- Email: [email protected]
Complications related to off-pump bypass grafting
Complications Related to Off-Pump Bypass Grafting Todd M. Dewey and James R. Edgerton Surgical myocardial revascularization is a technique undergoing critical reevaluation in an attempt to reduce operative morbidity and mortality. As the age and number of comorbidities in the surgical population presenting for bypass increases, improved strategies to lessen operative risk have evolved. The use of off-pump bypass grafting to avoid the detrimental effects of extracorporeal circulation demonstrates great promise in reducing operative complications. However, with new techniques come new challenges. Avoidance of the cardiopulmonary bypass circuit has been linked to the development of a hypercoagulable state postoperatively. Complications related to the unique management of the ascending aorta and target vessels during the performance of beating heart surgery are also being reported. Moreover, despite increasing experience in a number of centers, hemodynamic collapse does occur during off-pump bypass, thereby requiring rapid institution of cardiopulmonary bypass. Continued scientific investigation and research should provide the tools to manage the unique obstacles encountered with off-pump coronary artery bypass. © 2003 Elsevier Inc. All rights reserved. Key words: Off-pump coronary artery bypass, beating heart surgery, hypercoagulability, aortic dissection, conversion to cardiopulmonary bypass.
Great deeds are usually wrought at great risk —Herodotus he success of catheter-based techniques for treating ischemic coronary syndromes has profoundly influenced the population of patients presenting for surgical revascularization. Patients presenting for coronary artery bypass grafting are older with more diffuse and advanced atherosclerotic disease and increased comorbidity. By definition, this population of patients is at increased risk for surgical complications.1 Off-pump bypass grafting has experienced striking growth as surgeons attempt to reduce morbidity and mortality in these patients. The elimination of the extracorporeal circuit has fostered the growth of new techniques to safely provide complete myocardial revascularization. As with any major change in surgical approach, new complications and associated morbidity are identified. This article serves to highlight
T
From the Cardiopulmonary Research Science and Technology Institute (CRISTI), Dallas, TX. Address reprint requests to Todd Dewey, Cardiopulmonary Research Science and Technology Institute (CRISTI), 7777 Forest Lane, Suite A323, Dallas, TX 75230. © 2003 Elsevier Inc. All rights reserved. 1043-0679/03/1501-00008-X$30.00/0 doi:10.1016/S1043-0679(03)00008-X
unique problems encountered with off-pump bypass grafting.
Hypercoagulability and Thrombotic Complications Cardiopulmonary bypass (CPB) results in a coagulopathic state by altering the body’s normal hemostatic equilibrium. Multiple mechanisms combine to account for the coagulopathy observed with CPB (Table 1). Exposure of blood to the foreign surface of the extracorporeal circuit results in platelet activation and initiation of the intrinsic coagulation and fibrinolytic cascades.2 Additionally, the extrinsic loop of the coagulation system becomes activated by tissue phospholipids liberated during the course of surgery. In contradistinction, normal intact vascular endothelium exhibits both passive and active antithrombotic properties as a result of the release of prostacyclin and endothelium-derived relaxing factor, as well as activation of protein C and inactivation of thrombin. Other factors contributing to the alteration in hemostasis associated with bypass surgery include length of time on bypass, hemodilution of clotting factors, hypothermia, endocrine stress response, surgical trauma, and a generalized inflammatory response. This inflammatory re-
Seminars in Thoracic and Cardiovascular Surgery, Vol 15, No 1 ( January), 2003: pp 63-70
63
64
Dewey and Edgerton
Table 1. Hemostatic Changes Associated With Cardiopulmonary Bypass Thrombocytopenia Reduced platelet adhesion and aggregation Hypofibrogenemia secondary to fibrinolysis Reduced Factor II Reduced Factor V Reduced Factor VIII:c Activation of Factor XII
sponse is potentiated by leukocyte and platelet activation with associated cytokine release, which in turn activates complement.3,4 Comparison of platelet counts before and after bypass shows a striking reduction in absolute numbers.5,6 Furthermore, bleeding times as a measure of impaired hemostasis demonstrate a strong linear correlation with the length of time on bypass.7 The significance of these changes is a propensity for postoperative bleeding in the patient undergoing conventional bypass grafting with CPB. However, diminished hemostasis may also be advantageous in regards to protecting anastomotic patency. In fact, CPB has previously been shown to result in abnormal hemostasis and platelet function for at least 24 hours postoperatively.8 One can easily infer that impaired hemostasis may prevent early graft loss from thrombosis until a postoperative regimen of antiplatelet therapy can be instituted Off-pump bypass grafting eliminates the extracorporeal circuit, thereby maintaining the hemostatic integrity of the patient. Multiple reports in the literature, including two large randomized trials, attest to the significantly reduced transfusion requirements in patients having beating heart surgery.9,10 Most authors attribute this to preserved platelet function and avoidance of the coagulopathy generated by CPB. However, preserved clotting function comes at a price, which may very well be the development of a hypercoagulable state. Postoperative hypercoagulability is a well-recognized state in the general surgical and orthopedic literature and can result in significant morbidity. Tissue injury during dissection releases significant amounts of tissue factor into the circulation. Thrombin generation occurs as this procoagulant traverses the capillary bed of the lungs leading to enhanced clotting ability.11 Sternotomy induces the release of thrombin-antithrom-
bin complexes and increases tissue plasmin activator. Bleeding from the sternal bone results in the local seeding of procoagulant factors across the surgical field. Additionally, there is evidence to suggest that elevated thrombin generation and increased tissue factor levels may be associated with the development of coronary artery disease.12 Taken in totality, these factors can combine to create a hypercoagulable environment in the off-pump patient. A recent report by Mariani and colleagues13 from the Netherlands seems to support this hypothesis. They prospectively examined the coagulation profiles of 22 patients undergoing offpump bypass grafting by measuring prothombin factor 1 and 2 (F1 ⫹ 2), factor VII, and fibrin degradation products. Measurements were taken before the start of the operation (baseline), after heparin administration, at the end of the procedure, and 24 hours postoperatively. Additionally, platelet counts, -thromboglobulin, and Von Willebrands factor were recorded at baseline and 24 hours. They demonstrated that platelet counts and platelet activation remained at preoperative levels after the procedure, thereby confirming intact platelet function. Procoagulant activity was evidenced by an increase in F1 ⫹ 2 and fibrin degradation products and by the decrease in factor VII. Enhanced hemostasis that can be attributable to endothelial activation was indicated by an increase in Von Willebrands factor on postoperative day number one. Although no clinical sequelae were reported during follow-up, these findings strongly suggest the development of a hypercoagulable state early after beating heart surgery. Mariani and colleagues concluded that an aggressive perioperative anticoagulant regimen be instituted in all off-pump patients. Complications related to a potential procoagulant effect with off-pump surgery have been described in the literature. Cartier and Robitaille recently published a small series of thrombotic complications in off-pump patients.14 In a series of 500 patients revascularized off-pump, thromboembolic complications occurred in 1%. These events were characterized by two iliofemoral deep venous thromboses and three episodes of pulmonary emboli, one of which was fatal. Correspondingly, in a contemporary cohort of 1,476 patients revascularized using CPB there were only five episodes of deep venous thrombosis and three episodes of pulmonary emboli, for an inci-
Complications Related to Off-Pump Bypass Grafting
Figure 1. Contrast-enhanced spiral computed tomography scan of the chest showing large saddle type pulmonary embolus in a patient 11 days after off-pump bypass grafting.
