Anesthesia for endovascular therapy for acute ischemic stroke in adults

INTRODUCTION — For patients with either anterior or posterior circulation acute ischemic stroke (AIS), timely restoration of cerebral blood flow using reperfusion therapy is the most effective maneuver for salvaging ischemic brain tissue that is not already infarcted. Intravenous (IV) tissue plasminogen activator (tPA) is first line thrombolytic therapy for eligible patients with AIS, including those with AIS affecting the posterior circulation; the most commonly used tPA preparation is alteplase. Mechanical thrombectomy is indicated for patients with acute ischemic stroke due to a large artery occlusion in the anterior or posterior circulation who can be treated within 24 hours of the time last known to be well, whether or not they receive IV tPA for the same ischemic stroke event. Patients may also receive directed intra-arterial administration of thrombolytics with or without mechanical thrombectomy. Hence, we use the term "endovascular therapy," to indicate mechanical thrombectomy and/or directed intra-arterial administration of thrombolytics. (See "Approach to reperfusion therapy for acute ischemic stroke", section on 'Intravenous thrombolysis'.)

Endovascular therapy (EVT) can be performed with general anesthesia (GA) or with monitored anesthesia care (MAC). For EVT, MAC involves local anesthesia for vascular access, with or without sedation. In the authors' opinion, all patients who undergo EVT should be cared for by anesthesia clinicians, who are capable of managing the airway and rapidly inducing GA if necessary. This is supported in a retrospective review [1].

This topic will discuss anesthetic management for EVT for ischemic stroke, including the choice of GA versus MAC. Initial management of AIS, the choice of reperfusion therapy for AIS, and epidemiology of stroke are discussed separately. (See "Initial assessment and management of acute stroke" and "Approach to reperfusion therapy for acute ischemic stroke" and "Mechanical thrombectomy for acute ischemic stroke" and "Stroke: Etiology, classification, and epidemiology".)

PREANESTHESIA EVALUATION — Preanesthesia evaluation should be focused and concise to minimize delays in care. Relevant medical history will have been obtained by the admitting service and should be available to the anesthesia team; time should not be spent duplicating questions. Electronic records including medications, laboratory results (including blood glucose), electrocardiogram, and imaging results can often be reviewed while waiting for the patient to arrive in the interventional suite. The initial evaluation of patients with acute stroke is discussed separately. (See "Initial assessment and management of acute stroke", section on 'Initial assessment'.)

Whenever possible, a personal and family history of problems with anesthesia should be elicited, and in all cases an airway assessment should be rapidly performed. (See "Airway management for induction of general anesthesia", section on 'Airway assessment'.)

Preoperative assessment and preparation should be performed in parallel with other caregivers. Preanesthesia concerns specific to patients who undergo EVT include the following:

●Neurologic status – The patient's prior and current neurologic status, including level of consciousness and focal deficits, and course of evolution of deficits, are important considerations when deciding whether to perform general anesthesia (GA) or monitored anesthesia care (MAC), and may affect the choice of anesthetic agents. Importantly, patients should be quickly assessed for the likely ability to lie flat and still protect their airway and to follow commands during the procedure with a light level of sedation. (See 'Choice of anesthetic technique: General anesthesia versus monitored anesthesia care' below.)

●Cardiovascular comorbidity – Many patients who undergo EVT for stroke have multiple cardiovascular risk factors that may affect anesthetic care, including hypertension, diabetes, smoking, peripheral vascular disease, and atrial fibrillation. In addition, stroke itself can cause cardiac complications, including myocardial infarction, heart failure, neurogenic cardiac damage, arrhythmias, and cardiac arrest. (See "Overview of secondary prevention of ischemic stroke", section on 'Major risk factors' and "Complications of stroke: An overview", section on 'Cardiac complications'.)

Anesthetic considerations for patients with these conditions are discussed separately in multiple topics on those conditions.

●Potential cervical spine injury – If there is a suspicion of fall related to a stroke, cervical spine precautions should be considered, particularly for older patients and those with altered level of consciousness. Such precautions are especially important during airway management for GA and are discussed separately. (See "Cervical spinal column injuries in adults: Evaluation and initial management" and "Anesthesia for adults with acute spinal cord injury", section on 'Airway management'.)

STEPS OF THE PROCEDURE — EVT is usually performed by inserting a catheter via the femoral artery, though occasionally, radial, brachial, axillary or carotid artery may be used if necessary. Cannulation of the artery for access is not painful after local anesthesia is infiltrated. However, intracranial catheter manipulation can cause pain, and can result in patient agitation or movement [2,3].

The following are the usual steps for EVT procedures.

●The skin over the vascular access site (usually right groin or radial artery) is prepared with antiseptic solution, drapes are positioned to create a sterile field, and local anesthetic is infiltrated in the skin over the access site.

●The artery is cannulated, an introducer is placed, and the catheter is passed through the major arteries into extra- and intracranial arteries. If an arterial catheter has not yet been placed, the side port can be used to transduce arterial pressure, though the waveform may not be optimal. (See 'Monitoring' below.)

●An angiogram is performed to identify site of arterial blockage. A brief period of apnea is typically requested during angiography to minimize movement (ie, ventilator pause during general anesthesia [GA], breath hold in cooperative patients during monitored anesthesia care [MAC]).

