Chapter 48B: Critical Care of Cerebrovascular Disease

Stroke is the leading cause of disability and fourth greatest cause of death in the United States. Of the 800,000 annual strokes in the United States, 85% are acute ischemic strokes (AIS). The remaining 15% are intracerebral hemorrhage (ICH) or subarachnoid hemorrhage (SAH). Intensive care issues pertinent to the care of AIS include the recognition and diagnosis of stroke, the provision of fibrinolytics and/or endovascular management, as well the medical management in the 24 hours posttreatment, management of cerebellar or large hemispheric infarcts (LHIs), and considerations in the management of a critical stenosis of a cervical arterial vessel.

History and Physical Examination

AIS is the result of a sudden loss of blood flow within a vascular region of the brain, retina, or spinal cord due to arterial vessel occlusion. The flow is most often compromised by the development of an atherosclerotic clot (ie, thrombosis) or sudden occlusion by a clot from a distant source (ie, embolism). The contemporary definition includes imaging findings of an acute infarction, even if clinical symptoms are not observed. Clinical symptoms of stroke are highly heterogeneous and variable (Table 48B–1) and are objectively scored using the National Institutes of Health Stroke Scale (NIHSS) (Table 48B–2). Presentations associated with a delay in stroke diagnosis included mild symptoms (NIHSS < 4), severe symptoms (NIHSS > 25), strokes affecting multiple vascular territories, isolated aphasia, posterior circulation symptoms, and young age. Symptoms typically begin suddenly, but a stuttering presentation can be seen in the setting of a partially occluded large vessel. The patient's last known well time (LKWT, ie, when they did not have any stroke symptoms) must be aggressively, thoroughly, and creatively sought (eg, using last social media interaction).

The physical examination is focused on cataloging a very accurate NIHSS, exploring for etiologies of stroke mimics, and findings suggestive of contraindications to intravenous (IV) thrombolysis (eg, scar from recent surgery).

Diagnostics

The ideal diagnostic approach utilizes lean processing, with a prioritization of diagnostics that are critical for the decision to offer thrombolysis with IV recombinant tissue plasminogen activator (IV-rtPA) (see treatment, later). A list of comprehensive diagnostics is found in Table 48B–3. Critical immediate diagnostics are a blood glucose and noncontrast head computed tomography (NCHCT) to rule out hypoglycemia and an intracranial hemorrhage, respectively. The remaining diagnostics should be immediately obtained, but stroke specialists will not uncommonly make decisions to offer IV-rtPA based on patient-related factors, rather than waiting for the results. This is to hasten the provision of thrombolysis, as a 15-minute delay in treatment is associated with worsened outcomes, even if within 3 hours of symptom onset and within an hour of hospital presentation.

The timing of CT angiography (CTA) of the head and neck, CT perfusion, and/or magnetic resonance imaging (MRI) is dependent on local protocols and patient specific conditions. The acquisition of these images should never delay the administration of thrombolysis. A diagnosis of AIS is most commonly confirmed using MRI through interpretation of the diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) sequences. An acute stroke (< 7 days) has occurred when a specific region of the brain is bright (hyperintense) on DWI and dark on ADC. Magnetic resonance angiography (MRA) is also able to characterize the cerebral and cervical vasculatures.

Treatment

Acute treatment of AIS consists of restoring blood flow to the compromised neuronal tissue. The first step is to determine eligibility for thrombolysis with IV-rtPA, which should be administered as soon as possible to all patients that meet inclusion/exclusion criteria (Table 48B–4) at a dose of 0.9 mg/kg, maximum dose 90 mg, with 10% bolused over 1 minute, and the remainder dripped over 1 hour. If possible, consent should be obtained, although if the patient is unable to consent and a legal representative is not available, IV-rtPA should be administered, as the benefits outweigh the risks of treatment and the degree of benefit is time-dependent. If a stroke mimic (ie, neurologic change concerned for AIS) is suspected, but cannot be definitely determined, and no contraindications exist, IV-rtPA administration should not be delayed.

AIS secondary to large vessel occlusions (LVOs), particularly those with large clots (> 8 mm), respond poorly to IV-rtPA alone. Recent randomized controlled trials (RCTs) have concluded that endovascular treatment (EVT) of AIS improves outcomes in those with anterior circulation LVOs (ie, internal carotid artery [ICA] terminus and middle cerebral artery [MCA] M1). Identification and selection of candidates for this therapy can involve clinical assessment, CTA, or MRA. The administration of IV-rtPA should never be delayed for the acquisition of angiography. Based on these new data, American Heart Association/American Stroke Association (AHA/ASA) AIS guidelines now make the class 1 recommendation that EVT should be provided to AIS patients with an ICA or M1 occlusion who are more than 17 years of age, receiving IV-rtPA and have a prestoke modified Rankin Score (mRS) of 0 to 1 (see Table 48B–5), NIHSS greater than 5 (Table 48B–2), and Alberta stroke program early CT score (ASPECTS) greater than 5 (see Table 48B–6), and who can have EVT initiated (ie, groin puncture) within 6 hours of their LKWT. Patients should never be observed for a response to IV-rtPA who meet the indications for EVT. As with IV-rtPA, time to endovascular recanalization has been shown to be a predictor of outcomes.

Of note, the first 3 endovascular trials did not find a benefit, which has been attributed to the poor recanalization rates with the early generation devices. The 5 most recent RCTs, all of which were positive, used newer generation stent retrievers that achieve a much higher rate of vessel recanalization. Although the strongest recommendations were made for those with a mRS of 0 to 1, NIHSS of greater than 5, and ASPECTS of greater than 5, those that fail to meet these criteria should not be interpreted as having an absolute contraindication to EVT. Rather, a class IIb recommendation was issued that it may be reasonable to acutely revascularize some patients with a M1 or ICA occlusion within 6 hours of LKWT, who do not meet all of the ASPECTS, NIHSS, and mRS criteria. More distal MCA (M2, M3), anterior cerebral artery (ACA), basilar artery, and posterior cerebral artery (PCA) occlusions can also be considered for EVT, but the degree of benefit is less certain, as they were either less prevalent or not included in the recent RCTs. Lastly, patients with contraindications to IV-rtPA, but within the 6-hour time window, can be considered for acute EVT (eg, international normalized ratio [INR] > 1.7).

During and after the administration of reperfusion therapy (IV-rtPA or endovascular) the patient requires frequent reassessment, with careful attention to their neurologic status and blood pressure (BP). BP parameters must be vigilantly maintained (< 180/105). All antiplatelet (AP) and anticoagulant (AC) medications are held for at least the first 24 hours. Aspirin, which has been shown to reduce stroke reoccurrence and mortality, should be provided 24 hours after IV-rtPA or EVT if the patient is neurologically stable and neuroimaging does not demonstrate hemorrhagic conversion of the infarct. In some circumstances, BP targets maybe more strict after EVT with full revascularization (thrombolysis in cerebral infarction 3 or 2b flow).

BP targets in patients not receiving IV-rtPA are less clear. Current guidelines support gentle BP reduction if the systolic BP (SBP) greater than 220 mm Hg or the diastolic BP is greater than 120 mm Hg. If performed, a modest goal should be set, so as to avoid expanding the infarct by hypoperfusing penumbral tissue, particularly in those with untreated LVOs.

Large Hemispheric Infarction

AIS secondary to the occlusion of a cervical or large proximal intracranial vessel, such as the carotid terminus (ICA-t) or MCA, is termed LHI. Although LHIs only account for 10% of all AIS, their mortality rate is 15% to 80% and the majority of survivors have significant disability. Clinical manifestations include hemiparesis, homonymous hemianopsia, ipsilateral gaze deviation, aphasia (usually with left LHIs), and agnosia and left-sided neglect (with right LHIs).

The large area of infarcted brain tissue produces substantial clinical defects at ictus. Over the following 24 to 96 hours (and less commonly up to 10 days) cytotoxic edema (CE) develops, causing the tissue to enlarge, which then compresses, distorts, and herniates neighboring uninfarcted tissues. Unaffected cerebral vessels can be pinched or kinked by the herniating infarcted brain, expanding the territory of infarction.

The first clinical sign of cerebral edema is a decline in the level of arousal. This is followed by ipsilateral or contralateral leg weakness (due to compression of the anterior cerebral artery in the setting of subfalcine herniation). Alternatively or later, findings of uncal herniation (see Chapter 48A—Principles of Neurosciences Critical Care) are seen, during which time the PCA may be compressed expanding the area of infarction. The lateral forces may also compress the third ventricle, producing an obstructive hydrocephalus and its associated symptoms.

The management of LHI includes therapies aimed to minimize the development of CE and identifying candidates for decompressive hemicraniectomy (DC). Strategies that mitigate CE include keeping the head of bed elevated to 30° and the neck in a neutral position, avoiding hypervolemia, and maintaining normothermia, normonatremia, normoglycemia, and normocarbia. The infusion of dextrose containing and hypoosmotic solutions must not occur.

Empiric administration of hyperosmotic solutions is not an uncommon practice, but evidence is lacking to support its efficacy. Aggressively maintaining the serum blood glucose below 125 mg/dL may lead to an increase in infarct size, so a goal of 140 to 180 mg/dL is recommended. Seizure prophylaxis is not indicated, but continuous video electroencephalogram (cvEEG) monitoring can be considered, particularly in patients with fluctuations in mental status. Corticosteroids do not have a proven role in the management of LHI. Dual AP therapy and therapeutic anticoagulation should be held, but subcutaneous heparin or low-molecular-weight heparin should be used for venous thromboembolism (VTE) prophylaxis.

Unfortunately, the available evidence is not able to fully predict which patient is going to have cerebral edema. Intracranial pressure (ICP) monitoring is of limited utility and is not typically performed. Instead the neurologic examination is performed at frequent, regular intervals. Older patients have some degree of cerebral atrophy; therefore they are less likely to deteriorate even if cerebral edema occurs, while the opposite holds true for the younger patient. Risk factors for cerebral edema and subsequent neurologic deterioration include female gender, nausea and vomiting on presentation, congestive heart failure, history of hypertension, leukocytosis, and initial NIHSS greater than 20 in dominant LHI or greater than 15 in nondominant LHI. Neuroimaging findings suggesting increased risk are summarized in Table 48B–7.

In the event that hemispheric swelling occurs causing symptomatic neurologic deterioration, the initial response is to provide therapies that reduce cerebral edema, such as bolus hyperosmolar therapy (eg, 30 mL of 23.4% hypertonic saline or 1-1.5 mg/kg of 20% mannitol). See Chapter _____ for further guidance on the management of cytotoxic cerebral edema/intracranial hypertension. Ultimately, 70% to 80% of patients will die with maximal medical therapy in the absence of early surgical intervention (ie, DC).

Five randomized control trials have explored the efficacy of early DC in LHI. Overall results demonstrate a reduction in mortality (from ~70% to ~20%) and an improvement in functional outcomes in those 18 to 60 years of age. A recent trial demonstrated a similar reduction in mortality in older patients (61-82 years of age), although functional outcomes were not improved. When a DC is performed it should create a bony window of greater than or equal to 12 cm in the anterior-posterior dimension, with a medial border 1 cm lateral to the superior sagittal sinus and inferior border at the floor of the middle cranial fossa. The dura is then opened with cruciate incision (ie, durotomy). Although the timing of DC was not uniform among the published data, a consistent benefit was seen if it was performed within 48 hours of stroke onset in those experiencing neurologic deterioration. It is uniformly felt that once deterioration occurs, DC should be performed promptly, as unreversed herniation will produce irreversible brainstem injury. Although the evidence for its efficacy is not clear more than 48 hours after stroke onset, it is similarly felt that if deterioration occurs, benefits are likely to be seen.

When considering DC, patients and families must be aware that the procedure is not restorative. That is, it is a lifesaving procedure that may help reduce the degree of loss in functionality in the young patient (< 60 years of age), while strictly striving to save the life of the older patient, who will most likely be left moderately to severely disabled at 12 months (ie, mRS of 4 or 5, see Table 48B–5.) It is best to inform the patient and family of their possible candidacy for the procedure on admission to the ICU. That way if deterioration does occur, they will not be forced to make a hasty decision under duress. Furthermore, neurosurgery should be engaged on admission to avoid unnecessary procedural delays in the event of neurologic deterioration.

Cerebellar Infarctions

Similar to LHI, cerebellar strokes are subject to swelling and mass effect leading to herniation. Although they are smaller infarcts, their infratentorial location and juxtaposition to the fourth ventricle lends less room for their growth. Small increases in volume can compress the fourth ventricle, producing obstructive hydrocephalus. Further swelling will directly compress the brainstem.

As with LHI, similar measures to reduce the likelihood of swelling apply. If neurologic deterioration occurs, ICP lowering therapies can temporarily reverse the mass effect of the swelling, but ultimately surgical intervention is required. Obstructive hydrocephalus from fourth ventricular obstruction is often addressed with the placement of an external ventricular drain (EVD), but this will cause the swollen cerebellum to herniate upward, compressing and kinking the brainstem, producing local ischemia and potentially infarction of the midbrain and pons. Therefore, a suboccipital craniectomy (SOC) should be performed immediately after or simultaneously with EVD placement, which will allow for the swollen cerebellum to temporarily herniate extracranially. In contrast to LHI, nearly 3/4 of patients with large cerebellar infarcts will have very good outcomes.

ICH, also referred to as “hemorrhagic stroke,” constitutes approximately 15% of all strokes and has an incidence of approximately 25 cases per 100,000 person-years. When compared with AIS of similar presenting severity, the mortality and long-term morbidity of ICH is greater. Management of ICH includes measures to prevent further hematoma expansion, mitigate and treat cerebral edema, prevent further neurologic deterioration, and provide excellent supportive care. Although ICH carries a high morbidity and mortality, decisions on the withdrawal of life sustaining therapy (WLST) and early do-not-resuscitate (DNR) have been found to contribute greatly to outcomes. Prognostication is difficult and ambivalent cases with supportive and willing surrogate decision makers should be allowed time to declare their outcomes, in lieu of the early implementation of WLST.

History and Physical Examination

Those presenting with a headache, unexplained nausea and vomiting, or neurologic change may be harboring an ICH. Clinical features commonly seen in, but not specific to ICH, include nausea, vomiting, impaired level of consciousness, severe hypertension, and headache. As with any neurologic emergency, the LKWT must be expeditious ascertained. Patients presenting within 6 hours of symptom onset are much more likely to experience further enlargement of the hematoma. Hematoma volume is predictive of poor outcomes, therefore those presenting quickly after onset may respond favorable to treatments targeting hemostasis. A history of AC or AP use must be obtained.

Diagnostics

Any patient suspected of having an ICH should immediately receive a NCHCT. MRI can also detect an ICH, particularly when T2*, gradient-recalled echo (GRE), or susceptibility weighted images (SWI) are obtained, but a NCHCT is typically much quicker to obtain. Simultaneous information should be obtained about their hemostatic capacity by obtaining a complete blood count, basic metabolic panel, prothrombin time (PT), INR, and activated partial thromboplastin time (aPTT). Some centers routinely obtain viscoelastic tests, such as thromboelastography, and platelet function assays, but the value of these diagnostics and subsequent therapies targeting their “correction” is as yet unclear.

A critical component of ICH care is determining the etiology of the ICH. Subcortical and pontine ICHs commonly result from hypertension-induced lipohyalinosis of the small perforating arteries in the pons, thalamus, and basal ganglia. Cerebral amyloid angiopathy (CAA) is more common in those more than 70 years of age and involves the cortical regions (lobar ICH). ICHs due to ruptured vascular malformations (arteriovenous malformation [AVM], cavernous, aneurysms, dural arteriovenous fistula [dAVf]) must be promptly identified promptly to allow for prompt surgical interventions that will mitigate rebleeding. Other etiologies are listed in Table 48B–8. CT or MRA is highly sensitive for underlying vascular malformations. Younger patients, those without a history of hypertension or coagulopathy, and females are at greater risk for vascular malformations. The secondary ICH (sICH) score maybe helpful in deciding who should undergo angiography to rule out an underlying vascular malformation (see Table 48B–9).

CTA may also identify a “spot sign” or evidence of contract extravasation into the hematoma, implying that the hemorrhage is ongoing and the hematoma is enlarging. The clinical utility of this information is as yet unclear. Ongoing trials are exploring the value of administering a prohemostatic agent (either activated factor VIIa [aFVIIa] in STOP-IT and SPOTLIGHT or tranexamic acid [TXA] in the SPOT-AUST) in those with a “spot sign.” Other features on the NCHCT suggestive of hematoma expansion include a heterogeneous appearance of the ICH, larger volumes, and those with ventricular extension.

The volume of ICH can be estimated on NCHCT by applying the (A × B × C)/2 method, where A represents the maximal diameter on an axial image, B is the diameter perpendicular to A, and C is the height of the ICH, as determined by the number of axial slices the ICH is seen on times the width of the slices. Applying this method on serial images provides a quasiobjective was of assessing for hematoma growth. The severity of the ICH is often quantified by determining the ICH score (see Table 48B–10). This score should not be used to assign outcomes, as its correlation with outcomes worsens when excluding those that have undergone WLST.

Treatment

An immediate treatment goal is to stop further hemorrhaging from occurring, as hematoma size is predictive of poor outcome. In parallel, those with acute neurologic deterioration due to cerebral herniation or mass effect may require craniectomy and/or clot evacuation, while those with symptomatic hydrocephalus due to intraventricular hemorrhage (IVH) may undergo ventricular catheterization.

BP control and reduction is a must to mitigate hematoma expansion. Two large randomized trials have been completed in this realm. Second Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial (INTERACT-2) randomized ICH patients presenting within 6 hours of onset and a SBP of 150 to 220 mm Hg to a target SBP of less than 140 mm Hg or less than 180 mm Hg (without a prescribed approach). No difference in mortality or adverse effects were seen, but an ordinal analysis found a statistically significant reduction in the mRS (Table 48B–5) in the group targeted to a SBP of less than 140 mm Hg. Based on these data, the subsequent AHA/ASA ICH guidelines published in 2015 were updated to state “for ICH patients presenting with SBP between 150 and 220 mm Hg and without contraindication to acute BP treatment, acute lowering of SBP to 140 mm Hg is safe (class I; level of evidence A) and can be effective for improving functional outcome (class IIa; level of evidence B).” The most recent RCT of SBP control in ICH, Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATACH-2), failed to find any difference in functional outcomes or hematoma growth, but there were more renal adverse events in the intensive control, compared to the conventional control arm (SBP < 140 mm Hg vs < 180 mm Hg, 9% vs 4%, number needed to harm of 20). These conflicting findings maybe explained by how the patients in the each arm of ATACH-2 and INTERACT-2 were actually treated, as the conventional treatment arm in ATACH-2 had SBPs lower than the intensive arm in INTERACT-2. In the absence of updated guidelines, the best recommendation based on available data is to lower the SBP to 140 to 150 mm Hg in those presenting with a SBP of 180 to 240 mm Hg. If the SBP is greater than 240 mm Hg, a 25% reduction is a reasonable goal, followed by reassessment for tolerance of a lower SBP. Additionally, a post-hoc analysis of the INTERACT-2 trial found that increasing degrees of BP variability were associated with worse outcomes, suggesting that continuous SBP management is preferred to reactionary maneuvers by using a continuous infusion of an antihypertensive, such as nicardipine, labetalol, or clevidipine.

Acute ICH complicated by ACs or coagulopathy must be provided reversal or prohemostatic therapy immediately and without delay. The time to “reversal” is predictive of outcome, with those achieving their target BP and antithrombotic reversal within 4 hours of onset having the best outcomes. Table 48B–11 provides a summary of ACs and suggested reversal strategies. The impact of APs on outcomes and how they should be managed in ICH is less clear. A recently completed RCT of platelet transfusion for reversal of AP activity in nonvascular supratentorial ICH patients presenting within 6 hours of onset, a history of AP use (aspirin, P2Y₁₂ inhibitor, or adenosine reuptake inhibitor), and not requiring an intraventricular catheter (IVC) or surgery found that not only was platelet transfusion ineffective in reducing hematoma expansion, but it was also observed to worsen outcomes and increase the risk of neurologic deterioration. Therefore, platelet transfusion should not be performed in this subcategory of ICH. Table 48B–12 provides a recommended approach to ICH complicated by AP agents.

After the initial several hours, secondary intracranial and systemic effects are the next layer of complications that are to be managed. Perihematomal edema develops early, and commonly peaks in the following 4 to 7 days, but atypically courses are seen and can last for upward of a month. Management includes the maintenance of normoglycemia, normothermia, normonatremia, and avoidance of elevated central venous pressures. Fever is associated with worsened outcomes, but temperature control has not been proven to improve outcomes, and maybe associated with longer hospital stays and increased tracheostomy.

Seizures, both convulsive and nonconvulsive, are common are ICH, particularly in those with larger more symptomatic and cortical hemorrhages. Prophylaxis with phenytoin has been associated with, but not causational of, worse outcomes. Guidelines recommend against seizure prophylaxis with antiepileptic drugs, but it is not unreasonable to do so in those at higher risk. A low threshold should be kept for evaluation for nonconvulsive seizures with continuous electroencephalography, particularly in those whose neurologic examination is worse than expected.

Up to one quarter of intubated ICH patients will develop acute respiratory distress syndrome (ARDS), therefore lung protective ventilation should be provided after intubation, as long as normocarbia and ICP can be maintained. Hyperoxia is associated with worsen outcomes in ICH, among other acute brain injuries, therefore normoxia should be targeted at all times, sans the peri-intubation period to prevent hypoxia during airway establishment.

The timing of VTE prophylaxis is not clearly guided by available evidence. Best practice at this time is to initiate pharmacologic VTE prophylaxis with unfractionated or low-molecular-weight heparin 24 hours after demonstrated ICH stability (ie, no further hematoma expansion), assuming there is not underlying coagulopathy. Sequential compression devices should be placed immediately. In the case of a newly diagnosed VTE during the course of care, there is a lack of evidence to advise on when it is safe to initiate therapeutic anticoagulation. A common practice is to wait at least 14, if not 30 days after an ICH. Depending on the etiology of the ICH, the patient may never be a candidate for therapeutic anticoagulation (eg, unsecured vascular etiology or CAA).

Surgical options in the management of ICH include craniotomy with or without clot evacuation and, in the setting of obstructive hydrocephalus due to IVH, the placement of an IVC. Two randomized trials (Surgical Trial in Lobar Intracerebral Hemorrhage [STICH] and STICH-2) explored the value of craniotomy for supratentorial ICHs and failed to find a benefit, possibly due to the increased cerebral injury caused by the open craniotomy approach. As such, at present, craniotomy with surgical evacuation is largely considered in those with a deteriorating neurologic examination, symptomatic herniation syndromes, or refractory intracranial hypertension not amendable to medical therapy and diversion of cerebrospinal fluid (CSF) (when IVH complicates the ICH). The failure of open craniotomy led to an interest in minimally invasive stereotactic approaches to aspirate the hematoma. An approach currently under investigation is to stereotactically place a catheter into the hemorrhage, aspirate hematoma, and then leave the catheter in place to instill rtPA every 8 hours for 9 total doses (Minimally Invasive Surgery Plus rt-PA for Intracerebral Hemorrhage Evacuation [MISTIE III]).

ICH complicated by IVH causes obstructive hydrocephalus, which maybe aided by enhanced clearance of the ventricular clot with intraventricular rtPA. Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage Phase III (CLEAR-III), a randomized trial of intraventricular rtPA in IVH, failed meet its primary outcome of increasing the proportion of patients with a good outcome (mRS 0-3), but there was a 10% reduction in mortality and a subgroup benefit in those with a large clots or if the catheter was inserted directly into the clot.

SAH can occur spontaneously, which is often the result of a vascular malformation, such as an arterial aneurysm (aSAH), or from traumatic injury to bridging veins. The incidence of spontaneous SAH, which this section will focus on, is poorly quantified, but is estimated to occur in 20 out of every 100,000 adults in the United States. A specific subtype of nonaneurysmal SAH is perimesencephalic SAH (PMSAH), which is thought to be due to a spontaneous venous rupture in the basilar cisterns, producing a radiographic pattern of SAH similar to that seen with aSAH, but it has a much more benign course. Table 48B–13 outlines etiologies of spontaneous SAH.

Risk factors for aSAH include hypertension, smoking, alcohol abuse, use of sympathomimetic drugs, such as cocaine, female sex, family history, collagen vascular diseases, such as type IV Ehlers-Danlos syndrome, and polycystic kidney disease. Cerebral arterial aneurysms are present in 3.6% to 6% of the population and more common in the anterior circulation. Rupture risk increases with aneurysm size and posterior circulation aneurysms.

When an aneurysm ruptures, the pressure in the subarachnoid space rapidly increases and approximates the arterial pressure. This can lead to a temporary decrease or cessation in blood flow to the brain, resulting clinically in syncope. During this period of reduced cerebral blood flow, a clot is able to tenuously form in the aneurysm, stopping further hemorrhaging. This sudden increase in ICP, combined with a brief period of cerebral ischemia can lead to the development of cerebral edema in the first several hours after the aneurysm rupture. Additionally, subarachnoid blood can obstruct the flow and absorption of CSF through the foramen of Luschka and Magendie and arachnoid granulations respectively, producing an obstructive and/or communicating hydrocephalus. The tenuous clot in the recently ruptured aneurysm is a high risk for dislodgement, making early surgical obliteration of the aneurysm a priority. Beginning at about day 4 after the aneurysm bleed, the risk for cerebral vasospasm (VSP) resulting in delayed cerebral ischemia (DCI) begins. This typically lasts for 14 days, with the peak period being 7 to 10 days postbleed, but can be seen up to 28 days from ictus.

Outcomes after aSAH are highly associated with the severity of presenting symptoms, quantity of SAH, comorbidities, and age. Overall mortality rate ranges from 30% to 40%, with nearly half of survivors experiencing some type of permanent disability. More subtle deficits in memory, executive function, and language ability, as well as psychiatric symptoms, such as anxiety or depression are common. Despite the high morbidity and mortality, even patients with severe presentations can experience outcomes that allow them to return to independent living or a high quality of life.

History and Physical Examination

The most common presenting symptom of aSAH is headache. Less common presentations are syncope or new onset lethargy or confusion. Deciding which headaches are concerning for aSAH can be challenging, both for health care providers and patients themselves. Classically, the patient will describe the worst headache of their life that came on very suddenly (ie, “thunderclap”). A careful neurologic examination should be performed with particular attention to the cranial nerve examination and performance of an assessment of concentration (eg, reciting months of the year backward). Unfortunately, in the presence of a normal neurologic examination there are not any highly specific historical features of aSAH. Given the extremely high morbidity and mortality of a missed aSAH, we must be persuaded to seek out highly sensitive tools at the expense of specificity.

A sentinel headache, or milder headache or headache without neurologic change, can occur a few weeks prior in up to 40% those who experience an aSAH. It represents an initial hemorrhage from the aneurysm of less severity.

There are 2 wildly used systems that score the severity of the aSAH and are based on clinical symptoms. Hunt and Hess (H/H) and World Federation of Neurosurgeons (WFNS) are outlined in Table 48B–14.

Diagnostics

The initial diagnostic test is a NCHCT. If performed without 6 hours of symptom onset, it is a modern day CT scanner, and the patient is not anemic or coagulopathic, the sensitivity approaches 100%. There are reports of reduced sensitivity in this time frame on modern scanners when the head CT is read by nonneuroradiologists in the community setting. After this time frame, a lumber puncture (LP) may need to be performed to full exclude the possibility of SAH, depending on the pretest probability/clinical suspicion, as the SAH will begin to degrade and may no longer be hyperdense on a NCHCT. Of note, the LP should be performed at least 12 hours after symptom onset, as the intracerebral subarachnoid blood may not have adequately circulated into the lumbar cistern, from which the CSF will be collected. An aSAH is excluded by the absence of red blood cells and xanthochromia in the CSF. An MRI of the head can also be performed, with particular attention to the GRE, T2*, or SWI images, but LP remains the gold standard when performed more than 12 hours after symptom onset.

Once SAH is diagnosed, an etiology must be pursued, which begins with vascular imaging. A CTA of the head is most common performed, but if contraindications exist to its performance, an MRA can be performed, although it has a lower sensitivity for aneurysms than CTA. If the CTA is nondiagnostic, a digital subtraction angiogram (DSA) should be performed.

The risk of VSP, DCI, and delayed neurologic deterioration (DND) are proportional to the amount of SAH. Two scoring systems are in use to help stratify this risk, based on the initial head CT. The original system was the Fischer Scale, which was modified in 2001 by Claassen et al and then revalidated by Frontera in 2006. Table 48B–15 summarizes each of these scales and their associated risk of DND.

Treatment

Early (the first 6 hours) treatment goals include BP control, cardiopulmonary stability, correction of symptomatic hydrocephalus, treatment of intracranial hypertension, reversal of herniation syndromes, and consideration of antifibrinolytics. Aneurysmal rebleeding is a catastrophic event. It is associated with a more than 50% mortality rate and none of the survivors recover to a mRS of 0 to 2 (good outcome). It occurs in 8% to 23% of aSAH, with 83% of the rebleeds occurring in the first 6 hours. The risk can be mitigated with rapid, smooth, consistent BP control, minimization of patient agitation, and the rapid correction of any coagulopathy. Commonly selected short acting continuous infusion antihypertensives include nicardipine, clevidipine, labetalol, and esmolol. Robust vasodilators, such as nitroglycerin, hydralazine, and sodium nitroprusside are typically avoided, as they may cause significant increases in cerebral blood volume and blood flow dynamics. Antifibrinolytics such as TXA or epsilon-aminocaproic acid (EACA) appear to reduce the risk of rebleeding, but have not been definitively shown to improve functional outcomes. If administered, they should be discontinued 4 hours prior to DSA to minimize the risk of iatrogenic ischemic strokes during the procedure from catheter thrombi. If the patient has a depressed level of consciousness or demonstrated intracranial hypertension, a mean arterial pressure of 80 mm Hg (assuming no intracranial monitoring) or cerebral perfusion pressure of 60 mm Hg should be maintained, while controlling the SBP. An epidemiologic BP target has not been clearly established, but many centers use a target SBP of less than 140 mm Hg. The value of seizure prophylaxis is unclear. Similar observations about phenytoin have been made in aSAH. Many centers will provide 7 days of prophylactic levetiracetam in poor grade aSAH and those undergoing open surgical clipping, but the value of this approach is unclear. Any patient with poor neurologic examination or one that is worse than expected for the severity of the hemorrhage should undergo cvEEG monitoring.

Neurogenic stunned myocardium and/or neurogenic pulmonary edema are not uncommon after aSAH, particularly in those with poor grade (ie, high H/H or WFNS scores) aSAHs. Cardiac complications include dramatic (self-limited) reduction in ejection fraction (Takotsubo cardiomyopathy), electrocardiographic changes (ST/T wave, QT prolongation, U waves) and supraventricular and ventricular arrhythmias. In fact, some aSAH will present in cardiac arrest. Ventilatory support with positive pressure along with optimization of serum potassium (> 4.0) and magnesium (> 2.0) are cornerstones in the management of these cases. If an airway needs to be obtained, be sure to minimize patient agitation, physiologic fluctuations, and BP spikes during the procedure to minimize the risk of causing a rebleed.

