| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1999;68:2273-2278
© 1999 The Society of Thoracic Surgeons
a Department of Cardiothoracic Surgery, Boston University School of Medicine, Boston, Massachusetts, USA
Address reprint requests to Dr Shapira, Department of Cardiothoracic Surgery, Boston Medical Center, 88 East Newton St, Boston, MA 02118
e-mail: oshapira{at}bu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentReferences |
|---|
Methods. One hundred fifty-one consecutive patients undergoing first-time CABG with CE between 1991 and 1997 were compared with a concurrent group of 757 patients undergoing CABG without CE (Control).
Results. Age, gender, left ventricular ejection fraction, percent nonelective were similar in both groups. Compared with control, the CE group had a higher incidence of hypertension (80% versus 71%, p = 0.02), diabetes (42% versus 32%, p = 0.01), prior myocardial infarction (MI) (68% versus 59%, p = 0.05), peripheral vascular disease (36% versus 16%, p < 0.001), renal failure (15% versus 4%, p < 0.001), and three-vessel coronary disease (81% versus 70%, p = 0.007), resulting in higher Society of Thoracic Surgeons database predicted mortality (4.9 ± 5.9% versus 3.9 ± 4.6%, p = 0.05). Despite the higher risk profile of the CE group, hospital mortality (CE 2.0%, Control 1.2%) and the incidence of major complications such as cerebrovascular accident (CVA) (0.7% versus 1.5%), major respiratory complications (8% versus 5%), and postoperative MI (3% versus 1.4%) were similar between the groups (all p = NS). In a multiple logistic regression analysis, prolonged cardiopulmonary bypass time was an independent predictor of postoperative MI (odds ratio 1.2, CI 1.05 to 1.39, p < 0.01) and the use of heparin-bonded cardiopulmonary bypass circuits of reduced MI rate (odds ratio 0.25, CI 0.08 to 0.76, p < 0.01). Mean follow-up for 94% of patients was 30 ± 19 months (range 1 to 83 months). Five-year survival after CE was 70 ± 5%, with 96% of patients in Canadian Cardiovascular Society class I/II.
Conclusions. In a contemporary series of carefully selected patients, mortality and major complications after CE are now similar to CABG without CE. CE itself is not an independent predictor of postoperative MI. Functional class of hospital survivors is excellent.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
However, many surgeons are still reluctant to use this technique because of previously reported increased postoperative mortality in the range of 3.2% to 10%, and myocardial infarction (MI) rates of 4% to 15%, as well as uncertainty regarding its long-term clinical efficacy [3–15].
The purpose of this study was to determine the impact of recent advances in practice on early and long-term clinical outcomes of patients undergoing CABG with CE, and to identify independent risk factors of adverse outcomes by comparison with a concurrent cohort of patients undergoing CABG without CE.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Surgical technique
All operations were performed using near-normothermic (34°C to 35°C) cardiopulmonary bypass. In the early phase of the study, conventional cardiopulmonary bypass circuits and standard systemic anticoagulation (heparin dose of 3 mg/kg, target activated clotting time 480 seconds) were used. In 1994, we switched to "tip-to-tip" heparin-bonded cardiopulmonary bypass circuits (Duraflo II; Baxter, Irvine, CA) and low systemic anticoagulation (heparin dose of 1 mg/kg, target activated clotting time 280 seconds), as previously described [16]. Distal anastomoses were constructed during a single aortic cross-clamp period.
Myocardial protection was achieved using cold antegrade and retrograde blood cardioplegia supplemented by topical cooling with cold (4°C) saline. A terminal warm bolus of cardioplegia was administered before removal of the aortic cross-clamp.
