Decision Making in Emergency Critical Care

SECTION 6 - Neurological Critical Care

Subarachnoid and Intracerebral Hemorrhage

A.L.O. Manoel, Cappy Lay, D. Turkel-Parrella, Joshua Stillman, and Alberto Goffi


Cerebrovascular disease is the fourth leading cause of death in North America and accounts for approximately 130,000 deaths per year. Eighty percent of strokes are ischemic; the remaining 20% are hemorrhagic. Hemorrhagic stroke is subdivided into spontaneous intracranial hemorrhage (ICH) (15%) and subarachnoid hemorrhage (5%). Although less common, hemorrhagic stroke has markedly worse outcomes than ischemic stroke, including higher mortality and poorer functional outcomes. This chapter reviews the management of both spontaneous ICH and aneurysmal subarachnoid hemorrhage.


The estimated incidence of spontaneous ICH worldwide is 24.6/100,000 person-years; 67,000 cases are reported annually in the United States.1 Among strokes, ICH carries the poorest prognosis for survival and functional recovery, with high rates of early mortality (median 30-day mortality 40.4%, with 50% of these deaths occurring within the first 2 days); poor long-term survival2; and moderate to severe persistent deficits among survivors (<40% ever achieve independent function).1 Recent population-based studies, however, suggest that more than half of all patients present with small ICHs, where excellent and timely medical care can have a powerful, positive impact on morbidity and mortality.3 In fact, observational reports suggest that misguided prognostic pessimism has led to withdrawal of life support in patients who would have had acceptable clinical outcomes if properly managed.46 ICH must therefore be considered an acute neurologic emergency with potential interventions that may significantly mitigate primary and subsequent secondary brain injury. The following discussion focuses exclusively on spontaneous (i.e., not traumatic) ICH.

Etiology and Risk Factors for ICH

An increased incidence of spontaneous ICH is associated with many underlying conditions, including hypertension, advanced age, and male gender. Other conditions associated with a poorer prognosis—after controlling for age and gender—include diabetes mellitus and a posterior fossa location (Table 21.1). The most common risk factor associated with spontaneous ICH is chronic arterial hypertension, which is present in approximately 75% of all patients with ICH and is associated with deep hemorrhage. The most common sites for hypertensive bleeds are deep perforator arteries in the pons, midbrain, thalamus, basal ganglia, and the deep cerebellar nuclei.7 The lobar region is the second most common location for ICH (45%). It is more common in the elderly and is associated with cerebral amyloid angiopathy. Posterior fossa hemorrhage accounts for the remaining 10% of ICH and carries the worst prognosis.

TABLE 21.1 Etiology and Risk Factors for ICH

Important risk factors for secondary ICH are myriad: coagulopathies (resulting from the use of antithrombotic or thrombolytic agents or from congenital or acquired factor deficiencies); systemic diseases such as thrombocytopenia; lymphoproliferative disorders; and hepatic and renal failure. The increasing use of oral anticoagulants, especially vitamin K inhibitors (such as warfarin) and newer oral anticoagulants (such as dabigatran), has resulted in a surge of coagulopathy-associated ICH in recent years and now accounts for more than 15% of all cases of ICH.8 Other identified risk factors for ICH are advanced age, high alcohol intake, low cholesterol, and low triglyceride levels.9,10 Socioeconomic and ethnic factors also appear to play a role in the prevalence of cerebral hemorrhage. ICH is twice as common in low-income and middle-income countries when compared with high-income countries; Asians, African Americans, and Hispanics are at higher risk than Caucasians.11,12

Causes of ICH include intracranial aneurysms and arteriovenous malformations (AVMs). Aneurysms most commonly rupture into the subarachnoid space but may also cause intraparenchymal hematomas. AVMs typically remain asymptomatic; however, ICH is their most common presentation (60% of AVMs present with intraparenchymal hemorrhage).13 Hemorrhage due to an AVM may occur at any location within the cerebrum, brainstem, or cerebellum.

Brain tumors are a rare cause of intracerebral hemorrhage and account for <5% of all cases.14 These may be primary tumors, most commonly glioblastoma multiforme (GBM) or oligodendrogliomas, or they may be metastatic brain tumors. Lung cancer, because of its high prevalence, is the most common source for brain metastases causing ICH. Other sources of brain metastasis causing ICH include melanoma, renal cell carcinoma, thyroid carcinoma, and choriocarcinoma.15

Less frequent causes of secondary ICH include infections, vasculitis, sinus venous thrombosis, carotid endarterectomy, Moyamoya disease, and drug use (e.g., cocaine). Finally, it should be noted that hemorrhagic transformation of acute ischemic stroke is relatively common, but in the absence of anticoagulation or thrombolytic therapy, is most often asymptomatic.

Mechanisms of Brain Injury

Acute neurologic injuries cause immediate damage (primary brain injury) and delayed damage (secondary brain injury). In ICH, primary injury is defined by local tissue destruction, which results from the rupture of a blood vessel into the brain parenchyma and ensuing ischemia and elevated intracranial pressure (ICP). In more than one-third of patients, substantial expansion of the hemorrhage is observed during the first few hours, resulting in further mechanical injury and early clinical deterioration.16 It is thought that much of this initial damage cannot be reversed.

Primary brain injury initiates a cascade of biochemical events at the cellular level, including ischemic and apoptotic cell injury cascades, edema, and excitotoxicity, resulting in delayed and often progressive secondary brain injury. Unlike primary injury, secondary brain injury is considered preventable or reversible in the first hours to days following the initial hemorrhagic event. If present, conditions that decrease cerebral oxygen and glucose delivery (e.g., hypotension, hypoxia, anemia, and hypoglycemia) or increase cerebral metabolic demand (e.g., fever, seizures, and hyperglycemia) exacerbate secondary brain injury.17 Minimization of secondary brain injury requires an early, aggressive, and well-structured approach to patient care and may result in improved long-term functional outcomes.

History and Physical Exam

Classically, ICH presents as a sudden onset of a focal neurologic deficit that evolves over minutes to hours. Clinical assessment, however, cannot reliably distinguish intracerebral hemorrhage from ischemic stroke.18 Neurologic signs and symptoms can help indicate the location of the hemorrhage: (1) hemiplegia/hemiparesis, hemisensory loss, or homonymous hemianopsia suggest putaminal and thalamic ICH; (2) ataxia, vomiting, headache, and coma indicate brainstem compression in cerebellar bleeding; (3) deep coma, total paralysis, and pinpoint pupils suggest pontine bleeding.

Common symptoms for all types of ICH include headache (~40%), nausea and vomiting (~40% to 50%), and alteration in level of consciousness (LOC) (~50%), particularly for large ICH. Seizures occur in up to one-third of patients and often reflect an expanding hemorrhage, an underlying vascular or neoplastic etiology, or a lobar hemorrhage affecting cortical tissue.19

Blood pressure (BP) is typically elevated in ICH. Nonspecific EKG abnormalities are common (e.g., prolonged QT interval, depressed ST segments, flat or inverted T waves) and are thought to result from a centrally mediated release of catecholamines. Ventricular arrhythmias have also been described with brainstem compression.

Progression of neurologic deficits with deterioration of LOC during the first 48 hours after hospital admission has been described in 22% to 50% of patients with ICH.20,21

Diagnostic Evaluation

Recently published Emergency Neurological Life Support (ENLS) protocols22 emphasize the following aspects of emergent clinical assessment for patients presenting with suspicion of ICH: (1) a concise and targeted assessment of the patient's clinical condition and (2) rapid and accurate diagnosis using neuroimaging to define ICH characteristics (i.e., location, volume, and possible etiology). Clinical assessment proceeds as follows:

1.ABCs. Immediate assessment and stabilization of airway, breathing, and circulation.

2.Evaluate all vital signs, oxygen saturation, and blood glucose. Almost any alteration in vital signs can contribute to secondary brain injury.

3.Perform and document a standardized neurologic stroke severity scale during the initial encounter. This allows for easy communication about the initial level of disability and for comparison over time. The most common rating scales include the National Institutes of Health Stroke Scale (NIHSS)—appropriate for patients who are awake or drowsy—and the Glasgow Coma Scale (GCS)—for the obtunded or comatose patient. Often, both scales are used.

4.Evaluate for bleeding disorders. Investigate current anticoagulant use and any history of coagulopathy. Determine when the last dose of antithrombotic medication was taken. Measure the platelets count, partial thromboplastin time (PTT), and international normalized ratio (INR).

5.Perform frequent neurologic assessments. Ideally every 15 to 30 minutes, for rapid detection of clinical deterioration and signs of increased ICP.