Early decisions should be made regarding whether to place an IVC to treat a hydrocephalus, particularly if the patient is going to DSA for diagnostics and possible aneurysmal intervention, where they will be laying flat for a prolonged period of time and difficult to monitor. Cerebral edema may require treatment with hyperosmolar agents, such as mannitol or hypertonic saline.

Early surgical or EVT of the aneurysm is recommended as it improves outcome, most commonly with 24 hours or, as identified in a meta-analysis, within 72 hours. The European International Subarachnoid Aneurysm Trial (ISAT; 2002) randomized good neurologic grades (WFNS I-III) aSAH to surgical clipping (n = 1070) or coiling (n = 1073) with death and dependency at 1 year as the primary outcome measure. The authors reported better outcome for coiling group (23.5%) compared to surgical clipping group (30.9%); however, the rebleeding rate at 1 year was 2.6% in the coiling versus 1% in the clipping group. However, at 5-year follow-up a reduced death rate for coiling (11%) compared to clipping (14%) was observed while the percentage of independent survivors did not differ making the overall death rate in the coiled group 3% at 5 years. Bias in recruitment strategy, aneurysm location, aneurysm size, selection of good grades, treatment time window, operator experience, age exclusion, presurgical evaluations, choice of techniques and questionnaire follow up are cited in order to understand the limitation of this landmark study. A follow up trial (ISAT II) is underway. Currently, most investigators identify equipoise with respect to longer term outcomes and treatment approach and identify that certain patients are not suitable for either surgical or EVT. For example, coiling is the less preferred option in patients with aneurysms of large size, with large mass effect, wide neck (neck-to-dome ratio > 0.5), fusiform appearance, and at arterial bifurcations. In contrast, posterior circulation aneurysms, especially basilar tip aneurysms, locations within or between the cavernous sinuses, or those that can more readily be accessed by an endovascular approach. Age is a relevant factor in the decision making as younger patients have better long-term protection from recurrent SAH with clipping while coiling is preferred in elderly with multiple comorbidities.

As with all critical brain injuries, normoglycemia, normothermia, normonatremia, and normovolemia must be maintained. Fluid and electrolyte abnormalities are common in aSAH patients. Cerebral salt wasting syndrome and syndrome of inappropriate secretion of antidiuretic hormone (SIADH) are common, and may even coexist in the same patient. Fluid balance must be vigilantly maintained, as negative fluid balances in aSAH are associated with worsened outcomes.

As stated aforementioned, VSP occurs commonly during period following rupture of aSAH, with a peak incidence on postbleed days 7 to 10 (with the day of the bleed being day 0). Those with a normal neurologic examination should have serial assessments vigilantly performed. If the neurologic examination is impaired, serial angiographic assessments with transcranial Dopplers (TCDs), CTAs, CT perfusion studies, and/or DSAs. TCDs noninvasively assess for the velocity of blood flow through the large cerebral arteries (MCA, ACA, Basilar), with increase in velocity indicative of VSP. MCA velocities of greater than 200 cm/s or a ratio between the MCA and the extracranial ICA of greater than 6 (Lindegaard ratio [LR]) are suggestive of severe VSP. TCDs also provide a pulsatility index (PI), which is a marker of distal arteriolar resistance, as would be seen in small vessel VSP.

Examination changes related to VSP should be treated immediately by ensure euvolemia and raising the BP, often pharmacologically. If the neurologic change does not resolved, then more aggressive should be expeditiously performed, such as DSA where intra-arterial verapamil or milrinone can be administered or angioplasty can be performed. Medical measures that have been demonstrated to have some effect with the treatment of VSP include IV milrinone, intrathecal nicardipine, IV magnesium, high-dose statins, and endothelin-receptor antagonists. Unfortunately RCTs exploring the value of the later 3 classes failed to meet their primary functional outcomes, despite evidence suggesting that they are effective at treating angiographic VSP.

Most recently, DCI is being better understood as an arteriolar VSP that results in microvascular thrombosis and perfusion mismatch and neurovascular uncoupling, complicated by spreading depolarizations, which begin at the time of the aneurysm rupture and ultimately lead to cortical infarctions. Nimodipine is a stalwart in the management of aSAH. It should be started on day of arrival and continued for 21 days. It has been shown to improve functional outcomes, but interestingly, it failed to produce a statistically significant reduction in VSP. Reasons for this discrepancy are related to other suspected benefits of calcium channel blockage, including the prevention of small vessel VSP, stabilization of the neurovascular unit, prevention of intracellular calcium influx as part of apoptosis, and mild profibrinolysis that mitigates microvascular thrombosis.