●The catheter is advanced into the intracranial arterial vasculature. For mechanical thrombectomy, retrieval catheters are advanced and clot is removed. Patient movement should be avoided during clot retrieval. Mechanical thrombectomy can be performed with or without directed intra-arterial administration of thrombolytics. Alternatively, the proceduralist may administer directed intra-arterial thrombolytics without mechanical thrombectomy.

●After successful clot retrieval or dissolution, another angiogram is performed to assess vessel recanalization.

●Catheters are removed and pressure is applied to the vascular access site. The patient is kept flat (or reverse Trendelenburg) for up to 24 hours, with the cannulated leg straight and immobilized for four to six hours.

●Follow-up head computed tomography (CT) scan is performed to rule out intracranial hemorrhage, after which the patient is transported to the neurointensive care or stroke unit. Patients who have GA are extubated after the CT scan, or later in the neurointensive care unit, at the discretion of the anesthesia team. (See 'Emergence and extubation' below.)

MONITORING — Standard American Society of Anesthesiologists (ASA) monitors (ie, electrocardiogram, pulse oximeter, noninvasive blood pressure, temperature) should be used during all anesthetics (table 1). Capnography is a standard monitor during general anesthesia (GA) and moderate or deep sedation, and should be used for all patients who undergo EVT with GA or monitored anesthesia care (MAC). During MAC, capnography is used to avoid hypercarbia related to sedation induced respiratory depression, to monitor respiratory rate and regularity, and to help diagnose airway obstruction (see "Monitored anesthesia care in adults", section on 'Monitoring during monitored anesthesia care'). Once the arterial circulation has been accessed by the proceduralist, we measure blood gases and correlate the end-tidal carbon dioxide (ETCO₂) with the arterial partial pressure of carbon dioxide (PaCO₂). (See "Basic patient monitoring during anesthesia", section on 'Capnography'.)

Continuous invasive blood pressure should be monitored with an intra-arterial catheter during EVT regardless of anesthetic technique employed. Arterial catheters are often placed during initial stroke evaluation, and are optimally placed prior to induction of GA, but treatment should not be delayed for placement of an arterial catheter. Direct arterial pressure can be transduced from the arterial introducer placed for the procedure.

VENOUS ACCESS — We place at least two peripheral intravenous catheters (optimally, at least one large bore [ie, ≥18 gauge]) prior to EVT procedures, to facilitate rapid administration of fluid or blood products and/or dedicated administration of vasoactive medications. Placement of additional vascular access by the anesthesia team should not delay progress in completing the EVT procedure, if possible.

PREMEDICATION — The need for premedication should be individualized based on patient's neurologic examination and level of anxiety or pain. We avoid premedication in most patients to minimize impairment of neurologic assessment.

CHOICE OF ANESTHETIC TECHNIQUE: GENERAL ANESTHESIA VERSUS MONITORED ANESTHESIA CARE — Endovascular therapy can be performed with general anesthesia (GA) or with monitored anesthesia care (MAC) by anesthesia clinicians, or with local anesthesia with or without sedation by other providers. The choice between GA and MAC with or without sedation should be individualized based on patient and procedural factors and resource availability, in close communication with the interventionalist. For patients in whom either GA or MAC would be appropriate, we suggest GA because it provides a still patient, a secure airway, the ability to institute controlled apnea, and the ability to fully control procedural pain. However, practice varies, and others prefer MAC because it allows neurologic examination during and after the procedure.

The best available evidence suggest that GA with optimal hemodynamic control may improve technical success [4] and functional outcome [5]. As reviewed below, retrospective analyses of very large numbers of patients support MAC, whereas well-done single-center randomized controlled trials with much smaller numbers show no difference or modestly improved outcomes with GA. (See 'Literature comparing general anesthesia with monitored anesthesia care or conscious sedation' below.)

Issues to consider when choosing MAC or GA include the following:

●MAC may be appropriate for patients who can lie flat (potentially for several hours), cooperate, communicate, and protect their airways, and who do not have increased risk of respiratory depression or airway obstruction with sedation (eg, due to obstructive sleep apnea, or stroke induced dysphagia). Sleep-related respiratory dysrhythmias may be present in up to 95 percent of stroke patients, which can lead to hypoxemia, vibration of the radiologic view due to snoring, or respiratory failure with sedation [6-9].

●GA is preferred for patients with poor neurologic examination, have inattention or disinhibition, who are hemodynamically unstable, who cannot lie flat, are at high risk of aspiration or high risk of seizures, or who require endotracheal intubation for impending respiratory failure or airway protection.

Potential advantages of GA (with endotracheal intubation) include the following:

●Less risk for patient movement during the critical components of the EVT, with potentially lower risk of vascular injury, such as perforation or dissection, and potentially faster road mapping in device placement.

●Secure airway, with controlled oxygenation and ventilation (PaCO₂) and reduced risk for aspiration.

●Improved patient and neuro-interventionalist comfort.

●Evidence of improved functional outcome in randomized trials, compared with MAC [5].

