- Email: [email protected]
Nonreversed saphenous vein grafts for coronary artery bypass grafting
Nonreversed Saphenous Vein Grafts for Coronary Artery Bypass Grafting J. Ernest0 Molina, MD, PhD Division of Cardiovascular and Thoracic Surgery, University of Minnesota Medical School, Minnesota Heart and Lung Institute, Minneapolis, Minnesota
Between August 1985 and December 1988, valvotomized saphenous vein grafts were used in 365 patients undergoing coronary artery bypass grafting (CABG). In this operation, the femoral end of the vein is attached to the aorta and the pedal end is attached to the coronary artery. Vein diameters measured 8 f 2 mm at the femoral end, 4.5 f 1.2 mm at the knee level, and 3.5 f 1.3 mm at the ankle. Ratios between levels were as follows: knee to femoral end, 0.56, and ankle to femoral end, 0.43. The ratio of knee to femoral end was 0.42 in cases with vein midthigh bifurcation. There were 1,310 grafts implanted (3.6 per patient). In 341 patients, CABG alone was performed, and 24 patients had combined procedures: 11 had CABG with mitral valve replacement, 9 had CABG with aortic valve replacement, 2 had CABG with repair
of postinfarct ventricular septal defect, and 2 had CABG with automatic defibrillator implantation. Follow-up (up to 3.5 years) was attained in 97% of patients. For various reasons, 34 patients had a second angiogram between 3 and 41 months postoperatively. Of 120 vein grafts, 108 (90%) were patent. At autopsy, 11 patients with 45 vein grafts had 43 patent and clean grafts and two thrombosed. Use of nonreversed saphenous vein for coronary bypass is recommended. It assures a large proximal anastomosis, natural vein bifurcations can be used with fewer proximal anastomoses, better vein-coronary artery size matching is obtained, and the patency rate is satisfactory.
hen an anastomosis is performed between saphenous veins and coronary arteries, a mismatch exists: the vein is substantially larger than the recipient vessel. Moreover, if the vein is positioned in a reversed manner (ie, with the pedal end of the vein anastomosed to the ascending aorta, and the femoral end to the coronary artery), the mismatch is even greater. Among the hemodynamic considerations affecting the reversed position is the likelihood of developing stenosis at the proximal suture line between the aorta and the vein, particularly when the vein is small and is stretched or flattened to reach and cover the aortotomy site. At the distal end, more turbulence may occur, causing the blood to flow from a small opening in the aorta to a wider distal end connected to a small vessel. This may encourage thrombosis. Nonreversed saphenous vein has been used to bypass the branched renal artery and the hepatic artery [l-31; in the lower extremities, it has been used in that position for the in situ saphenous vein bypass procedure that offers better results than the reversed operation [MI. In about 5% of patients, the saphenous vein bifurcates in the upper part of the thigh, giving origin to two veins that usually run parallel in the thigh down to the knee level. Using the vein in a nonreversed position allows this naturally occurring bifurcation to bypass more coronary branches with only one proximal anastomosis. With time, fibrotic changes that occur in the veins usually lead to a thickening of the intima, with reduction
of its lumen. If the proximal anastomosis is large enough to compensate for this change, fewer possibilities exist of causing early obstruction of the vein. The only potential risk factor in valvotomizing the saphenous vein graft is the question of intimal damage that may induce thrombosis.
Accepted for publication April 12, 1989. Address reprint requests to Dr Molina, University of Minnesota, Box 182 UMHC,420 Delaware St, SE, Minneapolis, MN 55455. 0 1989 by
The Society of Thoracic Surgeons
(Ann Thoruc Surg 1989;48:624-7)
Material a n d Methods Between August 1985 and December 1988, 365 patients were operated on at the University of Minnesota Hospitals with valvotomized saphenous vein grafts used in a nonreversed position for coronary bypass. Of this group, 341 underwent coronary artery bypass grafting (CABG) alone. In 24 patients, combined procedures were done: 11 had CABG with mitral valve replacement, 9 had CABG with aortic valve replacement, 2 had CABG with repair of an acute ventricular septal defect caused by infarction or a ruptured ventricle, and 2 had CABG with automatic defibrillator implantation. We operated on 47 patients with an acute infarction and postinfarction angina (14%).