The clinical presentation of ICH is indistinguishable from ischemic stroke, but its management can be very different; therefore, rapid neuroimaging is essential. Noncontrast computed tomography (CT) is the most commonly used imaging modality for emergency diagnosis and characterization of ICH (location and extent of the hematoma). Noncontrast CT is highly sensitive and specific for acute bleeding, which will appear hyperdense, then, over weeks, become isodense, and may have a ring-enhancing appearance. In addition to the location of the primary hematoma, the degree of bleeding (including volume, the presence of intraventricular hemorrhage [IVH], and signs of increased ICP or herniation) is among the strongest predictors of long-term outcome.

A rapid estimate of ICH volume helps determine stroke severity and delineate treatment options. A simple and validated method that can be used in the emergency department (ED) is the ABC/2 formula,23 where A is the greatest hemorrhage diameter on the CT slice with the largest area of ICH, B is the largest perpendicular diameter on the same CT slice, and C is the approximate number of CT slices with hemorrhage multiplied by the slice thickness in centimeters, which is often 0.5 cm. For calculation of C, a slice is counted as 1 if the hemorrhage area is >75% of the largest hematoma area on the reference slice; as 0.5 if the hemorrhage area is approximately 25% to 75%; and not counted if the area is <25%. ABC/2 gives the ICH volume in cm3. In children, the ABC/XYZ has been proposed, where X, Y, and Z are perpendicular measures of the supratentorial intracranial space (% of total brain volume).24

Recently, it has been suggested that identification of active extravasation of intravenous contrast into the hematoma, called the “spot sign,” during contrast-enhanced CT and/or CT angiography (CTA) may predict hematoma expansion.25,26

In patients with confirmed acute ICH, CT or MR angiography, or catheter angiography is recommended to exclude an underlying lesion such as an aneurysm, AVM, or tumor. However, in hypertensive patients with a well-circumscribed hematoma in a typical location for hypertensive bleeding (thalamus, basal ganglia, pons, and cerebellum), the yield of such studies is extremely low, and a decision not to proceed with these additional diagnostic tests is reasonable.27 At the other extreme, young, nonhypertensive patients with isolated intraventricular hemorrhage (IVH) deserve aggressive workup.

Risk Stratification and Prognostication

Several clinical grading scales have been developed to assist with risk stratification and prognostication. An easy-to-use and well-validated model is the ICH score28 (Table 21.2), which is based on patient demographics (age), clinical condition (GCS), and neuroimaging findings (ICH volume, presence of IVH, and supratentorial or infratentorial origin of ICH). The ICH score has been validated for stratification of 30-day mortality28 and 12-month functional outcome29; each point increase is associated with increased mortality risk and poorer functional outcome. However, it has been shown that withdrawal of life support in patients likely to have a poor outcome may significantly bias these predictive models. In the ED, clinical grading scales should be used only for communication about a patient's condition, or for research purposes, and not to limit interventions in the initial management of patients with ICH.22

TABLE 21.2 ICH Score

From Hemphill JC III, Bonovich DC, Besmertis L, et al. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke. 2001;32(4):891–897.

Management Guidelines

Emergency Department management of patients with acute ICH entails (1) initial stabilization of airway and hemodynamics, (2) minimization of primary injury and (3) prevention of secondary brain injury. The most recent American Heart Association/American Stroke Association guidelines30 and the recently published ENLS protocols22 are reviewed in the following sections.

Initial Stabilization

Management of ICH begins by ensuring adequate patient airway, breathing, and circulation. Early endotracheal intubation is essential for patients with a depressed LOC who are unable to protect their airway. Classically, a GCS ≤ 8, rapidly deteriorating LOC, and uncontrolled seizures are indications for intubation. Stable patients requiring transfer to another medical facility should be carefully assessed for the possibility of airway compromise in the the near term, and, if the risk is deemed high, be intubated prior to leaving the referring center. Whenever possible, a rapid and concise neurologic assessment should precede intubation in order to document the patient's baseline functioning before the exam is confounded by use of sedative or paralytic drugs.

Maintenance of both brain perfusion and oxygenation is critical for prevention of secondary brain injury. To this end, steps should be taken to prevent elevations in ICP, including minimization of airway manipulation and use of ICP lowering medications. Oxygen saturation should be maintained >94% and carbon dioxide (PaCO2) levels should be kept in the normal range (35–45 mm Hg). In mechanically ventilated patients, use of lung-protective ventilation strategies (pressure- and volume-limited mechanical ventilation) is appropriate. In a setting of increased ICP and/or signs of acute brain herniation, hyperventilation to a goal PaCO2 of 28 to 32 mm Hg may be used. Hyperventilation is not a definitive treatment for elevated ICP because of the risk of increased brain ischemia and rebound elevations in ICP; a normal PaCO2 should be reinstituted as soon as definitive treatments to control ICP are in place.31

Minimization of Primary Injury

Blood Pressure Management

Arterial blood pressure is elevated in the majority of patients who present with ICH. Mean arterial pressure (MAP) is >120 mm Hg in over two-thirds of ICH patients and >140 mm Hg in over one-third.32Such acute elevations in BP have been implicated as a cause of bleeding and as a normal physiologic response to maintain cerebral perfusion pressure (CPP). Although there is general agreement that low BP levels are associated with poorer outcome and must be corrected, it is not clear at this time whether this observation simply reflects the fact that low BP levels occur more often in severe cases.30

Current guidelines30 recommend the following BP targets in patients with spontaneous ICH:

  • SBP > 200 mm Hg or MAP > 150 mm Hg: Aggressive reduction of BP with target MAP of 110 mm Hg or BP 160/90 mm Hg
  • SBP > 180 mm Hg or MAP > 130 mm Hg and no clinical evidence of elevated ICP: Target MAP of 110 mm Hg or BP 160/90 mm Hg
  • SBP > 180 mm Hg or MAP > 130 mm Hg with clinical evidence of ICP elevation on exam, CT, or ICP monitor; If ICP monitoring is available, target a CPP of ≥ 60 mm Hg (50 to 70 mm Hg); if ICP monitoring is not available, target a MAP of 80 to 90 mm Hg (assuming an ICP of 20 to 30 mm Hg)

The evidence underlying these guidelines is controversial. In the recent, large multicenter trial “Intracerebral Hemorrhage Acutely Decreasing Arterial Pressure Trial 2” (INTERACT 2), 2,839 patients with spontaneous ICH were randomized to rapid blood pressure lowering with a target SBP = 140 mm Hg within 1 hour; or to the standard guideline-recommended target SBP of 180 mm Hg. Analysis of a composite outcome of death and severe disability on the modified Rankin scale (mRS = 3 to 6) showed an 8% benefit in the more aggressive treatment group; however, the result was not statistically significant. Although the safety of this lower-BP target has been demonstrated, an evidence-based benefit in clinical outcome has yet to be confirmed. More answers are expected from the Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATACH) II trial.

If a decision is made to lower blood pressure, management should be started immediately. A short-acting, titratable, intravenous agent should be used to achieve the target quickly and with minimal risk for overshoot. Labetalol (initial bolus dose 5 to 20 mg titrated every 10 minutes to effect) is a reasonable agent if there are no contraindications. Nicardipine is another excellent option (initial dose 5 mg/hour, with titration by 2.5 mg/hour every 15 minutes as needed; maximum dose 15 mg/hour). Angiotensin-converting enzyme inhibitors (e.g., enalapril) and hydralazine may be used. Sodium nitroprusside and nitroglycerin increase ICP and lower cerebral blood flow and should be avoided.

Twenty-four to forty-eight hours following brain injury, oral/enteral antihypertensive medications should be initiated to help achieve individualized blood pressure targets for secondary stroke prevention.

Correction of Coagulopathy

Coagulopathy in patients with ICH is most commonly due to use of therapeutic anticoagulation; other risk factors include acquired or congenital coagulation factor deficiencies and qualitative or quantitative platelet abnormalities. Coagulopathies in ICH are associated with poor prognosis because of prolonged bleeding and hematoma expansion; whenever possible, these deficits should be immediately corrected.

Specific Anticoagulants

1.Vitamin K antagonists (VKAs, e.g., warfarin) are currently the most commonly prescribed oral anticoagulants. ICH occurs 8 to 10 times more frequently in VKA anticoagulated patients than in non–anticoagulated patients, with a twofold higher mortality rate. Therapy includes withholding anticoagulants and treating to rapidly normalize the INR with IV vitamin K (5 to 10 mg) and replacement of vitamin K–dependent factors. Debate continues over the optimal strategy for replacing vitamin K–dependent factors; currently both fresh frozen plasma (FFP; 10 to 15 mL/kg) and prothrombin complex concentrates (PCCs—25 to 50 IU/kg) are used. AHA/ASA guidelines recommend PCCs because of their smaller infusion volume and subsequently lower risk of volume overload and pulmonary edema.30 PCCs have the added advantages of rapid reconstitution and administration and result in the correction of INR within minutes. The most recent American College of Chest Physicians (ACCP) Evidence-Based Clinical Practice Guidelines recommend using PCCs rather than FFP33 to reverse significant warfarin-associated ICH.

2.Novel oral anticoagulants (direct thrombin inhibitors, e.g., dabigatran, and Xa inhibitors, e.g., rivaroxaban) have also been associated with ICH. Clinical experience in reversing coagulopathy from these agents is limited, and no specific reversal protocols or agents currently exist; inhibitors for dabigatran and rivaroxaban are under development, but not yet commercially available. There is some evidence that hemodialysis may be effective in dabigatran-associated bleeding, and, within 2 hours of ingestion, there may be a role for oral activated charcoal (also suggested for rivaroxaban).34 PCCs may have a role in treating ICH related to rivaroxaban, but not to dabigatran. In the case of patients treated with one of these newer oral anticoagulants, urgent hematologic consultation is recommended.

3.For patients receiving unfractionated heparin (UFH), protamine sulfate is the reversal agent of choice. Standard dosing is 1 mg of protamine for every 100 units of heparin administered (maximum dose 50 mg). When UFH is given as continuous infusion, only the UFH given in the preceding 2 hours should be considered when estimating the quantity of heparin to be reversed. If more than 4 hours have elapsed since the last dose of UFH, reversal is unlikely to be necessary (PTT should still be documented). With low molecular weight heparin (LMWH), full reversal is not possible, although protamine may still be used in an attempt at partial reversal (provides a maximum of 60% to 75% inhibition of the anti-Xa activity).

Antiplatelet Agents

Conflicting results have been published regarding the impact of antiplatelet agents on hematoma expansion and clinical outcomes. There is a small increased risk of ICH with the use of antiplatelet agents (0.2 events per 1,000 patient-years).35,36 Some centers support empiric use of platelet transfusion, while others discourage this practice, or suggest assaying for platelet function to guide transfusion.22Current guidelines highlight a lack of evidence and consider platelet transfusion in ICH patients with a history of antiplatelet use as experimental.30 Additionally, some authors suggest the use of desmopressin (DDAVP, 0.3 mcg/kg), as has been used in the treatment of uremia-associated bleeding.22

Fibrinolytic Agents

Symptomatic ICH is one of the most life-threatening complications of thrombolytic therapy. The incidence of symptomatic ICH following recombinant tissue plasminogen activator (rt-PA) therapy for ischemic stroke is approximately 6%; of interest, symptomatic ICH following thrombolysis for myocardial infarction (MI), for which a higher dose of rt-PA is used than in stroke (1.1 mg/kg in MI vs. 0.9 mg/kg in stroke), is quite rare (0.4% to 1.3%). The difference is thought to reflect the fact that healthy cerebral vessels do not readily bleed from thrombolysis. Management of suspected ICH during or after fibrinolytic infusion begins with immediate cessation of the infusion, clinical stabilization (ABCs), and emergent noncontrast CT head. The NINDS rt-PA study37 protocol recommends empiric treatment in these cases with 6 to 8 units of cryoprecipitate or FFP and 6 to 8 units of platelets; however, evidence on the most effective treatment in this situation is lacking.

Even patients without evidence of coagulopathy may experience hematoma expansion, especially in the first 24 hours. Because hematoma expansion is one of the major risk factors for poor outcome, it has been hypothesized that use of procoagulant agents could improve outcomes after ICH. Five randomized trials tested this hypothesis using recombinant factor VIIa (rFVIIa) (NovoSeven® RT) in non-coagulopathic patients with ICH (spontaneous and traumatic ICH). A meta-analysis38 of these studies showed significant reduction in hematoma growth, but an increased rate of thromboembolic events and no overall net difference in mortality or long-term disability. Current guidelines do not recommend the use of rFVIIa in the treatment of ICH.30 However, rFVIIa might benefit specific subsets of patients in whom the risk of hematoma expansion outweighs the risk of thromboembolic events. Two ongoing trials address this question in patients thought to be at high risk for hematoma expansion. The SPOTLIGHT trial (Spot Sign Selection of Intracerebral Hemorrhage to Guide Hemostatic Therapy) and the STOP-IT trial (Spot Sign for Predicting and Treating ICH Growth Study) are both addressing the role of rFVIIa in patients identified on CTA as having a positive “spot sign,” a finding indicative of extravasation of contrast into the hematoma and suggestive of significant risk for imminent hematoma expansion.39

Surgical Interventions

Based on current evidence and guidelines, surgical intervention may be considered in the following conditions.

Infratentorial ICH

Although no randomized controlled trials (RCTs) of cerebellar hematoma evacuation have been undertaken, several case series suggest that surgical evacuation with cerebellar decompression is associated with improved outcomes in patients with ICH > 3 cm in diameter and clinical deterioration, or radiographic evidence of either brainstem compression or hydrocephalus. Treatment with external ventricular drainage (EVD) alone without posterior fossa decompression is not recommended because of the theoretical risk of upward herniation. Patients with cerebellar hemorrhage should be always referred for urgent neurosurgical consultation.

Supratentorial ICH

Current guidelines suggest that surgical evacuation of supratentorial ICH should be considered only in patients presenting with lobar clots >30 mL that are within 1 cm of the surface.30,40,41 The recently published Surgical Trial in Intracerebral Hemorrhage (STICH) II41 did not show any difference in unfavorable outcomes at 6 months when comparing early surgery to conservative treatment in this specific subgroup of patients. The trial showed a slight survival advantage (OR = 0.86) for surgery within a few hours of the onset of hemorrhage in conscious patients with a modestly decreased GCS (9 to 12) and with lobar hematomas, but the survival advantage was far from achieving statistical significance.42 Expert consensus is that surgery should be considered as a life-saving procedure for treatment of refractory increased ICP, especially in patients with ongoing clinical deterioration, recent onset of hemorrhage, involvement of the nondominant hemisphere, and relatively accessible hematomas.

Intraventricular Hemorrhage and Hydrocephalus

IVH is quite common in spontaneous ICH (45% of patients), especially in patients with hypertensive hemorrhages involving the basal ganglia and the thalamus.43 Acute hydrocephalus may develop after ICH, either in association with IVH or because of direct mass effect on ventricles. Patients with acute hydrocephalus require urgent neurosurgical consultation for possible EVD placement. Unfortunately, ventriculostomy in the setting of IVH is difficult to manage because of frequent obstruction secondary to blood clots. Flushing the catheter helps remove the thrombus but may cause ventriculitis. Recently, use of intraventricular thrombolytic agents has been suggested as adjunct to EVD for accelerating blood clearance and clot lysis. The safety phase 2 trial of the CLEAR-IVH trial (Clot Lysis: Evaluating Accelerated Resolution of IVH) prospectively evaluated the safety of intraventricular use of 3 mg rt-PA versus placebo in 48 patients. Results from this study suggest that intraventricular rt-PA is safe and can have a significant benefit on clot clearance. However, pending results of the ongoing phase III CLEAR-IVH trial, current guidelines consider this treatment experimental.30

Prevention of Secondary Injury

Although this chapter focuses on the initial evaluation and management of patients with ICH, it is reasonable for the emergency physician to implement early intrventions that can help minimize secondary injury in the ensuing 24 to 72 hours.22

Intracranial Pressure Monitoring

Few studies have addressed the incidence, management, and impact of elevated ICP on outcomes of ICH patients. Current guidelines are based on the principles and goals of traumatic brain injury (TBI) management.22,30,44

  • Indications for ICP monitoring: GCS ≤ 8, large hematoma with mass effect suggestive of elevated ICP, or hydrocephalus
  • Goals: ICP < 20 mm Hg, CPP 50 to 70 mm Hg (if possible, adjustments based on the patient's cerebral autoregulatory status)
  • Interventions: Initial measures: elevate the patient's head (30 to 45 degrees), drain cerebral spinal fluid (CSF) using an EVD; provide analgesia and sedation to achieve a motionless state, and maintain normal body temperature
  • Advanced measures:hypertonic solutions (e.g., mannitol and hypertonic saline); hyperventilation (as bridge to further management); neuromuscular blockade; hematoma evacuation/decompressive craniectomy; mild hypothermia; barbiturate coma

Seizure Prophylaxis

Seizures frequently complicate ICH; however, their incidence varies widely depending on diagnostic criteria, duration of follow-up, and the population studied. The estimated incidence of clinical seizures in patients with ICH is 4.2% to 20%, subclinical seizures 29% to 31%, and status epilepticus 0.3% to 21.4%. About 50% to 70% of seizures will occur within the first 24 hours, and 90% in the first 3 days.39Predisposing factors include ICH with a lobar location (typically nonoccipital and subcortical hemorrhages), large hematoma size, hydrocephalus, midline shift, and low GCS. Although seizures theoretically may exacerbate brain injury, conflicting results have been reported on seizure association with clinical outcome and mortality. No RCTs exist to guide decision making for seizure prophylaxis or treatment specifically in patients with ICH.

As in traumatic brain injury, prophylactic anticonvulsants in patients with lobar ICH may reduce the risk of early seizures but do not affect long-term risk of developing epilepsy. In addition, two recent studies found their use to be associated with worse functional outcomes.45,46 Based on available data, current guidelines do not recommend routine use of prophylactic anticonvulsants.30

However, if a patient with ICH develops clinical seizures, or there is a change in mental status associated with EEG evidence of seizures, experts recommend initiation of treatment with antiepileptic agents. The choice of initial drug should depend on individual patient characteristics (i.e., medical comorbidities, concurrent drugs, and contraindications). Initial treatment typically begins with an intravenous benzodiazepine (e.g., lorazepam 0.05 to 0.10 mg/kg), followed by a loading dose of an IV agent (e.g., phenytoin 15 to 20 mg/kg, valproic acid 15 to 45 mg/kg, levetiracetam 500 to 1,500 mg, or phenobarbital 10 to 20 mg/kg).

Glycemic Control

A high proportion of patients with ICH (~60%) will develop stress hyperglycemia in the first 72 hours, even in the absence of a previous history of diabetes mellitus.40 Multiple studies have associated increased serum glucose in the acute phase of ICH with higher risk of poor outcome (hematoma expansion, increased edema, and death or severe disability).41 However, clear causality between hyperglycemia and poor outcome and, more interestingly, evidence of improved outcome with glycemic control have not been proven. Recent microdialysis studies have demonstrated increased cerebral hypoglycemic events in patients treated with tight glucose control strategy, and a large multicenter RCT in a general ICU population found increased mortality with intensive glucose control.47 Current guidelines recommend close glucose monitoring and avoidance of both hypoglycemia (<70 mg/dL) and hyperglycemia (>180 mg/dL); most experts agree that an insulin infusion should aim for a serum glucose of 140 to 180 mg/dL.39 By contrast, tight control (80 to 110 mg/dL) has been shown to increase mortality.47

Temperature Control

Fever is relatively common in patients with ICH (up to 40%), and it has been independently associated with poor outcome. However, no RCT has yet demonstrated improved clinical outcome with induced normothermia.39 Despite a lack of evidence, there is general agreement that the presence of fever should prompt an appropriately broad workup; infectious sources should be identified and treated, and hyperthermia should be corrected (target core temperature below 38°C–37.5°C).

Venous Thromboembolism Prophylaxis

Patients with ICH are at high risk of venous thromboembolism (VTE). Independent risk factors for thromboembolic disease in patients with ICH include greater severity of stroke, prolonged immobilization, advanced age, female gender, African–American ethnicity, and thrombophilia. Discontinuation of antithrombotic agents is itself, of course, associated with an increased risk of deep vein thrombosis (DVT).39 Guidelines suggest the use of intermittent pneumatic compression (IPC) devices in addition to elastic stockings in patients admitted for ICH, based on an RCT showing a reduced occurrence of asymptomatic DVT (4.7% vs. 15.9%).48 Evidence regarding use of prophylactic UFH or LMWH is less definitive. Based on small studies showing safety of pharmacologic prophylaxis (no increased risk of hematoma expansion or further bleeding), current guidelines suggest consideration of LMWH starting 1 to 4 days after ICH, provided follow-up imaging has documented cessation of bleeding.30


Patients with ICH are frequently medically and neurologically unstable and are at significant risk for sudden clinical deterioration, particularly in the immediate poststroke period. Care of ICH patients in highly specialized stroke or neurointensive intensive care units has been associated with lower mortality and better functional outcome,49 and admission to such a unit is considered standard of care.30 An institutional algorithm for referral protocol and/or transfer to centers with higher levels of care is recommended.


The challenge of emergency medicine lies in identifying those patients who can be treated and released and those patients whose complaints represent a life-threatening process requiring urgent intervention. Among the diseases with the greatest potential for catastrophic consequence when undiagnosed is subarachnoid hemorrhage from a ruptured cerebral aneurysm. When an emergency department patient presents with headache due to aneurysmal rupture, timely diagnosis by the emergency physician is the best chance for avoiding the devastating effects of rebleeding that so often result in severe disability or death.

Aneurysmal subarachnoid hemorrhage (aSAH) accounts for only a small proportion of patients who present to the ED with a complaint of headache. Unfortunately, despite our awareness of the severity of this disease, 12% of aSAH patients are misdiagnosed on initial presentation. Misdiagnosed patients are more likely to have normal mental status, to present more than a day after the onset of symptoms, be unmarried, less educated, and speak English as a second language.50


Subarachnoid hemorrhage (SAH) is classified as either traumatic or spontaneous. Ruptured intracranial aneurysms are the leading cause of spontaneous SAH, followed by AVMs and nonaneurysmal “perimesencephalic” bleeding (characterized by a typical CT pattern and a benign clinical course). Cerebral aneurysms are vascular outpouchings that occur most frequently in the circle of Willis, where they typically form at branch points of the major cerebral arteries. Although cerebral aneurysms may be found at any arterial location in the cerebral circulation, the most common sites are the anterior communicating artery (30%), the posterior communicating artery (25%), the middle cerebral artery (20%), internal carotid bifurcation (7.5%), basilar tip (7%), and the posterior–inferior cerebellar artery (3%).51

Autopsy studies have shown that 6% to 8% of the general population harbors a cerebral aneurysm. The risk of rupture depends on many factors, including aneurysm location, size, and previous history of rupture.52 In the United States, aSAH affects 30,000 persons per year and is twice as common in women (average age of 55 years old).5357 Although aSAH accounts for only 5% of all types of stroke, it is responsible for 27% of productive years of life lost from cerebrovascular diseases.58

Hypertension and smoking have a causative role in both aneurysm formation and rupture.59 A recent study reported that smoking increased the odds of aneurysm rupture threefold.60 Several heritable conditions are associated with the development of cerebral artery aneurysms, including a first-degree relative with aSAH, autosomal dominant polycystic kidney disease (PKD), neurofibromatosis type I, Marfan syndrome, multiple endocrine neoplasia (MEN) type I, pseudoxanthoma elasticum, hereditary hemorrhagic telangiectasia, and Ehlers-Danlos syndrome type II and IV.61 Family history and PKD account for 10% and 1% of all cases of aSAH, respectively.

History and Physical Exam

Aneurysmal SAH patients typically present with a sudden onset of severe headache. It is commonly described as the “worst headache of life,” but unfortunately, this description is given by more than 78% of all patients with headache of any etiology who present to the ED.62 The development of pain from aSAH is almost always rapid, though not instantaneous, and will usually reach peak intensity within 30 minutes of onset. Pain can be accompanied by loss of consciousness, vomiting, and neck pain or stiffness. Clinical grading scales have been developed to classify the severity and to predict the long-term outcome of aSAH (Table 21.3).63,64

TABLE 21.3 Clinical Grading Scales

WFNS, World Federation of Neurosurgical Societies.

Although SAH represents only 2% of acute headaches in the ED, its potential for devastating outcomes makes accurate diagnosis essential.6567 A clinical decision rule was recently developed to rule out aSAH in patients with acute headache (Ottawa SAH rule).68 In patients presenting to the ED with an acute headache and normal neurologic exam, any of the following factors raises the likelihood of aSAH and mandates additional workup (Table 21.4): age ≥ 40 years, neck pain or stiffness, witnessed loss of consciousness, onset during exertion, thunderclap headache (instantly peaking pain), and limited neck flexion on examination. This rule showed a sensitivity of 100% for detecting spontaneous SAH.

TABLE 21.4 Ottawa SAH Rule

SAH, Subarachnoid Hemorrhage.

From Perry JJ, Stiell IG, Sivilotti ML, et al. Clinical decision rules to rule out subarachnoid hemorrhage for acute headache. JAMA. 2013;310(12):1248–1255.

Diagnostic Evaluation

In a series of 482 patients with aSAH admitted to a tertiary hospital between 1996 and 2001, 56 patients (12%) of cases were initially misdiagnosed.50 In 43% of cases, the misdiagnosis occurred in the ED. Most commonly, these patients received the diagnosis of tension headache or migraine (36%). The most common diagnostic error was the failure to acquire a head CT prior to discharge (73%). Three factors that were independently associated with misdiagnosis were normal mental status, small aSAH volume, and right-sided aneurysm location. Patients presenting with a normal mental status had higher rates of misdiagnosis (19%) than those with altered mental status.

The diagnostic workup for aSAH has traditionally included emergent noncontrast CT imaging followed by a lumbar puncture (LP) to evaluate for red blood cells or xanthochromia in the CSF if CT imaging is negative.69

Early studies of CT for detection of SAH demonstrated a sensitivity of 93% to 95% in the first 24 hours following onset of symptoms, dropping to 85% 3 days after, and 50% a week after symptom onset.70 More recent studies have reported sensitivities close to 100% in the first 72 hours using more advanced CT technology, raising question of whether lumbar puncture is always required to rule out the diagnosis.71,72 A recent prospective study of 3,132 patients with nontraumatic acute headache reported the sensitivity and negative predictive value of CT for the detection of SAH in the first 6 hours after symptom onset to be 100%.73 All studies were performed on third-generation CT scanners and were interpreted by a trained radiologist. These results suggest that lumbar puncture may not be necessary to rule out the diagnosis of aSAH when a patient presents to an ED within 6 hours of ictus. Another less invasive diagnostic approach that has been proposed is noncontrast CT followed by CTA. The CT/CTA approach excludes aSAH with a >99% post-test probability.74,75 The disadvantages of this last approach lie in the radiation dose and the need for iodinated contrast.

On CT, acute SAH appears as hyperdense material, most often filling the suprasellar, ambient, quadrigeminal, and prepontine cisterns, with extension into the sylvian fissures and interhemispheric fissure. IVH is common and is a risk for the development of communicating hydrocephalus. Thicker cisternal clots and IVH have been associated with the development of delayed cerebral ischemia (DCI) in the course of aSAH (Table 21.5).

TABLE 21.5 CT Grading Scales

SAH, subarachnoid hemorrhage; IVH, intraventricular hemorrhage.

Less frequently, aneurysmal rupture can occur directly into brain parenchyma, resulting in intracerebral hemorrhage in addition to SAH and IVH. Depending on the location of the ICH, this is often accompanied by clinical hemiplegia or hemiparesis. Global cerebral edema may also be present on initial head CT and is more commonly seen in patients with Hunt and Hess scores of 4 or 5.

Management Guidelines

Approximately 12% of aSAH patients will die immediately.78 For patients who survive to reach medical attention, rebleeding is the most life-threatening entity, with mortality rates close to 70%.79,80Traditionally, the risk of rebleeding after SAH has been quoted as 4% in the first 24 hours, 1% to 2% per day for the next 14 days, 50% risk during the initial 6 months after ictus, and 3% yearly thereafter. This is now believed to be an underestimate,81 with ultra-early rebleeding occurring in up to 17% of cases.79 Proper initial management therefore includes taking steps to prevent rebleeding and to ensure transfer of patients to high-volume centers for definitive treatment. Other key interventions include implementation of strategies to prevent secondary complications, such as DCI. The following sections reference the most recent American Heart Association/American Stroke Association guidelines,82 the recommendations from the Neurocritical Care Society's Multidisciplinary Consensus Conference,83 and the recently published ENLS protocols.84

Initial Stabilization

As with any medical emergency, initial management focuses on the ABCs. Cardiopulmonary complications are not uncommon following aSAH and are likely related to catecholamine discharge. Troponin elevation, arrhythmias (prolonged QT, ventricular arrhythmias, ST-segment changes), and wall motion abnormalities on echocardiography (stress-induced cardiomyopathy) are observed in 25% to 35% of patients. Severe cardiac compromise can occur and results in sudden death, cardiogenic shock, and pulmonary edema. Neurogenic pulmonary edema has also been described. These manifestations are usually transient and tend to resolve during the first week after hospitalization.85

Prevention of Rebleeding

Blood Pressure Management

Blood pressure control is one of the most important early interventions in patients with aSAH. However, unlike for ICH, limited data exist to guide BP management in acute aSAH patients with an unsecured aneurysm (i.e., prior to either neurosurgical clipping or endovascular coiling). Lowering BP may decrease the risk of rebleeding but increases the risk of cerebral infarction in patients with impaired autoregulation. In a series of 134 patients with aSAH, a lower incidence of rebleeding (15% vs. 33%) but a higher incidence of infarction (43% vs. 22%) was reported in patients given antihypertensive therapy.86 Randomized controlled studies are lacking. Current guidelines acknowledge the paucity of data and recommend balancing the risk of hypertensive-induced rebleeding with the risk of ischemia from reduced CPP. Maintaining an SBP below 160 mm Hg or a MAP below 110 mm Hg is considered reasonable.82,84 Labetalol and nicardipine, both fast-acting and titratable drugs, are the preferred agents; as in ICH patients, the use of nitroprusside or nitroglycerine should be avoided because of the risk of increased ICP secondary to increased cerebral blood volume.87

Pain and Anxiety Management

Pain (especially headache) and anxiety are common complaints after aSAH. Management of pain or anxiety in this population is challenging because of the difficult balance between effective management and avoidance of oversedation. There is no medication of choice; acetaminophen (1 g orally every 6 hours) along with an opioid agent (e.g., fentanyl, morphine, or hydromorphone) is a common strategy. Prior to the aneurysm being secured, NSAIDS should be avoided given their anti-platelet activity. Once the aneurysm is secured, the use of nonsteroid anti-inflammatory drugs (NSAIDs) as adjunctive opioid-sparing therapy may be considered. However, NSAID use has to be carefully considered because of potential detrimental effect on CPP and brain tissue hypoxia.88 Small doses of benzodiazepines may help in a significantly anxious patient. However, agitation, confusion, and delirium can be insidious signs of symptomatic DCI and have to be carefully addressed in this population.89

Antifibrinolytic Agents

Definitive aneurysm treatment often requires transfer to specialized centers (see Transfer to High-Volume Center); this strategy, however, can be associated with delay and potential increased risk of rebleeding. Antifibrinolytic therapy (e.g., tranexamic acid, aminocaproic acid) has therefore gained interest for its potential role in this group of patients at risk of ultraearly rebleeding. In one randomized trial, 254 patients with ruptured aneurysms received tranexamic acid (1 g IV immediately after CT diagnosis, followed by 1 g IV every 6 hours until aneurysm obliteration—for a maximum of 72 hours); 251 patients served as controls. Patients receiving tranexamic acid—70% of whom had the aneurysm secured within 24 hours—showed a reduction in the rebleeding rate from 10.8% to 2.4% and an inferred 80% reduction in the mortality rate due to early rebleeding.90 Current recommendations suggest consideration of early (at diagnosis) and short (<72 hours) course of antifibrinolytic therapy (tranexamic acid or aminocaproic acid) for prevention of early rebleeding when definitive treatment of the aneurysm is unavoidably delayed and no risk factors for VTE are identified.82,83 Delayed (>48 hours after the ictus) or prolonged (>72 hours) treatment with these agents is not recommended because of the associated risk of complications (VTE and cerebral ischemia).83

Correction of Coagulopathy

The same principles of coagulopathy management discussed in the ICH section apply to spontaneous SAH. Although there are limited data available to support this guideline, most experts recommend reversal of all antithrombotic agents in patients with aSAH until definitive obliteration of the aneurysm has been achieved.84

Monitoring of Neurologic Status

Acute Hydrocephalus

Acute hydrocephalus is one of the most common complications of aSAH (seen in 9% to 67% of patients), causing increased ICP and rapid neurologic deterioration.91 Hydrocephalus in aSAH develops as a result of accumulation of subarachnoid blood on the arachnoid granulations, preventing the reabsorption of CSF. In addition, blood can obstruct the ventricular system, causing obstructive or noncommunicating hydrocephalus. Patients with aSAH and good neurologic grade (e.g., a WFNS of 1 to 3) require only frequent neurologic assessments. If clinical deterioration occurs (usually within 72 hours), an emergent noncontrast CT should be performed and, if hydrocephalus is confirmed, an EVD should be inserted. Patients with poor-grade (WFNS 4 and 5) and CT evidence of hydrocephalus require immediate EVD placement. Approximately 30% of these patients will demonstrate clinical improvement after EVD insertion.92 While EVD insertion theoretically can cause rebleeding in unsecured ruptured aneurysms, observational studies have not confirmed this concern.91

Elevated ICP

Elevated ICP secondary to acute hydrocephalus and reactive hyperemia/cerebral edema is common in patients suffering from high-grade aSAH, and is associated with poor outcomes. Definitive evidence is lacking in this population, and most of strategies, as in ICH, are derived from TBI management (see ICH section).93


Seizures are uncommon after aSAH (<20%), usually follow aneurysm re-rupture and are associated with poor outcome. Risk factors for seizures at onset are presence of intraparenchymal clot, middle cerebral artery aneurysm, and surgical clipping. There is, however, disagreement among experts regarding the routine use of anticonvulsants, and observational studies have shown worse cognitive and functional outcomes with prophylactic use of phenytoin.82,83,89,94 If a decision to use seizure prophylaxis is undertaken, a very short course (3 to 7 days) with an agent other than phenytoin is advised.83 For patients with documented clinical or electrographic seizures, treatment with an anticonvulsant is advised.

Transfer to High-Volume Center

Once a patient has been stabilized, transfer to a specialized high-volume center (>35 aSAH/year), with experienced neurovascular surgeons, endovascular specialists, and multidisciplinary neurointensive care services, is recommended.82 Unfortunately, despite evidence of improved outcome, only a minority of aSAH patients are managed in these centers.95

Definitive Treatment of a Ruptured Aneurysm

Definitive treatment of a ruptured aneurysm involves either neurosurgical clipping or endovascular coiling to mechanically secure the lesion and isolate it from the intracranial circulation. The International Subarachnoid Aneurysm Trial (ISAT)96 compared these two interventions in patients with aSAH, and showed that endovascular coiling resulted in significantly better disability-free survival at 1 year. However, debate still persists as to the superiority of one treatment over another.97 There is general agreement that, regardless of modality, early treatment confers a clinical benefit.82

Postobliteration Management

Prevention of Delayed Cerebral Ischemia

In the first 2 weeks following bleeding, patients with aSAH are at risk of deterioration as a result of cerebral vasospasm and DCI. The use of nimodipine (60 mg orally every 4 hours), started on ICU admission and continued for 21 days, is the only strategy currently available to decrease the risk of DCI and to improve functional outcomes.98 Interestingly, oral nimodipine does not decrease the incidence of angiographic vasospasm, traditionally considered the primary cause of DCI. The most common complication of nimodipine use is hypotension, considered detrimental in aSAH patients because of the risk of cerebral hypoperfusion. In case of hypotension related to nimodipine, the dose can be changed to 30 mg every 2 hours.

Maintenance of euvolemia and normonatremia is fundamental in the management of patients suffering from aSAH. Typically, after aSAH, patients experience increased natriuresis and urine output, with subsequent hyponatremia and hypovolemia, respectively. Both entities are associated with increased risk of DCI and worse functional outcomes.99 Unfortunately, routine fluid balance and vital signs are poor markers of intravascular fluid status in this population, and advanced hemodynamic monitoring may be required.100 Strategies currently advocated are avoidance of hypotonic solutions; the use of isotonic (e.g., normal saline) or hypertonic solutions (e.g., 3% saline—especially if hyponatremia is present); and consideration of fludrocortisone in patients with persistent negative fluid balance. Finally, once the ruptured aneurysm has been secured, BP should not be reduced in the subsequent weeks.

Monitoring and Management of Symptomatic Vasospasm/Delayed Cerebral Ischemia

More than 60% of patients with aSAH will demonstrate vasospasm on CT or ultrasound, but only about 30% will become symptomatic.89,101 Many monitoring techniques, including frequent neurologic examinations, daily transcranial Doppler, CTA and CT perfusion, and multimodal physiologic monitoring (brain tissue oxygenation, microdialysis, jugular oximetry, continuous EEG), are currently available and should be implemented during the first 2 weeks after aSAH.89

In case of acute neurologic deterioration (decrease in two or more GCS points or increase in two or more NIHSS points), confounding factors—such as fever, hyponatremia, infection, or seizures—should be immediately ruled out, and the patient should be promptly treated for presumptive DCI. Historically, triple-H therapy (hypertension, hypervolemia, and hemodilution) was considered standard of care for these patients; recent studies, however, have shown no additional benefit, and an increased complication rate, with hypervolemia when compared to euvolemia. Therefore, current guidelines suggest maintenance of euvolemia followed by induced hypertension with vasopressors (hemodynamic augmentation). No specific BP level has been defined; each patient should be managed in a stepwise approach with assessment of neurologic function at each SBP or MAP level. If the neurologic deficit does not reverse with aggressive hemodynamic augmentation, urgent angiography should be considered for angioplasty and/or intra-arterial infusion of vasodilators.

Identification and Management of Medical Complications

Finally, systemic complications are very common in the aSAH population, including fever (54%), anemia (36%), hyperglycemia (30%), pneumonia (20%), and pulmonary edema (14%). Hyperglycemia, fever, and anemia are significantly associated with higher mortality and worse functional outcome.102 Interestingly, there is considerable uncertainty regarding anemia management in aSAH. Some studies suggest a risk of worsened outcomes with packed RBC transfusion, and there is no agreement on optimal transfusion threshold. Transfusion criteria for general medical patients (Hgb < 7 g/dL) are, however, considered inadequate, and guidelines support packed RBC transfusion to maintain hemoglobin concentration above 8 g/dL.112


ICH and SAH are diseases that both result in a high rate of morbidity and mortality. In the case of ICH, early, aggressive, and structured management of factors that cause secondary brain injury is essential for optimizing outcomes. Although there is much that is unknown, optimal outcomes result from management in experienced ICUs and, in particular, ICUs dedicated to neurocritical care. In the case of aSAH, the greatest challenges are avoiding misdiagnosis and preventing complications of vasospasm. Accurate diagnosis has improved with advanced imaging (CTA and MRI) but still requires a low threshold for lumbar puncture. Perhaps, someday there will be a “troponin” for SAH, but until then, vigilance and aggressive pursuit of the diagnosis are essential.


CI, confidence interval; OR, odds ratio.


1.van Asch CJ, Luitse MJ, Rinkel GJ, et al. Incidence, case fatality, and functional outcome of intracerebral haemorrhage over time, according to age, sex, and ethnic origin: a systematic review and meta-analysis. Lancet Neurol. 2010;9(2):167–176.

2.Saloheimo P, Lapp TM, Juvela S, et al. The impact of functional status at three months on long-term survival after spontaneous intracerebral hemorrhage. Stroke. 2006;37(2):487–491.

3.Zahuranec DB, Gonzales NR, Brown DL, et al. Presentation of intracerebral haemorrhage in a community. J Neurol Neurosurg Psychiatry. 2006;77(3):340–344.

4.Zahuranec DB, Morgenstern LB, Sánchez BN, et al. Do-not-resuscitate orders and predictive models after intracerebral hemorrhage. Neurology. 2010;75(7):626–633.

5.Becker KJ, Baxter AB, Cohen WA, et al. Withdrawal of support in intracerebral hemorrhage may lead to self-fulfilling prophecies. Neurology. 2001;56(6):766–772.

6.Tirschwell DL, Becker KJ, Creutzfeldt CJ, et al. Propensity score matching to estimate supported outcomes in intracerebral hemorrhage patients with withdrawal of life support. In: AHA, ed. International Stroke Conference; Honolulu, Hawaii. Stroke. 2013;44:ATMP83.

7.Rordorf G, McDonald C. Spontaneous intracerebral hemorrhage: pathogenesis, clinical features, and diagnosis. In: Kasner SE, ed. Waltham, MA: UpToDate; 2013.

8.Flaherty ML, Kissela B, Woo D, et al. The increasing incidence of anticoagulant-associated intracerebral hemorrhage. Neurology. 2007;68(2):116–121.

9.Ariesen MJ, Claus SP, Rinkel GJ, et al. Risk factors for intracerebral hemorrhage in the general population: a systematic review. Stroke. 2003;34(8):2060–2065.

10.Sturgeon JD, Folsom AR, Longstreth WT Jr, et al. Risk factors for intracerebral hemorrhage in a pooled prospective study. Stroke. 2007;38(10):2718–2725.

11.Labovitz DL, Halim A, Boden-Albala B, et al. The incidence of deep and lobar intracerebral hemorrhage in whites, blacks, and Hispanics. Neurology. 2005;65(4):518–522.

12.Feigin VL, Lawes CM, Bennett DA, et al. Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol. 2009;8(4):355–369.

13.Brown RD Jr, Wiebers DO, Torner JC, et al. Frequency of intracranial hemorrhage as a presenting symptom and subtype analysis: a population-based study of intracranial vascular malformations in Olmsted Country, Minnesota. J Neurosurg. 1996;85(1):29–32.

14.Licata B, Turazzi S. Bleeding cerebral neoplasms with symptomatic hematoma. J Neurosurg Sci. 2003;47(4):201–210; discussion 210.

15.Katz JM, Segal AZ. Incidence and etiology of cerebrovascular disease in patients with malignancy. Curr Atheroscler Rep. 2005;7(4):280–288.

16.Brott T, Broderick J, Kothari R, et al. Early hemorrhage growth in patients with intracerebral hemorrhage. Stroke. 1997;28(1):1–5.

17.Hemphill JC, Andrews P, De Georgia M. Multimodal monitoring and neurocritical care bioinformatics. Nat Rev Neurol. 2011;7(8):451–460.

18.Vincent JL. Textbook of Critical Care. 6th ed. Philadelphia, PA: Elsevier/Saunders; 2011: xli, 1698.

19.Claassen J, Jette N, Chum F, et al. Electrographic seizures and periodic discharges after intracerebral hemorrhage. Neurology. 2007;69(13):1356–1365.

20.Sahni R, Weinberger J. Management of intracerebral hemorrhage. Vasc Health Risk Manag. 2007;3(5):701–709.

21.Leira R, Davalos A, Silva Y, et al. Early neurologic deterioration in intracerebral hemorrhage: predictors and associated factors. Neurology. 2004;63(3):461–467.

22.Andrews CM, Jauch EC, Hemphill JC III, et al. Emergency neurological life support: intracerebral hemorrhage. Neurocrit Care. 2012;17(suppl 1):S37–S46.

23.Kothari RU, Brott T, Broderick JP, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke. 1996;27(8):1304–1305.

24.Beslow LA, Ichord RN, Kasner SE, et al. ABC/XYZ estimates intracerebral hemorrhage volume as a percent of total brain volume in children. Stroke. 2010;41(4):691–694.

25.Goldstein JN, Fazen LE, Snider R, et al. Contrast extravasation on CT angiography predicts hematoma expansion in intracerebral hemorrhage. Neurology. 2007;68(12):889–894.

26.Wada R, Aviv RI, Fox AJ, et al. CT angiography “spot sign” predicts hematoma expansion in acute intracerebral hemorrhage. Stroke. 2007;38(4):1257–1262.

27.Zhu XL, Chan MS, Poon WS. Spontaneous intracranial hemorrhage: which patients need diagnostic cerebral angiography? A prospective study of 206 cases and review of the literature. Stroke. 1997;28(7):1406–1409.

28.Hemphill JC III, Bonovich DC, Besmertis L, et al. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke. 2001;32(4):891–897.

29.Hemphill JC III, Farrant M, Neill TA Jr. Prospective validation of the ICH Score for 12-month functional outcome. Neurology. 2009;73(14):1088–1094.

30.Morgenstern LB, Hemphill JC III, Anderson C, et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2010;41(9):2108–2129.

31.Seder DB, Riker RR, Jagoda A, et al. Emergency neurological life support: airway, ventilation, and sedation. Neurocrit Care. 2012;17(suppl 1):S4–S20.

32.Carlberg B, Asplund K, Hagg E. The prognostic value of admission blood pressure in patients with acute stroke. Stroke. 1993;24(9):1372–1375.

33.Guyatt GH, Akl EA, Crowther M, et al. Executive summary: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(Suppl 2):7S–47S.

34.Degos V, Westbroek EM, Lawton MT, et al. Perioperative management of coagulation in nontraumatic intracerebral hemorrhage. Anesthesiology. 2013;119(1):218–227.

35.He J, Whelton PK, Vu B, et al. Aspirin and risk of hemorrhagic stroke: a meta-analysis of randomized controlled trials. JAMA. 1998;280(22):1930–1935.

36.Gorelick PB, Weisman SM. Risk of hemorrhagic stroke with aspirin use: an update. Stroke. 2005;36(8):1801–1807.

37.Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. N Engl J Med. 1995;333(24):1581–1587.

38.Yuan ZH, Jiang JK, Huang WD, et al. A meta-analysis of the efficacy and safety of recombinant activated factor VII for patients with acute intracerebral hemorrhage without hemophilia. J Clin Neurosci. 2010;17(6):685–693.

39.Balami JS, Buchan AM. Complications of intracerebral haemorrhage. Lancet Neurol. 2012;11(1):101–118.

40.Godoy DA, Pinero GR, Svampa S, et al. Hyperglycemia and short-term outcome in patients with spontaneous intracerebral hemorrhage. Neurocrit Care. 2008;9(2):217–229.

41.Qureshi AI, Palesch YY, Martin R, et al. Association of serum glucose concentrations during acute hospitalization with hematoma expansion, perihematomal edema, and three month outcome among patients with intracerebral hemorrhage. Neurocrit Care.2011;15(3):428–435.

42.Mendelow AD, Gregson BA, Rowan EN, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas (STICH II): a randomised trial. Lancet. 2013;382(9890):397–408.

43.Hallevi H, Albright KC, Aronowski J, et al. Intraventricular hemorrhage: anatomic relationships and clinical implications. Neurology. 2008;70(11):848–852.

44.Brain Trauma Foundation; American Association of Neurological Surgeons; Congress of Neurological Surgeons. Guidelines for the management of severe traumatic brain injury. J Neurotrauma 2007;24(suppl 1):S1–S106.

45.Messe SR, Sansing LH, Cucchiara BL, et al. Prophylactic antiepileptic drug use is associated with poor outcome following ICH. Neurocrit Care. 2009;11(1):38–44.

46.Naidech AM, Garg RK, Liebling S, et al. Anticonvulsant use and outcomes after intracerebral hemorrhage. Stroke. 2009;40(12):3810–3815.

47.Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360(13):1283–1297.

48.Lacut K, Bressollette L, Le Gal G, et al. Prevention of venous thrombosis in patients with acute intracerebral hemorrhage. Neurology. 2005;65(6):865–869.

49.Diringer MN, Edwards DF. Admission to a neurologic/neurosurgical intensive care unit is associated with reduced mortality rate after intracerebral hemorrhage. Crit Care Med. 2001;29(3):635–640.

50.Kowalski RG, Claassen J, Kreiter KT, et al. Initial misdiagnosis and outcome after subarachnoid hemorrhage. JAMA. 2004;291(7):866–869.

51.Brisman JL, Song JK, Newell DW. Cerebral aneurysms. N Engl J Med. 2006;355(9):928–939.

52.Morita A, Kirino T, Hashi K, et al. The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med. 2012;366(26):2474–2482.

53.Juvela S. Prehemorrhage risk factors for fatal intracranial aneurysm rupture. Stroke. 2003;34(8):1852–1857.

54.Frontera JA, Fernandez A, Claassen J, et al. Hyperglycemia after SAH: predictors, associated complications, and impact on outcome. Stroke. 2006;37(1):199–203.

55.Ohkuma H, Tsurutani H, Suzuki S. Incidence and significance of early aneurysmal rebleeding before neurosurgical or neurological management. Stroke. 2001;32(5):1176–1180.

56.Kim HC, Nam CM, Jee SH, et al. Comparison of blood pressure-associated risk of intracerebral hemorrhage and subarachnoid hemorrhage: Korea Medical Insurance Corporation study. Hypertension. 2005;46(2):393–397.

57.Wijdicks EF, Kallmes DF, Manno EM, et al. Subarachnoid hemorrhage: neurointensive care and aneurysm repair. Mayo Clin Proc. 2005;80(4):550–559.

58.Johnston SC, Selvin S, Gress DR. The burden, trends, and demographics of mortality from subarachnoid hemorrhage. Neurology. 1998;50(5):1413–1418.

59.Juvela S, Porras M, Poussa K. Natural history of unruptured intracranial aneurysms: probability of and risk factors for aneurysm rupture. J Neurosurg. 2008;108(5):1052–1060.

60.Vlak MH, Rinkel GJ, Greebe P, et al. Independent risk factors for intracranial aneurysms and their joint effect: a case–control study. Stroke. 2013;44(4):984–987.

61.Caranci F, Briganti F, Cirillo L, et al. Epidemiology and genetics of intracranial aneurysms. Eur J Radiol. 2013;82(10):1598–1605.

62.Perry JJ, Stiell IG, Sivilotti ML, et al. High risk clinical characteristics for subarachnoid haemorrhage in patients with acute headache: prospective cohort study. BMJ. 2010;341:c5204.

63.Hunt WE, Hess RM. Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg. 1968;28(1):14–20.

64.Report of World Federation of Neurological Surgeons Committee on a Universal Subarachnoid Hemorrhage Grading Scale. J Neurosurg. 1988;68(6):985–986.

65.Edlow JA, Panagos PD, Godwin SA, et al. Clinical policy: critical issues in the evaluation and management of adult patients presenting to the emergency department with acute headache. Ann Emerg Med. 2008;52(4):407–436.

66.Perry JJ, Stiell I, Wells G, et al. Diagnostic test utilization in the emergency department for alert headache patients with possible subarachnoid hemorrhage. CJEM. 2002;4(5):333–337.