Potential advantages of MAC include the following:

●Shorter time to treatment initiation

●Reduced risk for intraoperative hypotension

●Reduced risk for postoperative respiratory complications

●Allows neurologic assessment during and sooner after EVT

Importantly, if MAC is chosen, clinicians must always have a plan for quickly changing course, inducing GA and securing the airway during the procedure if necessary. The reported rate of conversion from MAC to GA varies widely, with a range of 0 to 29.5 percent [10-16]. The reported conversion rate may not be precise, since it may be unclear when sedation transitions to GA with increasing doses of sedatives, whether or not an endotracheal tube is placed. Notably, posterior circulation EVT associates with higher conversion from MAC to GA, likely due to issues with airway protection related to the stroke location [10].

Literature comparing general anesthesia with monitored anesthesia care or conscious sedation — Some retrospective and prospective observational studies have reported worse outcomes with GA for EVT. However, observational studies [17,18] and several randomized trials have reported either no difference or a small improvement in infarct size or other clinical outcomes with GA [4,11-13,17,19-36]. A 2023 meta-analysis of seven randomized trials (980 patients) that compared GA with non-GA techniques (local anesthesia, conscious sedation [CS]) found higher recanalization rates and improved functional recovery at three months in patient who had GA [5]. A subsequent randomized multi-institution trial of 273 patients who underwent mechanical thrombectomy for large vessel anterior circulation acute ischemic stroke, functional independence at 90 days and major complications within 7 days were similar in patients who had general anesthesia (both intravenous and inhalational) versus mild to moderate procedural sedation [37]. Conclusions from this study are limited by lack of any control for the type of anesthetic medications used and significant variability in intraoperative blood pressure control. (See 'Anterior circulation stroke' below.)

Hemodynamic and other physiologic management (ie, ventilation, temperature, blood glucose control) may be more important determinants of outcome than the choice of anesthetic technique [14]. Most observational studies have reported few details of anesthetic management and higher rates of hypotension during GA. By contrast, the existing randomized trials have involved protocolized blood pressure management for all patients.

In much of the observational literature, local anesthesia with or without sedation is referred to as conscious sedation (CS), whether or not an anesthesia clinician is involved. Many studies have defined GA as the presence of an endotracheal tube, regardless of the drugs administered or depth of sedation/anesthesia. Thus, patients who arrived intubated, potentially moribund, to the procedure would be included in the GA group, while patients who received GA without intubation would not be included in that group.

Most of the nonrandomized studies are limited by potential selection bias, as the choice of anesthetic technique was at the discretion of the anesthesiologist or interventionalist, with higher admission National Institutes of Health Stroke Scale (NIHSS) scores in patients who received GA [38-42].

Anterior circulation stroke — Most of the literature comparing anesthetic techniques has involved anterior circulation stroke. A 2023 meta-analysis included seven randomized trials consisting of 980 patients who underwent endovascular thrombectomy for anterior circulation ischemic stroke with GA versus non-GA techniques (local anesthesia, CS) [5]. GA improved recanalization rate (84.6 versus 75.6 percent, odds ratio [OR] 1.75, 95% CI 1.26-2.42). GA also increased the rate of functional independence (modified Rankin Scale 0 to 2) at three months (44.6 versus 36.2 percent, OR 1.43, 95% CI 1.04-1.98). The quality of data was judged to be high for recanalization rates and moderate for functional status at three months.

Posterior circulation stroke — Only a few observational studies [43-46] and a small randomized trial with parallel groups [47] have compared GA with CS for posterior AIS, with inconsistent results. In the Choice of Anesthesia for Endovascular Treatment of Acute Ischemic Stroke (CANVAS II) trial that included 87 patients who underwent EVT for posterior circulation stroke in two hospitals in China, functional independence was similar in patients who were randomly assigned to GA versus CS [47]. Successful reperfusion was more common in patients who had GA (95.3 versus 77.3 percent, risk ratio 1.17, 95% CI 1.04-1.31). CS was converted to GA in 13 of the 44 patients randomized to CS, a higher conversion rate than in studies of patients with anterior circulation stroke. Mechanical thrombectomy for posterior circulation stroke is discussed in detail separately. (See "Mechanical thrombectomy for acute ischemic stroke", section on 'Basilar artery occlusion'.)

GENERAL ANESTHESIA

Induction — The overarching goal for induction of anesthesia in patients with AIS is to avoid secondary ischemic insult (ie, as a result of hypoxemia, excessive hyperventilation, or hypotension). Systolic blood pressure should be maintained in the range of 140 to 180 mmHg prior to recanalization. (See 'Hemodynamic management' below.)

Full stomach precautions, including rapid sequence induction and intubation (RSII), are indicated for most patients with AIS, since last oral intake is often uncertain, gastric emptying may be slowed, and risk of aspiration with induction of anesthesia may be increased. If difficulty with airway management is anticipated, awake intubation, inhalation induction, or a modified RSII may be appropriate (see "Rapid sequence induction and intubation (RSII) for anesthesia", section on 'Airway evaluation'). This concern merits caution when increasing dose of sedation during monitored anesthesia care (MAC) or conscious sedation (CS).

RSII and the choice of anesthetic induction agents are discussed separately. (See "Rapid sequence induction and intubation (RSII) for anesthesia" and "General anesthesia: Intravenous induction agents".)

