- Email: [email protected]
Sequential Venous Bypass Grafts: Results 10 Years Later
Sequential Venous Bypass Grafts: Results 10 Years Later Jan T. Christenson, MD, PhD, and Martin Schmuziger, MD Cardiovascular Surgery Unit, Hoˆpital de la Tour, Meyrin-Geneva, Switzerland
Background. To evaluate the long-term outcome of the sequential vein bypass grafting technique, we studied 92 patients with coronary artery disease undergoing coronary artery bypass grafting in 1984 by one surgeon and receiving at least one sequential vein bypass graft (total of 170 sequential bypass grafts). Methods. There was one hospital death and 1 patient was lost to follow-up. The remaining 90 patients were followed up by clinical evaluation, and 80% of the patients underwent coronary angiography within 1 year from the end point of the follow-up (June 1995), or before recurrence of symptoms or death. Results. All patients except 3 had improvement of their angina class (Canadian Cardiovascular Society) at the end of the follow-up. Twelve patients did not have improvement of their New York Heart Association functional class postoperatively, but only 1 deteriorated. The mean left ventricular ejection fraction remained unchanged at the end of the follow-up period, and ergometry results were satisfactory during the follow-up pe-
riod. The 10-year survival rate was 74%, and the cardiac event-free survival rate was 72%. Only 37% of the deaths occurring during the follow-up were cardiac-related deaths. In 56 patients with angiographic routine control 9 to 10 years postoperatively, 76 of 89 sequential vein grafts were found patent. Conclusions. It is thought that the optimal long-term results of sequential bypass grafts may be dependent on where the terminal anastomosis of the sequence (the end-to-side anastomosis) is placed. The technique of sequential grafting with the reversed saphenous vein is easier to employ than the single grafting technique, and in the present study has been demonstrated to have good long-term results. Furthermore, it allows for a more complete revascularization of the myocardium, which is particularly important in patients with diffuse coronary artery disease.
S
grafts have been demonstrated to retain their patency longer than single vein grafts [6 – 8]. However, a recent study showed no differences in 10-year patency rates between sequential and single bypass grafts [9]. Nevertheless, large series tend to span lengthy time periods, where variations in anesthesia, cardiopulmonary bypass, and surgical techniques may influence the results. In the present study we have evaluated the results of 170 sequential bypass grafts 10 years postoperatively. All operations were performed by the same surgeon, during a short time period (1 year), using standardized surgical, cardiopulmonary bypass, and anesthesia techniques.
equential coronary artery bypass grafts have been used with increasing frequency for coronary artery revascularization since the beginning of the 1970s [1, 2]. Their primary advantages include saving bypass material and the number of proximal anastomoses. The technical advantages provided by the sequential grafting technique are well known, but earlier studies have suggested that this technique also offers a hemodynamic advantage [3, 4]. In these studies, intraoperative measurements demonstrated a higher blood flow in the proximal segment of the sequential graft than in single coronary artery grafts. Moreover, postoperative angiograms revealed higher patency rates in the proximal (or side-toside) anastomoses than in the distal (or end-to-side) anastomoses. The distal anastomoses, however, had patency rates comparable with those of single bypass grafts. In 1981 O’Neill and associates [5], after performing intraoperative velocity studies, confirmed the results of previous studies, and concluded that the proximal segment of the sequential bypass graft has a higher velocity of blood flow than that seen in a single bypass graft. In numerous clinical series the sequential vein bypass
Accepted for publication July 24, 1996. Address reprint requests to Dr Christenson, Cardiovascular Surgery Unit, Hoˆpital de la Tour, 1 Av, J.-D. Maillard, CH-1217 Meyrin-Geneva, Switzerland.
© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
(Ann Thorac Surg 1997;63:371– 6) © 1997 by The Society of Thoracic Surgeons
Patients and Methods During 1984, 92 patients with coronary artery disease underwent elective myocardial revascularization by one surgeon (M.S.) at Hoˆpital de la Tour, Meyrin-Geneva, Switzerland, each receiving at least one sequential coronary vein bypass graft. All preoperative patient data and operative data as well as postoperative complications were entered into a computer data base (Digital; Digital Equipment Corporation, Geneva, Switzerland). The mean patient age was 56.0 6 7.8 years (ranging from 39 to 79 years), with only 14% of them older than 65 years at the time of the operation. There were 82 men (89%) and 10 women. In 3 patients the operation was a reoperative 0003-4975/97/$17.00 PII S0003-4975(96)01059-4
372
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
coronary artery bypass grafting (CABG). The preoperative risk factors were arterial hypertension, 49%; hyperlipidemia, 58%; smoking, 45%; insulin-dependent diabetes mellitus, 4%; generalized atherosclerosis (multifocal vascular disease), 4%; preoperative renal insufficiency, 3%; and unstable angina at the time of operation, 3% of the patients. The mean preoperative New York Heart Association (NYHA) functional class was 3.2 6 0.8 (range, I to IV) and the mean preoperative Canadian Cardiovascular Society (CCS) angina class was 3.6 6 0.6 (range, 1 to 4). Preoperative coronary angiography revealed twovessel coronary artery disease in only 2 patients (2%), whereas 90 patients had three-vessel disease (98%). Left main stem stenosis greater than 70% was present in 14 patients (15%), and the mean preoperative left ventricular ejection fraction (LVEF), calculated from the ventriculogram, was 0.57 6 0.14 (range, 0.28 to 0.80). Only 12 patients (13%) had a preoperative LVEF less than 0.40.
Cardiopulmonary Bypass Technique At this time (1984) cardiopulmonary bypass was performed using a hollow-fiber membrane oxygenator with moderate hypothermic cardiopulmonary bypass (core temperature, 28°C). Cardiac arrest and myocardial preservation were achieved by cold (4° to 8°C) crystalloid cardioplegic solution (St. Thomas’ II), 1,000 mL, with 20 mmol KCl, together with topical myocardial cooling using slush. The cardioplegia was infused at low pressure into the aortic root using a cardioplegia cannula placed in the proximal aorta. Cardioplegia (500 mL, with 20 mmol KCl and a temperature of 4° to 8°C) was repeated every 30 minutes or whenever electrical activity resumed. Just before the removal of the aortic clamp, an infusion of 1,000 mL of warm blood cardioplegia (50% of venous blood) was given into the aortic root.
Operative Technique A median sternotomy was performed in all patients. The pericardium was opened, followed by general heparinization, aortic cannulation, and venous cannulation using a double-stage atrial cannula. Cardiopulmonary bypass was started. A sequential vein graft was used for revascularization of the posterolateral part of the heart and a second vein graft was used for anterolateroseptal revascularization. The proximal anastomoses were fashioned first, and the grafts were then routed through the transverse sinus. The posterolateral graft thus was routed first over the circumflex vessels, ending toward the right coronary artery. The distal coronary anastomoses were performed within a single aortic cross-clamping period. The side-to-side anastomoses were always fashioned using the diamond-shape technique and continuous suture. The routine was always trying to route the sequential bypass in such a manner that the proximal side-toside anastomosis was fashioned to the arterial system with the least flow and the terminal, end-to-side anastomosis was attached to the largest possible coronary vessel, with the highest flow, revascularized in the se-
Ann Thorac Surg 1997;63:371– 6
Table 1. Number of Distal Anastomoses per Sequential Bypass Graft No. of Anastomoses 2 3 4 5 6
RCA, CX
LAD, D
No.