Coronary endarterectomy was selected for vessels with a preendarterectomy outer diameter of 1.5 mm or greater (expecting a postendarterectomy luminal diameter 
1.5 mm), supplying viable muscle with evidence of reversible ischemia, and in which diffuse atherosclerosis prevented satisfactory distal anastomosis. The traction technique was used in most cases [12, 16]. A limited arteriotomy was performed, and a plan between the atheromatous plaque and the adventitia was developed using a fine spatula. Using continuous traction, the plaque was dissected proximally and distally and removed. In the right coronary artery (RCA), the plaque was endarterectomized proximally as far as possible, especially if there was total occlusion. In cases of moderate (50% to 75%) occlusion, care was taken to sharply divide the proximal plaque. In the left anterior descending (LAD) territory, the proximal extent of the endarterectomy was limited, and usually the plaque was divided 1 to 2 cm proximal to the anastomosis site [16]. The specimen was inspected carefully, ensuring a smooth tapering distal end. If the distal end was irregular, or a residual plaque was felt to remain in the distal vessel, the arteriotomy was enlarged as necessary to allow complete removal of the plaque, converting the endarterectomy to an "open" technique [12, 16]. The endarterectomized vessel was irrigated was heparanized Plasmalite (Travenol; Baxter) to remove residual small particles. In most cases, the bypass conduit (arterial or venous) was directly anastomosed to the endarterectomy site after being appropriately enlarged. In selected cases, with a very long open segment, a vein patch was used to close the endarterectomy site, and the distal anastomosis was constructed between the conduit and the patch. Antiplatelet agents (mostly aspirin) were started 6 hours postoperatively in all patients. Other anticoagulants were not routinely used.
Data collection
Hospital data
All clinical data were prospectively collected and entered into a database. Data collected included: (1) demographic information; (2) comorbid risk factors; (3) operative data; (4) postoperative complications: mortality and significant morbidity defined as reoperation for bleeding, mediastinal infection, pneumonia, respirator more than 24 hours, transient ischemic attack or cerebrovascular event, myocardial infarction (new Q wave, elevation of Creatinine Phosphokinase-MB fraction (CPKMB) 50 U, new wall motion abnormality in echocardiogram), low cardiac output state (a newly placed intraaortic balloon pump or the use of inotropes for more than 24 hours to maintain a cardiac index greater than 2.0 L/min/m2), and other major complications; (5) bleeding and transfusion requirements; and (6) total hospital and intensive care unit length of stays.
Long-term follow-up
All clinical records were reviewed. Direct telephone contact with the patient or family, and the primary care physician or the cardiologist, was made in 99% of patients. Data obtained included survival, functional status, and long-term medical management, readmissions and reinterventions, and any surgery-related complications.
Statistical analysis
Data are expressed as means ± SD. Median and range values were added for selected variables. Statistical analysis was performed using SPSS software (release 6.1.2 for Windows). Two-tailed Student t test was used to analyze continuous variables. Categorical data were analyzed using 
2 with Yate’s correction or Fisher’s exact test when appropriate. A p value of less than 0.05 was considered significant. Multivariate analyses were performed to identify independent predictors of adverse outcomes and total hospital length of stay. A stepwise backward selection method was utilized in order to select the variables in the models, with a significance level of 0.1 for entry into the model, and a significance level of 0.05 for staying in the model. The Kaplan-Meier method was used to analyze actuarial survival and postoperative events.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
|
|
Hospital outcomes
Hospital outcomes are summarized in Table 3. The operative mortality was similar between the groups (CE 2% versus Control 1.2%, p = 0.71). Causes of hospital mortality in the CE group included low cardiac output, multiorgan failure secondary to ischemic bowel, and pulmonary embolism (one each). Also, the incidence of major complications such as postoperative MI (CE 3% versus Control 1.4%, p = 0.11), CVA (CE 0.7% versus Control 1.5%, p = 0.67), respiratory complications (CE 8% versus Control 5%, p = 0.19), and reexplorations for bleeding (CE 1.3% versus Control 1.4%, p = 0.76) were similar between the groups. Of the 5 patients with postoperative MI in the CE group, 3 were localized to the territory supplied by the endarterectomized vessel and 2 were localized to a different territory. The difference between MI rate in occluded (3.1%) and nonoccluded vessels (2.3%) was not statistically significant. There was no difference in mortality and postoperative complication rates comparing the "traction" and the "open" techniques (data not shown), although the "open" technique was used in relatively few cases (22 of 196 vessels).