We typically perform RSII as follows, with doses of medications modified for patient factors:

●Lidocaine 1 to 2 mg/kg intravenous (IV)

●Fentanyl 1 to 3 mcg/kg IV

●Propofol 2 to 2.5 mg/kg IV followed immediately by phenylephrine 50 to 100 mcg IV to avoid hypotension or, for patients with hemodynamic compromise, etomidate 0.2 to 0.5 mg/kg IV

●Succinylcholine 1 to 1.5 mg/kg IV, or rocuronium 1 to 1.2 mg/kg IV, or vecuronium 0.2 mg/kg IV (see "Rapid sequence induction and intubation (RSII) for anesthesia", section on 'Neuromuscular blocking agents (NMBAs)')

Maintenance of general endotracheal anesthesia

Choice of anesthetic agents — The optimal anesthetic agents for maintenance of anesthesia for EVT has not been determined. Further research is required before confidently recommending specific anesthetic agents for EVT. Anesthetic agents should be chosen based on patient factors, with a primary goal of avoiding hypotension. Inhalation anesthetics (eg, sevoflurane, isoflurane, desflurane, nitrous oxide), intravenous anesthetics (eg, propofol, typically with opioids), or a combination of the two can be used.

The literature comparing anesthetic agents for general anesthesia (GA) for EVT is limited, and both inhaled and intravenous anesthetics have been used in the studies described above. There are no randomized trials comparing inhaled and intravenous anesthesia for EVT. Examples of relevant studies, all of which are single center retrospective studies, include the following:

●In a review of records for 84 patients who underwent EVT, modified Rankin scores (mRS) at discharge were better in patients who had inhalation anesthesia (mostly desflurane), compared with IV anesthesia or a combination of the two [48].

●In contrast, another study reported a beneficial effect of propofol, versus volatile anesthesia, on neurologic outcome after EVT [49].

●Another study of 93 patients who underwent EVT compared neurologic outcomes in patients who had volatile versus intravenous anesthesia with propofol [50]. Neurologic outcomes were similar in the two groups. Intravenous anesthesia was associated with lower 90 day mortality.

It is probably reasonable to titrate anesthetic depth using a processed electroencephalography (EEG) monitor (eg, bispectral index [BIS] or Sedline), aiming to avoid excessive administration of anesthetic agents. There is no literature on the use of such monitors in patients with acute stroke, but it seems most appropriate to extrapolate from data in patients without cerebral ischemia. Several studies have reported an association between the use of processed EEG monitors and reduced early delirium or reduced longer term cognitive dysfunction in postoperative surgical patients without stroke. However, the mechanism by which monitoring might be beneficial is unclear, and a causative relationship has not been determined [51].

The physiologic effects of anesthetic agents, including effects on cerebral blood flow, are discussed separately. (See "Anesthesia for craniotomy in adults", section on 'Maintenance of anesthesia' and "Maintenance of general anesthesia: Overview".)

Oxygenation and ventilation during EVT — We agree with recommendations from the Society of Neuroscience in Anesthesiology and Critical Care [52] regarding oxygenation and ventilation, as follows:

●Fraction of inspired oxygen (FiO₂) should be titrated to maintain peripheral arterial oxygen saturation (SpO₂) >92 percent and partial pressure of oxygen (PaO₂) >60 mmHg.

●Minute volume should be adjusted to maintain normocapnia (partial pressure of carbon dioxide [PaCO₂] between 35 to 45 mmHg) during general endotracheal anesthesia.

End-tidal carbon dioxide (ETCO₂) goals may need to be modified after vessel recanalization to avoid or attenuate hyperperfusion injury. Hyperventilation should be avoided, as the vasoconstriction that accompanies hypocapnia can result in ischemia, particularly for ischemic brain tissue [53]. On the other hand, hypercapnia may lead to cerebral arterial vasodilation with development of intracranial hypertension.

Neuromuscular blockade — We maintain neuromuscular blockade with neuromuscular blocking agents during EVT, aiming for one to two twitches using a train-of-four twitch monitor, particularly during passage of intracranial catheters.

Emergence and extubation — Patients should be extubated as soon as possible after EVT. We awaken and extubate most patients after the post procedure CT scan, though this decision is individualized after discussion with the interventionalist. In some institutions, patients are routinely transported to the intensive care unit (ICU) before emergence and extubation. Protocols for early extubation in the ICU should be in place.

For patients who arrive for EVT intubated for medical indications, it is reasonable to delay extubation until more extensive evaluation has been performed. This is particularly true for patients with posterior AIS. In general, we prefer to leave patients with posterior AIS intubated until a more comprehensive neurologic examination is performed in the intensive care unit.

The ideal emergence should be rapid and smooth, with avoidance of cough, straining, and hypertension, with the patient awake enough for an adequate neurologic examination soon after extubation. Vasoactive medications should be continued as necessary to maintain blood pressure goals.

When the patient is slow to emerge from anesthesia, the cause may be related to procedural, anesthetic, preexisting, or physiologic factors (see "Anesthesia for craniotomy in adults", section on 'Delayed emergence'). In this setting, delayed emergence may result from hemorrhage into the reperfused brain tissue, which would require repeat computed tomography (CT) for diagnosis. Delayed emergence can also be caused by posterior circulation ischemia, which can produce unconsciousness or a locked in state.

MONITORED ANESTHESIA CARE

Supplemental oxygen — We administer supplemental oxygen for all patients who undergo EVT with monitored anesthesia care (MAC). Nasal cannulae with a low flow of oxygen (1 to 2 L/minute) is sufficient to maintain adequate oxygenation (peripheral arterial oxygen saturation [SpO₂] >92 percent and partial pressure of oxygen [PaO₂] >60 mmHg) in most patients. Nasal cannula with ETCO₂ monitoring capability should be employed.