dence of 0.5%. Although this difference did not reach statistical significance, it does serve to alert surgeons of the potential for thrombotic complications not generally associated with surgical revascularization. By way of example, the authors recently admitted a patient with a large nonfatal pulmonary embolus 12 days after off-pump bypass grafting (Fig 1). Duplex scanning revealed extensive deep venous thrombosis in both popliteal veins despite an aggressive postoperative antiplatelet regimen. We would concur with the recommendations of Cartier and Robitaille that an increased index of suspicion for thrombotic complications should be maintained in the offpump population. The overriding concern with off-pump bypass grafting clearly lies in the possibility of reduced graft patency due to a prothrombotic state. However, the reported clinical experience does not reveal such a pattern. Numerous reported angiographic series demonstrate early patency rates equivalent to previously published data for on pump patients.15-17 This information appears to obviate the concern regarding acute graft closure as the result of a hypercoagulable state. Unfortunately, long-term patency remains largely an unknown. Saphenous vein patency is a function of the development of intimal hyperplasia and graft atherosclerosis. Inciting events during surgery, such as intimal injury with subsequent platelet
65
activation and thrombus deposition, can set the stage for reduced long-term graft patency.18 Only long-term comparative reviews documenting patency equivalent to conventional revascularization will fully answer the question. In light of the well-documented and frequent unresponsiveness of patients to aspirin, many surgeons have empirically chosen to augment their standard postoperative antiplatelet regimen with the addition of clopidogril.19 Indeed, the authors provide patients with a loading dose of clopidogril within the first 12-24 hours and continue a maintenance dose for at least 6 weeks. To date no bleeding complications attributable to the addition of Clopidogril have been identified, and thrombotic complications have been rare. Ultimately, larger studies are needed to definitively characterize the hemostatic alterations associated with beating heart surgery and delineate its potential impact on early and long term results.
Vascular Complications Vascular complications, although rare, are a potentially fatal complication of coronary artery bypass grafting. Aortic dissection has been described with a frequency of approximately 0.12% in patients undergoing revascularization using CPB.20,21 It can occur intraoperatively or in the postoperative period, including after discharge. There are numerous risk factors related to the development of this complication, including aortic cannulation, aortic cross clamping, use of a partial occluding clamp, intimal injury or tear during the suturing of grafts to the aorta, or failure to achieve intima to intima approximation (Table 2). Predisposing patient factors include hypertension, atherosclerosis of the aorta, thin dilated aortic walls, or inherited connective tissue Table 2. Risk Factors for Aortic Dissection Hypertension Aortic atherosclerosis Connective tissue disorders Thin friable aortic wall Aortic Trauma Arterial and aortic cannulation for Cardiopulmonary bypass Aortic cross clamping and partial occlusion clamping Proximal graft anastomosis
66
Dewey and Edgerton
disorders, such as Marfans disease. Off-pump bypass grafting requires that a partial occluding clamp be placed on the pulsatile and pressurized ascending aorta to construct the proximal vein anastomosis. This raises the theoretical concern over an increased risk of aortic dissection in this population. Evidence illustrating this potential was recently reported by Chavanon and colleagues describing an increased incidence of aortic dissection in their off-pump patients.22 A retrospective analysis revealed three episodes of aortic dissection among 308 patients undergoing beating heart surgery for an incidence of 0.97%. This is compared with one episode in 2,723 patients revascularized using CPB (0.04%, P ⬍ 0.00001). The authors did not attempt to reduce arterial pressure before applying the side-biting clamp in their first 100 patients and then only lowered the blood pressure to 100 mm Hg for the rest of the series. In their early experience, both distal and proximal anastomosis were performed sequentially requiring multiple applications of the partial occluding clamp to the aorta. Additionally, no mention of the use of epiaortic scanning or transesophageal echocardiography to detect atherosclerosis of the ascending aorta is described in the report. Chavanon and colleagues22 recommend pharmacologic reduction of blood pressure below 100 mm Hg before application of the clamp and maintaining this reduction during the performance of the anastomoses. Additionally, all proximal grafts should be performed under a single application of the partial occluding clamp. Concern over the increased potential for aortic dissection with off-pump bypass grafting appears valid, and highlights the need to reduce manipulation of the ascending aorta and maintain tight control over the patients’ hemodynamics. Alternative techniques to partial occlusion, such as total arterial revascularization, mechanical anastomotic connectors, and placing proximal grafts on the brachiocephalic artery or off the internal mammary artery, should be considered in high-risk situations. Off-pump bypass grafting differs greatly from arrested heart revascularization in the need for temporary proximal occlusion of the target arteries. Construction of a quality distal anastomosis on a beating heart requires meticulous placement of suture bites and a bloodless field. Methods for obtaining optimal anastomotic conditions include the use of intravascular shunts and oc-
cluders, snares using polypropylene sutures or silicone loops, and the use of gas insufflation. Reports have implicated each of these techniques with causing endothelial injury to the target vessel. Burfeind and colleagues reported a series of animal experiments examining the effects of high-flow gas insufflation on coronary endothelium.23 Five pigs had an arteriotomy of the left anterior descending artery (LAD) that was then exposed to a direct stream of carbon dioxide at 15 L/minute for 20 minutes. Two control pigs underwent a similar arteriotomy that was only exposed to room air. Sectioning and processing of the segments revealed that all LAD segments exposed only to room air demonstrated normal contiguous endothelium, whereas segments exposed to high flow gas demonstrated near-complete loss of endothelium. Although limited by the use of higher flow rates than clinically used, nonhumidification of the gas, and normal vascular endothelium rather than atherosclerotic vessels, this report raises questions regarding gas insufflation for visualization. Early patency problems related to gas insufflation have not been clinically identified. However, exposure of denuded endothelium to blood elements may promote smooth muscle cell migration and proliferation, thereby setting the stage for late graft failure. A common method for securing a bloodless operative field involves the use of extravascular snaring of the target coronary artery. This technique uses either silicone loops or polypropylene sutures placed proximally to transiently impede blood flow through the vessel. Distal snares are generally not placed to avoid potential injury to the downstream arterial bed.24 Hangler and associates demonstrated by scanning electron microscopy severe endothelial injury in the segment of artery subjected to occlusion than in the nonmanipulated segments.25 Additionally, in the region of local occlusion, plaque rupture was noted in three of 34 vessels, injury to side branches in two of 44 vessels, and local microthrombus formation was evident in six of 44 segments. Several clinical case reports also document complications related to snaring of the target coronary artery. Izzat and associates reported the distal embolization of atherosclerotic debris resulting in occlusion of an internal mammary to LAD anastomosis.26 This complication was identified by on-table angiography and required reconstruction of the
Complications Related to Off-Pump Bypass Grafting
anastomosis. Demaria and colleagues described the development of snare related flow limiting lesions both proximally and distally to patent grafts early postoperatively in a diabetic patient.27 These lesions were located at the sites of snare application. They theorize that because patients with diabetes usually have severe and extensive calcification of the arterial media, an increased risk of marked intimal hyperplasia and smooth muscle proliferation may occur in response to plaque rupture or injury. Clearly, the potential for vascular injury exists when circumferential force is applied to susceptible areas. Perrault and associates reported in animals with normal vascular endothelium that occlusive snaring does not cause coronary endothelial dysfunction.28 However, caution must be used when applying this technique to atherosclerotic arteries that may poorly tolerate manipulation. In the series by Hangler, a trend toward reduced injury was identified when occlusion was provided by silicone loops rather than polypropylene sutures. Acceptable visualization should not be achieved at the expense of local coronary injury, which may ultimately cause reduced graft patency.