Operative Technique Measurements were obtained of the vein diameter. The cross-sectional area was calculated at three levels: the femoral end, the knee level, and the ankle (Fig 1). In patients with bifurcation or duplication (Fig 2) of the saphenous vein in the thigh, only two levels were measured, the femoral end and the knee level. We tested the vein for leaks by inserting a blunt, beaded-tipped needle at the femoral end and injecting heparinized blood to distend it (Fig 3). A Mills valvulotome (Fig 4) was inserted Ooo3-4975/89/!§3.50
MOLINA NONREVERSED SAPHENOUS VEIN G M S
Ann Thorac Surg 1989;48:624-7
625
midthigh Fig 1. Dimensions found in a routine segment of greater saphenous vein graft with calculated cross-sectional areas (in millimeters).
through the distal end and advanced to the proximal end, severing the valves when it was withdrawn. A bifurcated vein was used every time if the bifurcation was more than 2 cm away from the femoral vein and the branches were long enough to reach the coronary arteries. This situation allowed positioning of the bifurcated graft either in front of or behind the great vessels in the transverse sinus. Whenever possible, sequential graftings were performed with each of the limbs. In 120 patients, the proximal ends were anastomosed to the aorta first, before the patient was placed on cardiopulmonary bypass. The patients selected for this technique had stable angina, no left main coronary obstruction, and no recent infarction. The proximal saphenous vein anastomosis to the aorta was usually constructed on the right side of the ascending aorta, tailored with a bevel oriented toward the back to pass the saphenous vein in the transverse sinus behind the ascending aorta and with the pulmonary artery exiting over the left atrial appendage. On removal of the side-biting clamp in the aorta, the vein was filled immediately with the blood under systemic pressure, preventing any twisting of the graft itself. When the vein is bifurcated, this maneuver allows optimal positioning of the graft over the vessels to be bypassed and provides measurement of the exact length to the prospective coronary arteries while the heart is still normally distended. The actual flow in the graft can be precisely measured under the patient’s normal hemodynamic conditions. In the remaining 245 patients, we connected the distal ends first under cardioplegic arrest and the proximal ends later. With either technique, all the proximal anastomoses to the aorta were performed to a circular orifice made with a disposable punch-instrument
B Fig 2 . A case of normally occurring bifurcation of the saphenous vein shown in situ ( A ) and after excision and valvotomy of all the vein values and being tested by injecting fluid at the femoral end (B).
(Hancock) 6 mm in diameter, using 6-0 monofilament suture. The sequential anastomoses were performed by making a longitudinal incision in the vein and sewing it crossfashion with the coronary artery using 7-0 monofilament suture material. Because the saphenous vein valves are destroyed during preparation of the graft, the air is bled either proximally or distally, without the need to puncture the graft at any level.
Adjuvant Drug Therapy All patients received Persantine (dipyridamole, 75 mg orally three times a day) and Ascriptin (aspirin plus magnesium hydroxide and dried aluminum hydroxide, 325 mg every day) postoperatively for at least 1 year.
Follow-up No routine postoperative angiograms were obtained, but 34 patients have had a second angiogram for various reasons in the 3.5 years of postoperative follow-up. Twelve patients (3.2%)were lost to follow-up.
626
MOLINA NONREVERSED SAPHENOUS VEIN GRAFTS
Ann Thorac Surg 1989;48:624-7
A
Fig 3. Saphenous vein being injected at its femoral end, showing positioning and technique of performing the valvotomy with a Mills valvulotome.