Percent
No.
Percent
9 36 40 6 1
10% 39% 43% 7% 1%
51 22 5 ... ...
66% 28% 6% ... ...
CX 5 circumflex; D 5 diagonal; RCA 5 right coronary artery.
LAD 5 left anterior descending;
quence or to the right coronary artery. The policy of complete revascularization was adopted. There were a total of 170 sequential bypass grafts and 14 single grafts. The mean number of distal anastomoses/ patient was 5.7 6 0.8 (ranging from 5 to 8 anastomoses). Seventy-eight patients had two sequential bypass grafts, 14 had one sequential bypass graft, and 14 had a single bypass graft. All patients had sequential bypass to the right coronary or circumflex artery, and 78 (85%) had sequential bypass to the left anterior descending or diagonal artery. The number of distal anastomoses per sequential bypass graft is shown in Table 1. The mean aortic cross-clamp time was 69.1 6 17.6 minutes (6 standard deviation) with a range of 48 to 96 minutes, and the mean cardiopulmonary bypass time was 82.2 6 21.8 minutes, with a range of 57 to 164 minutes. Standard anesthesia techniques were employed. All patients received prophylactic antibiotics, cephalosporine at induction of anesthesia and for 48 hours postoperatively. Antacids and H2-blockers were routinely administered for prophylaxis against stress ulcerations. The patients were routinely extubated within 24 hours, and oral fluids were started. Oral anticoagulation drugs (dicumarol) were given to all patients for 2 months, followed by antiplatelet therapy, 100 mg acetylsalicylic acid daily (Aspirin, Bayer AG, Leverkusen, Germany).
Follow-up All hospital survivors were followed up clinically for 10 years (until June 15, 1995) or until death. Only 1 hospital survivor was lost to follow-up (1.1%; 1/91), leaving 90 patients available for complete follow-up. The functional results were evaluated by the NYHA and CCS classifications as well as ergometry. Left ventricular function during the follow-up was evaluated by echocardiography, measured in four-chamber view (area-length method). Fifty-six patients at random (62%) had control coronary angiography within 1 year of the end of the follow-up period. Sixteen patients (18%) underwent coronary angiography during the follow-up period due to renewed cardiac symptoms, and the remaining 18 patients (20%) had no renewed cardiac symptoms and excellent ergometry results, NYHA class, and CCS class, thus not indicating any major graft problems (symptomatic graft occlusions or stenoses).
Ann Thorac Surg 1997;63:371– 6
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
Table 2. Causes of Death in 16 Patients Who Died During the Follow-up Period After Myocardial Revascularization Using Sequential Bypass Grafts Cause of Death Cardiac After myocardial infarction After redo CABG After PTCA Cerebrovascular accidents Cancer Pulmonary carcinoma Gastric carcinoma Pneumonia and old age Sepsis after perforated cholecystitis Multitrauma after road traffic accident Renal failure, dialysis (7 years after CABG) CABG 5 coronary artery bypass grafting; transluminal coronary angioplasty.
No.
Percent
6 3 2 1 2 3 2 1 2 1 1 1
37
13 19
13 6 6 6
PTCA 5 percutaneous
Statistics Student’s t test (one-sample paired test) was employed to assess differences between groups for statistical significance where appropriate. Differences were considered significant at a probability level less than 0.05. Actuarial techniques using the life-table method by Kaplan-Meier were employed to assess time-dependent events (mortality and cardiac events). Results in the test are expressed as mean 6 standard deviation, where appropriate.
Results The hospital mortality was 1% (1 patient). The patient died in the operating room due to a possible intraoperative myocardial infarction, causing ventricular fibrillation and low cardiac output. He was treated with massive inotropic pharmacologic support and intraaortic balloon pump support unsuccessfully. There were no other nonfatal myocardial infarctions, and other postoperative complications were few.
Mortality During the follow-up 16 patients died (17.8%; 16/90). The causes of death are listed in Table 2. Interestingly the majority of the patients (63%; 10/16) died of causes unrelated to the coronary artery disease.
Cardiac Events Six patients had myocardial infarction during the followup. Two patients had a small myocardial infarction, caused by occlusion of one sequential bypass graft (2 and 4 years postoperatively), without further sequelae. One patient had a myocardial infarction after a single bypass graft occlusion 10 years postoperatively and underwent successful redo CABG. Three patients died after acute myocardial infarction (4, 10, and 10 years postoperatively), all caused by single graft occlusion to the left anterior descending artery. Five patients underwent redo CABG, 4 due to occlu-
373
sion or severe stenosis of their single bypass graft (1, 8, 9, and 10 years postoperatively) and 1 due to occlusion of the left anterior descending coronary artery after percutaneous transluminal coronary angioplasty (PTCA). All patients had satisfactory outcome of the reoperation. Two patients with occluded sequential bypass grafts underwent redo CABG 6 years after the primary CABG operation. Both died postoperatively, 1 immediately and the other 10 months after the redo CABG. Nine patients underwent PTCA (1, 5, 7, 8, 9, 9, 9, 10, and 10 years postoperatively). There were seven single bypass graft stenoses and two sequential bypass graft occlusions. Five patients had successful PTCA, whereas 1 died immediately after the PTCA and another patient had occlusion of the dilated vessel but underwent urgent redo CABG with successful outcome (both the latter PTCAs were performed 10 years after the primary CABG operation). Two patients with extremely diffuse coronary artery disease, regarded as ultima ratio at operation, had multiple PTCAs during the follow-up, without improvement, and subsequently underwent cardiac transplantation (1 year and 9 years after the CABG). Actuarial survival and cardiac event-free survival are presented in Figure 1. Fifty-seven patients (63%) had no mortality or cardiac events during the 10 years after CABG with sequential bypass grafts.
Functional Results The mean CCS angina class at the end of follow-up or before death was improved compared with preoperatively (1.0 6 0.9 [range, 0 to 4] versus 3.6 6 0.6 [range, 2 to 4]; p , 0.001), and the same was observed regarding NYHA functional class (1.4 6 0.9 versus 3.2 6 0.8; p , 0.001). Three patients had no improvement in their CCS angina class, and none had an increase in their CCS class (Fig 2). The mean ergometry result at the end of the follow-up was 125 6 31 W (range, 50 to 200 W). The mean left ventricular ejection fraction at the end of follow-up or before death was 0.59 6 0.10 (range, 0.29 to 0.82). This does not differ from the mean preoperative LVEF. Ten years after the operation 26% of the patients had approximately the same LVEF as preoperatively (LVEF 6 0.04), 26% had an ejection fraction at least 0.05 less than the preoperative LVEF, 48% of the patients had increased their preoperative LVEF by 0.05 or more. Fewer patients had a LVEF less than 0.40 postoperatively (2/90; 2.2%) compared with preoperatively (12/92; 13.0%).
