INTRODUCTION — Perioperative stroke is defined as stroke occurring within 30 days following surgery. Although the incidence of clinically recognized perioperative stroke is very low after noncardiac and nonneurologic surgery, it is a potentially devastating postoperative complication and an important cause of perioperative mortality and morbidity [1-3]. This topic will discuss perioperative stroke after noncardiac, noncarotid, nonneurologic surgery, including risk factors, strategies for risk reduction, and what is known about mechanisms of stroke in these settings. In 2021 the American Heart Association and the American Stroke Association developed a joint scientific statement on the cerebrovascular complications of noncardiac surgery, highlighting the importance of perioperative stroke [4,5].
Stroke after cardiac and carotid surgery is discussed separately. (See "Neurologic complications of cardiac surgery", section on 'Cerebrovascular disease' and "Complications of carotid endarterectomy", section on 'Perioperative stroke'.)
Assessment and management of acute stroke are also discussed separately. (See "Overview of the evaluation of stroke" and "Initial assessment and management of acute stroke".)
INCIDENCE AND RISK FACTORS — The overall incidence of clinically recognized stroke after noncardiac, nonneurologic surgery is reported to be between 0.1 and 0.8 percent in registry and database studies [1,3,6-8]. Whereas the incidence of acute myocardial infarction (MI) and death after noncardiac surgery has decreased over time, the incidence of perioperative stroke may be increasing . Perioperative stroke is significantly more common in certain patient populations, and after particular surgical procedures, even within the category of noncardiac, nonneurologic procedures. Patient and surgical risk factors for perioperative stroke are shown in a table (table 1).
The incidence of clinically unrecognized stroke ("covert stroke") after major noncardiac nonneurologic surgery may be as high as 7 percent in patients 65 years and older  and may have implications for long-term outcomes (see 'Outcomes after perioperative stroke' below). In the large studies described below, the incidence of stroke may be underestimated as signs and symptoms of stroke may be difficult to recognize in the postoperative patient because of comorbid factors, sedative medications, and examination by non-neurologists.
Examples of contemporary studies on the incidence of perioperative stroke and risk factors are as follows:
●In a study of data on 523,059 patients who had noncardiac, nonneurologic surgery from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database, the overall incidence of perioperative stroke was 0.1 percent . Independent predictors of stroke in decreasing order of risk included age ≥62 years, history of MI within six months prior to surgery, acute renal failure, history of stroke, dialysis dependence, hypertension, history of transient ischemic attack (TIA), chronic obstructive pulmonary disease, and current tobacco use. Body mass index 35 to 40 kg/m2 was slightly protective, with an odds ratio (OR) of 0.7. The incidence of perioperative stroke ranged from 0 in patients who underwent arthroscopy, to 0.9 percent in patients ≥65 years of age who underwent abdominal exploration. In this study patients with five or more risk factors had a 21-fold increased risk of perioperative stroke with an incidence of 1.9 percent .
●In another study, records were analyzed from the National Inpatient Sample for approximately 372,000 patients who underwent one of three major intraabdominal, orthopedic, and noncardiac thoracic surgical procedures . The overall incidence of perioperative stroke was 0.7 percent after hemicolectomy, 0.2 percent after total hip replacement, and 0.6 percent after lobectomy/segmental lung resection. For patients >65 years of age, the incidences were 1 percent, 0.3 percent, and 0.8 percent, respectively. Identified patient risk factors for perioperative stroke included renal disease, atrial fibrillation, history of stroke, and cardiac valvular disease.
●An analysis of records for 10,581,621 major inpatient noncardiac surgical procedures (including neurosurgery) from the National Inpatient Sample in the United States from 2004 to 2013 found that the rate of acute MI and death declined over the time period studied, whereas the rate of perioperative stroke increased from 0.52 percent to 0.77 percent . The highest risks of stroke were associated with thoracic, vascular, and transplant surgery.
●NeuroVISION was a prospective, international cohort study of over 1100 patients ≥65 years of age who underwent elective major noncardiac, noncarotid surgery . Perioperative covert stroke (defined as acute infarction on brain MRI without clinical diagnosis of stroke before the MRI) occurred in 7 percent of patients (95% CI 6-9 percent), and occurred in all types of surgery.
Patient risk factors — Most of the identified patient risk factors for perioperative stroke include risk factors for cardiovascular disease in general, and many are non-modifiable (table 1).
Age — Advanced age has been identified as a risk factor for perioperative stroke in multiple studies [1-3,6,7,10,11], consistent with the age-associated decreases in cerebrovascular reserve and increases in risk factors for cardiovascular comorbidities. The age at which risk increases varies in the literature, but it is likely to be progressive from age 62 years onwards. In one prospective observational study of patients who underwent noncardiac surgery, compared with patients who were <50 years old, risk of perioperative stroke increased in patients 50 to 69 years (OR 4.7, 95% CI 1.8-12.4), and increased markedly for patients older than 70 years (OR 23.6, 95% CI 9.6-58.1) . In another study of patients who underwent vascular surgery, each year of additional age slightly increased the risk of perioperative stroke (OR 1.02, 95% CI 1.01-1.04) 
Cardiovascular disease — Existing cardiovascular diseases (eg, hypertension, recent MI, atrial fibrillation, congestive heart failure, cerebrovascular disease) are identified risk factors for perioperative stroke [1-3,6,7,10].