Sedation and analgesia — The choice of and need for sedation and analgesia should be individualized. Many patients undergoing EVT are comfortable with only local anesthesia injected at the cannulation site.

Level of sedation — For patients who require sedation for EVT, we aim for moderate sedation, which means that the patient responds purposefully to verbal or tactile stimuli and maintains the airway and ventilation without assistance (table 2). We repeatedly assess the level of sedation during MAC, cautiously during periods in which patient movement might be problematic, to avoid oversedation or general anesthesia (GA).

The utility of processed electroencephalogram (EEG) monitors during MAC is unclear, and we do not routinely use them during MAC for EVT. (See "Monitored anesthesia care in adults", section on 'Processed electroencephalography'.)

Choice of sedatives and analgesics for MAC — If sedatives and/or analgesics are administered, rapid-onset short-acting agents are preferred, to allow rapid titration of the depth of sedation, serial neurologic assessment, and early identification of neurologic deficits. In general, we avoid benzodiazepines and/or long-acting opioids. (See "Monitored anesthesia care in adults", section on 'Drugs used for sedation and analgesia for monitored anesthesia care'.)

Our strategy — Multiple medications can be administered by bolus or infusion for MAC for EVT. We typically administer small doses of fentanyl and lidocaine, followed by propofol infusion, to initiate MAC as follows, modified for patient factors:

●Fentanyl, 25 to 50 mcg intravenous (IV), titrated to effect

●Lidocaine, 20 to 40 mg IV to minimize pain related to propofol injection (see "General anesthesia: Intravenous induction agents", section on 'Disadvantages and adverse effects')

●Propofol infusion, 25 to 100 mcg/kg/minute, with or without boluses of propofol 10 to 20 mg IV to speed onset

●For additional pain control, if necessary, remifentanil 0.05 to 0.2 mcg/kg/min IV

Use of dexmedetomidine for EVT has been reported [32,54]. At doses used for sedation, dexmedetomidine lacks respiratory depression. However, the slow onset of effect, lack of titratability, and variable offset limit its usefulness. The effects of dexmedetomidine on cerebral blood flow and flow metabolism coupling are unclear. (See "Monitored anesthesia care in adults", section on 'Dexmedetomidine' and "Anesthesia for craniotomy in adults", section on 'Intravenous anesthesia'.)

Airway management during MAC — Most patients should receive minimal or moderate sedation during monitored anesthesia care (MAC) for EVT, which by definition means that there should be no airway intervention required and spontaneous ventilation should be adequate (table 2). If respiratory depression or airway obstruction occurs, either as a result of medications or neurologic compromise, the interventionalist should be notified immediately that there may be the need to secure the airway. Options for management include the following:

●Airway obstruction related to sedatives or opioids may be relieved with jaw thrust and/or placement of an oral or nasopharyngeal airway while waiting for the drug effect to wane. A nasopharyngeal airway should be avoided for patients who have received a thrombolytic or anticoagulant medication, due to risk of epistaxis.

Airway obstruction during sedation may mean the patient is transitioning into GA, and securing the airway or decreasing sedative doses should be considered.

●Opioid-induced respiratory depression may be reversed with naloxone, titrated in small doses (eg, 20 to 40 mcg IV increments) to avoid agitation or hypertension. (See "Perioperative uses of intravenous opioids in adults: General considerations", section on 'Prevention and management of adverse opioid effects'.)

●If oxygenation and ventilation are not improved with such maneuvers, MAC should be converted to GA and the airway secured.

Unexpected intubation during MAC — A plan must always be in place for an unexpected need to intubate. Once the decision is made to secure the airway, steps are as follows:

●If possible, stop intracranial intervention and retract intracranial catheters to a safe position, anticipating patient movement

●Call for help

●Ventilate by mask as needed, using low peak pressure to avoid gastric insufflation

●Perform rapid sequence induction and intubation (RSII) followed by maintenance of anesthesia as previously described (see 'General anesthesia' above)

HEMODYNAMIC MANAGEMENT — The optimal blood pressure goals during EVT have not been clearly defined. We follow the recommendations in the 2014 consensus statement from the Society for Neuroscience in Anesthesiology and Critical Care, which state that systolic blood pressure should be maintained between 140 and 180 mmHg prior to recanalization, whether or not the patient receives intravenous (IV) tissue plasminogen activator (tPA), and diastolic blood pressure should be maintained <105 mmHg [52].

We manage hemodynamics during EVT as follows:

●Most importantly, avoid sustained hypotension and hypertension. The exact definitions of hypotension and hypertension in this setting are somewhat ill-defined, with guidelines based on population-based data. There are no reliable technologies that allow hemodynamic titration based on individualized assessment of neurophysiology.

●Begin continuous blood pressure, heart rate, and cardiac rhythm monitoring as soon as diagnosis of AIS is confirmed.

●Place an arterial catheter for blood pressure monitoring, as long as placement does not delay the procedure. If necessary, blood pressure may be monitored using the arterial catheter placed by the interventionalist. If an arterial line is not available, noninvasive blood pressure cuff should be cycled at least once every three minutes. (See 'Monitoring' above.)