Conversion to Cardiopulmonary Bypass Inherent in the performance of off-pump bypass grafting is the potential for conversion to CPB for completion of the operation. Despite the growing enthusiasm for off-pump bypass grafting, few authors directly address the clinical situations that lead to the abandonment of a beating heart approach for pump-supported revascularization. An examination of the literature finds a reported incidence of conversion from beating heart surgery to cardiopulmonary bypass ranging anywhere from 1.1-16.3%.29,30 Altering the surgical strategy from off-pump to on-pump grafting during a procedure does not necessarily signify a problem, however, patient morbidity and mortality resulting from the conversion clearly indicates a complication due to the initial surgical approach. Circumstances leading to conversion may logically be grouped into two broad categories: 1) patient specific factors that do not enable the operation to be performed without cardiopulmonary support or 2) patient deterioration requiring conversion for hemodynamic support or salvage (Table 3).
67
Table 3. Circumstances Leading to Conversion from Off-pump to Cardiopulmonary Bypass Deeply intramyocardial vessels Calcified or difusely diseased targets Diminuitive vessels Need for extended endarterectomy of the target vessel Hemodynamic instability Ongoing ischemia Arrhythmias Cardiomegaly
With experience, many patient-related factors that may preclude an off-pump approach can be identified preoperatively. These include intramyocardial vessels, heavily calcified or diffusely diseased targets, diminutive arteries by angiography, and patients hemodynamically unstable before surgery. Additionally, there are rare patients in which adequate positioning of the heart to expose the target vessel may not be able to be achieved. This most frequently occurs with attempts to expose the posterior-lateral wall of the heart. These circumstances of conversion from off-pump grafting to pump supported grafting rarely translate into increased patient mortality. Anyanwu and colleagues described their experience with eight similar conversions of 285 off-pump procedures.31 No increased mortality was identified in this group. Edgerton and associates recently reported a large retrospective review of 1385 patients undergoing beating heart surgery over a 2-year period.32 The authors described a cohort of 48 patients converted from off-pump bypass grafting to bypass with CPB support. The authors classified the patients depending on whether their conversion took place electively versus urgently, and as to whether the conversion was early in the course of the procedure, ie, before the initiation of any anastomoses, or late in the operation. There were no deaths identified in the patients converted early in the procedure regardless of whether this was done electively or urgently. Likewise, Baumgartner and associates reported no mortality in a series of 141 patients with nine discretionary conversions.33 Therefore, it appears reasonable to conclude that patients converted from an off-pump approach to CPB before encountering significant instability demonstrate no increased morbidity or mortality than those patients undergoing conventional procedures.
68
Dewey and Edgerton
Hemodynamic deterioration during off-pump grafting contributes to a small but significant set of patients who are converted from beating heart surgery to CPB. Multiple factors contribute to the downward hemodynamic spiral that ultimately leads to rapid institution of cardiopulmonary support. Significant hemodynamic compromise may result from mechanical alterations of the normal geometry of the heart such as right ventricular compression, right ventricular outflow tract obstruction, induced mitral annular deformation leading to functional mitral stenosis or regurgitation, and impaired left ventricular filling.34 Coexistent ischemia compounds the effects of mechanical dysfunction resulting in the need for emergent institution of CPB. Grafting strategy plays an important role in preventing ischemia induced or potentiated hemodynamic compromise, especially in patients with poor ventricular function or acute ischemia preoperatively. In general, the anterior vessels should be revascularized and perfused first; this strategy facilitates the manipulation of the heart to graft the remaining targets. Additionally, collateralized vessels should be grafted before collateralizing vessels. Once patients become unstable to the extent that pharmacologic therapy and replacing the heart in an anatomic position fails to restore or improve hemodynamics, emergent transition to cardiopulmonary bypass is indicated. CPB is preferred over isolated right or left heart support because of the ability to simultaneously support both ventricles and completely decompress and unload the heart. Grafting can then continue either with the heart beating utilizing off-pump techniques or under cardioplegic arrest. Several series illustrate the increased morbidity and mortality associated with urgent or emergent transition to CPB. Vassiliades and associates recently reported a series of 23 cases converted to CPB after sustaining hemodynamic collapse during elective off-pump bypass grafting.35 The resulting mortality rate was 8.7% versus 1.7% in patients not converted to CPB (P ⫽ 0.08). Additionally, there was a significant increase in the incidence of perioperative myocardial infarction, renal failure, transfusion rate, inotrope use, and requirement for an intraaortic balloon pump to leave the operating room. The converted patients also had a significantly longer length of hospital stay. Soltoski and colleagues compared the tendency toward conversion and
outcomes between two groups of surgeons in their practice with differing levels of experience with off-pump bypass grafting.36 Surgeons comprising group A approached all patients as candidates for off-pump grafting, where as surgeons comprising group B were selective in offering patients beating heart surgery. Surgeons in group A performed 248 off-pump procedures with an 11% conversion rate. Surgeons in group B had a 26% conversion rate in 90 off-pump procedures. An increased mortality of 8.3% was identified for both groups in the converted patients. Patients converted for hemodynamic instability had a 23% complication rate, and a 6.25% mortality rate. A subgroup of patients with graft failure leading to conversion had a mortality rate of 28.5%. Increased surgeon experience with off-pump bypass clearly reduced the incidence of conversion between the two groups, but did not affect the outcome of the patients once conversion took place. Edgerton and associates32 in their series reported a mortality rate of 14% in seven patients with discretionary conversion late in the procedure, but a mortality of 41% in 17 patients converted late in the operation because of hemodynamic instability. Analogous to Soltoski, they examined the influence of surgeon experience on the rate and outcome of conversion. Once again, surgeons with limited off-pump experience had the highest rate of conversion. However, unlike the report form Soltoski, this cohort did not have an increased mortality. This suggests that with experience, surgeons can complete more procedures, but they also may delay conversion longer than inexperienced colleagues thereby encountering increased mortality and morbidity. Edgerton also demonstrated increased complications in the converted group when compared with a computer-matched group of on-pump patients. Converted patients had significantly higher rate of postoperative cardiac arrest, multi-system organ failure, pneumonia, and vascular complications. Identification of patients at increased risk of conversion from hemodynamic collapse has been difficult because of the small numbers of cases reported in the literature. Vassiliades found that patients converted for hemodynamic instability had a higher New York Heart Association score (3.25 versus 2.7, P ⫽ 0.001), an increased incidence of preoperative myocardial infarction (73.9% versus 25.4%, P ⫽ 0.001), more cardio-
Complications Related to Off-Pump Bypass Grafting
megaly (73.4% versus 49.9%, P ⫽ 0.043) and a smaller body surface area (1.9% versus 2.1%, P ⫽ 0.001). Edgerton found by logistic regression analysis that previous bypass grafting, the presence of a preoperative intraaortic balloon pump, and limited surgeon experience were significant risk factors for conversion. However, until larger series are reported, firm conclusions regarding risk of conversion based on preoperative variables cannot be made.