Graft Patency Three sequential bypass grafts and one single bypass graft were lost to follow-up (hospital mortality, patient lost to follow-up). A total of 15 angiographically confirmed sequential bypass graft occlusions occurred during the follow-up (9.0%; 15/167), whereas five single bypass grafts occluded (39%; 5/13). There were ten single bypass graft stenoses treated either by PTCA or by redo CABG. In 5 cases, one out of two or more anastomoses of sequential bypass grafts occluded; however, they did not require renewed surgical intervention or interventional
374
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
Ann Thorac Surg 1997;63:371– 6
Fig 1. Actuarial survival (diamonds) and cardiac event-free survival (squares) in patients undergoing myocardial revascularization with at least one sequential bypass graft. The 10-year survival rate was 74%, and the cardiac event-free survival was 72% at 10 years.
cardiology (PTCA). In 4 of these cases it was the distal, end-to-side anastomosis that had failed. Retrospective analysis of preoperative angiograms revealed that the terminal anastomosis was to a very small coronary artery with poor quality. A separate analysis of a random sample of 56 patients who survived 10 years and who had routine control angiogram within 1 year of the 10-year follow-up period revealed that there were 46 sequential bypass grafts placed to the posterolateral coronary circulation and 43 to
the anterior descending and diagonal system. There was no difference in age and sex distribution, preoperative CCS and NYHA class, preoperative LVEF, or concomitant diseases between those patients with and those without routine angiography. Forty-one of the posterolateral sequential bypass grafts were found patent on angiography. Ten (24%) of them showed angiographic signs of atherosclerosis but without significant stenoses (.50%). One graft had an isolated proximal occlusion, whereas the rest of the graft including all the distal anastomoses were remaining patent. Four grafts had occlusion of one distal anastomosis only (the terminal end-to-side anastomosis). The 9- to 10-year patency rate was 89%. Thirty-five of the sequential bypass grafts to the left anterior descending artery and diagonal artery were patent on angiographic examination 9 to 10 years after the operation. Nine showed diffuse wall irregularities and two of them had additional identifiable stenotic lesions (.50%). Six grafts to the left anterior descending artery and diagonal artery were angiographically occluded, and two grafts had the distal anastomosis (endto-side anastomosis) occluded at the time of angiography. The 9- to 10-year patency rate in this group was 81%. Of 89 sequential venous bypass grafts studied angiographically at random in 56 patients 9 to 10 years postoperatively, 76 remained patent (85%).
Comment
Fig 2. Canadian Cardiovascular Society (CCS) angina class preoperatively and at 10 years postoperatively or before death in patients undergoing myocardial revascularization with at least one sequential bypass graft.
Although it is generally agreed that the internal mammary artery implant is a better bypass conduit than the saphenous vein, the vein is still often used for a variety of reasons. Sequential bypass technique has been used widely as one of the useful alternative bypass methods of CABG [3, 4]. Its advantages include saving bypass conduits and the number of proximal anastomoses and increasing the total graft flow to the minimum required to keep the graft patent. This stems from a summing of the flow to several small coronary branches, thus resulting in potentially higher graft patency. This technique is especially advan-
Ann Thorac Surg 1997;63:371– 6
tageous with extremely poor coronary run-off. Furthermore, the operating time and ischemic arrest time can be shortened and the revascularization can be more complete because anastomoses can be made on smaller coronary arteries [2, 3, 10]. The disadvantage of sequential bypass grafting, frequently voiced, has been that the entire revascularization depends on one proximal segment. It is thought that a closure of the proximal segment would result in a catastrophic decrease in myocardial blood supply. The fear of a closure occurring in the proximal segment has led in many instances to a return to the single-graft technique as advocated by Kieser and associates [10]. However, this is not true. In our experience, a proximal occlusion of a sequential bypass to the posterolateral part of the heart may result in the reappearance of angina, but without infarction in most instances. If a proximal occlusion occurs the sequential bypass graft will function as a large collateral vessel, provided that the terminal anastomosis is connected to a large coronary vessel with high flow. Sequential grafting has been controversial. A major problem has been the difference between patency rates of end-to-side and side-to-side anastomoses described in the earlier years [3, 4]. Kieser and associates [10] found a clear superiority of side-to-side over end-to-side and a questionable superiority of control single end-to-side over sequential end-to-side anastomoses. However, the study period was long, with many surgeons involved and only 50% angiographic controls at 5 years, which lowers the strength of these results [10]. Eschenbruch and colleagues [11] described 243 patients having coronary bypass by the “jump-graft” technique. Patency of 361 anastomoses at angiographic follow-up was 87.8% compared with 82.2% for 248 single control grafts. Eschenbruch and colleagues emphasized that the “sequence of the jumpgraft anastomoses should always have a major flow capacity in the most distal position” [11]. In a recent study regarding patency rate of 7,551 coronary artery bypass grafts, Coltharp and colleagues [9] have reported a 10-year graft patency of 53% for single bypass grafts and 49% patency for sequential bypass grafts. Drawbacks of previous studies on follow-up include a long time span where changes in anesthetic, extracorporeal circulation, and surgical techniques could have occurred which would skew the results. Most of the studies have been retrospective, have had the operation performed by several surgeons, and have had follow-up, eg, angiographic evaluation, at variable times postoperatively. The present series is unusual because all patients were operated on by the same surgeon during a short time period (1 year) using the same anesthesia and cardiopulmonary bypass techniques, including techniques for myocardial protection. Furthermore, the same surgical tactics and techniques were employed in all the studied patients. In Coltharp and colleagues’ series [9] control coronary angiography ranged from 7 days to 20.2 years postoperatively (mean, 5.3 years), whereas control angiography in our series was within 1 year of the 10-year
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
375
end point of the study or close to the time of the occurrence of cardiac events during the follow-up period. In 56 patients who had routine angiography 9 to 10 years postoperatively, 76 of 89 sequential vein bypass grafts were patent (85%). All our patients had either three-vessel disease or left mainstem stenosis, they all had anginal symptoms, and a large number of the patients had depressed left ventricular function. These patients could therefore be regarded as a high-risk group. The importance of blood flow velocity in autogenous vein grafts was suggested by studies by Rittgers and colleagues [12]. They found an inverse correlation between velocity of flow in chronic venous grafts and the amount of intimal proliferation observed on histologic examination. A double sequential coronary artery bypass graft has nearly half the coronary vascular resistance of a single bypass graft [5]. Because the vein graft resistance in both single and sequential grafts is quite small compared with the distal coronary resistance, coronary vascular resistance is the more important determinant of blood flow through the graft. O’Neill and associates [5] measured coronary blood flow at operation in 106 single and 35 double sequential grafts and found that flow velocity was higher in proximal segments of sequential