67.Morgenstern LB, Huber JC, Luna-Gonzales H, et al. Headache in the emergency department. Headache. 2001;41(6):537–541.

68.Perry JJ, Stiell IG, Sivilotti ML, et al. Clinical decision rules to rule out subarachnoid hemorrhage for acute headache. JAMA. 2013;310(12):1248–1255.

69.Perry JJ, Spacek A, Forbes M, et al. Is the combination of negative computed tomography result and negative lumbar puncture result sufficient to rule out subarachnoid hemorrhage? Ann Emerg Med. 2008;51(6):707–713.

70.van Gijn J, Kerr RS, Rinkel GJ. Subarachnoid haemorrhage. Lancet. 2007;369(9558):306–318.

71.Cortnum S, Sorensen P, Jorgensen J. Determining the sensitivity of computed tomography scanning in early detection of subarachnoid hemorrhage. Neurosurgery. 2010;66(5):900–902; discussion 903.

72.Byyny RL, Mower WR, Shum N, et al. Sensitivity of noncontrast cranial computed tomography for the emergency department diagnosis of subarachnoid hemorrhage. Ann Emerg Med. 2008;51(6):697–703.

73.Perry JJ, Stiell IG, Sivilotti ML, et al. Sensitivity of computed tomography performed within six hours of onset of headache for diagnosis of subarachnoid haemorrhage: prospective cohort study. BMJ. 2011;343:d4277.

74.McCormack RF, Hutson A. Can computed tomography angiography of the brain replace lumbar puncture in the evaluation of acute-onset headache after a negative noncontrast cranial computed tomography scan? Acad Emerg Med. 2010;17(4):444–451.

75.Agid R, Andersson T, Almqvist H, et al. Negative CT angiography findings in patients with spontaneous subarachnoid hemorrhage: when is digital subtraction angiography still needed? AJNR Am J Neuroradiol. 2010;31(4):696–705.

76.Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery. 1980;6(1):1–9.

77.Claassen J, Bernardini GL, Kreiter K, et al. Effect of cisternal and ventricular blood on risk of delayed cerebral ischemia after subarachnoid hemorrhage: the Fisher scale revisited. Stroke. 2001;32(9):2012–2020.

78.Huang J, van Gelder JM. The probability of sudden death from rupture of intracranial aneurysms: a meta-analysis. Neurosurgery. 2002;51(5):1101–1105; discussion 1105–1107.

79.Fujii Y, Takeuchi S, Sasaki O, et al. Ultra-early rebleeding in spontaneous subarachnoid hemorrhage. J Neurosurg. 1996;84(1):35–42.

80.Naidech AM, Janjua N, Kreiter KT, et al. Predictors and impact of aneurysm rebleeding after subarachnoid hemorrhage. Arch Neurol. 2005;62(3):410–416.

81.Starke RM, Connolly ES Jr. Rebleeding after aneurysmal subarachnoid hemorrhage. Neurocrit Care. 2011;15(2):241–246.

82.Connolly ES Jr, Rabinstein AA, Carhuapoma JR, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2012;43(6):1711–1737.

83.Diringer MN, Bleck TP, Claude Hemphill J, et al. Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the Neurocritical Care Society's Multidisciplinary Consensus Conference. Neurocrit Care. 2011;15(2):211–240.

84.Edlow JA, Samuels O, Smith WS, et al. Emergency neurological life support: subarachnoid hemorrhage. Neurocrit Care. 2012;17(suppl 1):S47–S53.

85.Bruder N, Rabinstein A. Cardiovascular and pulmonary complications of aneurysmal subarachnoid hemorrhage. Neurocrit Care. 2011;15(2):257–269.

86.Wijdicks EF, Vermeulen M, Murray GD, et al. The effects of treating hypertension following aneurysmal subarachnoid hemorrhage. Clin Neurol Neurosurg. 1990;92(2):111–117.

87.Suarez JI, Tarr RW, Selman WR. Aneurysmal subarachnoid hemorrhage. N Engl J Med. 2006;354(4):387–396.

88.Schiefecker AJ, Pfausler B, Beer R, et al. Parenteral diclofenac infusion significantly decreases brain-tissue oxygen tension in patients with poor-grade aneurysmal subarachnoid hemorrhage. Crit Care. 2013;17(3):R88.

89.Rabinstein AA, Lanzino G, Wijdicks EF. Multidisciplinary management and emerging therapeutic strategies in aneurysmal subarachnoid haemorrhage. Lancet Neurol. 2010;9(5):504–519.

90.Hillman J, Fridriksson S, Nilsson O, et al. Immediate administration of tranexamic acid and reduced incidence of early rebleeding after aneurysmal subarachnoid hemorrhage: a prospective randomized study. J Neurosurg. 2002;97(4):771–778.

91.Hellingman CA, van den Bergh WM, Beijer IS, et al. Risk of rebleeding after treatment of acute hydrocephalus in patients with aneurysmal subarachnoid hemorrhage. Stroke. 2007;38(1):96–99.

92.Ransom ER, Mocco J, Komotar RJ, et al. External ventricular drainage response in poor grade aneurysmal subarachnoid hemorrhage: effect on preoperative grading and prognosis. Neurocrit Care. 2007;6(3):174–180.

93.Mak CH, Lu YY, Wong GK. Review and recommendations on management of refractory raised intracranial pressure in aneurysmal subarachnoid hemorrhage. Vasc Health Risk Manag. 2013;9:353–359.

94.Naidech AM, Kreiter KT, Janjua N, et al. Phenytoin exposure is associated with functional and cognitive disability after subarachnoid hemorrhage. Stroke. 2005;36(3):583–587.

95.Vespa P, Diringer MN. High-volume centers. Neurocrit Care. 2011;15(2):369–372.

96.Molyneux A, Kerr R, Stratton I, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet. 2002;360(9342):1267–1274.

97.Spetzler RF, McDougall CG, Albuquerque FC, et al. The Barrow Ruptured Aneurysm Trial: 3-year results. J Neurosurg. 2013;119(1):146–157.

98.Dorhout Mees SM, Rinkel GJ, Feigin VL, et al. Calcium antagonists for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev. 2007(3):CD000277.

99.Rabinstein AA, Bruder N. Management of hyponatremia and volume contraction. Neurocrit Care. 2011;15(2):354–360.

100.Gress DR. Monitoring of volume status after subarachnoid hemorrhage. Neurocrit Care. 2011;15(2):270–274.

101.Rowland MJ, Hadjipavlou G, Kelly M, et al. Delayed cerebral ischaemia after subarachnoid haemorrhage: looking beyond vasospasm. Br J Anaesth. 2012;109(3):315–329.

102.Wartenberg KE, Schmidt JM, Claassen J, et al. Impact of medical complications on outcome after subarachnoid hemorrhage. Crit Care Med. 2006;34(3):617–623; quiz 624.

103.Anderson CS, Huang Y, Wang JG, et al. Intensive blood pressure reduction in acute cerebral haemorrhage trial (INTERACT): a randomised pilot trial. Lancet Neurol. 2008;7(5):391–399.

104.Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATACH) investigators. Antihypertensive treatment of acute cerebral hemorrhage. Crit Care Med. 2010;38(2):637–648.

105.Butcher KS, Jeerakathil T, Hill M, et al. The intracerebral hemorrhage acutely decreasing arterial pressure trial. Stroke. 2013;44(3):620–626.

106.Anderson CS, Heeley E, Huang Y, et al. Rapid blood-pressure lowering in patients with acute intracerebral hemorrhage. N Engl J Med. 2013;368(25):2355–2365.

107.Mayer SA, Brun NC, Begtrup K, et al. Efficacy and safety of recombinant activated factor VII for acute intracerebral hemorrhage. N Engl J Med. 2008;358(20):2127–2137.

108.Mendelow AD, Gregson BA, Fernandes HM, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. Lancet.2005;365(9457):387–397.

109.Pickard JD, Murray GD, Illingworth R, et al. Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. BMJ. 1989;298(6674):636–642.

110.Lennihan L, Mayer SA, Fink ME, et al. Effect of hypervolemic therapy on cerebral blood flow after subarachnoid hemorrhage: a randomized controlled trial. Stroke. 2000;31(2):383–391.

111.Egge A, Waterloo K, Sjoholm H, et al. Prophylactic hyperdynamic postoperative fluid therapy after aneurysmal subarachnoid hemorrhage: a clinical, prospective, randomized, controlled study. Neurosurgery. 2001;49(3):593–605; discussion 605–596.

112.Le Roux PD. Anemia and transfusion after subarachnoid hemorrhage. Neurocrit Care. 2011;15(2):342–353.