Conclusion The goal of off-pump bypass surgery is to reduce operative morbidity and mortality compared with conventional surgical revascularization. However, surgeons need to be aware of the unique challenges and risks inherent in any new surgical approach. Vascular injuries can be minimized by depressurizing the aorta when placing the partial occluding clamp, and by reduced manipulation and instrumentation of the target vessel. Lowpressure gas insufflation and proximal silicone snares can provide adequate visualization for construction of the anastomosis without increased risk of injury to the coronary artery. The concern over a hypercoagulable state following off-pump surgery requires diligence with antiplatelet inhibition postoperatively, and a low index of suspicion for potential thrombotic events. Impending hemodynamic collapse requires prompt recognition and corrective action to avoid increased mortality and morbidity. Avoidance of potential hemodynamic problems requires 1) careful patient selection, 2) an appropriate grafting sequence based on the coronary anatomy, 3) careful cardiac manipulation and positioning, 4) early recognition of mechanically induced cardiac impairment or ischemia before initiation of an anastomosis, and 5) prompt response to impending collapse. Many situations may be salvaged by the appropriate use of intravascular shunts, cardiac pacing, returning the heart to its anatomic position, or on rare occasions isolated right or left ventricular support. In the future, increasing experience with the management of theses unique patients promises to provide improved strategies to prevent adverse events and thus a safer procedure.
69
References 1. Katz NM, Gersh BJ, Cox JL: Changing practice of coronary bypass surgery and its impact on early risk and long-term survival. [Review] Curr Opin Cardiol 13:465475, 1998 2. Salmenpera MT, Levy JH, Harker LA: Hemostasis and cardiopulmonary bypass, in Cardiopulmonary Bypass: Principles and Techniques of Extracorporeal Circulation. New York, NY, Springer-Verlag, 1995, pp 88-113 3. Butler J, Rocker GM, Westaby S: Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 55:552-559, 1993 4. Edmunds LH: Why cardiopulmonary bypass makes patients sick: strategies to control the blood-synthetic surface interface. Adv Card Surg 6:131-167, 1995 5. Jung G. Razafindranaibe F, Elkouby A, et al: Modification in platelet shape and change in ATP release during CPB. Haemostasis 25:149-157, 1995 6. Wahba A, Rothe G, Schmitz G, et al: Heparin effect on platelets and fibrinolysis. Ann Thorac Surg 61:1590-1591, 1996 7. Bertolino G, Locatelli A, Noris P, et al: Platelet composition and function in patients undergoing CPB for heart surgery. Haemotologica 81:116-120, 1996 8. Tabuchi N, de Haan J, van Oeveren W, et al: Rapid recovery of platelet function after cardiopulmonary bypass. Blood 82:2930-2931, 1993 9. Van Dijk D, Nierich AP, Jansen EWL, et al: Early outcome after off-pump versus on-pump coronary bypass surgery. Circulation 104:1761-1766, 2001 10. Ascione R, Williams S, Lloyd CT, et al: Reduced postoperative blood loss and transfusion requirement after beating heart coronary operations: A prospective randomized study. J Thorac Cardiovasc Surg 121:689-696, 2001 11. Dahl OE: Mechanisms of hypercoagulability. Thromb Haemost 82:9902-9906, 1999 12. Falciani M, Gori AM, Fedi S, et al: Elevated tissue factor and tissue factor pathway inhibitor circulating levels in ischaemic heart disease. Thromb Haemost 79:495-499, 1998 13. Mariani MA, Gu J, Boonstra PW, et al: Procoagulant activity after off-pump coronary operation: Is the current anticoagulation adequate? Ann Thorac Surg 67:13701375, 1999 14. Cartier R, Robitaille D: Thrombotic complications in beating heart operations. J Thorac Cardiovasc Surg 121: 920-922, 2001 15. Puskas JD, Thourani VH, Marshall JJ, et al: Clinical outcomes, angiographic patency, and resource utilization in 200 consecutive off pump coronary bypass patients. Ann Thorac Surg 71:1477-1484, 2001 16. Calafiore AM, Teodori G, Di Giammarco G, et al: Multiple arterial conduits without cardiopulmonary bypass: Early Angiographic results. Ann Thorac Surg 67:450-456, 1999 17. Izzat MB, Khaw KS, Atassi W, et al: Routine intraoperative angiography improves the early patency of coronary grafts performed on the beating heart. Chest 115:987990, 1999
70
Dewey and Edgerton
18. Motwani JG, Topol EL: Aortocoronary saphenous vein graft disease: Pathogenesis, predisposition and prevention. Circulation 97:916-931, 1998 19. Steinhubl SR: Antiplatelet agents in cardiology: The choice of therapy. Ann Thorac Surg 70:S3-8, 2000 20. Blakeman BM, Pifarre R, Sullivan HJ, et al: Perioperative dissection of the ascending aorta: Types of repair. J Card Surg 3:9-14, 1988 21. Still RJ, Hilgenberg AD, Akins CW, et al: Intraoperative aortic dissection. Ann Thorac Surg 53:374-379, 1992 22. Chavanon O, Carrier M, Cartier R, et al: Increased incidence of acute ascending aortic dissection with off pump aortocoronary bypass surgery? Ann Thorac Surg 71:117121, 2001 23. Burfeind WR, Duhaylongsod FG, Annex BH, et al: Highflow gas insufflation to facilitate MIDCABG: Effects on coronary endothelium. Ann Thorac Surg 66:1246-1249, 1998 24. Gundry S, Romano MA, Shattuck OH, et al: Seven-year follow up of coronary artery bypasses performed with and without cardiopulmonary bypass. J Thorac Cardiovasc Surg 115:1273-1278, 1998 25. Hangler HB, Pfaller K, Antretter H, et al: Coronary endothelial injury after local occlusion on the human beating heart. Ann Thorac Surg 71:122-127, 2001 26. Izzat MB, Yim APC, El-Zufari H: Snaring of a coronary artery causing distal atheromas embolization. Ann Thorac Surg 66:1806-1808, 1998 27. Demaria RG, Fortier S, Carrier M, et al: Early multifocal stenosis after coronary artery snaring during off-pump coronary artery bypass in a patient with diabetes. J Thorac Cardiovasc Surg 122:1044-1045, 2001 28. Perrault LP, Menasche´ P, Bidouard JP, et al: Snaring of the target vessels in less invasive bypass operations does
29.
30.
31.
32.
33.
33.
34.
35.
not cause endothelial dysfunction. Ann Thorac Surg 63: 751-755, 1997 Calafiore AM, Mauro MD, Teodori G, et al: Impact of aortic manipulation on incidence of cerebrovascular accidents after surgical myocardial revascularization. Ann Thorac Surg 73:1387-1393, 2002 Novick RJ, Fox SA, Stitts LW, et al: Cumulative sum failure analysis of a policy change from on-pump to offpump coronary artery bypass. Ann Thorac Surg 72:S10161021, 2001 Anyanwu AC, Al-Ruzzeh SA, George SJ, et al: Conversion to off-pump coronary bypass without increased morbidity or change in practice. Ann Thorac Surg 73:798-802, 2002 Edgerton JR, Dewey TM, Herbert M, et al: Conversion in off-pump coronary artery bypass grafting (OPCAB): An analysis of predictors and outcomes. Ann Thorac Surg (in press) Baumgartner FJ, Gheissari A, Capouya ER, et al: Technical aspects of total revascularization in off-pump coronary bypass via sternotomy approach. Ann Thorac Surg 67:1653-1658, 1999 George SJ, Al-Ruzzeh S, Amrani M: Mitral annulus distortion during beating heart surgery: A potential cause for hemodynamic disturbance- A three-dimensional echocardiography reconstruction study. Ann Thorac Surg 73: 1424-1430, 2002 Vassiliades TA, Nielsen JL, Lonquist JL: Hemodynamic collapse during off-pump coronary artery bypass grafting. Ann Thorac Surg 73:1874-1879, 2002 Soltoski P, Salerno T, Levinsky L, et al: Conversion to cardiopulmonary bypass in off-pump coronary artery bypass grafting: Its effect on outcome. J Cardiol Surg 13: 328-334, 1998
Great deeds are usually wrought at great risk —Herodotus he success of catheter-based techniques for treating ischemic coronary syndromes has profoundly influenced the population of patients presenting for surgical revascularization. Patients presenting for coronary artery bypass grafting are older with more diffuse and advanced atherosclerotic disease and increased comorbidity. By definition, this population of patients is at increased risk for surgical complications.1 Off-pump bypass grafting has experienced striking growth as surgeons attempt to reduce morbidity and mortality in these patients. The elimination of the extracorporeal circuit has fostered the growth of new techniques to safely provide complete myocardial revascularization. As with any major change in surgical approach, new complications and associated morbidity are identified. This article serves to highlight
T
From the Cardiopulmonary Research Science and Technology Institute (CRISTI), Dallas, TX. Address reprint requests to Todd Dewey, Cardiopulmonary Research Science and Technology Institute (CRISTI), 7777 Forest Lane, Suite A323, Dallas, TX 75230. © 2003 Elsevier Inc. All rights reserved. 1043-0679/03/1501-00008-X$30.00/0 doi:10.1016/S1043-0679(03)00008-X
unique problems encountered with off-pump bypass grafting.
Hypercoagulability and Thrombotic Complications Cardiopulmonary bypass (CPB) results in a coagulopathic state by altering the body’s normal hemostatic equilibrium. Multiple mechanisms combine to account for the coagulopathy observed with CPB (Table 1). Exposure of blood to the foreign surface of the extracorporeal circuit results in platelet activation and initiation of the intrinsic coagulation and fibrinolytic cascades.2 Additionally, the extrinsic loop of the coagulation system becomes activated by tissue phospholipids liberated during the course of surgery. In contradistinction, normal intact vascular endothelium exhibits both passive and active antithrombotic properties as a result of the release of prostacyclin and endothelium-derived relaxing factor, as well as activation of protein C and inactivation of thrombin. Other factors contributing to the alteration in hemostasis associated with bypass surgery include length of time on bypass, hemodilution of clotting factors, hypothermia, endocrine stress response, surgical trauma, and a generalized inflammatory response. This inflammatory re-
Seminars in Thoracic and Cardiovascular Surgery, Vol 15, No 1 ( January), 2003: pp 63-70
63
64
Dewey and Edgerton
Table 1. Hemostatic Changes Associated With Cardiopulmonary Bypass Thrombocytopenia Reduced platelet adhesion and aggregation Hypofibrogenemia secondary to fibrinolysis Reduced Factor II Reduced Factor V Reduced Factor VIII:c Activation of Factor XII
sponse is potentiated by leukocyte and platelet activation with associated cytokine release, which in turn activates complement.3,4 Comparison of platelet counts before and after bypass shows a striking reduction in absolute numbers.5,6 Furthermore, bleeding times as a measure of impaired hemostasis demonstrate a strong linear correlation with the length of time on bypass.7 The significance of these changes is a propensity for postoperative bleeding in the patient undergoing conventional bypass grafting with CPB. However, diminished hemostasis may also be advantageous in regards to protecting anastomotic patency. In fact, CPB has previously been shown to result in abnormal hemostasis and platelet function for at least 24 hours postoperatively.8 One can easily infer that impaired hemostasis may prevent early graft loss from thrombosis until a postoperative regimen of antiplatelet therapy can be instituted Off-pump bypass grafting eliminates the extracorporeal circuit, thereby maintaining the hemostatic integrity of the patient. Multiple reports in the literature, including two large randomized trials, attest to the significantly reduced transfusion requirements in patients having beating heart surgery.9,10 Most authors attribute this to preserved platelet function and avoidance of the coagulopathy generated by CPB. However, preserved clotting function comes at a price, which may very well be the development of a hypercoagulable state. Postoperative hypercoagulability is a well-recognized state in the general surgical and orthopedic literature and can result in significant morbidity. Tissue injury during dissection releases significant amounts of tissue factor into the circulation. Thrombin generation occurs as this procoagulant traverses the capillary bed of the lungs leading to enhanced clotting ability.11 Sternotomy induces the release of thrombin-antithrom-
bin complexes and increases tissue plasmin activator. Bleeding from the sternal bone results in the local seeding of procoagulant factors across the surgical field. Additionally, there is evidence to suggest that elevated thrombin generation and increased tissue factor levels may be associated with the development of coronary artery disease.12 Taken in totality, these factors can combine to create a hypercoagulable environment in the off-pump patient. A recent report by Mariani and colleagues13 from the Netherlands seems to support this hypothesis. They prospectively examined the coagulation profiles of 22 patients undergoing offpump bypass grafting by measuring prothombin factor 1 and 2 (F1 ⫹ 2), factor VII, and fibrin degradation products. Measurements were taken before the start of the operation (baseline), after heparin administration, at the end of the procedure, and 24 hours postoperatively. Additionally, platelet counts, -thromboglobulin, and Von Willebrands factor were recorded at baseline and 24 hours. They demonstrated that platelet counts and platelet activation remained at preoperative levels after the procedure, thereby confirming intact platelet function. Procoagulant activity was evidenced by an increase in F1 ⫹ 2 and fibrin degradation products and by the decrease in factor VII. Enhanced hemostasis that can be attributable to endothelial activation was indicated by an increase in Von Willebrands factor on postoperative day number one. Although no clinical sequelae were reported during follow-up, these findings strongly suggest the development of a hypercoagulable state early after beating heart surgery. Mariani and colleagues concluded that an aggressive perioperative anticoagulant regimen be instituted in all off-pump patients. Complications related to a potential procoagulant effect with off-pump surgery have been described in the literature. Cartier and Robitaille recently published a small series of thrombotic complications in off-pump patients.14 In a series of 500 patients revascularized off-pump, thromboembolic complications occurred in 1%. These events were characterized by two iliofemoral deep venous thromboses and three episodes of pulmonary emboli, one of which was fatal. Correspondingly, in a contemporary cohort of 1,476 patients revascularized using CPB there were only five episodes of deep venous thrombosis and three episodes of pulmonary emboli, for an inci-
Complications Related to Off-Pump Bypass Grafting
Figure 1. Contrast-enhanced spiral computed tomography scan of the chest showing large saddle type pulmonary embolus in a patient 11 days after off-pump bypass grafting.