The presence of a patent foramen ovale (PFO) is associated with cryptogenic stroke, and may be associated with perioperative stroke [12,13]. PFO may represent a mechanism for stroke after intraoperative venous air embolism, or perioperative thrombosis. In a large retrospective database study, the risk of perioperative stroke in patients with PFO was lower in those who received perioperative anti-thrombotic therapy . The association between PFO and stroke is discussed separately. (See "Atrial septal abnormalities (PFO, ASD, and ASA) and risk of cerebral emboli in adults" and "Patent foramen ovale".)
Carotid stenosis — The risk of perioperative stroke associated with extracranial carotid artery stenosis is unclear, and there are only limited data pertaining to noncardiac surgery. One of the larger studies was a retrospective report of over 2000 patients who had carotid duplex ultrasound performed within six months before and one month after noncardiac, noncarotid surgery . The study found no association between carotid artery stenosis and perioperative stroke, which occurred in 2.6 percent, or 30-day mortality, which affected 5.3 percent. It is likely that the subjects included in this study had carotid duplex studies because of cardiovascular risk factors or neurologic symptoms, suggesting a higher risk of stroke than the general population.
Some patients without a history of carotid stenosis are found to have a carotid bruit on a preoperative physical examination. However, further evaluation of the carotid arteries in a patient with an asymptomatic carotid bruit is not usually warranted prior to surgery; the presence of a carotid bruit by itself does not correlate with the severity of underlying carotid stenosis [15-17], and has not been shown to increase the risk of perioperative stroke [18,19]. As an example, a prospective single center study of 735 patients who underwent general and cardiac surgery found that the incidence of postoperative stroke was similar in patients with and without carotid bruits .
Most of the available literature involves cardiac surgery patients and suggests that the risk of perioperative stroke with cardiac surgery is increased in patients with symptomatic carotid stenosis, and in some patients with high grade asymptomatic carotid stenosis as well. These issues are discussed in detail separately. (See "Coronary artery bypass grafting in patients with cerebrovascular disease", section on 'Carotid stenosis'.)
There is a strong association between carotid artery disease and coronary heart disease; the presence of known carotid artery disease or a carotid bruit  should be a signal that the patient may have coexisting coronary heart disease. (See "Overview of established risk factors for cardiovascular disease", section on 'Noncoronary atherosclerotic disease' and "Management of asymptomatic extracranial carotid atherosclerotic disease", section on 'Risk of stroke and cardiovascular events'.)
The overall risk of stroke associated with carotid artery stenosis is discussed separately. (See "Management of asymptomatic extracranial carotid atherosclerotic disease" and "Management of symptomatic carotid atherosclerotic disease".)
History of prior stroke — A history of stroke has been consistently identified as a risk factor for perioperative stroke [1-3,6,7,21] and for other postoperative major adverse cardiac events . Indeed, history of stroke or TIA is one of the risk factors in the widely used Revised Cardiac Risk Index. (See "Evaluation of cardiac risk prior to noncardiac surgery", section on 'Revised cardiac risk index' and "Evaluation of cardiac risk prior to noncardiac surgery", section on 'Risk assessment'.)
The risk of recurrent stroke unrelated to surgery is high in the first months after a TIA or ischemic stroke  and declines over time, but the increase in risk is significantly greater after surgery . For patients with any prior stroke, observational data suggest that the risk of recurrent ischemic stroke within 30 days of surgery may be approximately 4- to 16-fold higher compared with patients with no prior stroke [6,21]. (See "Definition, etiology, and clinical manifestations of transient ischemic attack", section on 'Risk of recurrent stroke'.)
Timing of elective surgery after stroke is discussed below. (See 'Timing of surgery after ischemic stroke' below.)
Systemic disease — Preexisting renal failure, dependence on dialysis, and diabetes have been associated with an increased risk of perioperative stroke [1,3,7].
Perioperative beta blockers — Preoperative and intraoperative administration of beta blockers may increase the risk of perioperative stroke. The Perioperative Ischemic Evaluation (POISE) trial and subsequent observational studies have found an increased risk of stroke and death associated with acute preoperative administration of beta blockers, specifically metoprolol, and in patients who take metoprolol chronically [23-25]. These results relating to metoprolol should be applied cautiously in clinical practice as they have not been substantiated in prospective trials. (See "Choice of drug therapy in primary (essential) hypertension".)
The etiologic role of beta blockers in perioperative stroke is often conflated with the role of hypotension, since beta blockers are commonly used to lower blood pressure and may result in hypotension. In the POISE trial, hypotension occurred more commonly in patients who were randomized to receive high dose metoprolol; however hypotension was defined as the lowest blood pressure recorded during hospitalization and was not evaluated in a systematic fashion . Subsequent observational studies have found an association between perioperative stroke and administration of metoprolol, independent of hypotension .