Results Coronary arteriography performed in 34 patients, with 120 grafts, between 3 and 41 months postoperatively showed 90% or 108 grafts to be patent and functioning well. The mean time to angiography was 19.3 months. Angiography showed no evidence of vein valve abnormalities in the grafts. Twelve patients (3.2%)died in the postoperative period, most in the acutely infarcted group (6 of 47, or 12.7%).Their vein grafts, 45 in all, were examined at autopsy: 43 (95%)were found patent without any evidence of thrombosis or stenosis at any level. In 28 specimens, the site of the former vein valve was identified, with small curled remnants of tissue without thrombus or any other pathology. In the remaining 15 specimens, the valve site could not be distinguished from the rest of the vein wall. Only two grafts were thrombosed. Of the remaining 353 patients, 76 underwent treadmill exercise tests postoperatively; in 2 who had suggestions of ischemia, coronary arteriography was performed. In both, however, the grafts were patent. At various times, ranging from 18 months to 3 years, 7 patients required reoperation due to stensosis in coronary arteries beyond the anastomosis or at the suture line itself or to development of new lesions since the first operation. In 4 of these patients, part of the old saphenous vein graft was removed: in 2 patients at 2.5 years after the original operation and in 2 others more than 3 years after the original operation. The veins showed atherosclerosis and cholesterol deposits in the wall, but determination of the site where the vein valves had been located was impossible.
Comment Saphenous veins are not the perfect or ideal type of grafts for aortocoronary bypass. Pedicle arterial grafts are supe-
Fig 4 . ( A ) Insertion of valvulotome at distal end of the Saphenous vein graft. (B)After valvotomy is almost completed.
rior, but because so few arteries are available for such use and because most patients need implantation of multiple grafts, the saphenous vein will continue to be the most commonly used aortocoronary conduit and the standard with which to compare any new modification. Because the attrition rate of vein occlusion increases with time, many patients will eventually require reoperation when their saphenous vein grafts become nonfunctional. Although atherosclerosis is most commonly blamed for late occlusion of saphenous vein grafts, fibrosis and destruction of the normal morphology of the saphenous vein wall also causes occlusion. Such degeneration may be inherent to removing the vein from its natural position and devascularizing and implanting it in a new position. The vein wall dies either before a new vasa vasorum system can develop or because the diffusion process of cellular nutrition is not sufficient. Since its inception in the late 1960s, the aortocoronary bypass grafting operation has used the saphenous vein in a reverse position. This was probably an acceptable technique, previously used for femoral-popliteal bypass grafting, because the vein valves could not be destroyed unless the vein was everted totally and the valves were removed with scissors. There was, however, considerable risk of
MOLINA NONREVERSED SAPHENOUS VEIN GRAFTS
Ann Thorac Surg
1989;48:624-7
intimal damage, and the maneuver was tedious and time-consuming. Since the design of valvulotomes of the Mills type, this maneuver can be done very easily with minimal or no trauma and within a few seconds [I, 61. Chin and colleagues [7] showed that vein valves, even in the reversed position, reduce the effective orifice and flow in small veins. When the veins were tested under pulsatile flow, the flow increased only if the vein valves had been destroyed. The smaller the vein diameter, the more significant these changes were. Although the changes observed were significant when the vein was smaller than 2.5 mm in diameter, they may occur in veins somewhat larger but are not detectable by the method of analysis used. It appears that a valvotomized vein, even in the reversed position, has some advantages over its nonvalvotomized counterpart. There is no information in the literature as to whether valves in the reversed position play any role when they are used in aortocoronary bypass. During systole when the left ventricle contracts forcefully, the coronary flow stops within the myocardium and the blood remaining in the epicardial portion of the coronary artery may reflux freely back into the aorta. In this manner, a “washing mechanism” may help to keep a graft open. Whether the presence of a valve in the vein prevents this movement is unknown, but it might play a role in long-term patency. The valvotomized vein used as a bypass graft functions more like a normal coronary artery. The main advantage of using the vein in a nonreversed position is that the proximal orifice in the aorta can be made larger, thereby preventing fibrosis leading to obstruction at the aortic anastomosis. Second, the vein tapers gradually to a smaller size that connects to the coronary artery. Therefore, the mismatch between vein and artery is diminished and rheology may be improved (81. In our study’s long-term follow-up, there was no evidence that saphenous veins used in a nonreversed position occlude earlier or more frequently than those in a reversed position. In fact, the contrary appears to be the case. Therefore, the minimal trauma associated with the valvotomy is probably of no importance. A clear advan-
627
tage of the method is the use of normally occurring bifurcations in the greater saphenous vein system. A single proximal coronary anastomosis to the aorta may feed two or more branches of saphenous vein, which can be sequentially attached to the coronary arteries, allowing for multiple grafting and shortening the operating time. We believe the advantages of using saphenous vein grafts in a nonreversed position for coronary artery bypass are as follows: (1) large proximal orifice; (2) use of natural bifurcation; (3) better size-matching of vein and artery; (4) retrograde air venting; (5) less chance of turbulent flow; (6) equal or better patency rate; and (7) shorter operation. The use of nonreversed saphenous veins is therefore recommended. No disadvantages have been found in 3.5-year follow-up, and the advantages are quite substantial.