dence of 0.5%. Although this difference did not reach statistical significance, it does serve to alert surgeons of the potential for thrombotic complications not generally associated with surgical revascularization. By way of example, the authors recently admitted a patient with a large nonfatal pulmonary embolus 12 days after off-pump bypass grafting (Fig 1). Duplex scanning revealed extensive deep venous thrombosis in both popliteal veins despite an aggressive postoperative antiplatelet regimen. We would concur with the recommendations of Cartier and Robitaille that an increased index of suspicion for thrombotic complications should be maintained in the offpump population. The overriding concern with off-pump bypass grafting clearly lies in the possibility of reduced graft patency due to a prothrombotic state. However, the reported clinical experience does not reveal such a pattern. Numerous reported angiographic series demonstrate early patency rates equivalent to previously published data for on pump patients.15-17 This information appears to obviate the concern regarding acute graft closure as the result of a hypercoagulable state. Unfortunately, long-term patency remains largely an unknown. Saphenous vein patency is a function of the development of intimal hyperplasia and graft atherosclerosis. Inciting events during surgery, such as intimal injury with subsequent platelet
65
activation and thrombus deposition, can set the stage for reduced long-term graft patency.18 Only long-term comparative reviews documenting patency equivalent to conventional revascularization will fully answer the question. In light of the well-documented and frequent unresponsiveness of patients to aspirin, many surgeons have empirically chosen to augment their standard postoperative antiplatelet regimen with the addition of clopidogril.19 Indeed, the authors provide patients with a loading dose of clopidogril within the first 12-24 hours and continue a maintenance dose for at least 6 weeks. To date no bleeding complications attributable to the addition of Clopidogril have been identified, and thrombotic complications have been rare. Ultimately, larger studies are needed to definitively characterize the hemostatic alterations associated with beating heart surgery and delineate its potential impact on early and long term results.
Vascular Complications Vascular complications, although rare, are a potentially fatal complication of coronary artery bypass grafting. Aortic dissection has been described with a frequency of approximately 0.12% in patients undergoing revascularization using CPB.20,21 It can occur intraoperatively or in the postoperative period, including after discharge. There are numerous risk factors related to the development of this complication, including aortic cannulation, aortic cross clamping, use of a partial occluding clamp, intimal injury or tear during the suturing of grafts to the aorta, or failure to achieve intima to intima approximation (Table 2). Predisposing patient factors include hypertension, atherosclerosis of the aorta, thin dilated aortic walls, or inherited connective tissue Table 2. Risk Factors for Aortic Dissection Hypertension Aortic atherosclerosis Connective tissue disorders Thin friable aortic wall Aortic Trauma Arterial and aortic cannulation for Cardiopulmonary bypass Aortic cross clamping and partial occlusion clamping Proximal graft anastomosis
66
Dewey and Edgerton
disorders, such as Marfans disease. Off-pump bypass grafting requires that a partial occluding clamp be placed on the pulsatile and pressurized ascending aorta to construct the proximal vein anastomosis. This raises the theoretical concern over an increased risk of aortic dissection in this population. Evidence illustrating this potential was recently reported by Chavanon and colleagues describing an increased incidence of aortic dissection in their off-pump patients.22 A retrospective analysis revealed three episodes of aortic dissection among 308 patients undergoing beating heart surgery for an incidence of 0.97%. This is compared with one episode in 2,723 patients revascularized using CPB (0.04%, P ⬍ 0.00001). The authors did not attempt to reduce arterial pressure before applying the side-biting clamp in their first 100 patients and then only lowered the blood pressure to 100 mm Hg for the rest of the series. In their early experience, both distal and proximal anastomosis were performed sequentially requiring multiple applications of the partial occluding clamp to the aorta. Additionally, no mention of the use of epiaortic scanning or transesophageal echocardiography to detect atherosclerosis of the ascending aorta is described in the report. Chavanon and colleagues22 recommend pharmacologic reduction of blood pressure below 100 mm Hg before application of the clamp and maintaining this reduction during the performance of the anastomoses. Additionally, all proximal grafts should be performed under a single application of the partial occluding clamp. Concern over the increased potential for aortic dissection with off-pump bypass grafting appears valid, and highlights the need to reduce manipulation of the ascending aorta and maintain tight control over the patients’ hemodynamics. Alternative techniques to partial occlusion, such as total arterial revascularization, mechanical anastomotic connectors, and placing proximal grafts on the brachiocephalic artery or off the internal mammary artery, should be considered in high-risk situations. Off-pump bypass grafting differs greatly from arrested heart revascularization in the need for temporary proximal occlusion of the target arteries. Construction of a quality distal anastomosis on a beating heart requires meticulous placement of suture bites and a bloodless field. Methods for obtaining optimal anastomotic conditions include the use of intravascular shunts and oc-
cluders, snares using polypropylene sutures or silicone loops, and the use of gas insufflation. Reports have implicated each of these techniques with causing endothelial injury to the target vessel. Burfeind and colleagues reported a series of animal experiments examining the effects of high-flow gas insufflation on coronary endothelium.23 Five pigs had an arteriotomy of the left anterior descending artery (LAD) that was then exposed to a direct stream of carbon dioxide at 15 L/minute for 20 minutes. Two control pigs underwent a similar arteriotomy that was only exposed to room air. Sectioning and processing of the segments revealed that all LAD segments exposed only to room air demonstrated normal contiguous endothelium, whereas segments exposed to high flow gas demonstrated near-complete loss of endothelium. Although limited by the use of higher flow rates than clinically used, nonhumidification of the gas, and normal vascular endothelium rather than atherosclerotic vessels, this report raises questions regarding gas insufflation for visualization. Early patency problems related to gas insufflation have not been clinically identified. However, exposure of denuded endothelium to blood elements may promote smooth muscle cell migration and proliferation, thereby setting the stage for late graft failure. A common method for securing a bloodless operative field involves the use of extravascular snaring of the target coronary artery. This technique uses either silicone loops or polypropylene sutures placed proximally to transiently impede blood flow through the vessel. Distal snares are generally not placed to avoid potential injury to the downstream arterial bed.24 Hangler and associates demonstrated by scanning electron microscopy severe endothelial injury in the segment of artery subjected to occlusion than in the nonmanipulated segments.25 Additionally, in the region of local occlusion, plaque rupture was noted in three of 34 vessels, injury to side branches in two of 44 vessels, and local microthrombus formation was evident in six of 44 segments. Several clinical case reports also document complications related to snaring of the target coronary artery. Izzat and associates reported the distal embolization of atherosclerotic debris resulting in occlusion of an internal mammary to LAD anastomosis.26 This complication was identified by on-table angiography and required reconstruction of the
Complications Related to Off-Pump Bypass Grafting
anastomosis. Demaria and colleagues described the development of snare related flow limiting lesions both proximally and distally to patent grafts early postoperatively in a diabetic patient.27 These lesions were located at the sites of snare application. They theorize that because patients with diabetes usually have severe and extensive calcification of the arterial media, an increased risk of marked intimal hyperplasia and smooth muscle proliferation may occur in response to plaque rupture or injury. Clearly, the potential for vascular injury exists when circumferential force is applied to susceptible areas. Perrault and associates reported in animals with normal vascular endothelium that occlusive snaring does not cause coronary endothelial dysfunction.28 However, caution must be used when applying this technique to atherosclerotic arteries that may poorly tolerate manipulation. In the series by Hangler, a trend toward reduced injury was identified when occlusion was provided by silicone loops rather than polypropylene sutures. Acceptable visualization should not be achieved at the expense of local coronary injury, which may ultimately cause reduced graft patency.