grafts than in single grafts. They postulated that this phenomenon was responsible for the differences between side-to-side and end-to-side patency in sequential grafts, possibly by producing a greater shear stress at the vessel wall and decreasing the tendency for intimal proliferation. Based on these findings they therefore suggested that to obtain maximum hemodynamic advantage with a possibility of improving long-term patency rates, it is advisable to use the smaller coronary artery for the proximal sequential anastomosis [5]. The present series followed the principle that whenever possible the coronary system with the least flow was attached proximally on a sequential graft, with the larger vessel, with the greatest flow, used for the distal anastomosis [5, 11, 13]. Routine flow measurements in the 70s and 80s showed that good flow conditions existed in left anterior descending artery grafts (80 to 120 mL/min), whereas poorer flow, 30 to 50 mL/min, was found in anterolateral and posterolateral grafts. It also showed that the right coronary artery had high flow (70 to 100 mL/min) whenever a good flow condition toward the posterior descending branch persisted. These findings led us to adopt a two-graft sequential bypass tactic: one graft with the left anterior descending artery as the end point and the other with either the right coronary artery or the posterior descending branch as the end point. This is the most probable explanation of the good long-term results in the present series. As the yearly mortality rate increases in our series, progression of the atherosclerotic process seems to be the major problem. In a report by Meeter and associates [14], both univariate and multivariate analyses showed that patients with a sequential bypass graft are not at an additional risk of dying or of having to undergo a second
376
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
revascularization procedure compared with those patients for whom the single graft technique was chosen. Overall there is advantage in favor of total combined anastomotic patency of sequential grafts. Kieser and associates [10], even though critical of the use of sequential grafts, stated “unquestionably the side-to-side anastomosis of a double graft has much better patency rate than end-to-side anastomosis of sequential grafts.” However, they found that a single graft had a better 5-year patency rate than the end-to-side anastomosis of sequential grafts [10]. In a study by Fujiwara and associates [15] it was demonstrated that the prominent feature of the velocity profile just proximal to the side-to-side anastomosis in the sequential vein grafts was the skewing of the profile toward the anastomosis side wall, thus suggesting high velocity into the anastomosed vessel. Others have observed that the flow in a side-to-side anastomosis was less agitated than that at an end-to-side anastomosis [16]. It is thought that it is these flow configurations that may contribute to the better patency rate of the side-to-side anastomosis reported. The higher flow velocity in the proximal segment of the sequential grafts observed by Fujiwara and associates [15] is compatible with the data of O’Neill and associates [5]. O’Neill and associates also reported that the proximal segment of the double sequential graft had higher velocity of blood flow than that seen in a single bypass graft, although the diameters of the two graft types were nearly equal. The importance of higher velocity for longer patency of grafts has been suggested by earlier studies [17]. The distensibility of the vessel wall is an important factor in the hemodynamics of pulsatile blood flow. It has been reported that vein grafts are not as compliant as those of arteries, when arterial pressure is applied [17]. A compliant wall acts as an elastic reservoir and absorbs energy during systole, which is especially important in the coronary artery, where flow is predominantly diastolic. A less compliant vessel causes the pulse-wave velocity to increase in value, and more energy is lost in pulsatile form, thereby reducing the energy available for distal perfusion. Even though the sequential bypass grafting technique increases blood flow through the graft if may not prevent late degenerative changes in the vein graft, and these degenerative changes could have an impact on the longterm patency rate. Taking all this knowledge into consideration we believe that the optimal long-term results of sequential vein bypass grafts may be dependent on where the terminal anastomosis of the sequential (the end-to-side anastomosis) is placed. In the present series we have always, whenever possible, placed the end-toside anastomosis in the largest possible vessel with the highest flow, which may at least partly explain why the long-term results in the present series are better than those in many previous reports of sequential vein bypass grafts. We feel confident that the technique of sequential
Ann Thorac Surg 1997;63:371– 6
grafting with the reversed saphenous vein not only is easy to employ but also yields potentially promising long-term results. Furthermore, sequential bypass grafting allows for a more complete revascularization of the myocardium, which is particularly important in patients with diffuse coronary artery disease.
References 1. Flemma RJ, Johnson WD, Lepley D Jr. Triple aortocoronary vein bypass for coronary insufficiency. Arch Surg 1971;103: 82–3. 2. Bartley TD, Bigelow JC, Page US. Aortocoronary bypass grafting with multiple sequential anastomoses to a single vein. Arch Surg 1972;105:915–7. 3. Grondin CM, Limet R. Sequential anastomoses in coronary artery grafting: technical aspects and early and late angiographic results. Ann Thorac Surg 1977;23:1– 8. 4. Sewell WH, Sewell KV. Technique for the coronary snake graft operation. Ann Thorac Surg 1976;22:58– 65. 5. O’Neill MJ Jr, Wolf PD, O’Neill TK, Montesano RM, Waldhausen JA. A rationale for the use of sequential coronary artery bypass grafts. J Thorac Cardiovasc Surg 1981;81: 686–90. 6. Borst HG, Leitz KH, Liese W, Dragojevic D, Lichtlen P. Leistungsfa¨higkeit des sequentiellen Koronarvenenbypass. Thoraxchirurgie 1978;26:218–20. 7. Grondin CM, Lespe´rance J, Bourassa MG, Campeau L. Coronary artery grafting with the saphenous vein or internal mammary artery. Ann Thorac Surg 1975;20:605–18. 8. Meurala H, Valle M, Hekali P, Somer K, Frick MH, Harjola P-T. Patency of sequential versus single vein grafts in coronary bypass surgery. Thorac Cardiovasc Surg 1982;30: 147–51. 9. Coltharp WH, Decker MD, Stoney WS, et al. Long-term follow-up of 7551 coronary artery bypass grafts: factors influencing patency. Presented at the 74th Annual Meeting of The American Association for Thoracic Surgery, New York, NY, April 24 –27, 1994. 10. Kieser TM, FitzGibbon GM, Keon WJ. Sequential coronary bypass grafts. Long-term follow-up. J Thorac Cardiovasc Surg 1986;91:767–72. 11. Eschenbruch EM, Pabst F, Tollenaere P, Roskamm H, Schmuziger M. The significance of coronary topography for operative technique and tactics in multiple myocardial revascularization with jump-grafts. Thorac Cardiovasc Surg 1981;29:206–9. 12. Rittgers SE, Karayannacos PE, Guy JF, et al. Velocity distribution and intimal proliferation in autologous vein grafts in dogs. Circ Res 1978;42:792– 801. 13. Schmuziger M, Eschenbruch E. Operationstechnik. In: Schmuziger M, ed. Operative Therapie der koronaren Herzkrankheit. Erlangen, Germany: Perimed Fachbuch-Verlagsgesellschaft, 1982:80–7. 14. Meeter K, Veldkamp R, Tijssen JGP, van Herweden LL, Bos E. Clinical outcome of single versus sequential grafts in coronary bypass operations at ten years’ follow-up. J Thorac Cardiovasc Surg 1991;101:1076– 81. 15. Fujiwara T, Kajiya F, Kanazawa S, et al. Comparison of blood-flow waveforms in different coronary artery bypass grafts. Sequential saphenous vein grafts and internal mammary artery grafts. Circulation 1988;78:1210–7. 16. Watts KC, Marble AE, Sarwal SN, Kinley CE, Watton J, Mason MA. Simulation of coronary artery revascularization. J Biomech 1985;19:491–9. 17. Walden R, L’Italien GH, Megerman J, Abbott WM. Matched elastic properties and successful arterial grafting. Arch Surg 1980;115:1166–9.