Potential mechanisms for the apparent increase in perioperative stroke risk with non-cardioselective beta blockers supported by both human and animal studies include an attenuated beta-2 mediated cerebrovascular dilation with reduced oxygen delivery [25,26], increased blood-brain barrier penetration compared with cardioselective beta blockers , and the attenuated cardiac output response to hemorrhage or anemia .
Because of the risks associated with acute discontinuation of beta-blocker therapy, those patients on chronic beta-blocker therapy (including metoprolol) preoperatively should continue therapy through the perioperative period. The benefits of starting beta blockers preoperatively with the specific intent of reducing cardiovascular risks, however, are unclear, and this practice may increase the risk of major adverse cardiovascular events . Perioperative beta blockers and the related literature, including the POISE trial, are discussed separately. (See "Management of cardiac risk for noncardiac surgery", section on 'Beta blockers'.)
Intraoperative administration of beta blockers is discussed below. (See 'Intraoperative beta blockers' below.)
Others — There are conflicting data on the risk of perioperative stroke associated with sex and gender, tobacco use, and body mass index [1-3,7,30].
The effect of the patient’s biologic sex on the risk of perioperative stroke is unclear; studies have reported higher risk in females [3,7], lower risk in females [1,31], and similar risk in females and males . Conclusions from some studies are limited because they do not clearly define gender versus biologic sex in the patient populations.
The risk of perioperative stroke related to anemia is discussed below. (See 'Perioperative blood transfusion' below.)
Risk associated with the surgical procedure — Some noncardiac, nonneurologic surgical procedures are associated with a higher risk of perioperative stroke than others. In general, major intraabdominal procedures (eg, colectomy, hepatobiliary surgery, abdominal exploration), pulmonary resection, vascular surgery, transplant surgery, and arthroplasty are higher-risk procedures (table 1) [1,8,32].
A number of cases of strokes have been reported in patients who had shoulder surgery in the beach chair position because of the potential for low perfusion pressures to the brain. The incidence of such injuries is unknown. This issue is discussed in detail separately. (See "Patient positioning for surgery and anesthesia in adults", section on 'Physiologic effects of sitting position'.)
The incidence of perioperative stroke is much higher after surgical procedures involving the heart or great vessels, and may be as high as 4 percent after carotid endarterectomy, and as high as 14 percent after some cardiac procedures. (See "Complications of carotid endarterectomy", section on 'Perioperative stroke' and "Neurologic complications of cardiac surgery", section on 'Incidence'.)
Studies of the risk for stroke associated with noncarotid head and neck surgery are conflicting. One retrospective single institution study of 100 patients who underwent major noncarotid head and neck procedures reported a postoperative stroke rate of 4.8 percent . In contrast, larger studies have reported much lower stroke rates [34,35]. In a retrospective cohort study that used registry databases including 14,800 patients who underwent neck dissection over the course of 17 years, the incidence of perioperative stroke was 0.7 percent, similar to the stroke rate for patients who underwent colectomy or thoracic surgery .
Risk prediction tools — Several perioperative stroke risk prediction tools have been suggested, though all have limitations (eg, complexity of use, lack of external validation). We do not use such risk screening tools, and instead make an overall broad assessment of the patient's risk of perioperative stroke (low, moderate, or high) based on the risk factors discussed above.
An observational study of data on over 540,000 patients who had noncardiac surgery between 2009 and 2010 from the US National Surgical Quality Improvement Program database evaluated the ability of existing cardiovascular risk models to assess the risk for perioperative stroke . Notably, this analysis included patients undergoing neurosurgery, who may have distinct risk profiles. Of the six models compared, stroke risk scores derived from either the American College of Surgeons surgical risk calculator (ACS-SRC) or the Myocardial Infarction and Cardiac Arrest (MICA) models had the highest discriminatory ability, with the MICA having the additional benefit of simplicity.
In 2021, the American Heart Association and American Stroke Association suggested using the ACS-SRC for perioperative stroke risk prediction . However, it is not clear how the ACS-SRC would be used, since it does not provide a specific quantitative estimate of stroke risk, it is not clear how the clinician would calculate that risk, and the NSQIP database does not include all relevant variables, such as atrial fibrillation and timing of previous stroke.
A further perioperative stroke risk prediction tool derived from the NSQIP database has been proposed, with limitations similar to other tools . The more comprehensive Stroke After Surgery (STRAS) risk prediction tool derived from and validated with a United States administrative dataset of nearly 250,000 patients in Massachusetts identified 13 variables associated with ischemic stroke within one year after surgery. While promising, this tool requires further external validation.
MECHANISMS OF PERIOPERATIVE STROKE — Perioperative strokes are predominantly ischemic, rather than hemorrhagic [1,3,24,37-40]. These ischemic strokes can be embolic, thrombotic, or hemodynamic (ie, hypoperfusion of vulnerable areas) and may be a combination of these mechanisms (eg, hypoperfusion in the setting of embolic stroke). Ischemic stroke has rarely been reported following anterior cervical spine surgery involving retraction of the carotid artery, possibly due to combination of cerebral hypoperfusion and thrombosis in the setting of pre-existing carotid vascular disease . Aside from patients who have embolic stroke related to atrial fibrillation or fat or air embolism, the mechanisms for postoperative cerebral embolism or thrombosis are largely speculative. Additional mechanisms for perioperative stroke may include hypercoagulability after surgery and surgery-induced inflammatory responses, but at present there is no supporting evidence for these theories. (See "Pathophysiology of ischemic stroke".)