Conversion to Cardiopulmonary Bypass Inherent in the performance of off-pump bypass grafting is the potential for conversion to CPB for completion of the operation. Despite the growing enthusiasm for off-pump bypass grafting, few authors directly address the clinical situations that lead to the abandonment of a beating heart approach for pump-supported revascularization. An examination of the literature finds a reported incidence of conversion from beating heart surgery to cardiopulmonary bypass ranging anywhere from 1.1-16.3%.29,30 Altering the surgical strategy from off-pump to on-pump grafting during a procedure does not necessarily signify a problem, however, patient morbidity and mortality resulting from the conversion clearly indicates a complication due to the initial surgical approach. Circumstances leading to conversion may logically be grouped into two broad categories: 1) patient specific factors that do not enable the operation to be performed without cardiopulmonary support or 2) patient deterioration requiring conversion for hemodynamic support or salvage (Table 3).
67
Table 3. Circumstances Leading to Conversion from Off-pump to Cardiopulmonary Bypass Deeply intramyocardial vessels Calcified or difusely diseased targets Diminuitive vessels Need for extended endarterectomy of the target vessel Hemodynamic instability Ongoing ischemia Arrhythmias Cardiomegaly
With experience, many patient-related factors that may preclude an off-pump approach can be identified preoperatively. These include intramyocardial vessels, heavily calcified or diffusely diseased targets, diminutive arteries by angiography, and patients hemodynamically unstable before surgery. Additionally, there are rare patients in which adequate positioning of the heart to expose the target vessel may not be able to be achieved. This most frequently occurs with attempts to expose the posterior-lateral wall of the heart. These circumstances of conversion from off-pump grafting to pump supported grafting rarely translate into increased patient mortality. Anyanwu and colleagues described their experience with eight similar conversions of 285 off-pump procedures.31 No increased mortality was identified in this group. Edgerton and associates recently reported a large retrospective review of 1385 patients undergoing beating heart surgery over a 2-year period.32 The authors described a cohort of 48 patients converted from off-pump bypass grafting to bypass with CPB support. The authors classified the patients depending on whether their conversion took place electively versus urgently, and as to whether the conversion was early in the course of the procedure, ie, before the initiation of any anastomoses, or late in the operation. There were no deaths identified in the patients converted early in the procedure regardless of whether this was done electively or urgently. Likewise, Baumgartner and associates reported no mortality in a series of 141 patients with nine discretionary conversions.33 Therefore, it appears reasonable to conclude that patients converted from an off-pump approach to CPB before encountering significant instability demonstrate no increased morbidity or mortality than those patients undergoing conventional procedures.
68
Dewey and Edgerton
Hemodynamic deterioration during off-pump grafting contributes to a small but significant set of patients who are converted from beating heart surgery to CPB. Multiple factors contribute to the downward hemodynamic spiral that ultimately leads to rapid institution of cardiopulmonary support. Significant hemodynamic compromise may result from mechanical alterations of the normal geometry of the heart such as right ventricular compression, right ventricular outflow tract obstruction, induced mitral annular deformation leading to functional mitral stenosis or regurgitation, and impaired left ventricular filling.34 Coexistent ischemia compounds the effects of mechanical dysfunction resulting in the need for emergent institution of CPB. Grafting strategy plays an important role in preventing ischemia induced or potentiated hemodynamic compromise, especially in patients with poor ventricular function or acute ischemia preoperatively. In general, the anterior vessels should be revascularized and perfused first; this strategy facilitates the manipulation of the heart to graft the remaining targets. Additionally, collateralized vessels should be grafted before collateralizing vessels. Once patients become unstable to the extent that pharmacologic therapy and replacing the heart in an anatomic position fails to restore or improve hemodynamics, emergent transition to cardiopulmonary bypass is indicated. CPB is preferred over isolated right or left heart support because of the ability to simultaneously support both ventricles and completely decompress and unload the heart. Grafting can then continue either with the heart beating utilizing off-pump techniques or under cardioplegic arrest. Several series illustrate the increased morbidity and mortality associated with urgent or emergent transition to CPB. Vassiliades and associates recently reported a series of 23 cases converted to CPB after sustaining hemodynamic collapse during elective off-pump bypass grafting.35 The resulting mortality rate was 8.7% versus 1.7% in patients not converted to CPB (P ⫽ 0.08). Additionally, there was a significant increase in the incidence of perioperative myocardial infarction, renal failure, transfusion rate, inotrope use, and requirement for an intraaortic balloon pump to leave the operating room. The converted patients also had a significantly longer length of hospital stay. Soltoski and colleagues compared the tendency toward conversion and
outcomes between two groups of surgeons in their practice with differing levels of experience with off-pump bypass grafting.36 Surgeons comprising group A approached all patients as candidates for off-pump grafting, where as surgeons comprising group B were selective in offering patients beating heart surgery. Surgeons in group A performed 248 off-pump procedures with an 11% conversion rate. Surgeons in group B had a 26% conversion rate in 90 off-pump procedures. An increased mortality of 8.3% was identified for both groups in the converted patients. Patients converted for hemodynamic instability had a 23% complication rate, and a 6.25% mortality rate. A subgroup of patients with graft failure leading to conversion had a mortality rate of 28.5%. Increased surgeon experience with off-pump bypass clearly reduced the incidence of conversion between the two groups, but did not affect the outcome of the patients once conversion took place. Edgerton and associates32 in their series reported a mortality rate of 14% in seven patients with discretionary conversion late in the procedure, but a mortality of 41% in 17 patients converted late in the operation because of hemodynamic instability. Analogous to Soltoski, they examined the influence of surgeon experience on the rate and outcome of conversion. Once again, surgeons with limited off-pump experience had the highest rate of conversion. However, unlike the report form Soltoski, this cohort did not have an increased mortality. This suggests that with experience, surgeons can complete more procedures, but they also may delay conversion longer than inexperienced colleagues thereby encountering increased mortality and morbidity. Edgerton also demonstrated increased complications in the converted group when compared with a computer-matched group of on-pump patients. Converted patients had significantly higher rate of postoperative cardiac arrest, multi-system organ failure, pneumonia, and vascular complications. Identification of patients at increased risk of conversion from hemodynamic collapse has been difficult because of the small numbers of cases reported in the literature. Vassiliades found that patients converted for hemodynamic instability had a higher New York Heart Association score (3.25 versus 2.7, P ⫽ 0.001), an increased incidence of preoperative myocardial infarction (73.9% versus 25.4%, P ⫽ 0.001), more cardio-
Complications Related to Off-Pump Bypass Grafting
megaly (73.4% versus 49.9%, P ⫽ 0.043) and a smaller body surface area (1.9% versus 2.1%, P ⫽ 0.001). Edgerton found by logistic regression analysis that previous bypass grafting, the presence of a preoperative intraaortic balloon pump, and limited surgeon experience were significant risk factors for conversion. However, until larger series are reported, firm conclusions regarding risk of conversion based on preoperative variables cannot be made.
Conclusion The goal of off-pump bypass surgery is to reduce operative morbidity and mortality compared with conventional surgical revascularization. However, surgeons need to be aware of the unique challenges and risks inherent in any new surgical approach. Vascular injuries can be minimized by depressurizing the aorta when placing the partial occluding clamp, and by reduced manipulation and instrumentation of the target vessel. Lowpressure gas insufflation and proximal silicone snares can provide adequate visualization for construction of the anastomosis without increased risk of injury to the coronary artery. The concern over a hypercoagulable state following off-pump surgery requires diligence with antiplatelet inhibition postoperatively, and a low index of suspicion for potential thrombotic events. Impending hemodynamic collapse requires prompt recognition and corrective action to avoid increased mortality and morbidity. Avoidance of potential hemodynamic problems requires 1) careful patient selection, 2) an appropriate grafting sequence based on the coronary anatomy, 3) careful cardiac manipulation and positioning, 4) early recognition of mechanically induced cardiac impairment or ischemia before initiation of an anastomosis, and 5) prompt response to impending collapse. Many situations may be salvaged by the appropriate use of intravascular shunts, cardiac pacing, returning the heart to its anatomic position, or on rare occasions isolated right or left ventricular support. In the future, increasing experience with the management of theses unique patients promises to provide improved strategies to prevent adverse events and thus a safer procedure.