Background. To evaluate the long-term outcome of the sequential vein bypass grafting technique, we studied 92 patients with coronary artery disease undergoing coronary artery bypass grafting in 1984 by one surgeon and receiving at least one sequential vein bypass graft (total of 170 sequential bypass grafts). Methods. There was one hospital death and 1 patient was lost to follow-up. The remaining 90 patients were followed up by clinical evaluation, and 80% of the patients underwent coronary angiography within 1 year from the end point of the follow-up (June 1995), or before recurrence of symptoms or death. Results. All patients except 3 had improvement of their angina class (Canadian Cardiovascular Society) at the end of the follow-up. Twelve patients did not have improvement of their New York Heart Association functional class postoperatively, but only 1 deteriorated. The mean left ventricular ejection fraction remained unchanged at the end of the follow-up period, and ergometry results were satisfactory during the follow-up pe-
riod. The 10-year survival rate was 74%, and the cardiac event-free survival rate was 72%. Only 37% of the deaths occurring during the follow-up were cardiac-related deaths. In 56 patients with angiographic routine control 9 to 10 years postoperatively, 76 of 89 sequential vein grafts were found patent. Conclusions. It is thought that the optimal long-term results of sequential bypass grafts may be dependent on where the terminal anastomosis of the sequence (the end-to-side anastomosis) is placed. The technique of sequential grafting with the reversed saphenous vein is easier to employ than the single grafting technique, and in the present study has been demonstrated to have good long-term results. Furthermore, it allows for a more complete revascularization of the myocardium, which is particularly important in patients with diffuse coronary artery disease.
S
grafts have been demonstrated to retain their patency longer than single vein grafts [6 – 8]. However, a recent study showed no differences in 10-year patency rates between sequential and single bypass grafts [9]. Nevertheless, large series tend to span lengthy time periods, where variations in anesthesia, cardiopulmonary bypass, and surgical techniques may influence the results. In the present study we have evaluated the results of 170 sequential bypass grafts 10 years postoperatively. All operations were performed by the same surgeon, during a short time period (1 year), using standardized surgical, cardiopulmonary bypass, and anesthesia techniques.
equential coronary artery bypass grafts have been used with increasing frequency for coronary artery revascularization since the beginning of the 1970s [1, 2]. Their primary advantages include saving bypass material and the number of proximal anastomoses. The technical advantages provided by the sequential grafting technique are well known, but earlier studies have suggested that this technique also offers a hemodynamic advantage [3, 4]. In these studies, intraoperative measurements demonstrated a higher blood flow in the proximal segment of the sequential graft than in single coronary artery grafts. Moreover, postoperative angiograms revealed higher patency rates in the proximal (or side-toside) anastomoses than in the distal (or end-to-side) anastomoses. The distal anastomoses, however, had patency rates comparable with those of single bypass grafts. In 1981 O’Neill and associates [5], after performing intraoperative velocity studies, confirmed the results of previous studies, and concluded that the proximal segment of the sequential bypass graft has a higher velocity of blood flow than that seen in a single bypass graft. In numerous clinical series the sequential vein bypass
Accepted for publication July 24, 1996. Address reprint requests to Dr Christenson, Cardiovascular Surgery Unit, Hoˆpital de la Tour, 1 Av, J.-D. Maillard, CH-1217 Meyrin-Geneva, Switzerland.
© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
(Ann Thorac Surg 1997;63:371– 6) © 1997 by The Society of Thoracic Surgeons
Patients and Methods During 1984, 92 patients with coronary artery disease underwent elective myocardial revascularization by one surgeon (M.S.) at Hoˆpital de la Tour, Meyrin-Geneva, Switzerland, each receiving at least one sequential coronary vein bypass graft. All preoperative patient data and operative data as well as postoperative complications were entered into a computer data base (Digital; Digital Equipment Corporation, Geneva, Switzerland). The mean patient age was 56.0 6 7.8 years (ranging from 39 to 79 years), with only 14% of them older than 65 years at the time of the operation. There were 82 men (89%) and 10 women. In 3 patients the operation was a reoperative 0003-4975/97/$17.00 PII S0003-4975(96)01059-4
372
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
coronary artery bypass grafting (CABG). The preoperative risk factors were arterial hypertension, 49%; hyperlipidemia, 58%; smoking, 45%; insulin-dependent diabetes mellitus, 4%; generalized atherosclerosis (multifocal vascular disease), 4%; preoperative renal insufficiency, 3%; and unstable angina at the time of operation, 3% of the patients. The mean preoperative New York Heart Association (NYHA) functional class was 3.2 6 0.8 (range, I to IV) and the mean preoperative Canadian Cardiovascular Society (CCS) angina class was 3.6 6 0.6 (range, 1 to 4). Preoperative coronary angiography revealed twovessel coronary artery disease in only 2 patients (2%), whereas 90 patients had three-vessel disease (98%). Left main stem stenosis greater than 70% was present in 14 patients (15%), and the mean preoperative left ventricular ejection fraction (LVEF), calculated from the ventriculogram, was 0.57 6 0.14 (range, 0.28 to 0.80). Only 12 patients (13%) had a preoperative LVEF less than 0.40.
Cardiopulmonary Bypass Technique At this time (1984) cardiopulmonary bypass was performed using a hollow-fiber membrane oxygenator with moderate hypothermic cardiopulmonary bypass (core temperature, 28°C). Cardiac arrest and myocardial preservation were achieved by cold (4° to 8°C) crystalloid cardioplegic solution (St. Thomas’ II), 1,000 mL, with 20 mmol KCl, together with topical myocardial cooling using slush. The cardioplegia was infused at low pressure into the aortic root using a cardioplegia cannula placed in the proximal aorta. Cardioplegia (500 mL, with 20 mmol KCl and a temperature of 4° to 8°C) was repeated every 30 minutes or whenever electrical activity resumed. Just before the removal of the aortic clamp, an infusion of 1,000 mL of warm blood cardioplegia (50% of venous blood) was given into the aortic root.
Operative Technique A median sternotomy was performed in all patients. The pericardium was opened, followed by general heparinization, aortic cannulation, and venous cannulation using a double-stage atrial cannula. Cardiopulmonary bypass was started. A sequential vein graft was used for revascularization of the posterolateral part of the heart and a second vein graft was used for anterolateroseptal revascularization. The proximal anastomoses were fashioned first, and the grafts were then routed through the transverse sinus. The posterolateral graft thus was routed first over the circumflex vessels, ending toward the right coronary artery. The distal coronary anastomoses were performed within a single aortic cross-clamping period. The side-to-side anastomoses were always fashioned using the diamond-shape technique and continuous suture. The routine was always trying to route the sequential bypass in such a manner that the proximal side-toside anastomosis was fashioned to the arterial system with the least flow and the terminal, end-to-side anastomosis was attached to the largest possible coronary vessel, with the highest flow, revascularized in the se-
Ann Thorac Surg 1997;63:371– 6
Table 1. Number of Distal Anastomoses per Sequential Bypass Graft No. of Anastomoses 2 3 4 5 6
RCA, CX
LAD, D
No.