Stroke after noncardiac surgery often presents after a brief postoperative recovery period, and is only infrequently evident on emergence from anesthesia [38,42], which suggests that, in some cases, intraoperative events may be contributory rather than causal for stroke. In a small study of noncardiac and nonneurologic surgical patients only 10 percent of cases of perioperative stroke were recognized in the post-anesthesia care unit . Similarly in a large observational study of noncarotid vascular surgery patients only 15 percent of perioperative stroke was recognized in the first 48 hours after surgery . However, the precise time of stroke onset in the early perioperative period can be difficult to determine, and strokes that occur during surgery or shortly after may escape timely recognition. (See 'Recognition of perioperative stroke' below.)
Role of hypotension — The role of hypotension in the etiology of perioperative stroke is unclear. For noncardiac and nonvascular surgery, intraoperative hypotension may be associated with stroke, but causality has not been established [24,44-46]. Whereas it is commonly assumed that hypotension is the major cause of perioperative stroke, there are little data to support this assumption, partly because perioperative hypotension is endemic and difficult to study. A retrospective analysis of 358,391 noncardiac surgical patients found no significant association between intraoperative hypotension (both relative [mean arterial pressure <30 percent of baseline] and absolute [mean arterial pressure <55 mmHg], stratified by duration) and perioperative stroke within seven days after surgery, after adjusting for preoperative risk factors . Similarly, in a 2018 systematic review of retrospective observational studies of the relationship between hypotension and perioperative organ dysfunction, only insignificant associations were found between hypotension and perioperative stroke, based on four studies . In contrast, a retrospective case-control study of 126 patients with perioperative strokes matched to 500 control subjects found a positive association between cumulative duration of MAP of less than 55 mmHg and perioperative stroke (aOR 1.17, 95% CI 0.10-1.23) .
There is little doubt that a low enough blood pressure or a critically reduced blood pressure for long enough will result in a stroke. However, in individual patients these thresholds can only be determined in retrospect, after ischemia occurs.
●There is no standardized definition of hypotension in the anesthesia literature, and the reported incidence of hypotension ranges from 5 to 99 percent, depending on definition . In the published literature, hypotension is commonly defined as a 20 percent decrease in systolic pressure from a baseline value; a decrease below a systolic of 80 mmHg; or a combination of absolute value and percentage, eg, below 100 mmHg or a 30 percent decrease from baseline, whichever is greater.
●The temporal relationship between low blood pressure and the occurrence of a stroke is rarely known. As an example, perioperative hypotension (defined as the lowest documented blood pressure during hospitalization) was associated with perioperative stroke in the Perioperative Ischemic Evaluation (POISE) trial, but the timing of stroke related to this episode of hypotension was undocumented in this study .
●Cerebral hypoperfusion due to systemic hypotension may coexist with cerebral embolism, leading to impaired clearance of emboli from the microcirculation . Thus, hypotension is likely an important modulator of stroke, ensuring or worsening injury after an embolic or thrombotic event. After acute non-perioperative strokes, reductions in systolic blood pressure below 140 mmHg significantly increase morbidity and mortality, and best outcomes are associated with systolic blood pressure 140 to 180 mmHg [52,53], blood pressure goals that are rarely pursued during or after routine surgery. Hypotension in the postoperative period, when patients are less intensively monitored than they are during surgery, may be precipitated by dehydration or blood loss . While there is a growing body of evidence that hypotension is associated with an increased risk of mortality, renal injury and myocardial injury, evidence regarding a potential association between hypotension and stroke is inconsistent [47-49].
Role of hypocarbia and hypercarbia — The effect of intraoperative arterial and end tidal carbon dioxide on the risk of perioperative stroke is unclear. In theory, hypocarbia could lead to cerebral vasoconstriction and reduced cerebral blood flow, while hypercarbia could lead to cerebral vasodilation and vascular steal from areas with borderline perfusion. A retrospective case-control study (133 stroke cases, 500 controls) found associations between both intraoperative hypocarbia and hypercarbia and increased risk of stroke within 30 days of surgery . Using an area under the curve analysis, end-tidal carbon dioxide levels ≤30 mmHg or ≥45 mmHg were associated with increased risk of stroke of approximately 7 and 10 percent, respectively, per 10 mmHg - minute . However, a causal relationship could not be proven, and further study is required before recommending ventilation strategies to reduce the risk of stroke.
STRATEGIES FOR STROKE RISK REDUCTION — The most important strategies for reducing the risk of perioperative stroke in noncardiac patients at risk for stroke include the following:
●Maintaining blood pressure within 20 percent of baseline (See 'Hemodynamic management' below.)
●Appropriate management of anticoagulation for patients with atrial fibrillation and other conditions that predispose to thromboembolism (See 'Continuation of antiplatelet therapy and anticoagulation therapy' below.)
Timing of surgery after ischemic stroke — We suggest delaying truly elective surgery for at least three months after stroke, and if possible up to nine months, to reduce the risk of recurrent stroke [4,21]. The timing of surgery in patients who have a history of ischemic stroke should incorporate the risk of recurrent stroke and other cardiovascular events, patient factors, and the risks associated with delaying surgery. For patients with a history of ischemic stroke, the significant risks of recurrent perioperative stroke and death should be discussed with the patient and the surgeon, along with the risks associated with delaying surgery. (See 'History of prior stroke' above.)
Support for the suggestion to delay elective surgery after stroke comes from two separate analyses of large administrative health care databases in the United States and Denmark [6,21].
●In the Danish study, there was a 67-fold increase in the risk of perioperative ischemic stroke in patients who had surgery within three months of an ischemic stroke, compared with patients without prior stroke . The incidence of perioperative stroke was 12 percent in patients who had surgery within three months of a stroke. The risk of perioperative stroke decreased over time after stroke and leveled off at nine months, but remained elevated thereafter compared with patients without history of stroke. In response to concerns that these results merely reflected the normal temporal course for risk of recurrent stroke in nonoperative patients, a subsequent analysis of these data was performed . Patients with prior stroke but no surgery had a similar decline in the rate of recurrent stroke over time, but a markedly lower absolute risk of recurrent stroke than patients with prior stroke who underwent surgery. For example, patients who did not undergo surgery had a stroke recurrence rate of 3 to 4 percent at three months compared with 12 percent in surgical patients.
●An analysis of 5,841,539 Medicare patients aged 65 years and older in the United States also demonstrated that ischemic stroke prior to surgery increased the risk of perioperative stroke, with the highest risk within 30 days of surgery . Compared with patients who had never had a stroke, risk of ischemic stroke was increased 8-fold for those with a stroke within 30 days prior to surgery, 6-fold for patients with a stroke 31 to 60 days prior to surgery, and 5-fold in patients who had a stroke from 61 to 90 days prior to surgery. The incidence of stroke was 0.33 percent in patients with no prior stroke, 2.52 percent after surgery within 30 days of stroke, 1.99 percent within 30 to 60 days of stroke, and 1.61 percent within 61 to 90 days of stroke. The odds of stroke were similar in patients who had surgery at 61 to 90 days after surgery compared with the risk at six months and a year after surgery. Thus, in contrast with the analysis of the Danish database described above, elevated risk leveled off between 60 and 90 days after stroke. These results may have differed due to variation in the study cohort characteristics, including that the Medicare database contained only older patients.
One study addressed timing of emergency surgery after stroke. This study analyzed a cohort of patients from the Danish National Patient register for patients with prior ischemic stroke who underwent emergency noncardiac non-neurologic surgery, risk of major cardiovascular adverse events (ie, acute myocardial infarction, stroke, and cardiovascular death; MACE) and all-cause mortality were highest if surgery was in the first three months after stroke, leveled off until nine months after stroke, but remained elevated thereafter . In contrast with the composite outcome of MACE, the risk of ischemic stroke was similar for emergency surgery performed at any time in the first two weeks after stroke. The risk of MACE was lower for emergency surgery performed within three days after stroke compared with days 4 to 14.
Continuation of antiplatelet therapy and anticoagulation therapy — The decision to continue or stop ongoing antiplatelet or anticoagulant medication in the perioperative period represents a balance between reducing the risk of thromboembolism (and ischemic stroke) and increasing the risk of surgical bleeding during and after the planned procedure. Bleeding may be catastrophic during and after intracranial and neuraxial surgery, and discontinuation of antiplatelet and anticoagulant medications is usually warranted for these procedures. For other surgical procedures, medication management must be individualized. There may be a subset of patients at particular risk for perioperative stroke (recent ischemic stroke and/or transient ischemic attack [TIA], or existing intracranial stents) who warrant consideration for ongoing antiplatelet therapy perioperatively when the surgical procedure permits.
An approach to the management of perioperative anticoagulation and antiplatelet therapy, including the decision to interrupt or bridge anticoagulation, is discussed in detail separately. (See "Perioperative management of patients receiving anticoagulants" and "Perioperative medication management", section on 'Medications affecting hemostasis'.)
Statin therapy — Patients who take statin medications should continue them throughout the perioperative period whenever possible. In addition, patients who do not take a statin but are at high risk of cardiovascular disease should start a statin as early as possible before elective surgery, and when possible, before urgent or emergency surgery. The supporting evidence for starting statins preoperatively to reduce the risk of cardiovascular events, including perioperative stroke, is weak, but the risk associated with such therapy is very low. These issues are discussed in detail separately. (See "Management of cardiac risk for noncardiac surgery", section on 'Statins'.)
Intraoperative risk reduction
Choice of anesthetic technique — For most surgical procedures, the choice of general versus regional anesthesia should be based on the surgical procedure, patient comorbidities, and patient choice. There is very little evidence supporting one anesthetic technique over another for noncardiac, noncarotid surgery as a strategy to reduce the risk of stroke. For patients who undergo total hip arthroplasty, several observational studies have reported an association between neuraxial anesthesia and a lower risk of perioperative stroke, though a causal relationship has not been established [56,57]. (See "Anesthesia for total knee arthroplasty", section on 'General versus regional anesthesia'.)
There is no evidence that general anesthesia provides neuroprotection against perioperative stroke in noncardiac, noncarotid surgery patients.
Hemodynamic management — Despite the lack of high-quality evidence regarding the role of hypotension in the etiology of perioperative stroke, intraoperative hypotension should be avoided in patients at risk of stroke (see 'Patient risk factors' above). We suggest maintaining blood pressure within 20 percent of the patient's baseline blood pressure for most patients. The choice of fluids and vasoactive medications used to maintain blood pressure should be based on patient factors.
Management of blood pressure and blood pressure monitoring for patients who have surgery in the beach chair position are discussed separately. (See "Patient positioning for surgery and anesthesia in adults", section on 'Physiologic effects of sitting position'.)
Intraoperative beta blockers — Although patients chronically taking metoprolol should continue it perioperatively, for acute administration during anesthesia we suggest using cardioselective beta blockers (esmolol or labetalol) rather than non-cardioselective beta blockers (metoprolol) when possible. In a retrospective single center review of 55 strokes among over 57,000 patients who underwent noncardiac surgery, intraoperative administration of metoprolol was associated with increased risk of stroke (odds ratio [OR] 3.3, 95% CI 1.4-7.8); no association was found between the use of esmolol or labetalol and perioperative stroke . (See 'Perioperative beta blockers' above.)
Perioperative blood transfusion — The decision to transfuse blood during surgery is complex, and must consider patient factors and ongoing blood loss. There are retrospective data suggesting that patients who take beta blockers may be more sensitive to anemia and possibly at higher risk of perioperative stroke due to the attenuation of cerebral compensatory vasodilation by beta blockers . This appears to be particularly true for metoprolol versus more cardioselective beta blockers. (See "Intraoperative transfusion of blood products in adults" and "Indications and hemoglobin thresholds for red blood cell transfusion in the adult".)
The limited literature on the risks of perioperative stroke associated with anemia and transfusion is retrospective, includes small numbers of events, and is conflicting with respect to suggested management. Database studies cannot separate the relative contributions of anemia and hypotension, both of which often accompany the need to transfuse. The decision to transfuse must balance the effects of anemia against the known risks of blood transfusion.
●In a single center retrospective review of 44,000 patients who underwent noncardiac, nonneurologic surgery including 9800 who were taking beta blockers, hemoglobin <9 g/dL was associated with an increase in the risk of perioperative stroke, with a higher risk in patients who received atenolol or metoprolol compared with bisoprolol . Stroke occurred in 48 of 10,756 (0.44 percent) patients taking beta blockers and 40 of 33,336 (0.12 percent) patients who were not taking beta blockers, but a statistical analysis was not performed. This study suggests that a higher transfusion threshold may be appropriate for high-risk patients who take beta blockers.
●In contrast, a study of records from over 1,500,000 admissions in the Premier Perspective database for patients who underwent noncardiac, non-intracranial, non-major vascular surgery, transfusion of one unit of packed red blood cells during or on the day after surgery was associated with an adjusted OR of 2.33 for the risk of stroke or myocardial infarction (MI) . Similarly, an analysis of records from carotid endarterectomy from the American College of Surgeons National Surgical Quality Improvement Plan (ACS NSQIP) database reported that intraoperative transfusion of one or two units of packed red blood cells was associated with a fivefold increase in the risk of stroke .
Postoperative risk reduction — In the postoperative period, anticoagulation should be resumed as appropriate, taking into account the estimated thromboembolic risk, the risk of bleeding, and the type of anticoagulant. These issues are discussed in detail separately. (See "Perioperative management of patients receiving anticoagulants".)
In patients with risk factors for perioperative stroke (see 'Patient risk factors' above), postoperative blood pressure, similar to intraoperative blood pressure, should be maintained within 20 percent of baseline if possible.
RECOGNITION OF PERIOPERATIVE STROKE — Time is of the essence in recognizing stroke in order to preserve brain tissue and initiate stroke revascularization in appropriate patients. If there is any suspicion of a new neurologic deficit, an urgent neurology consultation should be obtained, and the institution's acute stroke protocol should be activated. Eligibility for acute endovascular intervention for patients who emerge from general anesthesia with new neurologic deficits may extend to 16 to 24 hours after stroke onset, as shown by the DEFUSE 3  and DAWN trials . These trials and eligibility for mechanical thrombectomy are discussed separately. (See "Mechanical thrombectomy for acute ischemic stroke", section on '6 to 24 hours'.)
Recognition of stroke-related neurologic deficits in the postoperative patient is often difficult, complicated initially by residual anesthetic medication, and later by the effects of opioids and other analgesics. Assessment of motor function may be hampered by surgical pain. The evaluation can also be confounded by postoperative delirium and cognitive dysfunction, particularly in older patients. These issues are of particular concern because mental status changes may be the only manifestation of stroke, without any focal deficits .
Acute stroke is in the differential diagnosis of delayed emergence after general anesthesia, and should always be considered while other possible etiologies (table 2) are ruled out (see "Delayed emergence and emergence delirium in adults", section on 'Delayed emergence'). Assessment using stroke screening tools such as the modified National Institutes for Health Stroke Scale early after emergence from anesthesia can be imprecise . Unilateral or focal neurologic deficits should never be considered normal following an anesthetic unless there is a clear surgical etiology.
In the perioperative period we recommend the use of a rapid screening tool for stroke, such as the Face Arm Anesthesia Speech Time (FAAST) scale, during vital sign checks in patients with risk factors for stroke . The FAAST tool evaluates patients with suspected stroke by assessing for the presence of a speech impairment, facial and/or arm weakness, and includes a question about the possibility of residual anesthesia effect and a reminder that time is of the essence particularly for those eligible for mechanical thrombectomy, as follows:
●Face – Uneven smile or facial droop
●Arm – Arm or leg numbness or weakness, not due to surgery (for limb procedures)
●Anesthesia – Residual anesthetic effect
●Speech – Slurring speech or difficulty with speaking or understanding speech
●Time – Get help immediately
Presence of any new focal deficits of the face, arm, or speech despite residual anesthetic effect, should prompt an urgent call for help for stroke management and neuroimaging. This or other available screening tools form the basis for implementation of a "code stroke," protocol whereby nurses can quickly activate a vascular neurology consultation  and consideration for transfer to a center for urgent mechanical thrombectomy, if appropriate. Stroke scales and grading systems are discussed in detail separately. (See "Use and utility of stroke scales and grading systems".)
The majority of noncardiac perioperative strokes occur during the first few postoperative days, though they may occur after the first postoperative week or after hospital discharge . Delayed recognition is common.
MANAGEMENT OF ACUTE PERIOPERATIVE STROKE — Management of acute stroke, including airway stabilization, blood pressure control and management of fluids, blood glucose, temperature are discussed separately. (See "Initial assessment and management of acute stroke".)
All patients with ischemic stroke should be evaluated for possible reperfusion therapy as soon as possible after diagnosis. Thrombolysis will be contraindicated or relatively contraindicated for many postsurgical patients (table 3). Recent intracranial or intraspinal surgery excludes patients from receiving tissue plasminogen activator (tPA, or alteplase). Other major surgery in the previous 14 days represents a relative contraindication; the increased risk of surgical site bleeding with alteplase should be weighed against the anticipated benefits of reduced stroke-related neurologic deficits.
Mechanical thrombectomy, which does not require systemic thrombolysis (eg, intravenous tPA), may be an option for patients with ischemic stroke caused by a proximal large artery occlusion in the anterior circulation who can be treated within 24 hours of the time they were last known to be at their neurologic baseline (algorithm 1). Selection of appropriate candidates for mechanical thrombectomy requires a neurologic evaluation and a neuroimaging study to confirm large vessel occlusion and salvageable brain.
The eligibility criteria for thrombolysis with tPA and for mechanical thrombectomy are discussed separately. (See "Approach to reperfusion therapy for acute ischemic stroke" and "Mechanical thrombectomy for acute ischemic stroke", section on 'Patient selection'.)
OUTCOMES AFTER PERIOPERATIVE STROKE — Outcome after perioperative stroke is often devastating; perioperative stroke is associated with high rates of morbidity and mortality and increased length of hospital stay. Reported 30-day mortality after perioperative stroke related to noncardiac, nonneurologic surgical procedures is as high as 46 percent after general surgery, and as high as 87 percent in patients with a prior stroke [1-3,6,7,37,65].
•In a study of data on 523,059 patients who had noncardiac, non-major vascular, nonneurologic surgery from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database, perioperative stroke was associated with an almost eightfold increase in mortality within 30 days of surgery, compared with patients who had no stroke .
•In another study of approximately 372,000 records from the National Inpatient Sample, in-hospital mortality after stroke occurred in 31 percent of patients who underwent hemicolectomy, 12 percent of patients who underwent total hip arthroplasty, and 33 percent of patients who underwent pulmonary resection . The odds ratios (ORs) for mortality after perioperative stroke, compared with patients without stroke, were 6 (95% CI 5.1-7.0) for hemicolectomy, 21.4 (95% CI 14.3-32.1) for total hip arthroplasty, and 13.3 (95% CI 10.0-17.5) for pulmonary resection. Perioperative stroke was also associated with increased risk of discharge to a location other than home.
•An analysis of data from the ACS NSQIP database for 47,750 patients who underwent noncarotid major vascular surgery found that perioperative stroke was associated with a threefold increase in 30-day all-cause mortality (OR 3.36, 95% CI 1.77-6.36) and an increase in surgical length of stay .
