INTRODUCTION — The treatment of coronary heart disease has evolved significantly over the past several years due in part to improvement in both surgical and percutaneous revascularization techniques. The majority of patients with chronic stable angina are still treated with medical therapy; however, revascularization with either coronary artery bypass graft surgery (CABG) or percutaneous coronary intervention should be considered in several subgroups. (See "Chronic coronary syndrome: Indications for revascularization".)
The major complications associated with CABG are death, myocardial infarction, stroke, wound infection, prolonged requirement for mechanical ventilation, acute kidney injury, and bleeding requiring transfusion or reoperation [1-4]. The short-term, particularly perioperative, noncardiac complications that can occur following conventional CABG (using cardiopulmonary bypass) will be reviewed here. Cardiac complications and perioperative mortality after CABG are discussed separately. (See "Early cardiac complications of coronary artery bypass graft surgery" and "Operative mortality after coronary artery bypass graft surgery".)
Off-pump and minimally invasive CABG are discussed separately. (See "Minimally invasive coronary artery bypass graft surgery: Definitions and technical issues" and "Off-pump and minimally invasive direct coronary artery bypass graft surgery: Clinical use".)
The risk and timing of surgery after COVID-19 infection are discussed separately. (See "COVID-19: Perioperative risk assessment and anesthetic considerations, including airway management and infection control", section on 'Risk of surgery with COVID-19' and "COVID-19: Perioperative risk assessment and anesthetic considerations, including airway management and infection control", section on 'Timing of surgery after COVID-19 infection'.)
PREVENTION OF COMPLICATIONS
Mechanisms — Many complications related to traditional techniques of cardiac surgery are primarily the result of cardiopulmonary bypass (CPB). An important factor is aortic instrumentation and manipulation, including cannulation, decannulation, and partial or complete clamping and unclamping, which can result in embolization of atherosclerotic debris. Technical errors in bypass graft construction that can lead to graft occlusion, primarily in saphenous vein grafts, also may be important [5,6]. (See "Thromboembolism from aortic plaque".)
Other factors that contribute to complications include:
●Global cardiac arrest
●Hypothermia
●Nonpulsatile bypass and artificial perfusion
●An intense "inflammatory" response to perfusion with artificial (nonendothelialized) surfaces
●The reintroduction of fat and particulate debris as well as procoagulant and proinflammatory factors from the pericardial surgical field into the systemic circulation via the use of cardiotomy (field) suction [7,8].
●The sternotomy and skin incision
●Conduit harvest
Minimal extracorporeal CPB — Off-pump coronary artery bypass graft surgery (OP CABG) procedures are considered to be less invasive than full CPB. Data from randomized trials and observational studies are conflicting as to whether or not there is an improvement in clinical outcomes such as mortality and stroke with OP CABG. (See "Off-pump and minimally invasive direct coronary artery bypass graft surgery: Clinical use", section on 'Off-pump CABG'.)
CPB systems designed to decrease the rates of the complications discussed above (particularly stroke) have also been evaluated. One such device, the mini-extracorporeal circulation (MECC), is a heparin bonded CPB circuit that requires a relatively small prime for the centrifugal pump, and does not include either an open venous reservoir (which minimizes blood-air contact) or direct reinfusion of cardiotomy suction from the pericardial space. This design has the potential to limit the systemic inflammatory response seen with traditional CPB.
In an initial prospective trial, 300 CABG patients were randomly assigned to MECC or OP CABG [9]. There were no differences between the two groups in operative mortality and morbidity, transfusion rates, hospital stay, or serum levels of creatine kinase and inflammatory markers including interleukin-6. In a study of 40 consecutive patients undergoing CABG who were randomly assigned to either MECC or standard CPB, the former was shown to be a safe alternative [10].
Improvements in surgical technique — Improvements in surgical technique have led to a steady reduction in morbidity after CABG, despite the fact that patients currently undergoing CABG have higher-risk profiles. This was illustrated in a report that compared 5051 patients undergoing CABG from 1986 to 1988 with 2793 patients undergoing CABG from 1993 to 1994 [1]. Although the patients in the latter period were at higher cardiovascular risk, the risk-adjusted morbidity rate decreased from 14.5 to 8.8 percent with no change in risk-adjusted hospital mortality (2.8 versus 2.9 percent). A later report that evaluated over 37,000 patients undergoing CABG for multivessel disease in New York state between 1997 and 2000 found an in-hospital mortality of only 1.8 percent [11].
The following are some of the improvements in technique that have reduced postoperative complications:
●Complications related to aortic manipulation can be limited by routine, careful evaluation of the ascending aorta using intraoperative transesophageal and epiaortic echocardiography to identify mobile and intramural atheromatous disease, and to help define safe areas for cannulation. Aortic manipulation also can be decreased using a "single-clamp" technique. By avoiding the application of an additional partial occlusion clamp to construct proximal anastomoses, neurologic complications can be reduced, although not eliminated [12].
●Advances have been made in the composition of cardioplegia fluid (eg, blood versus crystalloid) as well as substrate enhancement with aspartate and glutamate and superoxide radical scavengers [13]. These advances, used in conjunction with integration of different routes of cardioplegia administration (antegrade, retrograde, and down the newly constructed grafts) to effect homogeneous delivery to all parts of the heart beyond coronary artery blockages, have resulted in more optimal myocardial protection. As a result, the heart can be safely arrested for as long as two hours with minimal cardiac dysfunction, allowing precise repair of complex problems [14].
●Normothermic or near normothermic systemic perfusion is now routinely used. Avoiding systemic hypothermia decreases coagulation abnormalities and organ dysfunction.
●Many of the complications of CABG are related to the biologic response of the body to artificial perfusion and gas exchange through the nonendothelialized CPB circuit. Within seconds of CPB, formed and unformed blood elements come into contact with the large surface area of the CPB circuit. Despite anticoagulation with heparin, this interaction results in extensive activation of platelets, neutrophils, complement, cytokines, and the fibrinolytic system, producing a complex and intense "inflammatory" response. Although these responses are usually short-lived and leave no residual deficits, they can lead to long-lasting cardiac, pulmonary, renal, and neurologic dysfunction in a subset of patients.
●Advances in perfusion technology and techniques and research in biomaterial sciences have resulted in the application of more biocompatible CPB circuits, such as heparin-bonded circuits. The use of these circuits has reduced the need for homologous transfusion and decreased neutrophil and complement activation, resulting in reductions in thromboembolic complications (including neurologic dysfunction), myocardial and pulmonary dysfunction, and cost [15-20]. Elimination of cardiotomy suction is an important adjunct by minimizing inflammation, thrombin and platelet activation, and the increase in markers of neuronal injury [8].
●Use of leukoreduced blood and leukoreduction of autologous saved blood prior to reinfusion can reduce perioperative morbidity, particularly when more than three blood transfusions are required [21]. (See "Practical aspects of red blood cell transfusion in adults: Storage, processing, modifications, and infusion", section on 'Pre-storage leukoreduction' and "Surgical blood conservation: Blood salvage".)
●Additional pharmacologic interventions and perfusion surface modifications to further attenuate platelet, neutrophil, and complement activation and cytokine release are being actively investigated. As an example, direct C5 inhibition with a recombinant, humanized antibody during CPB blunts complement activation, proinflammatory byproducts, and leukocyte activation, and may result in decreased myocardial injury, blood loss, and new cognitive deficits [22]. (See "Regulators and receptors of the complement system".)
●Technical errors in bypass graft construction can lead to graft occlusion [5,6]. Such patients should undergo emergency surgical re-exploration/revision or coronary angiography. Hybrid operating room catheterization laboratories have been used to perform "completion angiography" of bypass grafts, and fluorescent angiographic techniques can be performed without angiography.
In a report of a novel technique, intraoperative angiography using fluorescent indocyanine green dye resulted in graft revision (all but one saphenous vein grafts) for technical problems in 4.2 percent of patients that would otherwise have gone unrecognized [5]. The test took 2.2 minutes to perform. Further studies are required.
Medical therapy — The 2004 American College of Cardiology/American Heart Association guidelines on bypass surgery issued general recommendations for preventive measures to minimize the risk of some of these complications [23]:
●Prophylactic antimicrobials to prevent surgical site infection
●Strict glycemic control (using an insulin infusion) during the perioperative period, which also may reduce sternal wound infection.
●An antifibrinolytic agent is used in most CABG procedures to minimize bleeding
●Beta blockers, aspirin, and statins are indicated in many patients
The evidence supporting the efficacy of these therapies is presented separately. (See "Coronary artery bypass surgery: Perioperative medical management" and "Atrial fibrillation and flutter after cardiac surgery", section on 'Prevention of atrial fibrillation and complications' and 'Sternal wound infection and mediastinitis' below and 'Antifibrinolytic agents' below.)
In addition to these medical therapies, carotid duplex ultrasound may be performed to identify severe carotid stenosis in patients with an audible bruit, severe peripheral artery disease, or a previous stroke or transient ischemic attack. (See "Neurologic complications of cardiac surgery", section on 'Risk factors'.)
Minimally invasive CABG — The preceding discussion applied to conventional CABG using CPB. OP CABG has been evaluated as a minimally invasive approach. Clinical trials comparing the two approaches have suggested that OP CABG may reduce morbidity but there is concern about long-term graft patency. These issues, as well as patient selection for OP CABG, are discussed elsewhere. (See "Off-pump and minimally invasive direct coronary artery bypass graft surgery: Clinical use", section on 'Off-pump CABG'.)
BLEEDING — Approximately 30 percent of patients require a blood transfusion after coronary artery bypass graft surgery (CABG) [24]. In addition, bleeding requiring reoperation is associated with large increases in transfusion requirements and intensive care unit and hospital stays. Rates of reoperation have ranged from 4 to 6 percent [25], although one study found that, during 1995 to 1997, there was a reduction in the rate of reoperation to 2 percent [26]. Although reoperation is usually performed in the operating room, reoperation in the intensive care unit is feasible [27].
Risk factors — In order to preoperatively identify patients at high risk for requiring blood, one study of 1007 patients undergoing a first CABG developed a prediction rule based upon data from two-thirds of the sample and prospectively applied it to the remaining one-third [24]. Independent factors predicting the need of transfusion were lower preoperative hemoglobin, lower weight, older age, and female gender.
Additional risk factors for perioperative transfusion include preoperative use of antiplatelet or antithrombotic drugs, reoperation, acquired or congenital clotting/coagulation abnormalities, complex procedures, and emergency operations [28].
Antiplatelet agents — Antiplatelet agents have a variable effect on bleeding risk.
●Aspirin – Evidence suggests that late (less than five days prior to surgery) discontinuation of aspirin did not decrease adverse cardiovascular events but did increase the risk of perioperative blood transfusion [29]. It is generally recommended that aspirin be started in the early postoperative period to reduce both mortality and morbidity [30]. However, with increasing experience, some surgeons recommend continuation of aspirin throughout the preoperative period. The possible risk of aspirin therapy in patients undergoing CABG, including recommendations for the timing of aspirin therapy, is discussed separately. (See "Coronary artery bypass surgery: Perioperative medical management", section on 'Aspirin' and "Perioperative medication management", section on 'Aspirin'.)
●Clopidogrel – Clopidogrel therapy within five days of CABG is associated with an increased bleeding risk. Two reports comparing the bleeding risk with or without recent clopidogrel use in (mostly) stable patients undergoing CABG are available [31,32]. The larger of the two is a retrospective observational study of 1572 consecutive patients who underwent nonemergent off-pump CABG; 281 (18 percent) had either received a 300 mg loading dose before PCI or had been on oral clopidogrel within seven days of surgery [32]. Patients with, compared to those without, recent clopidogrel use had a significantly increased need for packed red blood transfusion and of reoperation for bleeding (6.4 versus 1.4 percent, odds ratio 5.1 on multivariable analysis). There was no difference in operative mortality between the two groups (1.4 percent). (See "Off-pump and minimally invasive direct coronary artery bypass graft surgery: Clinical use".)
However, among those patients who receive clopidogrel within five to seven days of surgery, the risk of bleeding is more likely related to the residual antiplatelet effect of the drug than the number of days since the last dose. In a prospective study of 100 patients who received clopidogrel within five days of off pump CABG, blood loss (and transfusion requirement) was significantly greater in patients in the third tertile of the percentage of platelet inhibitory response to clopidogrel (greatest residual clopidogrel effect), as assessed by thromboelastography, compared to those in the first and second tertiles (914 versus 623 and 683 mL, respectively) [33]. In multivariate analysis, discontinuation date did not significantly predict either the degree of blood loss or transfusion requirement.
●Patients taking clopidogrel referred for CABG have generally been placed on it either for prevention of cardiovascular events after an acute coronary syndrome or for the prevention of stent thrombosis after stent placement. In these settings, discontinuation of clopidogrel prior to the recommended duration places the patient at increased risk. Thus, the timing of discontinuation of clopidogrel requires consideration of both the benefit (less bleeding) and risk (increased rate of ischemic events) of its discontinuation. The 2004 American College of Cardiology/American Heart Association guidelines on CABG and the 2008 American College of Chest Physicians guidelines on the primary and secondary prevention of coronary heart disease recommend five (preferably seven) days off clopidogrel before CABG [34,35]. (See "Long-term antiplatelet therapy after coronary artery stenting in stable patients".)
The decision whether to perform CABG within five days of clopidogrel use in an individual patient requires an analysis of the risks and benefits. Local institutional and surgeon preferences need to be taken into account.
●Potent P2Y12 receptor blockers – Bleeding at the time of CABG is also a concern for patients receiving prasugrel and ticagrelor, both of which are potent P2Y12 receptor blockers. In the major randomized trials comparing these agents to clopidogrel, the rates of major bleeding were similar to or higher with the more potent agent [36,37].
●Glycoprotein (GP) IIb/IIIa inhibitors – The effect of GP IIb/IIIa antagonists on bleeding risk is less clear. One report evaluated surgical outcomes in 85 patients in the EPILOG and EPISTENT trials who required emergency CABG after abciximab therapy for percutaneous coronary intervention [38]. These patients, compared to those not treated with abciximab, had no significant increase in major blood loss or red cell transfusion and no reduction in the use of optimal internal mammary artery grafts. However, abciximab-treated patients were more likely to require surgical re-exploration for bleeding (12 versus 3 percent).
Stopping and restarting P2Y12 receptor blocker — The decision as to when to discontinue platelet P2Y12 receptor blocker therapy prior to CABG must balance the perioperative risks of bleeding (discussed above) while on and acute ischemic events while off dual antiplatelet therapy.
The possible role of bridging therapy with cangrelor, an intravenous P2Y12 receptor blocker with a half-life of three to six minutes, was evaluated in the BRIDGE trial [39]. In BRIDGE, 210 patients awaiting CABG with an acute coronary syndrome or treated with a coronary stent and receiving either clopidogrel or prasugrel were randomly assigned to either cangrelor or placebo after discontinuation of the thienopyridine. Study drug was continued for at least 48 hours and was then discontinued one to six hours before CABG. As expected, cangrelor, compared to placebo, was associated with a significantly lower level of platelet reactivity (platelet reactivity units <240, 98.8 versus 19.0 percent). However, it was not associated with any significant difference in CABG surgery-related bleeding (11.8 versus 10. 4 percent, respectively). As BRIDGE did not test the hypothesis that bridging therapy would improve outcomes compared to either early discontinuation or continuation of P2Y12 receptor blocker therapy, we do not recommend such an approach.
There is no good evidence upon which a recommendation can be made as to when to restart P2Y12 receptor blocker therapy after CABG. Our reviewers generally wait until the risk of major bleeding has significantly fallen, perhaps 12 to 24 hours after uncomplicated surgery.
Prevention — The use of intraoperative activated clotting time-guided heparin dosing is appropriate in patients undergoing CABG [38]. Although some have suggested benefit from the selective use of platelet transfusions [40], a prospective report of over 5000 patients undergoing CABG noted increases in mortality and ischemic complications with platelet transfusions [30]. The latter observation and the benefit of postoperative aspirin therapy are consistent with a central role for platelet activation in the ischemic response to reperfusion injury.
Similar to recommendations made in the 2014 clinical practice guideline from the American Association of Blood Banks, we do not recommend prophylactic platelet transfusion for patients without thrombocytopenia who are scheduled to undergo cardiopulmonary bypass [41]. Patients with perioperative bleeding and thrombocytopenia are candidates for platelet transfusion.
Antifibrinolytic agents — The ability of the antifibrinolytic agents (aminocaproic acid, tranexamic acid, and aprotinin) to prevent bleeding following both on- and off-pump CABG is well established [42-46]. We use either aminocaproic acid or tranexamic acid in most cases. The relative efficacy and safety of these three antifibrinolytic drugs were addressed in a meta-analysis of 128 mostly small randomized controlled trials (published prior to August 2006) that compared the efficacy and safety of the three agents to placebo and to each other in patients undergoing CABG [42]. Compared to placebo, all agents were effective at reducing blood loss by 226 to 348 mL and at lowering the proportion of patients transfused with packed red blood cells. High-dose aprotinin was more effective at reducing blood loss than either aminocaproic acid or tranexamic acid. The meta-analysis found no statistically important worsening of the outcomes of mortality, stroke, myocardial infarction, or dialysis-dependent renal failure.
Tranexamic acid was evaluated in the more contemporary ATACAS trial [46]. (See "Coronary artery bypass surgery: Perioperative medical management", section on 'Aspirin'.) In ATACAS, 4631 patients undergoing coronary artery surgery, with or without valve surgery, were randomly assigned to tranexamic acid or placebo. There was no difference in the rate of the primary outcome, a composite of death and thrombotic complications (myocardial infarction, stroke, pulmonary embolism, renal failure, or bowel infarction [16.7 versus 18.1 percent; relative risk 0.92. 95% CI 0.81-1.05]) within 30 days after surgery. However, the use of the antifibrinolytic agent significantly decreased the total number of units of blood products that were transfused (4331 versus 7994). Major hemorrhage or cardiac tamponade leading to reoperation occurred in 1.4 and 2.8 percent (p = 0.001) of the two groups and seizures occurred in 0.7 and 0.1 percent (p = 0.002) of the two groups, respectively. At one year, there was no significant difference in the rate of death or disability, the primary one-year outcome, or the composite rate of myocardial infarction, stroke, and death [47].
Safety concerns exist with the use of aprotinin. In the meta-analysis discussed above, high-dose aprotinin significantly increased the risk of temporary kidney injury compared with the other agents, defined as an increase in serum creatinine (relative risk 1.47, 95% CI 1.12-1.94). An increased likelihood of acute kidney injury with aprotinin was also seen in observational studies [48-50]. A randomized trial and two large observational studies published after the meta-analysis provided evidence for an increase in both short- and long-term mortality in patients who received aprotinin compared to aminocaproic acid, tranexamic acid, or placebo [51-53]. The BART trial (Blood conservation using Antifibrinolytics in a Randomized Trial) was designed to further evaluate the safety of aprotinin [53]. After enrollment of 2331 of 3000 patients scheduled to be randomly assigned to either aprotinin, aminocaproic acid, or tranexamic acid, the trial was terminated early because of a significantly higher death rate in patients receiving aprotinin. Similar findings were noted in a retrospective analysis of data from over 33,000 aprotinin recipients and over 44,000 aminocaproic acid recipients in which the unadjusted risk of death within the first seven days after CABG was 4.5 and 2.5 percent, respectively [51]. After adjustment for 41 patient and hospital characteristics, the relative risk of death was significantly increased in the aprotinin group (relative risk 1.64, 95% CI 1.50-1.78). Another retrospective analysis found that the mortality risk with aprotinin remained significantly increased at one year [52]. Aprotinin is not available for routine use in the United States is used in Europe.
Glucocorticoids — A 2008 meta-analysis of 11 trials (436 patients) that evaluated the effects of prophylactic glucocorticoids found that their use resulted in a small but significant decrease in the rate of postoperative bleeding [54]. However, we do not recommend prophylactic glucocorticoid use for this purpose, as the results of the individual trials in the meta-analysis were heterogeneous, and only four studies reported transfusion rates.
Fresh frozen plasma — A 2015 meta-analysis of 15 randomized trials (755 patients) that principally compared the prophylactic use of fresh frozen plasma (FFP) with no FFP found no evidence to support its routine use to prevent blood loss [55].
Blood transfusion — For most patients with anemia during or after cardiac surgery, we transfuse red blood cells (RBCs) to maintain the hemoglobin level above 8 g/dL (hematocrit >24 percent). However, this threshold for transfusion should be individualized and should take into account clinical factors such as patient age, whether the person is entering the diuretic phase after surgery, and the presence and rate of active bleeding.
While anemia is an independent risk factor for morbidity and mortality after cardiac surgery, RBC transfusion used for the purpose of correcting anemia has been associated with increased rates of infection, ischemic complications, and death in observational studies [56,57]. In addition, RBC transfusion leads to increased financial cost, the consumption of limited resources, and risks such as infection or transfusion reaction. (See "Immunologic transfusion reactions" and "Transfusion-transmitted bacterial infection" and "The approach to the patient who declines blood transfusion" and "Hemolytic transfusion reactions" and "Transfusion-related acute lung injury (TRALI)" and "Blood donor screening: Medical history", section on 'Screening for infectious risks'.)
Thus, the issue of the optimal transfusion strategy is important for a variety of reasons, particularly as rates of transfusion between 25 and 75 percent (and as low as 8 or as high as 93 percent) have been reported [56,58-60]. Three randomized trials with more than 500 patients undergoing cardiac surgery have addressed the issue of whether a restrictive perioperative RBC transfusion strategy is as safe as liberal strategy:
●In the TITRe2 trial, 2007 patients who underwent cardiac surgery (60 percent CABG or CABG/valve) and had a postoperative hemoglobin level of <9 g/dL were randomly assigned to a restrictive transfusion threshold (hemoglobin <7.5 g/dL) or a liberal threshold (<9 g/dL) [59]. The rate of transfusion after randomization was 53.4 and 92.2 percent and the median number of units transfused was one and two, respectively. There was no difference in the rate of the primary outcome of serious infection or an ischemic event (permanent stroke, myocardial infarction, gut infarction, or acute kidney injury) within three months between two groups, respectively (35.1 and 33.0 percent, respectively; odds ratio 1.11, 95% CI 0.91-1.34). All-cause mortality at 90 days, a secondary outcome, was observed more frequently in the restrictive group (4.2 versus 2.6 percent; hazard ratio 1.64, 95% CI 1.00-2.67).
