization. J Thorac Cardiovasc Surg 1990;99: LiJscher TF, Diederich D, Siebenmann R, et al. Difference between endothelium-dependent relaxation in arterial and in venous coronary bypass grafts. N Engl J Med 1988;319: Subramanian VA, Hemandez Y, Tack-Goldman IG Grabowski EF. Prostacyclin production by internal mammary artery as a factor in coronary bypass grafts. Surgery 1986;100: Pearson PJ, Evora PRB, Schaff HV. Bioassay of EDRF from internal mammary arteries: implications for early and late bypass graft patency. Ann Thorac Surg 1992;54: Christie MI, Lewis MJ. Vascular smooth muscle sensitivity to endothelium-derived relaxing factor is different in different arteries. Br J Pharmacol 1988;95: O'Neil GS, Chester AH, Allen SP, et al. Endothelial function of human gastroepiploic artery. Implications for its use as a bypass graft. J Thorac Cardiovasc Surg 1991;102: Suma H, Wanibuchi Y, Furuta S, Takeuchi A. Does use of gastroepiploic artery graft increase surgical risk? J Thorac Cardiovasc Surg 1991;101: Loop FD, Lytle BW, Cosgrove DM, Golding LAR, Taylor PC, Stewart RW. Free (aorta-coronary) internal mammary artery graft: late results. J Thorac Cardiovasc Surg 1989;97: Landymore RW, Chapman DM. Anatomical studies to support the expanded use of the internal mammary artery graft for myocardial revascularization. Ann Thorac Surg 1987;44: Fisk RL, Brooks CH, Callaghan JC, Dvorkin J. Experience with the radial artery for coronary bypass. Ann Thorac Surg 1976;21: Louagie YAG, Haxhe J-P, Buche M, Schoevaerdts J-C. Intraoperative electromagnetic flowmeter measurements in coronary artery bypass grafts. Ann Thorac Surg 1994;57: Burlingame MW, Bonchek LI, Vazales BE. Left thoracotomy for reoperative coronary bypass. J Thorac Cardiovasc Surg 1988;95: Gandjbakhch I, Acar C, Cabrol C. Left thoracotomy approach for coronary artery bypass grafting in patients with pericardial adhesions. Ann Thorac Surg 1989;48: Marshall WG, Meng RL, Ehrenhaft JL. Coronary artery bypass grafting in patients with a tracheostoma: Use of bilateral thoracotomy incision. Ann Thorac Surg 1988;46: Peters WS. Minimally invasive cardiac surgery by cardioscopy [Invited letter]. Australas J Cardiac Thorac Surg 1993; 2:152-4.

6 Ann Thorac Surg JOHN ET AL ;59:190-5 IN SITU INTERCOSTAL ARTERY GRAFTS INVITED COMMENTARY This article is a fascinating anatomic study in cadavers in which John and associates clearly demonstrate that pedicled intercostal arteries can reach any coronary artery. Specifically, the third through eighth intercostal arteries on both the right and left chest wall could reach the heart through either the superior or inferior hilar route. In a recent histologic study of the intercostal artery [1], three combinations of histologic patterns were seen in the media of the vessel. These included, in respective sequence of prevalence, (1) a proximal elastic segment followed by subsequent elastomuscular and muscular segments, (2) a proximal elastomuscular segment with the remainder of the artery being muscular, and (3) a completely muscular pattern. In addition, the intima and media appeared to be thin. These properties, especially the elastic and elastomuscular content of the media, are favorable with regard to potential suitability of the intercostal artery as a conduit in myocardial revascularization. My major concern about clinical use of the intercostal artery consists of potentially inadequate luminal diameter and, therefore, inadequate inflow into the coronary arteries. In the aforementioned morphometric study of the intercostal artery, the mean luminal diameter of the fifth intercostal artery varied from mm at the origin to mm at 30 cm, with an estimated 30% decrease in diameter due to the flaccid state and rigor mortis. Therefore, as John and associates state, the probable diameter of the intercostal artery at the site of the anastomosis would be similar to that of a "small" internal mammary artery or right gastroepiploic artery. In summary, anatomically it is feasible to use the intercostal artery as an in situ graft for myocardial revascularization. The critical issue that remains is whether this conduit is of adequate diameter to completely or partially sustain the coronary circulation. Future studies should focus on this issue. Successful clinical use of the intercostal artery in myocardial revascularization would be a milestone after a long evolution of its use in cardiac surgery. Jacques A. M. van Son, MD, PhD Division of Cardiothoracic Surgery University of California, San Francisco 505 Parnassus Ave San Francisco, CA Reference 1. Van Son JAM, Smedts F, Korving J, Guyt A, de Kok LB. Intercostal artery: histomorphometric study to assess its suitability as a coronary bypass graft. Ann Thorac Surg 1993;56: