Coronary artery bypass grafting in patients with cerebrovascular disease

INTRODUCTION — Cerebrovascular complications are among the most feared consequences after coronary artery bypass graft surgery (CABG). Patients with concomitant cerebrovascular and coronary heart disease represent a subset with advanced atherosclerosis in whom other areas of the arterial system are also involved. In addition to a higher risk of perioperative stroke (see 'Risk factors' below), these patients also have a higher incidence of left main coronary disease and a reduced left ventricular ejection fraction compared with patients who have isolated coronary heart disease [1,2].

This topic will focus mainly on coexistent coronary and extracranial carotid atherosclerosis. Issues that will be discussed include the management of the patient with an asymptomatic carotid stenosis undergoing CABG, the role of combined or staged CABG and carotid revascularization in these patients, and which strategies will result in the lowest operative morbidity and mortality.

The indications for CABG are discussed elsewhere. (See "Revascularization in patients with stable coronary artery disease: Coronary artery bypass graft surgery versus percutaneous coronary intervention" and "Coronary artery bypass graft surgery in patients with acute ST-elevation myocardial infarction", section on 'Indications'.)

STROKE ASSOCIATED WITH CABG — Neurologic complications are among the most feared complications of coronary artery bypass graft surgery (CABG). Information from large databases published before 2002 suggested that a new clinical stroke or transient ischemic attack (TIA) occurred in approximately 3 percent of patients [3,4]. While data from large retrospective reports published in 2008 and 2011 suggested that the overall incidence of perioperative stroke had declined to 1.6 percent [5,6], a 2014 prospective study found a clinically apparent perioperative stroke rate of 3.1 percent [7]. Radiographically evident but clinically silent strokes occur much more frequently [7-9].

Approximately 40 percent of strokes occur intraoperatively and most of the remaining strokes occur during the first 48 hours postoperatively [6]. Perioperative strokes have significant impact on length of hospital stay and mortality outcome, with 10-fold higher hospital mortality rates in patients who suffered a perioperative stroke [5,6]. Other well-recognized neurologic complications of CABG include delirium, seizures, and neurocognitive dysfunction. (See "Neurologic complications of cardiac surgery", section on 'Encephalopathy'.)

Etiology — The mechanisms of stroke in patients undergoing CABG are discussed in detail separately. (See "Neurologic complications of cardiac surgery".)

Summarized briefly, the most common mechanism is embolism, as changes in hemodynamics and aortic manipulation such as cross-clamping, cannulation, and/or proximal graft anastomosis can cause embolization of thrombotic or atheromatous debris from complex plaques in the ascending aorta [5,10,11]. (See "Thromboembolism from aortic plaque", section on 'Cardiovascular procedures' and "Embolism from atherosclerotic plaque: Atheroembolism (cholesterol crystal embolism)".)

Atrial fibrillation is a common arrhythmia following CABG, occurring in 25 to 30 percent of patients, and is a frequent cause of postoperative embolic stroke as well.

Other causes of perioperative stroke include large and small vessel occlusive disease and hypoperfusion. As an example, a stenotic large artery can lead to focal cerebral hypoperfusion, resulting in a watershed or borderzone infarct between two cerebrovascular territories. Perioperative myocardial infarction (MI) and arterial dissection are also potential mechanisms of ischemic stroke. The development of postsurgical systemic inflammatory response and the withholding of antithrombotic therapy in the perioperative period may also contribute to perioperative stroke risk.

In one series of 76 patients with stroke post-CABG, 75 percent were attributed to a cardioembolic source (aortic arch atherosclerosis or atrial fibrillation), 13 percent to small-vessel disease, and 5 to percent large artery stenosis, including carotid artery disease [12].

Risk factors — There are multiple risk factors for perioperative stroke with CABG. The following have been described as independent risk factors in the literature [3,5,13-17]:

●Patient characteristics:

•Moderate to severe atherosclerosis of ascending aorta

•Atrial fibrillation

•Prior stroke or TIA

•Subcortical small vessel disease

•Moderate to severe carotid stenosis

•Peripheral vascular disease

•Diabetes

•Hypertension

•Pulmonary disease

•Heart failure

•Unstable angina

•Recent myocardial infarction

•Moderate to severe left ventricular dysfunction

•Prior cardiac surgery

•Older age

•Female sex

•Elevated pulse pressure

•Tobacco use

•Chronic kidney disease

●Intraoperative features:

•Severe hypotension

•Manipulation of atherosclerotic aorta

•Prolonged cardiopulmonary bypass time

•Use of intra-aortic balloon pump

•Air emboli during cardiopulmonary bypass

•Iatrogenic ascending aortic dissection

●Postoperative features:

•Atrial fibrillation

•Low cardiac output syndrome

The stroke risk associated with aortic atherosclerosis and carotid stenosis is elaborated in the sections that follow.

Aortic atherosclerosis — Atherosclerosis of the ascending aorta may be a more important cause of perioperative stroke than carotid artery stenosis [18]. In a study of over 900 patients undergoing cardiac surgery, the risk of perioperative stroke among patients with and without significant atherosclerosis of the ascending aorta was 9 versus 2 percent, respectively [19].

However, aortic stenosis may be a marker for high atherosclerotic burden and stroke risk and not a direct perioperative stroke mechanism in most patients. Predictors of aortic atherosclerosis are similar to predictors of generalized atherosclerosis and include older age, hypertension, hyperlipidemia, smoking, kidney disease, peripheral artery disease, and/or cerebrovascular disease. Aortic atheromas that are large (≥5 mm thick) or mobile carry a higher risk of stroke [20].

Carotid stenosis — The rate of stroke is elevated in patients with carotid stenosis who have CABG. However, similar to the presence of significant aortic atherosclerosis, carotid stenosis likely is a marker for high atherosclerotic burden and not often a direct perioperative stroke mechanism. The available data, summarized below, suggest that unilateral asymptomatic carotid stenosis of 50 to 99 percent is not an independent risk factor for ipsilateral ischemic stroke with CABG. By contrast, certain groups of patients with carotid artery disease appear to have an increased risk of stroke with CABG, including the following:

●Symptomatic carotid stenosis of 50 to 99 percent in men and 70 to 99 percent in women

●Bilateral asymptomatic stenosis of 80 to 99 percent

●Unilateral asymptomatic stenosis of 70 to 99 percent and contralateral carotid occlusion

However, the quality of the data is generally poor, since most studies are single-center, nonrandomized, and retrospective:

●A 2011 meta-analysis demonstrated that the risk of perioperative stroke after cardiac surgery was approximately 7 percent in those with ≥50 percent carotid stenosis and 9 percent in those with ≥80 percent stenosis, higher than the reported rate of 1.6 to 3 percent in the general population undergoing surgery [21]. The main stroke predictors were symptomatic carotid stenosis and bilateral carotid stenosis/occlusion. Exclusion of patients with prior stroke/TIA or those with complete occlusion of the carotid decreased the stroke risk to approximately 4 percent in the setting of 50 to 99 percent stenosis and 3 percent with 70 to 99 percent stenosis.

●The presence of a recently symptomatic carotid artery stenosis probably increases the risk of a postoperative stroke in patients undergoing CABG, but there are few data directly addressing this question. In one study, 28 patients with prior symptomatic unilateral carotid disease did not undergo prophylactic carotid endarterectomy, and ischemic stroke occurred in 4 (14 percent) [13]. However, only one of the four strokes was attributed to ipsilateral carotid stenosis and was therefore potentially preventable by prophylactic carotid revascularization.

●Asymptomatic carotid stenosis is not a proven independent risk factor for ipsilateral carotid territory ischemic stroke in patients having CABG [12,21]. The following represents a summary of the range of findings that have been noted in individual reports:

•One retrospective study of patients with preoperative carotid duplex ultrasound having CABG compared 117 patients who had severe asymptomatic carotid stenosis (≥75 percent) with 761 patients who did not have severe carotid stenosis [22]. Both groups had similar rates of in-hospital stroke (3.4 versus 3.6) and mortality (3.4 versus 4.2 percent).

•In another retrospective report, there were 18 patients with ≥50 percent carotid stenosis who had a perioperative ischemic stroke, but only 4 occurred in the territory of the stenotic or occluded carotid artery, and in 3 of these, the carotid was totally occluded on preoperative evaluation and not amenable to treatment [12]. Thus, only 1 of the 18 carotid territory strokes was potentially preventable by carotid intervention.

•Since 2005, four studies have reported patients with asymptomatic carotid stenosis of 70 to 99 percent (n = 156) or 50 to 99 percent (n = 42) who did not have prophylactic carotid revascularization; the rate of perioperative stroke with CABG in these patients was 0 percent [12,23-26].

While not high-quality data, these studies suggest that patients with asymptomatic unilateral carotid artery stenoses are at little or no increased risk of perioperative stroke during CABG. As such, while it remains unresolved whether patients with asymptomatic internal carotid artery stenosis would benefit from revascularization, it appears more likely that carotid stenosis is merely a surrogate stroke risk marker associated with multiple other potential stroke risk factors and mechanisms, in which case carotid revascularization would be expected to have little or no benefit.

Although there is little direct evidence, it is plausible that characteristics of the carotid lesion, such as plaque morphology and the presence of downstream microemboli on transcranial Doppler, may impact the risk of perioperative stroke with CABG and therefore inform the need for revascularization [27].

Prevalence and predictors of carotid stenosis — The reported prevalence of carotid artery disease in patients undergoing CABG has varied from 2 to 22 percent, with an average of approximately 8 percent [1,16,28-33]. This wide variation in the reported incidence of carotid artery disease is related to the populations studied, the methods used for screening of carotid disease, how frequently screening was performed, and the definition of a significant carotid stenosis.

The prevalence of carotid artery disease in patients having CABG increases with age. One series found that the prevalence of high-grade (≥75 percent) carotid artery stenosis was three times higher in patients 60 years of age or older compared with younger patients (11.3 versus 3.8 percent) [34]. Clinical predictors other than age for significant carotid artery stenosis in patients considered for CABG include the following [13,30]:

●Diabetes

●Peripheral vascular disease

●Left main coronary artery stenosis ≥60 percent

●Carotid bruit

●Prior stroke or TIA

●Prior vascular operation

●Smoking

●Female sex

PREVENTION OF PERIOPERATIVE STROKE — Strategies for prevention of stroke with coronary artery bypass graft surgery (CABG) include:

●Preoperative evaluation for identification and potential treatment of pre-existing stroke risk factors, including aortic atherosclerosis and carotid stenosis potentially associated with increased surgical risk (ie, symptomatic 50 to 99 percent stenosis, or bilateral asymptomatic stenosis of 80 to 99 percent, or unilateral asymptomatic stenosis of 70 to 99 percent and contralateral carotid occlusion) (see "Neurologic complications of cardiac surgery")

●Medical therapy, including the use of aspirin, antiarrhythmic drugs, and statins (see "Coronary artery bypass surgery: Perioperative medical management")

Identification of aortic atherosclerosis is recommended to reduce the risk of perioperative stroke. Current guidelines endorse evaluation of the aorta by routine intraoperative epiaortic ultrasound to determine the severity of ascending aortic plaque in an effort to reduce perioperative emboli [35-37]. Although there have been no randomized controlled trials, lower perioperative stroke rates (0 to 1.4 percent) were reported when epiaortic ultrasound was utilized to guide surgical decisions, such as determination of sites for aortic cross-clamping and cannula placement [38,39]. Surgical methods that reduce aortic manipulation may decrease the incidence of cerebral embolization with CABG. When performing on-pump CABG surgery, performing the distal and proximal anastomoses under one crossclamp period may decrease the incidence of perioperative strokes. (See "Off-pump and minimally invasive direct coronary artery bypass graft surgery: Clinical use" and "Early noncardiac complications of coronary artery bypass graft surgery", section on 'Improvements in surgical technique'.)

The evaluation for carotid disease and carotid revascularization is discussed below. (See 'Screening for carotid disease' below and 'Prophylactic carotid intervention' below.)

A separate issue is that intracranial atherosclerosis leading to stenosis of the distal intracranial internal carotid artery (ICA) or proximal large vessels of the circle of Willis is a common cause of stroke, particularly in people of Asian and African-American descent (see "Intracranial large artery atherosclerosis: Epidemiology, clinical manifestations, and diagnosis", section on 'Racial and ethnic differences'). In patients with a recent ischemic stroke or transient ischemic attack, aggressive medical management has been shown to be superior to endovascular stenting in patients with symptomatic intracranial 70 to 99 percent stenosis, due to the early risk of stroke following the procedure. (See "Intracranial large artery atherosclerosis: Treatment and prognosis", section on 'Secondary prevention'.)

For long-term stroke risk reduction, all patients with significant cerebrovascular and cardiac disease, regardless of surgical intervention, warrant aggressive medical management for control of vascular risk factors, including use of a statin (with a goal LDL <70 mg/dL) and assiduous blood pressure control. (See "Overview of secondary prevention of ischemic stroke".)

Screening for carotid disease — We agree with current guidelines from the American College of Cardiology Foundation and American Heart Association, which recommend selective pre-CABG screening for carotid artery disease by carotid duplex ultrasound in patients who are ≥65 years old, have left main coronary disease, peripheral vascular disease, history of tobacco use, history of prior stroke/transient ischemic attack (TIA), or a carotid bruit [35-37,40]. The data supporting this recommendation is summarized above. (See 'Prevalence and predictors of carotid stenosis' above.)

One analysis of 1138 patients concluded that screening patients with age ≥65 years, a carotid bruit, or a history of stroke or TIA would reduce the screening burden by 40 percent compared with unselected screening of all patients and would miss only 2 percent of all candidates with a carotid stenosis of ≥70 percent [30].

There are a number of noninvasive methods for establishing the presence and significance of a carotid artery stenosis, including carotid duplex ultrasound scanning (CDUS), magnetic resonance angiography, and computed tomographic angiography. A detailed discussion of these methods is found separately. (See "Evaluation of carotid artery stenosis".)

Prophylactic carotid intervention — In approximate agreement with major guidelines [35,40,41], we suggest carotid revascularization for patients needing CABG who have one of the following conditions (see 'Carotid stenosis' above):

●A recently symptomatic carotid stenosis (50 to 99 percent stenosis in men or 70 to 99 percent stenosis in women)

●Bilateral asymptomatic 80 to 99 percent carotid stenoses

●A unilateral asymptomatic stenosis of 70 to 99 percent and contralateral carotid occlusion

We suggest not performing prophylactic carotid revascularization for patients with isolated unilateral asymptomatic 50 to 99 percent carotid artery stenosis.

However, there are few randomized controlled trials addressing the effectiveness of prophylactic carotid revascularization in patients scheduled for CABG, and despite a large volume of retrospective data, there is no clear consensus on the optimal strategy [23,42,43]. The major considerations include the following issues:

●Patients with asymptomatic unilateral carotid artery stenoses, who account for the majority of patients with carotid disease undergoing cardiac surgery, appear to be at little or no increase in risk of perioperative stroke during CABG (see 'Carotid stenosis' above). Thus, any potential benefit of carotid intervention is likely to be offset by the procedural risks [21,43-45]. A systematic review suggested that at least one-half of perioperative strokes are not preventable by carotid intervention, as 50 percent of patients with stroke did not have significant carotid stenosis, and 60 percent of strokes identified radiographically or by autopsy could not be attributed to carotid disease [46]. A controlled trial randomly assigned 129 patients with asymptomatic high-grade internal carotid artery stenosis who required CABG surgery to synchronous (ie, combined) carotid endarterectomy (CEA) plus CABG or to CABG alone [44]. The study was stopped early due to slow recruitment and withdrawal of funding. At 30 days, the rate of any stroke or death was higher for the group assigned to combined CEA plus CABG group compared with the group assigned to CABG alone (18.5 versus 9.7 percent, 95% CI -3.2 to 20.8 percent) but the difference was not statistically significant. This difference persisted throughout follow-up, with a nonsignificant higher rate of stroke or death at five years for the CEA plus CABG group (40.6 versus 35.0 percent) [45]. All patients were examined by a neurologist, which may explain in part the relatively high overall event rates compared with historical data. Although not definitive, these results suggest that combined CEA plus CABG is not beneficial and may be harmful for patients with asymptomatic carotid stenosis.

●As noted above, the benefit of carotid revascularization with CEA or carotid stenting in patients requiring CABG has not been systematically addressed. In the general population (ie, patients who do not require concomitant cardiac surgery), large randomized clinical trials have demonstrated the benefit of CEA for patients who have asymptomatic carotid stenosis of 60 to 99 percent, provided that their perioperative risk is less than 3 percent and their life expectancy is at least five years. Similarly, major clinical trials have established clear benefit of CEA compared with medical treatment for patients who have symptomatic carotid disease, provided that their perioperative risk is less than 6 percent. (See "Management of asymptomatic extracranial carotid atherosclerotic disease" and "Management of symptomatic carotid atherosclerotic disease".)

One major caveat regarding these data is that optimal medical management of stroke risk factors has evolved since the major trials of CEA were conducted (generally the late 1980s through early 2000s), particularly with regard to use of statin therapy for management of hypercholesterolemia, more aggressive lipid and blood pressure goals, and the availability of newer antiplatelet agents. In reports published since the mid-2000s, rates of stroke in patients with significant carotid stenosis who are managed on medical therapy have declined when compared with those who were enrolled in the earlier surgical trials. Accumulating data suggest that aggressive medical therapy alone may be appropriate treatment for patients with asymptomatic carotid disease (see "Management of asymptomatic extracranial carotid atherosclerotic disease", section on 'Intensive medical therapy and follow-up' and "Management of symptomatic carotid atherosclerotic disease", section on 'Intensive medical management'). Moreover, the trials that established the benefit of CEA excluded patients who had a recent (within six months) history of myocardial infarction or unstable angina. As a result, the applicability of these data to patients undergoing CABG is uncertain. Ongoing randomized trials comparing contemporary revascularization techniques with advanced medical therapy may better inform these issues.

●Existing data are conflicting with regard to the benefit of carotid revascularization in patients with a severe (ie, 80 to 99 percent) carotid stenosis who also require cardiac surgery:

•In a nonrandomized study of patients with 80 to 99 percent carotid stenosis, there were no neurologic events in 53 patients undergoing CEA and CABG, compared with the development of a permanent neurologic defect in 3 of 15 patients undergoing CABG alone [47]. The results were similar in a second nonrandomized study of patients with prior stroke or TIA and 80 to 99 percent carotid stenosis; there were no carotid territory strokes among 114 who underwent staged or combined CEA and CABG, compared with carotid strokes in 4 of 12 patients who had CABG without prophylactic CEA [48].

•By contrast, several subsequent studies published since 2005 have reported no perioperative strokes with CABG among a total of over 200 patients with asymptomatic carotid stenosis of 50 to 99 percent who did not have prophylactic carotid revascularization [12,23-26]. (See 'Carotid stenosis' above.)

CAROTID TREATMENT OPTIONS — Options for patients requiring both carotid and cardiac revascularization include the following treatment decisions:

●Choice of carotid revascularization (ie, carotid endarterectomy [CEA] versus carotid artery stenting [CAS])

●Timing of revascularization (ie, combined carotid revascularization and coronary artery bypass graft surgery [CABG], staged carotid revascularization followed by CABG, or staged CABG prior to carotid revascularization)

These choices are inextricably linked for CAS because the use of dual antiplatelet therapy beginning at the time of stenting and lasting for weeks or months thereafter generally precludes concomitant cardiac surgery, as discussed in the sections that follow.

Highly selected patients undergoing CABG may be candidates for carotid revascularization (see 'Prophylactic carotid intervention' above). Given the available data, summarized in the sections that follow, and our clinical experience, we suggest staged carotid revascularization with CEA or CAS prior to CABG in patients with chronic stable angina in the absence of a recent myocardial infarction, severe left main coronary artery disease, or diffuse coronary heart disease without satisfactory collaterals. We suggest a combined procedure of CEA plus CABG for patients with severe left main coronary heart disease, diffuse coronary heart disease without satisfactory collaterals, or unstable angina. However, the optimal approach to carotid revascularization will be established only by future prospective randomized controlled trials.

Regardless of the choice and timing of revascularization, all patients with concomitant cardiac and carotid atherosclerosis should be treated with aggressive medical management unless contraindicated, including long-term antiplatelet treatment, statin therapy, aggressive blood pressure control, and tobacco cessation. (See "Overview of secondary prevention of ischemic stroke".)

Method of carotid revascularization — Accumulating evidence suggests that CEA and CAS provide similar long-term outcomes for patients with asymptomatic and symptomatic carotid occlusive disease, although the periprocedural risk of stroke and death may be higher with CAS, while the periprocedural risk of myocardial infarction may be higher with CEA. However, these data come from trials that excluded patients requiring cardiac revascularization. (See "Management of asymptomatic extracranial carotid atherosclerotic disease", section on 'Carotid stenting' and "Management of symptomatic carotid atherosclerotic disease", section on 'Patients appropriate for CAS'.)

An important concern in patients also slated to undergo coronary revascularization is that a dual antiplatelet regimen (aspirin plus clopidogrel) is typically recommended for one to three months following stenting to prevent stent thrombosis. As a result, the strategy of CAS immediately prior to CABG is not recommended, since the incidence of perioperative bleeding would be significantly increased. This risk is illustrated by results from the nonrandomized SHARP study of 101 patients who had same-day CAS and CABG [49]. All patients received aspirin beginning at least three days before carotid stenting, and clopidogrel was started within 6 to 10 hours after the end of CABG. At 30 days, major bleeding occurred in four patients (4 percent), a rate we consider four times higher than acceptable or expected for CABG. Furthermore, this major bleeding risk could exceed the risk of stroke and thus outweigh any potential benefit from the carotid procedure.

Thus, patients who need immediate CABG and who require a carotid intervention should have an open CEA rather than a stenting procedure. For patients with significant coronary heart disease, CAS can be considered only if the CABG is not urgent. Such patients could receive a carotid stent with dual antiplatelet therapy for a circumscribed amount of time, followed by CABG surgery.

A systematic review published in 2009 identified 11 studies with 760 patients who had staged CAS followed by CABG, which was done at different time intervals in individual studies, ranging from immediately after stenting to a mean of 70 days after stenting [50]. Eighty-seven percent of the patients had asymptomatic carotid stenosis; all of the studies were retrospective and most were single-center. Combined event rates, from CAS to 30 days after CABG, were as follows:

●Ipsilateral stroke, 3.3 percent

●Any stroke, 4.2 percent

●Myocardial infarction, 1.8 percent

●Death, 5.5 percent

●Death and any stroke, 9.1 percent

The investigators concluded that these event rates for CAS followed by CABG [50] were similar to those reported for patients who had staged CEA and CABG [51].

Timing of revascularization — Operating on the carotid lesion first might increase the risk of myocardial infarction, while operating on the coronary lesion first might increase the risk of perioperative stroke [52]. Furthermore, an additional episode of anesthesia might increase the risk of either of the staged procedures. Advocates of the combined procedure believe that there is more efficient use of the operating room facility and surgical personnel, resulting in shorter hospitalization and lower medical cost compared with staged procedures [53]. One concern with combined CEA and CABG is that the operative morbidity and mortality may be higher than for each procedure alone [54-56].

However, one of the problems in comparing observational series of patients who have carotid artery revascularization and CABG is selection bias; as an example, patients undergoing a combined procedure may have more high-risk features or differ in other important ways from patients who have staged procedures.

Comparative studies — The available evidence regarding the optimal strategy is conflicting, and there are few randomized trials comparing these strategies (ie, combined carotid revascularization and CABG, staged carotid revascularization followed by CABG, or staged CABG prior to carotid revascularization) with each other or with CABG alone.

●A systematic review published in 2005 by the American Academy of Neurology (AAN) identified studies with ≥50 patients that evaluated CEA before or simultaneous with CABG [57]. All were retrospective and observational in nature.

•Nine studies evaluated combined CEA and CABG and included a total of 1923 patients who had combinations of stable and unstable coronary heart disease as well as symptomatic and asymptomatic carotid disease, typically marked by a >70 percent carotid stenosis or by the presence of an ulcerated plaque. Overall average perioperative complication rates were stroke in 3 percent (range 0 to 9 percent), myocardial infarction in 2.2 percent (range 0 to 6 percent), and death in 4.7 percent (range 2.6 to 8.9 percent). Five- to six-year survival among 492 subjects in the three studies that reported long-term survival was 73 to 91 percent.

•One study evaluated CEA before CABG and included 297 patients with stable coronary heart disease. The perioperative complication rates were stroke in 1.9 percent, myocardial infarction in 4.7 percent, and death in 1.6 percent.

Based on these retrospective data, the AAN concluded that perioperative complication rates were probably similar with combined CEA and CABG compared with CEA before CABG.

●A 2013 single-center retrospective analysis evaluated the outcome of 350 patients with severe coronary and carotid stenosis who underwent combined CEA and cardiac surgery (n=195), staged CEA followed by cardiac surgery (n=45), or staged CAS followed by cardiac surgery (n=110) [58]. Most of the cardiac procedures (92 percent) involved CABG, either alone or combined with other cardiac interventions such as valve surgery. Carotid disease was asymptomatic in 81 percent; there was a contralateral carotid stenosis (80 to 99 percent) in 9 percent, and a contralateral carotid occlusion in 13 percent. The following outcomes were reported:

•There was a significantly lower risk of interstage myocardial infarction (MI) for the staged CAS/cardiac surgery group compared with the staged CEA/cardiac surgery group (3 versus 11 percent), although the rates were similar for stroke (1 versus 2 percent) and death (5 versus 7 percent).

•For the composite endpoint of death, MI, and stroke, the staged CAS/cardiac surgery group and the combined CEA/cardiac surgery group had lower rates compared with the staged CEA/cardiac surgery group at 30 days after cardiac surgery (10 versus 10 versus 31 percent, respectively) and at one year (16 versus 17 versus 40 percent, respectively). These results were driven largely by the higher interstage (ie, precardiac surgery) MI event rate for the staged CEA/cardiac surgery group.

•During the period beyond one year after cardiac surgery, the staged CAS/cardiac surgery group had a lower composite event rate compared with the combined CEA/cardiac surgery group and staged CEA/cardiac surgery group (12 versus 39 versus 27 percent, respectively).

These data suggest that staged CAS followed by CABG is the optimal strategy for patients with high-grade carotid stenosis who do not require urgent coronary revascularization, while combined CEA and cardiac surgery is optimal for those who do require urgent coronary intervention. One of the relative strengths of this study is that patients in the staged CAS/cardiac surgery group were evaluated by a neurologist before and after CAS, suggesting that the stroke event rate for this group is reliable; by contrast, patients in the CEA groups were examined by a neurologist on an as-needed basis, as is true of nearly all such studies. However, limitations to this study, particularly the lack of randomization and retrospective methodology, precluded definitive conclusions.

Surgical techniques of combined CEA and CABG — Once the decision has been made to perform a combined CEA and CABG procedure, does it make a difference whether the CEA is done before or during cardiopulmonary bypass? As noted previously (see 'Comparative studies' above), retrospective data suggest that perioperative complication rates are probably similar with combined CEA and CABG compared with CEA before CABG. In addition, performing CEA during cardiopulmonary bypass does not appear to increase the incidence of bleeding or prolong hospital stay [59-61].

Our preference is to perform the CEA before cardiopulmonary bypass. After general anesthesia is induced and the patient is prepared for operation and draped, the CEA team proceeds to expose the carotid artery. Simultaneously, the cardiac team is harvesting the saphenous vein or the radial artery, if such grafts are necessary. The CEA is performed using systemic heparinization and a shunt. Once the CEA is completed, the heparin is not reversed, and the neck wound is packed but not closed. The chest is opened, the internal mammary artery is harvested, and the patient is placed on cardiopulmonary bypass, and the revascularization is completed. After the heparin is reversed, the neck excision is carefully checked for bleeding and the wound is closed.

TIMING OF CABG AFTER STROKE — Patients with a prior history of stroke or transient ischemic attack (TIA) have an increased risk of perioperative stroke with coronary artery bypass graft surgery (CABG) (see 'Risk factors' above), which may be as high as 8.5 percent [46]. The optimal timing of CABG surgery following a stroke has not been systematically addressed and depends upon multiple factors, including how recently the stroke occurred, the size of the stroke, the risk of stroke recurrence (which is in turn dependent on the stroke mechanism), and the urgency of cardiac intervention. A study of 14,030 patients undergoing surgical aortic valve replacement in Denmark, including 616 with prior stroke, reported a high risk of adverse events if the surgery was performed early after stroke, but the risk declined over time and stabilized after three months [62]. Unless urgent cardiac revascularization is necessary, surgery should be delayed pending a thorough work-up for the stroke mechanism, including evaluation for large artery stenosis and any cardioembolic source, to clarify the risk of subsequent stroke and ensure appropriate stroke prevention therapies. (See "Overview of secondary prevention for specific causes of ischemic stroke and transient ischemic attack".)

Optimal timing of cardiac surgery should also include a sufficient delay to allow for poststroke recovery of autoregulatory capabilities of the cerebral vasculature prior to exposure to periprocedural hypotension, and for sufficient remodeling of the damaged parenchyma to decrease the risk of hemorrhagic transformation of the infarct. Deferral of cardiac surgery for several weeks is suggested, although patients with large strokes (ie, those with damage involving the majority of a vascular territory or a sizeable portion of multiple territories) would likely benefit from additional delay.

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: Stroke in adults".)

SUMMARY AND RECOMMENDATIONS

●The main mechanisms of ischemic stroke associated with coronary artery bypass graft surgery (CABG) are atheroembolization, particularly from the aortic arch, and cerebral hypoperfusion, which is less common. (See 'Stroke associated with CABG' above.)

●Atherosclerosis of the ascending aorta may be a more important cause of perioperative stroke than carotid artery stenosis. (See 'Risk factors' above and 'Aortic atherosclerosis' above.)

●The presence of a symptomatic carotid artery stenosis probably increases the risk of a perioperative stroke in patients undergoing CABG. In addition, the presence of bilateral 50 to 99 percent carotid stenoses or total carotid occlusion on one side combined with a 50 to 99 percent carotid stenosis on the other side may be associated with an increased stroke risk. However, patients with an asymptomatic unilateral carotid artery stenosis are likely to have little or no increased risk of carotid-related perioperative stroke during CABG and, therefore, we do not recommend carotid intervention in this population prior to CABG. (See 'Carotid stenosis' above.)

●Strategies for prevention of stroke with CABG include (see 'Prevention of perioperative stroke' above):

•Preoperative evaluation for identification and potential treatment of preexisting stroke risk factors, including aortic atherosclerosis and carotid stenosis

•Medical therapy, including the use of aspirin, antiarrhythmic drugs, and statins, and antihypertensive therapy

●For patients undergoing CABG who have carotid stenosis, we suggest carotid revascularization rather than no carotid intervention for the following subgroups only (Grade 2C) (see 'Prophylactic carotid intervention' above):

•A recently symptomatic carotid stenosis (50 to 99 percent stenosis in men or 70 to 99 percent stenosis in women)

•Bilateral asymptomatic 80 to 99 percent carotid stenoses

•A unilateral asymptomatic stenosis of 70 to 99 percent combined with a contralateral total (100 percent) carotid occlusion

●For patients undergoing CABG who have an isolated unilateral asymptomatic 50 to 99 percent carotid artery stenosis, we suggest not performing prophylactic carotid revascularization (Grade 2C). (See 'Prophylactic carotid intervention' above.)

●For patients undergoing CABG who are selected for carotid revascularization, we suggest a combined procedure with carotid endarterectomy (CEA) plus CABG, rather than a staged procedure, for those who have severe left main coronary artery disease, diffuse coronary heart disease without satisfactory collaterals, or unstable angina (Grade 2C). We suggest a staged carotid revascularization with CEA or carotid artery stenting (CAS) before CABG, rather than a combined procedure, for patients with chronic stable angina in the absence of a recent myocardial infarction (Grade 2C). (See 'Carotid treatment options' above.)

●For patients undergoing CABG who require carotid revascularization, we recommend not performing CAS immediately prior to CABG (Grade 1C). We make this recommendation because dual antiplatelet therapy is required following stenting, a factor that likely increases the perioperative risk of bleeding with the CABG procedure. However, CAS is an alternative to CEA if the CABG is not urgent. Such patients could receive a carotid stent with antiplatelet therapy for several weeks, followed by CABG surgery. (See 'Method of carotid revascularization' above.)

●Timing of cardiac surgery after a stroke should include sufficient delay to allow identification of the cause of stroke, restoration of cerebral autoregulatory mechanisms, and remodeling of the parenchymal damage to minimize the risk of hemorrhagic transformation. Unless emergent cardiac surgery is warranted, we suggest a delay of at least a month and preferably up to three months, particularly for strokes involving larger territories. (See 'Timing of CABG after stroke' above.)