●In the TRACS non-inferiority trial, 502 patients who underwent cardiac surgery were randomly assigned to a restrictive (hematocrit ≥24 percent) or a liberal (hematocrit ≥30 percent) strategy of blood transfusion [56]. Patients were transfused, at any time from the start of surgery until discharge, if the hematocrit was less than 24 percent in the restrictive group or less than 30 percent in the liberal group. The rates of transfusion were 47 and 78 percent in the two groups, respectively. There was no difference in the rate of the primary composite end point of all-cause mortality and severe morbidity (cardiogenic shock, acute respiratory distress syndrome, or acute renal injury requiring dialysis or hemofiltration) at 30 days (11 versus 10 percent). Hemoglobin concentrations were 9.1 and 10.5 g/dL. As pointed out by the authors, the absence of blood leukodepletion in this study may limit its generalizability. (See 'Leukoreduced blood' below.)
●In the TRICS III trial, 5243 high-risk (of death) adults undergoing cardiac surgery with cardiopulmonary bypass (45 percent CABG or CABG/valve) with a EuroSCORE I value of 6 or more (table 1) were randomly assigned to a restrictive red cell transfusion threshold (transfuse if hemoglobin <7.5 g/dL deciliter) or a liberal red cell transfusion threshold (transfuse if hemoglobin <9.5 g/dL in the operating room or intensive care unit [ICU] or <8.5 in the non-ICU ward) [60]. The following findings were noted:
●At day 28 or hospital discharge, whichever came first, the primary composite outcome (death from any cause, myocardial infarction, stroke, or new-onset renal failure with dialysis) occurred at a similar rate in both groups (11.4 versus 12.5 percent, respectively; absolute risk difference -1.11 percentage points, 95% CI -2.93 to 0.72; odds ratio 0.90, 95% CI 0.76-1.07). There was no significant difference in mortality (3.0 and 3.6 percent, respectively) and red cell transfusion occurred in 52.3 and 72.6 percent of the two groups, respectively.
●At six months, the primary composite outcome occurred at a similar rate in both groups (17.4 versus 17.1 percent, respectively; absolute risk difference before rounding, 0.22 percentage points, 95% CI -1.95 to 2.39; odds ratio 1.02, 95% CI 0.87-1.18) [61]. Mortality was similar in the two groups.
Leukoreduced blood — Leukoreduced blood may reduce complications in patients who require blood transfusion. The vast majority of blood used in the United States undergoes prestorage leukoreduction. (See "Practical aspects of red blood cell transfusion in adults: Storage, processing, modifications, and infusion", section on 'Pre-storage leukoreduction'.)
Donor leukocytes are associated with increased complications including alloimmunization, possible transmission of cytomegalovirus, febrile nonhemolytic transfusion reactions, and possibly impaired wound healing and postoperative infections. (See "Immunologic transfusion reactions", section on 'Febrile nonhemolytic transfusion reactions'.)
In one trial of 944 patients undergoing cardiac surgery, those receiving leukoreduced blood had a reduced incidence of infection and a nonsignificant reduction in mortality at 60 days (3.3 to 3.6 versus 7.8 percent; p = 0.15) [21]. This was particularly evident when more than three units of blood were required.
Duration of red cell storage and outcome — While longer RBC storage times are associated with structural and functional changes, several randomized trials have not demonstrated adverse outcomes from older (longer storage duration) versus younger (shorter storage duration) blood, including in patients undergoing cardiac surgery. This is discussed in detail separately. (See "Practical aspects of red blood cell transfusion in adults: Storage, processing, modifications, and infusion", section on 'RBC age/storage duration effect on clinical outcomes'.)
NEUROLOGIC COMPLICATIONS — Neurologic complications are an important source of morbidity and mortality after coronary artery bypass graft surgery (CABG). The major neurologic problems are stroke, neuropsychiatric abnormalities such as cognitive dysfunction, and peripheral neuropathy. The risk increases with patient age (figure 1) [62].
The mechanisms, risk factors, and possible therapies and preventive measures for neurologic complications after CABG, and issues related to CABG in patients with known carotid artery disease are discussed in detail elsewhere. (See "Neurologic complications of cardiac surgery" and "Coronary artery bypass grafting in patients with cerebrovascular disease".)
INFECTION
Sternal wound infection and mediastinitis — Issues related to postoperative sternal wound infection with mediastinitis are discussed in detail separately. The following discussion will be limited to issues related to coronary artery bypass graft surgery (CABG). (See "Postoperative mediastinitis after cardiac surgery" and "Surgical management of sternal wound complications".)
Mediastinitis after CABG has been reported in 0.9 to 1.3 percent of patients [63-66]. It is usually detected within the first two weeks (median about seven days), but the onset is delayed for more than one month in occasional patients. Virtually all patients have fever, tachycardia, chest pain or sternal instability, signs of sternal wound infection, and purulent discharge from the mediastinal area. Fever and systemic symptoms appear first in most patients. Streptococcus and Staphylococcus are the most frequent organisms.