|
Reflecting the higher preoperative risk profile, overall morbidity (48% versus 23%, p < 0.001) and the incidence and magnitude of allogeneic blood transfusions were higher in the CE group (Table 3). Also, use of inotropic drugs was more frequent in the CE group (16% versus 8%, p = 0.002), but the need for postoperative IABP support was low (less than 1%) and similar in both groups. The incidence of new-onset postoperative atrial fibrillation was similar between the groups. Independent predictors of overall postoperative morbidity in the multivariate logistic regression analysis included advanced age (OR 1.05, CI 0 to 1.07, p < 0.001), prior CVA (OR 2.04, CI 0.008 to 3.34, p = 0.004), prior MI (OR 1.63, CI 0.02 to 2.28, p = 0.004), preoperative atrial fibrillation (OR 2.87, CI 0.76 to 5.49, p = 0.001), and CE (OR 1.34, CI 0.009 to 2.90, p < 0.001). Use of heparin-bonded circuits was associated with reduced morbidity (OR 0.59, CI 0.42 to 0.89, p = 0.002).
Higher morbidity and blood transfusion requirement resulted in longer intensive care unit and total hospital length of stays in the CE group (2.7 ± 4 versus 1 ± 1 days, p < 0.001; and 7.7 ± 6 versus 6.5 ± 4 days, p = 0.003, respectively). Independent predictors of increased length of hospital stay in a multivariate linear regression analysis included nonelective surgery (Beta 0.0388, p = 0.002), history of hypertension (Beta 0.0524, p = 0.05), prolonged cardiopulmonary bypass time (Beta 0.0949, p = 0.03), postoperative respiratory complications (Beta 0.1838, p = 0.046), postoperative atrial fibrillation (Beta 0.2639, p < 0.001), and postoperative CVA (Beta 0.0371, p = 0.0002). CE by itself was not a predictor of prolonged length of stay (Beta 0.0520, p = 0.17).
Long-term results
Average follow-up for 94% (141 of 151 patients) of patients was 30 ± 19 months, (range 1 to 83 months). Twenty-eight patients expired during the follow-up period. Causes of late deaths included cardiac related (MI, congestive heart failure, arrhythmia), 8 patients; cancer, 8; CVA, 3; multiorgan failure, 3; pulmonary embolism, and 2; pneumonia and sepsis, 1 each. In 2 patients, the cause of death is unknown. Overall, 24 patients had cardiac-related events during the follow-up period, of which 7 had MI. Actuarial survival (including hospital mortality) at 1 and 5 years was 91 ± 2% and 70 ± 5%, respectively (Fig 1). Event-free survival at 1 and 5 years was 85 ± 3% and 61 ± 6%, respectively. Freedom from angina at 1 and 5 years was 88% ± 2% and 67 ± 5%, respectively. At the time of follow-up, 96% (115 of 120) of patients were in Canadian Cardiovascular Society angina class I/II, a significant improvement compared with preoperative status, where 76% (115 of 151) were class III/IV (p < 0.001).
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Selection criteria for endarterectomy were strict in this study. The technique was employed only for vessels with a preendarterectomy outer diameter of 1.5 mm or greater (expecting a postendarterectomy luminal diameter 1.5 mm), supplying viable myocardium with evidence of ischemia in stress testing and inability to graft the vessel using conventional techniques because of diffuse atherosclerosis. We believe that smaller vessels should not be endarterectomized. It is technically difficult to assure complete removal of the plaque, and the runoff in small-caliber nonendothelialized vessels is low, markedly increasing the risk of early graft occlusion. The improved outcomes observed in this study are, at least in part, related to strict adherence to these selection criteria.
The need for more grafts per patient as well as the time required to perform the endarterectomy itself resulted in longer cardiopulmonary bypass and myocardial ischemic times in the CE group compared with control. Similar observations were made by others [4–15], with reported duration of cardiopulmonary bypass and myocardial ischemic times as long as 6 and 3 hours, respectively, for extensive three-vessel direct vision endarterectomies [7]. Hence, optimal cardiopulmonary bypass technology and myocardial preservation strategies are of paramount importance. In the later part of the study, we have switched to "tip-to-tip" heparin-bonded cardiopulmonary bypass circuits with reduced systemic anticoagulation. We have previously demonstrated that the use of these biocompatible circuits was associated with dramatic improvement in clinical outcomes, including reduction of postoperative myocardial infarction [17]. These findings were confirmed in the present study, and in the multivariate analysis, the use of heparin-bonded cardiopulmonary bypass circuits was the only independent predictor of reduced postoperative MI rate. The value of the myocardial protection strategy of combined antegrade and retrograde cold blood cardioplegia with terminal warm bolus in this high risk patients is well documented [3, 11, 12, 18].