●Once euvolemia is achieved, vasoactive medications should be used to achieve and maintain blood pressure goals.

•The vasopressor of choice is usually a potent alpha-agonist, such as norepinephrine or phenylephrine, since the most likely cause of hypotension is anesthetic induced vasodilation. The drug should be chosen and adjusted based on the patient's overall hemodynamic profile and pre-existing cardiac function. (See "Use of vasopressors and inotropes".)

•For control of hypertension, nicardipine, labetalol, and clevidipine are the first choice agents in this setting (table 3). (See "Intravenous thrombolytic therapy for acute ischemic stroke: Therapeutic use", section on 'Management of blood pressure'.)

●Reperfused brain tissue may be at risk for hyperperfusion and hemorrhagic transformation with high blood pressure due to lack of cerebral autoregulation, whereas low blood pressure can lead to vessel reocclusion. Blood pressure targets after vessel recanalization with mechanical thrombectomy should be discussed with the neuro-interventionalist. Systolic blood pressure 120 to 140 mmHg is a reasonable target. (See "Mechanical thrombectomy for acute ischemic stroke", section on 'Blood pressure'.)

●For patients who undergo EVT with monitored anesthesia care (MAC) rather than GA, increasing the systolic blood pressure should be considered in patients who have worsening neurologic deficits or a decrease in the level of consciousness at a systolic blood pressure <160 mmHg. However, neurologic changes should be discussed with the interventionalist prior to blood pressure manipulation. Neurologic changes that occur after administration of sedatives may be an effect of the sedative rather than a change in brain perfusion. Subtle changes in the angiographic images might suggest a hemorrhage, in which case increasing blood pressure would be contraindicated and an urgent computed tomography (CT) scan should be obtained to guide management.

There are no prospective trials of blood pressure management during EVT. A secondary analysis of the General or Local Anesthesia in Intra Arterial Therapy (GOLIATH) trial, in which patients were randomly assigned to general anesthesia (GA) or conscious sedation (CS) for EVT, reported no associations between blood pressure-related variables and neurologic outcome [55]. In contrast, a subsequent analysis of data from 365 patients who were included in the GOLIATH, Sedation Versus Intubation for Endovascular Stroke Treatment (SIESTA), and Anesthesia during stroke (AnStroke) trials found that sustained hypotension and hypertension were associated with worse neurologic outcome [56]. Mean arterial pressure lower than 70 mmHg for more than 10 minutes, and greater than 90 mmHg for more than 45 minutes were associated with higher modified Rankin Scale (mRS) scores at 90 days, with a number needed to harm of 10 patients for each. Prospective randomized data on the timing of blood pressure abnormalities (eg, before or after recanalization) were not available. (See 'Literature comparing general anesthesia with monitored anesthesia care or conscious sedation' above.)

FLUID MANAGEMENT — Goals for fluid management during EVT include maintenance of normovolemia to achieve adequate cerebral perfusion, and avoidance of cerebral edema. Fluid should be administered at a rate and volume that achieves even fluid balance, taking into account the diuresis that results from contrast administration. When possible, volume status should be assessed with dynamic tests of fluid responsiveness, such as pulse pressure variation and/or stroke volume variation. (See "Intraoperative fluid management", section on 'Dynamic parameters to assess volume responsiveness'.)

For most patients, dextrose free isotonic (eg, plasmalyte), slightly hypotonic (eg, Ringer's lactate), or slightly hypertonic (eg, 0.9 percent sodium chloride [NaCl]) crystalloid solutions can be administered as maintenance fluid during EVT. Hypotonic fluids should be avoided to avoid cerebral edema. (See "Initial assessment and management of acute stroke", section on 'Fluids' and "Anesthesia for craniotomy in adults", section on 'Fluid management'.)

GLUCOSE MANAGEMENT — Blood glucose should be monitored closely in patients with AIS, as both hypoglycemia and hyperglycemia may worsen neurologic outcome.

Hyperglycemia is common in patients with stroke, whether or not they have diabetes. Serum glucose >140 mg/dL is associated with larger infarct size, inability to recanalize despite use of tissue plasminogen activator, lack of clinical improvement 24 hours after EVT, and increased risk of mortality [57-66]. However, tight glucose control is not beneficial and may lead to increased episodes of hypoglycemia [67,68]. Blood glucose management during EVT is generally based on guidelines for patients with acute stroke [52,69,70]. (See "Initial assessment and management of acute stroke", section on 'Hyperglycemia'.)

Emerging evidence suggests that continuous glucose monitoring may be beneficial for patients with AIS [71,72].

We manage blood glucose during EVT as follows:

●We measure glucose every 45 to 60 minutes during EVT or use a continuous glucose monitor.

●Low serum glucose (<60 mg/dL) should be corrected rapidly. It is reasonable to aim to maintain blood glucose between 140 and 180 mg/dL.

●Large swings in serum glucose should be avoided. An intravenous (IV) insulin infusion should be considered for easy titration intraoperatively. Most experts recommend protocol-driven IV insulin rather than intermittent subcutaneous dosing. (See "Perioperative management of blood glucose in adults with diabetes mellitus", section on 'IV insulin infusion'.)