A number of risk factors have been identified for the development of mediastinitis after CABG, although the same risk factors were not noted in all studies. (See "Surgical management of sternal wound complications", section on 'Risk factors for sternal wound complications'.)
These include:
●Obesity [64,66,67]
●Diabetes mellitus [63,64,68,69]
●Bilateral internal mammary artery grafts [63,64], a relationship that has not been confirmed in some studies [66,70].
●Prolonged duration of surgery [64,66]
●Prior cardiac surgery [66]
●Use of staples for skin closure [68]
●Underlying obstructive airways disease [67]
●Dual antiplatelet therapy with aspirin and clopidogrel (compared with aspirin alone) within five days of surgery [71].
Among patients with diabetes, strict perioperative glycemic control appears to reduce the risk of sternal wound infection [69].
Mediastinitis after CABG is associated with increases in both short- and long-term mortality. This was illustrated in a report of 6459 consecutive patients, 83 (1.3 percent) of whom developed mediastinitis [66]. The patients with mediastinitis compared to those without this complication had increased mortality at 90 days (11.8 versus 5.5 percent); the interval mortality remained high between one and two years after surgery (8.1 versus 2.3 percent).
The prevention and management of mediastinitis is discussed elsewhere. (See "Postoperative mediastinitis after cardiac surgery" and "Surgical management of sternal wound complications".)
Leg wound complications — The reported incidence of leg wound complications after saphenous vein graft (SVG) harvesting has varied widely, ranging from around 1 to 24 percent in older reports [72-77] and 18 percent in a 2004 report [78]. The most common manifestations are usually minor and do not require surgical intervention. These include dermatitis, cellulitis, greater saphenous neuropathy, chronic nonhealing ulcers, and lymphocele.
Major complications are rare. In a retrospective review of 3525 CABG procedures, lower extremity wound complications occurred in 4 percent of patients; however, only 0.65 percent required additional surgical intervention, including wound debridement, skin grafting, vascular procedure, amputation, or fasciotomy [74]. Significant predictors of major wound complications were female gender, peripheral artery disease, and postoperative intraaortic balloon pump.
For patients in whom a conventional (open) harvesting has been used, the wound may be closed with either staples or sutures. A 2010 Cochrane systemic review (updated in 2012) of three trials including over 300 patients found no significant difference in the rates of leg wound infection (10.8 versus 8.0 percent, respectively) or dehiscence (9.3 versus 8.8 percent) between these two techniques [79]. However, all included studies were felt to be of sub-optimal methodological quality and at risk of bias.
Endoscopic SVG harvesting was developed in the mid-1990s as a way to improve postoperative discomfort and to potentially reduce wound infection. It is estimated that nearly 90 percent of all procedures performed in the United States use the endoscopic technique [80]. Most studies have shown that the rate of leg wound complications is significantly reduced [75-77,81]. In REGROUP, a randomized trial comparing open to endoscopic vein-graft harvesting, there was a trend toward a higher rate of leg wound complications in the former group (3.1 versus 1.4 percent; relative risk 2.26, 95% CI 0.99-5.15) [82]. Other outcomes with these two approaches are discussed separately. (See "Coronary artery bypass graft surgery: Long-term clinical outcomes", section on 'Endoscopic vein-graft harvesting'.)
Post-venectomy cellulitis — The syndrome of post-venectomy lower extremity cellulitis can be a late complication of coronary artery bypass grafting, but is infrequent, particularly with the increased use of minimally invasive vein harvesting. If it occurs, it tends to present months to years after saphenous venectomy and recurrent episodes occur in most patients. Classically, patients present with the acute onset of high fever, often greater than 40ºC, systemic toxicity, and erythema and swelling of the lower extremity [83]. Erythema of the lower extremity usually begins along the medial aspect of the mid-tibial region at the saphenous venectomy site (picture 1). It then spreads posteriorly and anteriorly and can involve the dorsal aspect of the ankle and foot, and can extend proximally to involve the medial thigh. Many patients have maceration and erythema of interdigital spaces indicative of tinea pedis infection (athlete's foot), which may represent the source of the infection. Deep vein thrombosis and infections of underlying medical devices (eg, a prosthetic vascular graft) are the two major differential diagnoses of post-venectomy cellulitis.
The diagnosis is made clinically, based on the presence of lower extremity erythema and fever. Cultures of blood, skin biopsies, and lesion aspirates usually do not identify a pathogen. Toe web cultures may be useful in establishing a probable pathogen in patients with active cellulitis and tinea pedis. When pathogens are recovered, non-group A beta-hemolytic streptococci have been the most commonly isolated [84]. Other beta-hemolytic streptococci and Staphylococcus aureus (including methicillin-resistant S. aureus) are also possible pathogens. In the acute setting, antimicrobial therapy is directed against these organisms and usually results in the prompt resolution of systemic complaints and slower resolution of the lower extremity changes. Patients who are hospitalized for this syndrome should initially receive antibiotics intravenously. For patients with recurrent cellulitis, chronic suppressive antibiotic therapy and pre-emptive, patient-initiated therapy are options to prevent or manage additional episodes of cellulitis [85]. Treatment of tinea pedis, if present, in an effort to decrease the risk of recurrent episodes, is also thought to be an important element in the prevention of recurrent lower extremity cellulitis. These treatment issues are the same as those for cellulitis in general and are discussed in detail elsewhere. (See "Acute cellulitis and erysipelas in adults: Treatment" and "Suspected Staphylococcus aureus and streptococcal skin and soft tissue infections in children >28 days: Evaluation and management".)