Percent
No.
Percent
9 36 40 6 1
10% 39% 43% 7% 1%
51 22 5 ... ...
66% 28% 6% ... ...
CX 5 circumflex; D 5 diagonal; RCA 5 right coronary artery.
LAD 5 left anterior descending;
quence or to the right coronary artery. The policy of complete revascularization was adopted. There were a total of 170 sequential bypass grafts and 14 single grafts. The mean number of distal anastomoses/ patient was 5.7 6 0.8 (ranging from 5 to 8 anastomoses). Seventy-eight patients had two sequential bypass grafts, 14 had one sequential bypass graft, and 14 had a single bypass graft. All patients had sequential bypass to the right coronary or circumflex artery, and 78 (85%) had sequential bypass to the left anterior descending or diagonal artery. The number of distal anastomoses per sequential bypass graft is shown in Table 1. The mean aortic cross-clamp time was 69.1 6 17.6 minutes (6 standard deviation) with a range of 48 to 96 minutes, and the mean cardiopulmonary bypass time was 82.2 6 21.8 minutes, with a range of 57 to 164 minutes. Standard anesthesia techniques were employed. All patients received prophylactic antibiotics, cephalosporine at induction of anesthesia and for 48 hours postoperatively. Antacids and H2-blockers were routinely administered for prophylaxis against stress ulcerations. The patients were routinely extubated within 24 hours, and oral fluids were started. Oral anticoagulation drugs (dicumarol) were given to all patients for 2 months, followed by antiplatelet therapy, 100 mg acetylsalicylic acid daily (Aspirin, Bayer AG, Leverkusen, Germany).
Follow-up All hospital survivors were followed up clinically for 10 years (until June 15, 1995) or until death. Only 1 hospital survivor was lost to follow-up (1.1%; 1/91), leaving 90 patients available for complete follow-up. The functional results were evaluated by the NYHA and CCS classifications as well as ergometry. Left ventricular function during the follow-up was evaluated by echocardiography, measured in four-chamber view (area-length method). Fifty-six patients at random (62%) had control coronary angiography within 1 year of the end of the follow-up period. Sixteen patients (18%) underwent coronary angiography during the follow-up period due to renewed cardiac symptoms, and the remaining 18 patients (20%) had no renewed cardiac symptoms and excellent ergometry results, NYHA class, and CCS class, thus not indicating any major graft problems (symptomatic graft occlusions or stenoses).
Ann Thorac Surg 1997;63:371– 6
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
Table 2. Causes of Death in 16 Patients Who Died During the Follow-up Period After Myocardial Revascularization Using Sequential Bypass Grafts Cause of Death Cardiac After myocardial infarction After redo CABG After PTCA Cerebrovascular accidents Cancer Pulmonary carcinoma Gastric carcinoma Pneumonia and old age Sepsis after perforated cholecystitis Multitrauma after road traffic accident Renal failure, dialysis (7 years after CABG) CABG 5 coronary artery bypass grafting; transluminal coronary angioplasty.
No.
Percent
6 3 2 1 2 3 2 1 2 1 1 1
37
13 19
13 6 6 6
PTCA 5 percutaneous
Statistics Student’s t test (one-sample paired test) was employed to assess differences between groups for statistical significance where appropriate. Differences were considered significant at a probability level less than 0.05. Actuarial techniques using the life-table method by Kaplan-Meier were employed to assess time-dependent events (mortality and cardiac events). Results in the test are expressed as mean 6 standard deviation, where appropriate.
Results The hospital mortality was 1% (1 patient). The patient died in the operating room due to a possible intraoperative myocardial infarction, causing ventricular fibrillation and low cardiac output. He was treated with massive inotropic pharmacologic support and intraaortic balloon pump support unsuccessfully. There were no other nonfatal myocardial infarctions, and other postoperative complications were few.
Mortality During the follow-up 16 patients died (17.8%; 16/90). The causes of death are listed in Table 2. Interestingly the majority of the patients (63%; 10/16) died of causes unrelated to the coronary artery disease.
Cardiac Events Six patients had myocardial infarction during the followup. Two patients had a small myocardial infarction, caused by occlusion of one sequential bypass graft (2 and 4 years postoperatively), without further sequelae. One patient had a myocardial infarction after a single bypass graft occlusion 10 years postoperatively and underwent successful redo CABG. Three patients died after acute myocardial infarction (4, 10, and 10 years postoperatively), all caused by single graft occlusion to the left anterior descending artery. Five patients underwent redo CABG, 4 due to occlu-
373
sion or severe stenosis of their single bypass graft (1, 8, 9, and 10 years postoperatively) and 1 due to occlusion of the left anterior descending coronary artery after percutaneous transluminal coronary angioplasty (PTCA). All patients had satisfactory outcome of the reoperation. Two patients with occluded sequential bypass grafts underwent redo CABG 6 years after the primary CABG operation. Both died postoperatively, 1 immediately and the other 10 months after the redo CABG. Nine patients underwent PTCA (1, 5, 7, 8, 9, 9, 9, 10, and 10 years postoperatively). There were seven single bypass graft stenoses and two sequential bypass graft occlusions. Five patients had successful PTCA, whereas 1 died immediately after the PTCA and another patient had occlusion of the dilated vessel but underwent urgent redo CABG with successful outcome (both the latter PTCAs were performed 10 years after the primary CABG operation). Two patients with extremely diffuse coronary artery disease, regarded as ultima ratio at operation, had multiple PTCAs during the follow-up, without improvement, and subsequently underwent cardiac transplantation (1 year and 9 years after the CABG). Actuarial survival and cardiac event-free survival are presented in Figure 1. Fifty-seven patients (63%) had no mortality or cardiac events during the 10 years after CABG with sequential bypass grafts.
Functional Results The mean CCS angina class at the end of follow-up or before death was improved compared with preoperatively (1.0 6 0.9 [range, 0 to 4] versus 3.6 6 0.6 [range, 2 to 4]; p , 0.001), and the same was observed regarding NYHA functional class (1.4 6 0.9 versus 3.2 6 0.8; p , 0.001). Three patients had no improvement in their CCS angina class, and none had an increase in their CCS class (Fig 2). The mean ergometry result at the end of the follow-up was 125 6 31 W (range, 50 to 200 W). The mean left ventricular ejection fraction at the end of follow-up or before death was 0.59 6 0.10 (range, 0.29 to 0.82). This does not differ from the mean preoperative LVEF. Ten years after the operation 26% of the patients had approximately the same LVEF as preoperatively (LVEF 6 0.04), 26% had an ejection fraction at least 0.05 less than the preoperative LVEF, 48% of the patients had increased their preoperative LVEF by 0.05 or more. Fewer patients had a LVEF less than 0.40 postoperatively (2/90; 2.2%) compared with preoperatively (12/92; 13.0%).