TEMPERATURE MANAGEMENT — Patient temperature should be monitored during EVT, with the goal of maintaining normothermia. (See "Initial assessment and management of acute stroke", section on 'Fever'.)

Core temperature should be maintained between 35 and 37°C. Hyperthermia should be avoided, as fever has been associated with poor neurologic outcome after stroke. (See "Initial assessment and management of acute stroke", section on 'Fever'.)

Patients tend to become hypothermic during anesthesia; if necessary room temperature can be increased and warming devices may be used. It is often possible to cover the patient's head with a warm blanket to reduce heat loss. (See "Perioperative temperature management", section on 'Prevention and management'.)

Although therapeutic hypothermia may be beneficial in certain patients with neurologic injury, no evidence supports its routine use in AIS during EVT.

ANTICOAGULATION — Patients are not typically anticoagulated during mechanical thrombectomy, though they may receive heparin in the saline used by interventionalists to prevent the catheters from clotting. If additional heparin is administered, it is not reversed at the end of the procedure. In the event of arterial perforation or dissection, heparin may be reversed with protamine, and reversal agents for other anticoagulant or antithrombotic agents may be indicated. (See 'Intracranial arterial perforation' below.)

COMPLICATIONS — Intraoperative complications specific to EVT include bleeding or local vessel or nerve injury at the access site (eg, arterial dissection, pseudoaneurysm formation, retroperitoneal hemorrhage), or intracranial perforation or hemorrhage. (See "Mechanical thrombectomy for acute ischemic stroke", section on 'Adverse effects'.)

As always, anesthesia clinicians should be vigilant for signs of anaphylaxis or reactions resembling anaphylaxis, which can occur after administration of contrast agents, tissue plasminogen activator (tPA), or any other drugs. Perioperative anaphylaxis is discussed separately. (See "Perioperative anaphylaxis: Clinical manifestations, etiology, and management".)

Intracranial arterial perforation — Arterial perforation is likely the most feared and serious complication during EVT, and can lead to functional disability and poor long-term outcomes, especially in the setting of recent intravenous (IV) thrombolysis. Rates of perforation have been reported between 0 and 4.9 percent in clinical trials [40,73-78]. Extravasation of contrast material on post-thrombectomy computed tomography (CT) scans suggests arterial perforation [79] (but may also indicate blood brain barrier disruption).

Arterial perforation requires immediate management, in consultation with the neuro-interventionalist, as follows:

●Call for help, and for immediate neurosurgical consultation if the proceduralist is not a neurosurgeon

●Alert the operating room that emergency craniotomy may be necessary

●Control blood pressure with careful titration of short-acting medications (eg, nicardipine, nitroprusside), aiming for a systolic blood pressure ≤140 mmHg

●Discuss with the interventionalist discontinuation and immediate reversal of heparin, tPA, and any other anticoagulant or antiplatelet medications in effect. (See "Perioperative management of patients receiving anticoagulants", section on 'Urgent/emergency invasive procedure'.)

Our protocol for reversal of tPA includes administration of two units of platelets, two units of fresh frozen plasma, and two units of cryoprecipitate IV, with further treatment based on patient-specific factors and laboratory values. A protocol for management of intracranial hemorrhage in patients who have received tPA is discussed separately. (See "Intravenous thrombolytic therapy for acute ischemic stroke: Therapeutic use", section on 'Intracerebral hemorrhage'.)

●Treat elevated intracranial pressure as needed. (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Critical care management'.)

Extracranial arterial perforation — Arterial perforation in the chest, abdomen, or retroperitoneum may occur during the procedure, and may be indicated by an unexplained decrease in blood pressure. If such a perforation is suspected, the following steps are appropriate:

●Notify the interventionalist who may be able to confirm the diagnosis angiographically.

●If the diagnosis is confirmed call for help; a vascular neurosurgeon or other interventionalist may be required.

●Increase venous access as necessary.

●Alert the operating room if an invasive procedure is contemplated.

●Control blood pressure with careful titration of short-acting medications (eg, nicardipine, nitroprusside), aiming for a systolic blood pressure ≤140 mmHg.

POSTPROCEDURE CARE — We monitor patients continuously during transport from the interventional suite to the intensive care unit (ICU). Basic monitors should include electrocardiogram, invasive and/or noninvasive blood pressure, pulse oximetry, and if possible, respiratory rate and end-tidal carbon dioxide (ETCO₂). Emergency medications and airway equipment should be immediately available. (See "Handoffs of surgical patients".)

Postintervention patients, regardless of the presence or absence of invasive airway, should ideally be admitted to a dedicated ICU specialized in the care of stroke patients (ie, stroke unit, neurosciences ICU [NeuroICU]) [70]. Once in the ICU, the anesthesia provider must continue to monitor hemodynamics until a proper patient handoff is complete, since hemodynamic instability can occur during the ICU admission process.

CONSIDERATIONS REGARDING COVID-19 — UpToDate has information on many COVID-19 issues, including infection control, airway and other aspects of anesthetic management, intensive care, and neurologic disease, in topic reviews linked at the end of this section. Important considerations specific to anesthesia for endovascular therapy (EVT) include the following:

●COVID-19 is not itself an indication to use general anesthesia (GA) with endotracheal intubation rather than monitored anesthesia care (MAC) or conscious sedation (CS) for EVT. However, the threshold to perform GA may be lower in patients with COVID-19. In addition to other considerations, GA with endotracheal intubation may be preferable in hypoxemic patients, or those with active cough, who require high flow oxygen to maintain oxygenation, or who are likely to require conversion to GA during the procedure. (See 'Choice of anesthetic technique: General anesthesia versus monitored anesthesia care' above.)