Bloodstream infection — Bloodstream infection (BSI), defined as ≥1 positive blood culture for a known pathogen with additional criteria for positive cultures caused by potential skin contaminants, occurred in 3 percent of patients within 90 days of undergoing CABG [86]. Individuals with BSI had a significantly increased risk of death (HR 4.2), and the risk was highest among those with BSI due to gram-negative bacteria or Staphylococcus aureus.
ACUTE KIDNEY INJURY
Pathogenesis — Acute kidney injury (AKI, previously called acute renal failure) is a potential complication of coronary artery bypass graft surgery (CABG) that can arise from a variety of causes, including intraoperative hypotension, postoperative cardiac complications that impair renal perfusion, hemolysis, atheroemboli, and exposure to contrast media [87,88]. Reduced renal function due to transient hypoperfusion or contrast nephropathy usually resolves within a few days, but some patients develop more severe and persistent kidney injury with a requirement for dialysis. (See "Pathogenesis and etiology of ischemic acute tubular necrosis", section on 'Pathology and pathogenesis' and "Contrast-associated and contrast-induced acute kidney injury: Clinical features, diagnosis, and management".)
Incidence — One problem with the available data on the incidence of AKI after CABG is the variable definitions used [89]. The incidence is higher with smaller compared to larger reductions in estimated glomerular filtration rate (eg, 25 percent increase in serum creatinine compared to a 100 percent increase or the requirement for dialysis). In two studies of 843 and 649 patients undergoing cardiac surgery (mostly CABG), the incidence of AKI (defined as a rise in the serum creatinine of only 25 percent) was 17 and 24 percent [90,91]. Using this definition, a patient whose serum creatinine rose from 1.0 to 1.3 mg/dL (88 to 115 mmol/L) would be considered to have AKI.
Other contemporary studies that used a more restrictive definition noted a much lower rate of AKI. Two large series (one involving over 51,000 CABG procedures and a 2006 report from the Society of Thoracic Surgery [STS]) defined AKI as either an increase of serum creatinine to >2 mg/dL (177 mmol/L) with a minimum doubling of the preoperative value or a new requirement for dialysis [92,93]. The overall rate of AKI ranged from 3.6 to 5 percent and did not vary over time. In the STS report, a higher rate of 7.5 and 12.9 percent was noted when CABG was combined with aortic or mitral valve replacement, respectively [93].
A lower proportion of patients (0.9 to 1.7 percent in different large series) develops AKI severe enough to require dialysis [94-97] and a reduced estimated creatinine clearance is a major risk factor for the development of AKI requiring dialysis [94-96]. The magnitude of this effect was illustrated in a study using the Society of Thoracic Surgeons National Adult Cardiac Database of over 480,000 patients who underwent isolated CABG from 2000 to 2003 [94]. The rate of requirement of new dialysis was 0.2 percent in the 23 percent of patients with normal baseline renal function (estimated creatinine clearance ≥90 mL/min) compared to 0.5, 1.8, and 10.9 percent in patients with an estimated creatinine clearance of 60 to 89, 30 to 59, and less than 30 mL/min per 1.73 m2 (adjusted odds ratio 1.7, 4.7, and 20.4, respectively). (See "Assessment of kidney function", section on 'Assessment of GFR'.)
A study of over 2400 patients undergoing CABG after catheterization study found that the risk of AKI was highest in those undergoing preoperative coronary angiography within one day (24 percent) of CABG surgery compared to those having the test performed at least five days prior to surgery (15.8 percent) [98].
Risk factors — In addition to baseline kidney dysfunction, preoperative risk factors for AKI include New York Heart Association functional class IV (table 2), valve surgery, peripheral artery disease, emergent or urgent surgery, obesity, and the need for preoperative intraaortic balloon pump [95,96,99,100]. Perioperative factors such as anemia, red blood cell transfusions, prolonged cardiopulmonary bypass, and surgical reexploration are also associated with the development of AKI [99]. On the other hand, atherosclerotic renal artery stenosis does not appear to be a risk factor for a postoperative decrease in the glomerular filtration rate, need for renal replacement therapy, longer length of stay, or long-term mortality [101].
Multivariable risk models to predict cardiac surgery-associated AKI using either presurgical [96,102] or presurgical and intrasurgical clinical information have been developed [103]. We do not routinely use these risk prediction models.
Mortality — There is a strong relationship between the development of perioperative AKI and both short- and long-term mortality after CABG. This issue is discussed in detail elsewhere. (See "Operative mortality after coronary artery bypass graft surgery", section on 'Acute kidney injury'.)
Prevention — The methods that may prevent AKI, including the possible benefit with an off-pump approach in patients undergoing CABG, are discussed separately. (See "Possible prevention and therapy of ischemic acute tubular necrosis", section on 'Prevention'.)
OTHER COMPLICATIONS
Venous thromboembolism — Deep venous thrombosis (DVT) and pulmonary embolism (PE) may be difficult to recognize after coronary artery bypass graft surgery (CABG) and are therefore likely to be underdiagnosed [104]. In a series of 330 patients who underwent venous ultrasonography four to six days after CABG, DVT was present in 20 percent, and two patients had symptomatic PE [105]. Most such patients remain asymptomatic. This was illustrated in a review of hospital discharge records for 66,180 patients undergoing CABG in which only 736 (1.1 percent) were diagnosed with symptomatic DVT or PE within three months of surgery [106]. (See "Clinical presentation, evaluation, and diagnosis of the nonpregnant adult with suspected acute pulmonary embolism" and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)
Prophylactic measures, such as graduated compression stockings and anticoagulation, can reduce the incidence of venous thromboembolism after surgical procedures. Recommendations for thromboembolism prophylaxis are presented in detail separately. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)
Pulmonary — Pulmonary complications after cardiac surgery are discussed elsewhere. (See "Strategies to reduce postoperative pulmonary complications in adults" and "Postoperative complications among patients undergoing cardiac surgery", section on 'Pulmonary dysfunction'.)