References 1. Whittemore AD, Thompson RW, Ruby ST, Mannick JA. Branched non-reverse saphenous vein graft for complex renal artery stenosis. Arch Surg 1987;122:84&7. 2. Novick AC, Pohl MA. Atherosclerotic renal artery occlusion extending into branches: successful revascularization in situ with a branched saphenous vein graft. J Urol 1979;122:240-2. 3. Streem SB, Novick AC. Aortorenal bypass with a branched saphenous vein graft for in situ repair of multiple segmented renal arteries. Surg Gynecol Obstet 1982;155:855-9. 4. Leather RP, Powers SR, Karmody AM. A reappraisal of the in situ saphenous vein arterial bypass: its use in limb salvage surgery. Surgery 1979;86:45341. 5. Skagseth E, Hall KV. In situ vein bypass: experiences with new vein valve strippers. Scand J Thorac Cardiovasc Surg 1973;753-6. 6. Leather RP, Shah DM, Buchbinder D, Annest SJ, Karmody AM. Further experience with the saphenous vein used in situ for arterial bypass. Am J Surg 1981;142:506-10. 7. Chin AK, Mayer DN, Goldman RK, Lerman JA, Olcott C, Fogarty TJ. The effect of valvotomy on the flow rate through the saphenous vein graft: clinical implications. J Vasc Surg 1988;8:31&20. 8. Molina JE. The use of the non-reversed saphenous vein for aortocoronary bypass surgery. Symposium for current problems in coronary surgery, Freiburg, FRG, Sep 1986.
Between August 1985 and December 1988, valvotomized saphenous vein grafts were used in 365 patients undergoing coronary artery bypass grafting (CABG). In this operation, the femoral end of the vein is attached to the aorta and the pedal end is attached to the coronary artery. Vein diameters measured 8 f 2 mm at the femoral end, 4.5 f 1.2 mm at the knee level, and 3.5 f 1.3 mm at the ankle. Ratios between levels were as follows: knee to femoral end, 0.56, and ankle to femoral end, 0.43. The ratio of knee to femoral end was 0.42 in cases with vein midthigh bifurcation. There were 1,310 grafts implanted (3.6 per patient). In 341 patients, CABG alone was performed, and 24 patients had combined procedures: 11 had CABG with mitral valve replacement, 9 had CABG with aortic valve replacement, 2 had CABG with repair
of postinfarct ventricular septal defect, and 2 had CABG with automatic defibrillator implantation. Follow-up (up to 3.5 years) was attained in 97% of patients. For various reasons, 34 patients had a second angiogram between 3 and 41 months postoperatively. Of 120 vein grafts, 108 (90%) were patent. At autopsy, 11 patients with 45 vein grafts had 43 patent and clean grafts and two thrombosed. Use of nonreversed saphenous vein for coronary bypass is recommended. It assures a large proximal anastomosis, natural vein bifurcations can be used with fewer proximal anastomoses, better vein-coronary artery size matching is obtained, and the patency rate is satisfactory.