●Every effort should be made to minimize the delay in reperfusion time that results from modification of existing EVT protocols in response to COVID-19.

●Some institutions have established protocols that include intubation in a negative pressure airborne isolation room prior to transport to the interventional radiology (IR) procedure room, which is typically not a negative pressure room. The decision to intubate prior to transport to the IR room must weigh both the safety of the health care team and the documented value of avoiding delay of recanalization, and depends on local facility conditions. Thus, it may be reasonable to intubate in an emergency department with negative pressure ventilation if close to the IR procedure room, or in the IR procedure room with appropriate precautions. Transporting the patient to an intensive care unit or operating room solely to intubate in a negative pressure environment may represent an unacceptable delay.

The Society for Neuroscience in Anesthesiology and Critical Care (SNACC) has published a consensus statement on anesthesia for EVT during the pandemic, and more general recommendations for neuroanesthesia care [80], consistent with the approach above.

UpToDate topic reviews of COVID-19-related issues include the following:

●(See "COVID-19: Perioperative risk assessment, preoperative screening and testing, and timing of surgery after infection".)

●(See "Overview of infection control during anesthetic care", section on 'Infectious agents transmitted by aerosol (eg, COVID-19)'.)

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

●Preanesthesia assessment – Preanesthesia assessment in anticipation of endovascular stroke therapy should be focused and concise to minimize delays in care. (See 'Preanesthesia evaluation' above.)

●Monitoring – Standard American Society of Anesthesiologists (ASA) monitors (ie, electrocardiogram, pulse oximeter, noninvasive blood pressure, temperature) and capnography should be used during all anesthetics, regardless of the technique employed (table 1). Intra-arterial blood pressure monitoring should also be used for all patients who undergo endovascular therapy (EVT), though treatment should not be delayed for placement of an intra-arterial catheter. (See 'Monitoring' above.)

●Choice of anesthetic technique – The choice of general anesthesia (GA) or monitored anesthesia care (MAC) with or without sedation should be based on patient and procedural factors and resource availability, in close communication with the interventionalist. For patients in whom either general anesthesia (GA) or monitored anesthesia care (MAC) would be appropriate, we suggest GA because it provides a still patient, a secure airway, the ability to institute controlled apnea, and the ability to fully control procedural pain (Grade 2C). The best available evidence suggests that GA with optimal blood pressure control may improve technical success and functional outcome. However, practice varies and some prefer MAC because it allows neurologic examination during and after the procedure. (See 'Choice of anesthetic technique: General anesthesia versus monitored anesthesia care' above.)

•MAC may be appropriate for patients who can lie flat (potentially for several hours), cooperate, communicate, and protect their airways, without increased risk of respiratory depression or airway obstruction with sedation (eg, obstructive sleep apnea).

•GA is preferred for patients with poor neurologic examination, who have inattention or disinhibition, who are hemodynamically unstable, who cannot lie flat, who are at high risk of aspiration or high risk of seizures, or who require endotracheal intubation for impending respiratory failure or airway protection.

●Anesthetic management

•Avoid secondary insult – During induction and maintenance of GA, the overarching goal is avoidance of secondary ischemic insult (ie, as a result of hypoxemia, excessive hyperventilation, or hypotension). Anesthetic agents should be chosen based on patient factors. (See 'Induction' above and 'Maintenance of general endotracheal anesthesia' above.)

•Supplemental oxygen – We administer supplemental oxygen during GA and MAC to maintain peripheral arterial oxygen saturation (SPO₂) >92 percent and partial pressure of oxygen (PaO₂) >60 mmHg, and adjust ventilation during GA to maintain normocapnia (partial pressure of carbon dioxide [PaCO₂] between 35 to 45 mmHg). (See 'Oxygenation and ventilation during EVT' above and 'Supplemental oxygen' above.)

•Level of sedation during MAC – For patients who require sedation during MAC, we aim for moderate sedation, and repeatedly assess the level of sedation to avoid oversedation or GA. We prefer rapid onset short acting agents to allow rapid titration of the level of sedation. A plan for unexpected need to intubate must always be in place. (See 'Sedation and analgesia' above and 'Unexpected intubation during MAC' above.)

•Hemodynamic management – We suggest aiming for a systolic blood pressure between 140 and 180 mmHg prior to recanalization during EVT (Grade 2C). After recanalization blood pressure targets should be discussed with the interventionalist. (See 'Hemodynamic management' above.)

•Glucose management – Low serum glucose (<60 mg/dL) should be corrected rapidly. It is reasonable to aim to maintain blood glucose between 140 and 180 mg/dL. (See 'Glucose management' above.)

●Management of arterial perforation – Arterial perforation during EVT requires immediate management, in consultation with the neuro-interventionalist. Immediate management includes rapid preparation for possible emergency craniotomy; careful blood pressure control aiming for a systolic blood pressure ≤140 mmHg; and discontinuation and possible reversal of anticoagulants, thrombolytics, and antiplatelet agents. (See 'Intracranial arterial perforation' above.)