In this study, endarterectomy was performed in coronary arteries that were either totally occluded or had significant stenosis (> 50%, but not totally occluded) with similar postoperative MI rate. This is in contrast to Djalilian and colleagues, who reported increased MI rate in endarterectomized vessels that were not completely occluded [13].
The endarterectomy technique of choice is still a matter of controversy [7, 8, 12–16]. The traction technique is simpler, can be performed via a limited arteriotomy, and, therefore, is easier to reconstruct [12, 16]. The main risks associated with the traction technique are incomplete removal of the plaque in the distal vessel, and shearing-off the plaque in the side branches (the "snow-plow" effect). This may be particularly detrimental in the LAD territory, with impairment of flow in the septal perforators. The "open" technique avoids these problems and allows complete removal of the plaque from the main vessel and the side branches under direct vision [12, 16]. However, the procedure is more time consuming, and reconstruction often requires a patch to close the vessel before a distal anastomosis can be performed, resulting in a prolonged myocardial ischemic time. In this study, both techniques have been used. The traction technique was preferred in most cases. However, the endarterectomy specimen was inspected carefully, ensuring a smooth tapering distal end. If the distal end was irregular, or a residual plaque was felt to remain in the distal vessel, the arteriotomy was enlarged as necessary to allow complete removal of the plaque, converting the endarterectomy to an "open" technique. Conversion to the direct-vision technique was rarely necessary in the RCA endarterectomy, because there are only few side branches, and the angle of traction is optimal. In contrast, the angle of traction is often suboptimal and side branches are numerous in left-sided coronary arteries, and therefore the direct-vision technique was liberally used in these vessels, particularly the LAD. We did not observe differences in clinical outcomes between the two techniques, although the number of "open" endarterectomies was too small to allow definite conclusions.
In the present study, the internal mammary artery was used as a conduit to the endarterectomized vessel in 38%. In early reports, the internal mammary artery was sparingly used as a conduit to an endarterectomized vessel because of concerns regarding size mismatch and the potential for hypoperfusion [7]. More recently several authors reported satisfactory short- and long-term clinical outcomes, as well as good or even superior patency of the internal mammary graft to an endarterectomized vessel compared with saphenous vein [13–15]. This is in accordance with the well-documented superior biologic characteristics of the internal mammary artery, translating into superior long-term overall and event-free survival [19].
Several authors have recommended routine intravenous infusion of heparin, dipyridamole, or dextran in the immediate postoperative period, followed by warfarin for several months to minimize the risk of acute graft or native-vessel thrombosis, particularly for left-sided endarterectomies [8, 10, 13, 16]. In this study and others, only platelet-inhibitors (mostly aspirin) were used without untoward effects. It is emphasized that treatment with antiplatelet agents should be started within a few hours after surgery in order to be effective [12, 14].
In this study, the favorable hospital outcomes extended into the follow-up period, and mid-term clinical results were very encouraging. One- and 5-year survival rates were 91% and 70%, respectively. These figures are comparable with previous reports [4–15], and are particularly impressive considering the high-risk profile of the patient population. Of note is that only 8 of the 28 long-term deaths in this study were cardiac-related. Event-free survival was considerably less (only 61% at 5 years), reflecting the diffuse, advanced-stage coronary artery disease of these patients at the time of presentation. However, at the time of follow-up, 96% of patients were either angina free, or had very few symptoms, a substantial improvement compared with preoperative status.
One of the limitations of our study is the lack of follow-up angiographic data. Goldstein and colleagues reported satisfactory early patency rates. However, the vessels tended to show accelerated deterioration with time, resulting in low late patency rate [9]. In contrast, several authors documented that short- and long-term patency rates of bypass conduits (arterial or venous) grafted to an endarterectomized vessel were comparable conventional bypass grafting [3–5, 7, 13, 14].