Graft Patency Three sequential bypass grafts and one single bypass graft were lost to follow-up (hospital mortality, patient lost to follow-up). A total of 15 angiographically confirmed sequential bypass graft occlusions occurred during the follow-up (9.0%; 15/167), whereas five single bypass grafts occluded (39%; 5/13). There were ten single bypass graft stenoses treated either by PTCA or by redo CABG. In 5 cases, one out of two or more anastomoses of sequential bypass grafts occluded; however, they did not require renewed surgical intervention or interventional
374
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
Ann Thorac Surg 1997;63:371– 6
Fig 1. Actuarial survival (diamonds) and cardiac event-free survival (squares) in patients undergoing myocardial revascularization with at least one sequential bypass graft. The 10-year survival rate was 74%, and the cardiac event-free survival was 72% at 10 years.
cardiology (PTCA). In 4 of these cases it was the distal, end-to-side anastomosis that had failed. Retrospective analysis of preoperative angiograms revealed that the terminal anastomosis was to a very small coronary artery with poor quality. A separate analysis of a random sample of 56 patients who survived 10 years and who had routine control angiogram within 1 year of the 10-year follow-up period revealed that there were 46 sequential bypass grafts placed to the posterolateral coronary circulation and 43 to
the anterior descending and diagonal system. There was no difference in age and sex distribution, preoperative CCS and NYHA class, preoperative LVEF, or concomitant diseases between those patients with and those without routine angiography. Forty-one of the posterolateral sequential bypass grafts were found patent on angiography. Ten (24%) of them showed angiographic signs of atherosclerosis but without significant stenoses (.50%). One graft had an isolated proximal occlusion, whereas the rest of the graft including all the distal anastomoses were remaining patent. Four grafts had occlusion of one distal anastomosis only (the terminal end-to-side anastomosis). The 9- to 10-year patency rate was 89%. Thirty-five of the sequential bypass grafts to the left anterior descending artery and diagonal artery were patent on angiographic examination 9 to 10 years after the operation. Nine showed diffuse wall irregularities and two of them had additional identifiable stenotic lesions (.50%). Six grafts to the left anterior descending artery and diagonal artery were angiographically occluded, and two grafts had the distal anastomosis (endto-side anastomosis) occluded at the time of angiography. The 9- to 10-year patency rate in this group was 81%. Of 89 sequential venous bypass grafts studied angiographically at random in 56 patients 9 to 10 years postoperatively, 76 remained patent (85%).
Comment
Fig 2. Canadian Cardiovascular Society (CCS) angina class preoperatively and at 10 years postoperatively or before death in patients undergoing myocardial revascularization with at least one sequential bypass graft.
Although it is generally agreed that the internal mammary artery implant is a better bypass conduit than the saphenous vein, the vein is still often used for a variety of reasons. Sequential bypass technique has been used widely as one of the useful alternative bypass methods of CABG [3, 4]. Its advantages include saving bypass conduits and the number of proximal anastomoses and increasing the total graft flow to the minimum required to keep the graft patent. This stems from a summing of the flow to several small coronary branches, thus resulting in potentially higher graft patency. This technique is especially advan-
Ann Thorac Surg 1997;63:371– 6
tageous with extremely poor coronary run-off. Furthermore, the operating time and ischemic arrest time can be shortened and the revascularization can be more complete because anastomoses can be made on smaller coronary arteries [2, 3, 10]. The disadvantage of sequential bypass grafting, frequently voiced, has been that the entire revascularization depends on one proximal segment. It is thought that a closure of the proximal segment would result in a catastrophic decrease in myocardial blood supply. The fear of a closure occurring in the proximal segment has led in many instances to a return to the single-graft technique as advocated by Kieser and associates [10]. However, this is not true. In our experience, a proximal occlusion of a sequential bypass to the posterolateral part of the heart may result in the reappearance of angina, but without infarction in most instances. If a proximal occlusion occurs the sequential bypass graft will function as a large collateral vessel, provided that the terminal anastomosis is connected to a large coronary vessel with high flow. Sequential grafting has been controversial. A major problem has been the difference between patency rates of end-to-side and side-to-side anastomoses described in the earlier years [3, 4]. Kieser and associates [10] found a clear superiority of side-to-side over end-to-side and a questionable superiority of control single end-to-side over sequential end-to-side anastomoses. However, the study period was long, with many surgeons involved and only 50% angiographic controls at 5 years, which lowers the strength of these results [10]. Eschenbruch and colleagues [11] described 243 patients having coronary bypass by the “jump-graft” technique. Patency of 361 anastomoses at angiographic follow-up was 87.8% compared with 82.2% for 248 single control grafts. Eschenbruch and colleagues emphasized that the “sequence of the jumpgraft anastomoses should always have a major flow capacity in the most distal position” [11]. In a recent study regarding patency rate of 7,551 coronary artery bypass grafts, Coltharp and colleagues [9] have reported a 10-year graft patency of 53% for single bypass grafts and 49% patency for sequential bypass grafts. Drawbacks of previous studies on follow-up include a long time span where changes in anesthetic, extracorporeal circulation, and surgical techniques could have occurred which would skew the results. Most of the studies have been retrospective, have had the operation performed by several surgeons, and have had follow-up, eg, angiographic evaluation, at variable times postoperatively. The present series is unusual because all patients were operated on by the same surgeon during a short time period (1 year) using the same anesthesia and cardiopulmonary bypass techniques, including techniques for myocardial protection. Furthermore, the same surgical tactics and techniques were employed in all the studied patients. In Coltharp and colleagues’ series [9] control coronary angiography ranged from 7 days to 20.2 years postoperatively (mean, 5.3 years), whereas control angiography in our series was within 1 year of the 10-year
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
375
end point of the study or close to the time of the occurrence of cardiac events during the follow-up period. In 56 patients who had routine angiography 9 to 10 years postoperatively, 76 of 89 sequential vein bypass grafts were patent (85%). All our patients had either three-vessel disease or left mainstem stenosis, they all had anginal symptoms, and a large number of the patients had depressed left ventricular function. These patients could therefore be regarded as a high-risk group. The importance of blood flow velocity in autogenous vein grafts was suggested by studies by Rittgers and colleagues [12]. They found an inverse correlation between velocity of flow in chronic venous grafts and the amount of intimal proliferation observed on histologic examination. A double sequential coronary artery bypass graft has nearly half the coronary vascular resistance of a single bypass graft [5]. Because the vein graft resistance in both single and sequential grafts is quite small compared with the distal coronary resistance, coronary vascular resistance is the more important determinant of blood flow through the graft. O’Neill and associates [5] measured coronary blood flow at operation in 106 single and 35 double sequential grafts and found that flow velocity was higher in proximal segments of