69
References 1. Katz NM, Gersh BJ, Cox JL: Changing practice of coronary bypass surgery and its impact on early risk and long-term survival. [Review] Curr Opin Cardiol 13:465475, 1998 2. Salmenpera MT, Levy JH, Harker LA: Hemostasis and cardiopulmonary bypass, in Cardiopulmonary Bypass: Principles and Techniques of Extracorporeal Circulation. New York, NY, Springer-Verlag, 1995, pp 88-113 3. Butler J, Rocker GM, Westaby S: Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 55:552-559, 1993 4. Edmunds LH: Why cardiopulmonary bypass makes patients sick: strategies to control the blood-synthetic surface interface. Adv Card Surg 6:131-167, 1995 5. Jung G. Razafindranaibe F, Elkouby A, et al: Modification in platelet shape and change in ATP release during CPB. Haemostasis 25:149-157, 1995 6. Wahba A, Rothe G, Schmitz G, et al: Heparin effect on platelets and fibrinolysis. Ann Thorac Surg 61:1590-1591, 1996 7. Bertolino G, Locatelli A, Noris P, et al: Platelet composition and function in patients undergoing CPB for heart surgery. Haemotologica 81:116-120, 1996 8. Tabuchi N, de Haan J, van Oeveren W, et al: Rapid recovery of platelet function after cardiopulmonary bypass. Blood 82:2930-2931, 1993 9. Van Dijk D, Nierich AP, Jansen EWL, et al: Early outcome after off-pump versus on-pump coronary bypass surgery. Circulation 104:1761-1766, 2001 10. Ascione R, Williams S, Lloyd CT, et al: Reduced postoperative blood loss and transfusion requirement after beating heart coronary operations: A prospective randomized study. J Thorac Cardiovasc Surg 121:689-696, 2001 11. Dahl OE: Mechanisms of hypercoagulability. Thromb Haemost 82:9902-9906, 1999 12. Falciani M, Gori AM, Fedi S, et al: Elevated tissue factor and tissue factor pathway inhibitor circulating levels in ischaemic heart disease. Thromb Haemost 79:495-499, 1998 13. Mariani MA, Gu J, Boonstra PW, et al: Procoagulant activity after off-pump coronary operation: Is the current anticoagulation adequate? Ann Thorac Surg 67:13701375, 1999 14. Cartier R, Robitaille D: Thrombotic complications in beating heart operations. J Thorac Cardiovasc Surg 121: 920-922, 2001 15. Puskas JD, Thourani VH, Marshall JJ, et al: Clinical outcomes, angiographic patency, and resource utilization in 200 consecutive off pump coronary bypass patients. Ann Thorac Surg 71:1477-1484, 2001 16. Calafiore AM, Teodori G, Di Giammarco G, et al: Multiple arterial conduits without cardiopulmonary bypass: Early Angiographic results. Ann Thorac Surg 67:450-456, 1999 17. Izzat MB, Khaw KS, Atassi W, et al: Routine intraoperative angiography improves the early patency of coronary grafts performed on the beating heart. Chest 115:987990, 1999
70
Dewey and Edgerton
18. Motwani JG, Topol EL: Aortocoronary saphenous vein graft disease: Pathogenesis, predisposition and prevention. Circulation 97:916-931, 1998 19. Steinhubl SR: Antiplatelet agents in cardiology: The choice of therapy. Ann Thorac Surg 70:S3-8, 2000 20. Blakeman BM, Pifarre R, Sullivan HJ, et al: Perioperative dissection of the ascending aorta: Types of repair. J Card Surg 3:9-14, 1988 21. Still RJ, Hilgenberg AD, Akins CW, et al: Intraoperative aortic dissection. Ann Thorac Surg 53:374-379, 1992 22. Chavanon O, Carrier M, Cartier R, et al: Increased incidence of acute ascending aortic dissection with off pump aortocoronary bypass surgery? Ann Thorac Surg 71:117121, 2001 23. Burfeind WR, Duhaylongsod FG, Annex BH, et al: Highflow gas insufflation to facilitate MIDCABG: Effects on coronary endothelium. Ann Thorac Surg 66:1246-1249, 1998 24. Gundry S, Romano MA, Shattuck OH, et al: Seven-year follow up of coronary artery bypasses performed with and without cardiopulmonary bypass. J Thorac Cardiovasc Surg 115:1273-1278, 1998 25. Hangler HB, Pfaller K, Antretter H, et al: Coronary endothelial injury after local occlusion on the human beating heart. Ann Thorac Surg 71:122-127, 2001 26. Izzat MB, Yim APC, El-Zufari H: Snaring of a coronary artery causing distal atheromas embolization. Ann Thorac Surg 66:1806-1808, 1998 27. Demaria RG, Fortier S, Carrier M, et al: Early multifocal stenosis after coronary artery snaring during off-pump coronary artery bypass in a patient with diabetes. J Thorac Cardiovasc Surg 122:1044-1045, 2001 28. Perrault LP, Menasche´ P, Bidouard JP, et al: Snaring of the target vessels in less invasive bypass operations does
29.
30.
31.
32.
33.
33.
34.
35.
not cause endothelial dysfunction. Ann Thorac Surg 63: 751-755, 1997 Calafiore AM, Mauro MD, Teodori G, et al: Impact of aortic manipulation on incidence of cerebrovascular accidents after surgical myocardial revascularization. Ann Thorac Surg 73:1387-1393, 2002 Novick RJ, Fox SA, Stitts LW, et al: Cumulative sum failure analysis of a policy change from on-pump to offpump coronary artery bypass. Ann Thorac Surg 72:S10161021, 2001 Anyanwu AC, Al-Ruzzeh SA, George SJ, et al: Conversion to off-pump coronary bypass without increased morbidity or change in practice. Ann Thorac Surg 73:798-802, 2002 Edgerton JR, Dewey TM, Herbert M, et al: Conversion in off-pump coronary artery bypass grafting (OPCAB): An analysis of predictors and outcomes. Ann Thorac Surg (in press) Baumgartner FJ, Gheissari A, Capouya ER, et al: Technical aspects of total revascularization in off-pump coronary bypass via sternotomy approach. Ann Thorac Surg 67:1653-1658, 1999 George SJ, Al-Ruzzeh S, Amrani M: Mitral annulus distortion during beating heart surgery: A potential cause for hemodynamic disturbance- A three-dimensional echocardiography reconstruction study. Ann Thorac Surg 73: 1424-1430, 2002 Vassiliades TA, Nielsen JL, Lonquist JL: Hemodynamic collapse during off-pump coronary artery bypass grafting. Ann Thorac Surg 73:1874-1879, 2002 Soltoski P, Salerno T, Levinsky L, et al: Conversion to cardiopulmonary bypass in off-pump coronary artery bypass grafting: Its effect on outcome. J Cardiol Surg 13: 328-334, 1998