In summary, patients undergoing CABG with CE had higher risk profile with more extensive coronary artery disease and more frequent comorbid risk factors. Despite this higher risk profile, hospital mortality and major complications were similar to the control group. Higher overall morbidity and transfusion requirements resulted in longer intensive care unit and hospital length of stay in the CE group. The effect of CE is durable, and long-term clinical outcomes are encouraging. We conclude that in a contemporary series of carefully selected patients, mortality and major complications after coronary endarterectomy are now similar to CABG without CE, and CE itself is not an independent predictor of major adverse outcomes.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Related Article
Ann. Thorac. Surg. 1999 68: 2278.
This article has been cited by other articles:
|
|
 |
|
  B. Yilmazkaya, R. Circi, U. P. Circi, S. Gurkahraman, M. A. Yukselen, O. Z. Yondem, and O. Tasdemir Surgical Approaches in Left Anterior Descending Artery In-Stent Stenosis Ann. Thorac. Surg., May 1, 2008; 85(5): 1586 - 1590. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Schwann, A. Zacharias, C. J. Riordan, S. J. Durham, A. S. Shah, and R. H. Habib Survival and Graft Patency After Coronary Artery Bypass Grafting With Coronary Endarterectomy: Role of Arterial Versus Vein Conduits Ann. Thorac. Surg., July 1, 2007; 84(1): 25 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Nurozler, T. Kutlu, G. Kucuk, and C. Okten Off-Pump Coronary Endarterectomy in High-Risk Patients Asian Cardiovasc Thorac Ann, June 1, 2006; 14(3): 227 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Fukui, S. Takanashi, and Y. Hosoda Long Segmental Reconstruction of Diffusely Diseased Left Anterior Descending Coronary Artery With Left Internal Thoracic Artery With or Without Endarterectomy Ann. Thorac. Surg., December 1, 2005; 80(6): 2098 - 2105. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Nishi, S. Miyamoto, S. Takanashi, H. Minamimura, T. Ishikawa, Y. Kato, and Y. Shimizu Optimal Method of Coronary Endarterectomy for Diffusely Diseased Coronary Arteries Ann. Thorac. Surg., March 1, 2005; 79(3): 846 - 852. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Fukui, S. Takanashi, and Y. Hosoda Coronary Endarterectomy and Stent Removal in Patients With In-Stent Restenosis Ann. Thorac. Surg., February 1, 2005; 79(2): 558 - 563. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. G. Byrne, A. N. Karavas, T. Gudbjartson, M. Leacche, J. D. Rawn, G. S. Couper, R. J. Rizzo, L. H. Cohn, and S. F. Aranki Left anterior descending coronary endarterectomy: Early and late results in 196 consecutive patients Ann. Thorac. Surg., September 1, 2004; 78(3): 867 - 873. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Takanashi, T. Fukui, Y. Hosoda, and Y. Shimizu Off-pump long onlay bypass grafting using left internal mammary artery for diffusely diseased coronary artery Ann. Thorac. Surg., August 1, 2003; 76(2): 635 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. J. Woo and T. J. Gardner Myocardial Revascularization with Cardiopulmonary Bypass Card. Surg. Adult, January 1, 2003; 2(2003): 581 - 607. [Full Text]
|
|
|
|
 |
|
 F. Santini, G. Casali, M. Lusini, A. D'Onofrio, E. Barbieri, G. Rigatelli, G. Franco, and A. Mazzucco Mid-term results after extensive vein patch reconstruction and internal mammary grafting of the diffusely diseased left anterior descending coronary artery Eur. J. Cardiothorac. Surg., June 1, 2002; 21(6): 1020 - 1025. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Ruel, F. Wang, M. E. Bourke, J.-Y. Dupuis, J. A. Robblee, W. J. Keon, and F. D. Rubens Is tranexamic acid safe in patients undergoing coronary endarterectomy? Ann. Thorac. Surg., May 1, 2001; 71(5): 1508 - 1511. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