•A study of records for patients who underwent noncarotid vascular surgery in the Veterans Affairs National Surgery Quality Improvement Project and patient care records reported that perioperative stroke was associated with a sevenfold increase in 30-day mortality, and a mean increase in length of stay of five days .
•In an analysis of 100 patients with perioperative stroke identified from 351,531 noncardiac nonvascular nonneurologic surgeries, the following independent risk factors for in-hospital mortality were identified: preoperative atrial fibrillation (adjusted OR [aOR] 9.01, 95% CI 1.40-58.02), disturbance in consciousness as presenting symptom (aOR 5.56, 95% CI 1.52-20.33), no anticoagulant or antiplatelet therapy after stroke (aOR 8.20, 95% CI 1.02-66.06), diuretic treatment (aOR 4.94, 95% CI 1.23-19.82), and pulmonary infection (aOR 6.98, 95% CI 1.85-26.29). 
•In the NeuroVISION cohort study described above, there was a two-fold increase in the risk of cognitive decline (defined as a decrease of ≥2 points on the Montreal Cognitive Assessment [MOCA] from preoperative baseline to one-year follow-up) in patients who had covert stroke after surgery; cognitive decline developed in 42 percent of patients with versus 29 percent of patients without covert stroke (aOR 1.98, 95% CI 1.2-3.2) . Perioperative stroke was also associated with increased risk of postoperative delirium (10 versus 5 percent) and overt stroke or transient ischemic attack (TIA) at one year follow-up (4 versus 1 percent). (See 'Incidence and risk factors' above.)
•Of patients who suffered a nonfatal stroke in the Perioperative Ischemic Evaluation (POISE) trial, most required help with everyday activities or were incapacitated, and more than 50 percent were discharged to a long-term acute care facility .
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
●Incidence and risk factors
•The incidence of stroke after noncardiac, nonneurologic surgery is reported to be between 0.1 and 0.8 percent. Unrecognized stroke may be much more common. (See 'Incidence and risk factors' above.)
•Among noncardiac, nonneurologic surgical procedures, the risk of perioperative stroke is increased after major intraabdominal procedures, major vascular surgery, pulmonary resection, transplant surgery, and arthroplasty. Perioperative stroke has been reported after shoulder surgery in the beach chair position (table 1). (See 'Risk associated with the surgical procedure' above.)
•Patient risk factors include prior stroke or transient ischemic attack (TIA), advanced age, cardiovascular disease, renal disease, diabetes, chronic obstructive pulmonary disease, and female sex, and may include tobacco use, obesity, and carotid stenosis (table 1). (See 'Patient risk factors' above.)
•Preoperative and intraoperative administration of beta blockers, particularly metoprolol, may increase the risk of perioperative stroke. Patients chronically taking metoprolol should continue it perioperatively. (See 'Perioperative beta blockers' above.)
●Stroke risk reduction
•We suggest delaying truly elective surgery for at least three months, and if possible, up to nine months, after ischemic stroke, to reduce the risk of recurrent stroke (Grade 2C). Retrospective data suggest that the risk of perioperative stroke is substantially increased for at least three months after a prior stroke and decreases over time before leveling off between three to nine months later. The timing of surgery after ischemic stroke should be individualized and based on the risk of recurrent stroke and other cardiovascular events, patient factors, and the risks associated with delaying surgery. (See 'History of prior stroke' above and 'Timing of surgery after ischemic stroke' above.)
•In patients at risk for perioperative stroke or with a recent stroke, hypotension should be avoided throughout the perioperative period. We suggest maintaining blood pressure within 20 percent of baseline for patients at risk for perioperative stroke (Grade 2C). (See 'Patient risk factors' above and 'Hemodynamic management' above.)
•The decisions to continue, stop, and restart perioperative anticoagulation and antiplatelet therapy for patients with atrial fibrillation or at risk for thromboembolism must take into account the risk of stroke versus the risk of surgical bleeding. (See 'Continuation of antiplatelet therapy and anticoagulation therapy' above and "Perioperative medication management", section on 'Medications affecting hemostasis'.)
●Diagnosis of postoperative stroke
•Recognition of postoperative neurologic deficits is often difficult; perioperative stroke can present with only impaired consciousness. In the perioperative period we recommend the use of a rapid screening tool for stroke, such as the Face Arm Anesthesia Speech Time (FAAST) scale in patients with risk factors for stroke (See 'Recognition of perioperative stroke' above.).
•Unilateral or focal deficits are never normal during emergence unless explained by the surgical procedure. If there is any suspicion of neurologic deficit, urgent neurology consultation and neuroimaging should be obtained.
•Acute stroke is in the differential diagnosis of delayed emergence after general anesthesia and should always be considered while other possible etiologies are ruled out.
●Management – All patients with ischemic stroke should be evaluated for reperfusion therapy. Some postoperative patients with acute ischemic stroke may be candidates for reperfusion therapy including mechanical thrombectomy. (See 'Recognition of perioperative stroke' above and 'Management of acute perioperative stroke' above.)
●Prognosis – Perioperative stroke is associated with high rates of morbidity and mortality. (See 'Outcomes after perioperative stroke' above.)