hen an anastomosis is performed between saphenous veins and coronary arteries, a mismatch exists: the vein is substantially larger than the recipient vessel. Moreover, if the vein is positioned in a reversed manner (ie, with the pedal end of the vein anastomosed to the ascending aorta, and the femoral end to the coronary artery), the mismatch is even greater. Among the hemodynamic considerations affecting the reversed position is the likelihood of developing stenosis at the proximal suture line between the aorta and the vein, particularly when the vein is small and is stretched or flattened to reach and cover the aortotomy site. At the distal end, more turbulence may occur, causing the blood to flow from a small opening in the aorta to a wider distal end connected to a small vessel. This may encourage thrombosis. Nonreversed saphenous vein has been used to bypass the branched renal artery and the hepatic artery [l-31; in the lower extremities, it has been used in that position for the in situ saphenous vein bypass procedure that offers better results than the reversed operation [MI. In about 5% of patients, the saphenous vein bifurcates in the upper part of the thigh, giving origin to two veins that usually run parallel in the thigh down to the knee level. Using the vein in a nonreversed position allows this naturally occurring bifurcation to bypass more coronary branches with only one proximal anastomosis. With time, fibrotic changes that occur in the veins usually lead to a thickening of the intima, with reduction
of its lumen. If the proximal anastomosis is large enough to compensate for this change, fewer possibilities exist of causing early obstruction of the vein. The only potential risk factor in valvotomizing the saphenous vein graft is the question of intimal damage that may induce thrombosis.
Accepted for publication April 12, 1989. Address reprint requests to Dr Molina, University of Minnesota, Box 182 UMHC,420 Delaware St, SE, Minneapolis, MN 55455. 0 1989 by
The Society of Thoracic Surgeons
(Ann Thoruc Surg 1989;48:624-7)
Material a n d Methods Between August 1985 and December 1988, 365 patients were operated on at the University of Minnesota Hospitals with valvotomized saphenous vein grafts used in a nonreversed position for coronary bypass. Of this group, 341 underwent coronary artery bypass grafting (CABG) alone. In 24 patients, combined procedures were done: 11 had CABG with mitral valve replacement, 9 had CABG with aortic valve replacement, 2 had CABG with repair of an acute ventricular septal defect caused by infarction or a ruptured ventricle, and 2 had CABG with automatic defibrillator implantation. We operated on 47 patients with an acute infarction and postinfarction angina (14%).
Operative Technique Measurements were obtained of the vein diameter. The cross-sectional area was calculated at three levels: the femoral end, the knee level, and the ankle (Fig 1). In patients with bifurcation or duplication (Fig 2) of the saphenous vein in the thigh, only two levels were measured, the femoral end and the knee level. We tested the vein for leaks by inserting a blunt, beaded-tipped needle at the femoral end and injecting heparinized blood to distend it (Fig 3). A Mills valvulotome (Fig 4) was inserted Ooo3-4975/89/!§3.50
MOLINA NONREVERSED SAPHENOUS VEIN G M S
Ann Thorac Surg 1989;48:624-7
625
midthigh Fig 1. Dimensions found in a routine segment of greater saphenous vein graft with calculated cross-sectional areas (in millimeters).
through the distal end and advanced to the proximal end, severing the valves when it was withdrawn. A bifurcated vein was used every time if the bifurcation was more than 2 cm away from the femoral vein and the branches were long enough to reach the coronary arteries. This situation allowed positioning of the bifurcated graft either in front of or behind the great vessels in the transverse sinus. Whenever possible, sequential graftings were performed with each of the limbs. In 120 patients, the proximal ends were anastomosed to the aorta first, before the patient was placed on cardiopulmonary bypass. The patients selected for this technique had stable angina, no left main coronary obstruction, and no recent infarction. The proximal saphenous vein anastomosis to the aorta was usually constructed on the right side of the ascending aorta, tailored with a bevel oriented toward the back to pass the saphenous vein in the transverse sinus behind the ascending aorta and with the pulmonary artery exiting over the left atrial appendage. On removal of the side-biting clamp in the aorta, the vein was filled immediately with the blood under systemic pressure, preventing any twisting of the graft itself. When the vein is bifurcated, this maneuver allows optimal positioning of the graft over the vessels to be bypassed and provides measurement of the exact length to the prospective coronary arteries while the heart is still normally distended. The actual flow in the graft can be precisely measured under the patient’s normal hemodynamic conditions. In the remaining 245 patients, we connected the distal ends first under cardioplegic arrest and the proximal ends later. With either technique, all the proximal anastomoses to the aorta were performed to a circular orifice made with a disposable punch-instrument
B Fig 2 . A case of normally occurring bifurcation of the saphenous vein shown in situ ( A ) and after excision and valvotomy of all the vein values and being tested by injecting fluid at the femoral end (B).