Pleural effusions are common postoperatively occurring in up to 90 percent of patients who have undergone CABG. The effusions are usually small, left-sided, and do not require treatment. Early effusions (within 30 days of CABG) tend to be bloody, while late effusions are yellow exudates [107]. (See "Evaluation and management of pleural effusions following cardiac surgery".)
Aortic dissection — Ascending aortic dissection is a rare complication of CABG, occurring with both conventional on-pump CABG and, perhaps more often, with minimally invasive off-pump CABG (OP CABG) [108-111]. In a review from a single institution, ascending aortic dissection occurred in 1 of 2723 patients (0.04 percent) treated with conventional CABG and 3 of 308 undergoing OP CABG (1 percent) [110]. (See "Minimally invasive coronary artery bypass graft surgery: Definitions and technical issues".)
Dissection can occur intraoperatively or weeks to months after surgery. The site of aortic dissection is usually related to the partial occlusion clamp site, proximal vein graft anastomosis, and the site of cardioplegia delivery. Patients at increased risk are older adults and those with long-standing hypertension, severe atherosclerotic involvement of the ascending aorta, or ectasia of aorta. (See "Clinical features and diagnosis of acute aortic dissection".)
Protamine reactions — Protamine is administered intravenously to reverse the effect of heparin but may be associated with severe systemic reactions, including hypotension that may require inotropic support, an increase in pulmonary artery pressure, noncardiogenic pulmonary edema, and bronchospasm. In a series of 2069 patients, 2.6 percent had an adverse reaction within 10 minutes of protamine administration [112]. Independent risk factors for an adverse event were NPH insulin use, fish allergy, and a history of nonprotamine medication allergy.
Thrombocytopenia — Thrombocytopenia is common after CABG. The causes and management of this problem are presented elsewhere. (See "Management of heparin-induced thrombocytopenia (HIT) during cardiac or vascular surgery" and "Diagnostic approach to the adult with unexplained thrombocytopenia".)
Gastrointestinal — Single center reports have found serious gastrointestinal (GI) complications in 0.3 to 3 percent of patients undergoing cardiac surgery, primarily CABG [113,114]. Mortality in these patients has been as high as 33 percent.
The largest reported experience comes from the United States Agency for Healthcare Research and Quality, which assessed the incidence and impact of GI complications after 2.7 million CABG operations performed from 1998 to 2002 [115]. The most frequently reported complications were abscess, ileus, gastrointestinal ulcer (perforation/bleeding), and bleeding diverticulosis or diverticulitis in the colon. The following observations were noted:
●The incidence of GI complications was 4.1 percent. The largest increase in relative risk was seen in patients over age 65 and those on hemodialysis. In contrast, use of an internal mammary graft was associated with lower risk.
●The inpatient mortality was significantly increased in those with GI complications compared to those without (12.0 versus 2.5 percent). (See "Operative mortality after coronary artery bypass graft surgery".)
EARLY READMISSION — Rates of readmission within 30 days varied between 8.3 to 21.1 percent following coronary artery bypass graft surgery in an analysis of the New York State Registry. The most common reasons were postoperative infection (16.9 percent), heart failure (12.8 percent), and "other complications" (9.8 percent). The authors found that the following variables were independently associated with increased readmission risk: older age, female sex, African-American race, higher body mass index, comorbidities, renal failure, unplanned cardiac reoperation, United States Medicare or Medicaid status, discharge to a skilled nursing facility, saphenous vein grafts, and longer length of stay. Readmission was not found to correlate with mortality [116]. Other studies report that rates of readmission range between 13 to 16 percent at four to six weeks [117-120].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Coronary artery bypass graft surgery".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)
●Beyond the Basics topics (see "Patient education: Coronary artery bypass graft surgery (Beyond the Basics)" and "Patient education: Recovery after coronary artery bypass graft surgery (CABG) (Beyond the Basics)")
●Basics topic (see "Patient education: Coronary artery bypass graft surgery (The Basics)")
SUMMARY
●Many of the noncardiac complications related to coronary artery bypass graft (CABG) surgery are the result of either cardiopulmonary bypass or aortic instrumentation and manipulation. Preventative strategies are available. (See 'Prevention of complications' above.)
●Bleeding, neurologic complications, infection, and acute kidney injury are four of the most important adverse outcomes after CABG.
•Approximately 30 percent of patients require a blood transfusion after CABG. Risk factors include lower preoperative hemoglobin, lower weight, older age, and female gender. Additional risk factors for perioperative transfusion include preoperative use of antiplatelet or antithrombotic drugs, reoperation, acquired or congenital clotting/coagulation abnormalities, complex procedures, and emergency operations. (See 'Bleeding' above.)
•The major neurologic problems are stroke, neuropsychiatric abnormalities such as cognitive dysfunction, and peripheral neuropathy. (See 'Neurologic complications' above.)
•Sternal wound infection and mediastinitis after CABG occur in approximately 1 percent of patients. (See 'Sternal wound infection and mediastinitis' above.)
•Acute kidney injury is a potential complication of CABG that can arise from a variety of causes, including intraoperative hypotension, postoperative cardiac complications that impair renal perfusion, hemolysis, atheroemboli, and exposure to contrast media. Approximately 1 to 2 percent of patients require dialysis. (See 'Acute kidney injury' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff thank Dr. Julian M. Aroesty for his past contributions as an author to prior versions of this topic review.