sequential grafts than in single grafts. They postulated that this phenomenon was responsible for the differences between side-to-side and end-to-side patency in sequential grafts, possibly by producing a greater shear stress at the vessel wall and decreasing the tendency for intimal proliferation. Based on these findings they therefore suggested that to obtain maximum hemodynamic advantage with a possibility of improving long-term patency rates, it is advisable to use the smaller coronary artery for the proximal sequential anastomosis [5]. The present series followed the principle that whenever possible the coronary system with the least flow was attached proximally on a sequential graft, with the larger vessel, with the greatest flow, used for the distal anastomosis [5, 11, 13]. Routine flow measurements in the 70s and 80s showed that good flow conditions existed in left anterior descending artery grafts (80 to 120 mL/min), whereas poorer flow, 30 to 50 mL/min, was found in anterolateral and posterolateral grafts. It also showed that the right coronary artery had high flow (70 to 100 mL/min) whenever a good flow condition toward the posterior descending branch persisted. These findings led us to adopt a two-graft sequential bypass tactic: one graft with the left anterior descending artery as the end point and the other with either the right coronary artery or the posterior descending branch as the end point. This is the most probable explanation of the good long-term results in the present series. As the yearly mortality rate increases in our series, progression of the atherosclerotic process seems to be the major problem. In a report by Meeter and associates [14], both univariate and multivariate analyses showed that patients with a sequential bypass graft are not at an additional risk of dying or of having to undergo a second
376
CHRISTENSON AND SCHMUZIGER SEQUENTIAL VEIN CABG
revascularization procedure compared with those patients for whom the single graft technique was chosen. Overall there is advantage in favor of total combined anastomotic patency of sequential grafts. Kieser and associates [10], even though critical of the use of sequential grafts, stated “unquestionably the side-to-side anastomosis of a double graft has much better patency rate than end-to-side anastomosis of sequential grafts.” However, they found that a single graft had a better 5-year patency rate than the end-to-side anastomosis of sequential grafts [10]. In a study by Fujiwara and associates [15] it was demonstrated that the prominent feature of the velocity profile just proximal to the side-to-side anastomosis in the sequential vein grafts was the skewing of the profile toward the anastomosis side wall, thus suggesting high velocity into the anastomosed vessel. Others have observed that the flow in a side-to-side anastomosis was less agitated than that at an end-to-side anastomosis [16]. It is thought that it is these flow configurations that may contribute to the better patency rate of the side-to-side anastomosis reported. The higher flow velocity in the proximal segment of the sequential grafts observed by Fujiwara and associates [15] is compatible with the data of O’Neill and associates [5]. O’Neill and associates also reported that the proximal segment of the double sequential graft had higher velocity of blood flow than that seen in a single bypass graft, although the diameters of the two graft types were nearly equal. The importance of higher velocity for longer patency of grafts has been suggested by earlier studies [17]. The distensibility of the vessel wall is an important factor in the hemodynamics of pulsatile blood flow. It has been reported that vein grafts are not as compliant as those of arteries, when arterial pressure is applied [17]. A compliant wall acts as an elastic reservoir and absorbs energy during systole, which is especially important in the coronary artery, where flow is predominantly diastolic. A less compliant vessel causes the pulse-wave velocity to increase in value, and more energy is lost in pulsatile form, thereby reducing the energy available for distal perfusion. Even though the sequential bypass grafting technique increases blood flow through the graft if may not prevent late degenerative changes in the vein graft, and these degenerative changes could have an impact on the longterm patency rate. Taking all this knowledge into consideration we believe that the optimal long-term results of sequential vein bypass grafts may be dependent on where the terminal anastomosis of the sequential (the end-to-side anastomosis) is placed. In the present series we have always, whenever possible, placed the end-toside anastomosis in the largest possible vessel with the highest flow, which may at least partly explain why the long-term results in the present series are better than those in many previous reports of sequential vein bypass grafts. We feel confident that the technique of sequential
Ann Thorac Surg 1997;63:371– 6
grafting with the reversed saphenous vein not only is easy to employ but also yields potentially promising long-term results. Furthermore, sequential bypass grafting allows for a more complete revascularization of the myocardium, which is particularly important in patients with diffuse coronary artery disease.
References 1. Flemma RJ, Johnson WD, Lepley D Jr. Triple aortocoronary vein bypass for coronary insufficiency. Arch Surg 1971;103: 82–3. 2. Bartley TD, Bigelow JC, Page US. Aortocoronary bypass grafting with multiple sequential anastomoses to a single vein. Arch Surg 1972;105:915–7. 3. Grondin CM, Limet R. Sequential anastomoses in coronary artery grafting: technical aspects and early and late angiographic results. Ann Thorac Surg 1977;23:1– 8. 4. Sewell WH, Sewell KV. Technique for the coronary snake graft operation. Ann Thorac Surg 1976;22:58– 65. 5. O’Neill MJ Jr, Wolf PD, O’Neill TK, Montesano RM, Waldhausen JA. A rationale for the use of sequential coronary artery bypass grafts. J Thorac Cardiovasc Surg 1981;81: 686–90. 6. Borst HG, Leitz KH, Liese W, Dragojevic D, Lichtlen P. Leistungsfa¨higkeit des sequentiellen Koronarvenenbypass. Thoraxchirurgie 1978;26:218–20. 7. Grondin CM, Lespe´rance J, Bourassa MG, Campeau L. Coronary artery grafting with the saphenous vein or internal mammary artery. Ann Thorac Surg 1975;20:605–18. 8. Meurala H, Valle M, Hekali P, Somer K, Frick MH, Harjola P-T. Patency of sequential versus single vein grafts in coronary bypass surgery. Thorac Cardiovasc Surg 1982;30: 147–51. 9. Coltharp WH, Decker MD, Stoney WS, et al. Long-term follow-up of 7551 coronary artery bypass grafts: factors influencing patency. Presented at the 74th Annual Meeting of The American Association for Thoracic Surgery, New York, NY, April 24 –27, 1994. 10. Kieser TM, FitzGibbon GM, Keon WJ. Sequential coronary bypass grafts. Long-term follow-up. J Thorac Cardiovasc Surg 1986;91:767–72. 11. Eschenbruch EM, Pabst F, Tollenaere P, Roskamm H, Schmuziger M. The significance of coronary topography for operative technique and tactics in multiple myocardial revascularization with jump-grafts. Thorac Cardiovasc Surg 1981;29:206–9. 12. Rittgers SE, Karayannacos PE, Guy JF, et al. Velocity distribution and intimal proliferation in autologous vein grafts in dogs. Circ Res 1978;42:792– 801. 13. Schmuziger M, Eschenbruch E. Operationstechnik. In: Schmuziger M, ed. Operative Therapie der koronaren Herzkrankheit. Erlangen, Germany: Perimed Fachbuch-Verlagsgesellschaft, 1982:80–7. 14. Meeter K, Veldkamp R, Tijssen JGP, van Herweden LL, Bos E. Clinical outcome of single versus sequential grafts in coronary bypass operations at ten years’ follow-up. J Thorac Cardiovasc Surg 1991;101:1076– 81. 15. Fujiwara T, Kajiya F, Kanazawa S, et al. Comparison of blood-flow waveforms in different coronary artery bypass grafts. Sequential saphenous vein grafts and internal mammary artery grafts. Circulation 1988;78:1210–7. 16. Watts KC, Marble AE, Sarwal SN, Kinley CE, Watton J, Mason MA. Simulation of coronary artery revascularization. J Biomech 1985;19:491–9. 17. Walden R, L’Italien GH, Megerman J, Abbott WM. Matched elastic properties and successful arterial grafting. Arch Surg 1980;115:1166–9.