(Hancock) 6 mm in diameter, using 6-0 monofilament suture. The sequential anastomoses were performed by making a longitudinal incision in the vein and sewing it crossfashion with the coronary artery using 7-0 monofilament suture material. Because the saphenous vein valves are destroyed during preparation of the graft, the air is bled either proximally or distally, without the need to puncture the graft at any level.
Adjuvant Drug Therapy All patients received Persantine (dipyridamole, 75 mg orally three times a day) and Ascriptin (aspirin plus magnesium hydroxide and dried aluminum hydroxide, 325 mg every day) postoperatively for at least 1 year.
Follow-up No routine postoperative angiograms were obtained, but 34 patients have had a second angiogram for various reasons in the 3.5 years of postoperative follow-up. Twelve patients (3.2%)were lost to follow-up.
626
MOLINA NONREVERSED SAPHENOUS VEIN GRAFTS
Ann Thorac Surg 1989;48:624-7
A
Fig 3. Saphenous vein being injected at its femoral end, showing positioning and technique of performing the valvotomy with a Mills valvulotome.
Results Coronary arteriography performed in 34 patients, with 120 grafts, between 3 and 41 months postoperatively showed 90% or 108 grafts to be patent and functioning well. The mean time to angiography was 19.3 months. Angiography showed no evidence of vein valve abnormalities in the grafts. Twelve patients (3.2%)died in the postoperative period, most in the acutely infarcted group (6 of 47, or 12.7%).Their vein grafts, 45 in all, were examined at autopsy: 43 (95%)were found patent without any evidence of thrombosis or stenosis at any level. In 28 specimens, the site of the former vein valve was identified, with small curled remnants of tissue without thrombus or any other pathology. In the remaining 15 specimens, the valve site could not be distinguished from the rest of the vein wall. Only two grafts were thrombosed. Of the remaining 353 patients, 76 underwent treadmill exercise tests postoperatively; in 2 who had suggestions of ischemia, coronary arteriography was performed. In both, however, the grafts were patent. At various times, ranging from 18 months to 3 years, 7 patients required reoperation due to stensosis in coronary arteries beyond the anastomosis or at the suture line itself or to development of new lesions since the first operation. In 4 of these patients, part of the old saphenous vein graft was removed: in 2 patients at 2.5 years after the original operation and in 2 others more than 3 years after the original operation. The veins showed atherosclerosis and cholesterol deposits in the wall, but determination of the site where the vein valves had been located was impossible.
Comment Saphenous veins are not the perfect or ideal type of grafts for aortocoronary bypass. Pedicle arterial grafts are supe-
Fig 4 . ( A ) Insertion of valvulotome at distal end of the Saphenous vein graft. (B)After valvotomy is almost completed.
rior, but because so few arteries are available for such use and because most patients need implantation of multiple grafts, the saphenous vein will continue to be the most commonly used aortocoronary conduit and the standard with which to compare any new modification. Because the attrition rate of vein occlusion increases with time, many patients will eventually require reoperation when their saphenous vein grafts become nonfunctional. Although atherosclerosis is most commonly blamed for late occlusion of saphenous vein grafts, fibrosis and destruction of the normal morphology of the saphenous vein wall also causes occlusion. Such degeneration may be inherent to removing the vein from its natural position and devascularizing and implanting it in a new position. The vein wall dies either before a new vasa vasorum system can develop or because the diffusion process of cellular nutrition is not sufficient. Since its inception in the late 1960s, the aortocoronary bypass grafting operation has used the saphenous vein in a reverse position. This was probably an acceptable technique, previously used for femoral-popliteal bypass grafting, because the vein valves could not be destroyed unless the vein was everted totally and the valves were removed with scissors. There was, however, considerable risk of
MOLINA NONREVERSED SAPHENOUS VEIN GRAFTS
Ann Thorac Surg
1989;48:624-7
intimal damage, and the maneuver was tedious and time-consuming. Since the design of valvulotomes of the Mills type, this maneuver can be done very easily with minimal or no trauma and within a few seconds [I, 61. Chin and colleagues [7] showed that vein valves, even in the reversed position, reduce the effective orifice and flow in small veins. When the veins were tested under pulsatile flow, the flow increased only if the vein valves had been destroyed. The smaller the vein diameter, the more significant these changes were. Although the changes observed were significant when the vein was smaller than 2.5 mm in diameter, they may occur in veins somewhat larger but are not detectable by the method of analysis used. It appears that a valvotomized vein, even in the reversed position, has some advantages over its nonvalvotomized counterpart. There is no information in the literature as to whether valves in the reversed position play any role when they are used in aortocoronary bypass. During systole when the left ventricle contracts forcefully, the coronary flow stops within the myocardium and the blood remaining in the epicardial portion of the coronary artery may reflux freely back into the aorta. In this manner, a “washing mechanism” may help to keep a graft open. Whether the presence of a valve in the vein prevents this movement is unknown, but it might play a role in long-term patency. The valvotomized vein used as a bypass graft functions more like a normal coronary artery. The main advantage of using the vein in a nonreversed position is that the proximal orifice in the aorta can be made larger, thereby preventing fibrosis leading to obstruction at the aortic anastomosis. Second, the vein tapers gradually to a smaller size that connects to the coronary artery. Therefore, the mismatch between vein and artery is diminished and rheology may be improved (81. In our study’s long-term follow-up, there was no evidence that saphenous veins used in a nonreversed position occlude earlier or more frequently than those in a reversed position. In fact, the contrary appears to be the case. Therefore, the minimal trauma associated with the valvotomy is probably of no importance. A clear advan-
627
tage of the method is the use of normally occurring bifurcations in the greater saphenous vein system. A single proximal coronary anastomosis to the aorta may feed two or more branches of saphenous vein, which can be sequentially attached to the coronary arteries, allowing for multiple grafting and shortening the operating time. We believe the advantages of using saphenous vein grafts in a nonreversed position for coronary artery bypass are as follows: (1) large proximal orifice; (2) use of natural bifurcation; (3) better size-matching of vein and artery; (4) retrograde air venting; (5) less chance of turbulent flow; (6) equal or better patency rate; and (7) shorter operation. The use of nonreversed saphenous veins is therefore recommended. No disadvantages have been found in 3.5-year follow-up, and the advantages are quite substantial.
References 1. Whittemore AD, Thompson RW, Ruby ST, Mannick JA. Branched non-reverse saphenous vein graft for complex renal artery stenosis. Arch Surg 1987;122:84&7. 2. Novick AC, Pohl MA. Atherosclerotic renal artery occlusion extending into branches: successful revascularization in situ with a branched saphenous vein graft. J Urol 1979;122:240-2. 3. Streem SB, Novick AC. Aortorenal bypass with a branched saphenous vein graft for in situ repair of multiple segmented renal arteries. Surg Gynecol Obstet 1982;155:855-9. 4. Leather RP, Powers SR, Karmody AM. A reappraisal of the in situ saphenous vein arterial bypass: its use in limb salvage surgery. Surgery 1979;86:45341. 5. Skagseth E, Hall KV. In situ vein bypass: experiences with new vein valve strippers. Scand J Thorac Cardiovasc Surg 1973;753-6. 6. Leather RP, Shah DM, Buchbinder D, Annest SJ, Karmody AM. Further experience with the saphenous vein used in situ for arterial bypass. Am J Surg 1981;142:506-10. 7. Chin AK, Mayer DN, Goldman RK, Lerman JA, Olcott C, Fogarty TJ. The effect of valvotomy on the flow rate through the saphenous vein graft: clinical implications. J Vasc Surg 1988;8:31&20. 8. Molina JE. The use of the non-reversed saphenous vein for aortocoronary bypass surgery. Symposium for current problems in coronary surgery, Freiburg, FRG, Sep 1986.