10.1 Introduction

Despite of the advances in the diagnosis and treatment of acute stroke, it remains one of the leading causes of mortality and disability in the developed world. Ischemic stroke represents approximately 80% of all types of strokes. In the last two decades the acute management of stroke patients has evolved to stroke unit and stroke centers with improved outcomes, including reduced in-hospital length of stay and decreased mortality. ¹ Up to 20% of these patients benefit from neurocritical care. ² Several studies have demonstrated the beneficial effect of specialized neurocritical care (NCC) team intervention in these patients. ³

The critical care of the Acute Ischemic Stroke (AIS) patients must be developed by a multidisciplinary team, and is based on vital support, early recognition of neuroworsening, optimizing cerebral perfusion, minimizing the reperfusion injury, and finally avoid and treat neurologic and systemic complications. ⁴

In this chapter, we will review the clinical aspects of critical care of AIS patients, with emphasis in a practical and evidence based approach.

10.2 Indications for Admission to the Intensive Care Unit

Patients suffering an AIS require NCC due to several situations. ^{5 ,​ 6} Although there is variability in the criteria for intensive care unit (ICU) admission, the following indications are accepted for the majority of the authors (Table 10‑1):

1. 
   

   Severe neurological deficits with the need of artificial airway and mechanical ventilation due to depressed level of consciousness or reduced ability to protect their airway,

   
   
2. 
   

   Risk of neurological worsening, like: severe stroke score (NIHSS > 17), large hemispheric infarcts (>145 cc), significant infarcts involving the brainstem or cerebellum, significant mass effect or brain shift in the computed tomography.

   
   
3. 
   

   Need to specialized non-invasive or invasive neurologic monitoring or specific therapy for intracranial hypertension or cerebral edema.

   
   
4. 
   

   Generalized seizures or status epilepticus.

   
   
5. 
   

   Post-surgical interventions: decompressive craniectomy, evacuation of cerebellar infarct or ventriculostomy for hydrocephalus in posterior fossa infarcts.

   
   
6. 
   

   Patients with significant cardiovascular, respiratory or systemic dysfunction due to comorbidities or complications: pulmonary edema or embolism, acute heart failure, hypotension that requires vasopressors, severe arterial hypertension with the need of intravenous medications, arrhythmias, aortic dissection or metabolic abnormalities that require specific close monitoring.

   
   
7. 
   

   Immediate Post-thrombolysis or post-thrombectomy for hemodynamic and neurologic monitoring.

In these group of patients, as it was previously mentioned, specialized NCC units have shown to improve outcomes and decrease their length of stay. ^{7 ,​ 8} Another important topic of these patients is that delay in their admission to an ICU was significantly associated with a worse outcome, so they should be admitted as soon as possible. ⁹

10.3 Neuroworsening: Detection of Neurological Deterioration

Neurological deterioration or neuroworsening constitutes a cornerstone in the management of patients suffering AIS. Neurological examination is the most important tool to detect it, since intracranial pressure (ICP) monitoring, have limitations in detect and predict neurological deterioration after ischemic stroke. ^{10 ,​ 11} Neuroworsening after stroke commonly is due to one four mechanisms: 1) Brain edema with tissue shift, which does not necessarily cause elevated ICP but can cause an herniation syndrome, 2) Hemorrhagic transformation with worsening of mass effect and potentially elevated ICP, 3) Collateral cerebral blood flow failure with extension of the ischemic brain, and 4) seizures which can elicit a reduction in the level of consciousness with or without worsening brain edema and tissue shift.

The National Institutes of Health Stroke Scale (NHISS) is the most common standardized and efficient score used to evaluate the clinical status of AIS. An increase of two or more points is considered a threshold for significant neuroworsening. In patients with severe consciousness compromise, the specifically designed Glasgow Coma Scale (GCS) may be more useful. ^{5 ,​ 12} Additional score that is useful for neurological monitoring is the Full Outline of Unresponsiveness Score (FOUR Score), which permits a detailed assessment of language, motor, brainstem and breathing functions. The use of the FOUR score is now recommended for patients with stupor and/or coma.

The evaluation of pupillary size and reactivity to light is a fundamental part of the neurological examination, which is crucial in patients who are deteriorating due to cerebral midline shift or transtentorial herniation. The utilization of the infrared pupillometry, provide a non-invasive and bedside method, which eliminates limitations of the manual way of pupillary clinical assessment, like subjectivity, inconsistency and highly inter-observer variability. We use the portable infrared pupilometer (Forsite NeurOptics, Irvine, CA ®), which allows an automated, accurate, easy, quick, reproducible and quantitative measure of different pupillary parameters including: maximum and minimum apertures, percentage of aperture change following light stimulation, constriction and dilation velocities, and latency period ^{13 ,​ 14}

Routine ICP monitoring is not recommended in AIS, because significant tissue and brain herniation may occur in the absence of elevated ICP. Hence intracranial hypertension is a late sign of neuroworsening and ICP values are often normal even in the presence of large ischemic tissue volume. There are no randomized clinical trials evaluating ICP monitoring in stroke patients. ^{3 ,​ 15}

In a prospective study, Poca et al found that the majority of patients with malignant middle cerebral artery infarct had ICP values less than 20 mmHg. In this study, pupillary abnormalities, cerebral midline shift and brainstem compression were present despite normal ICP values. ¹¹ The relative value of ICP monitoring in predicting neuroworsening could be explained because the stroke is a focal lesion, and the compartmentalized ICP phenomenon may not reflect the pressure in the contralateral hemisphere or infratentorium. Other hypothesis could be that the distance between the ICP probe site and the brain herniation site, in generally the tentorial notch, is inversely proportional to the pressure gradient force. This is very important because we know now that there are morphometric anatomic variations of the structure of the tentorial notch, which could explain why some patients who have a narrow subtype of tentorial notch deteriorates earlier than others who have a large one. ¹⁶ We have recently developed a computed tomographic protocol to measure the tentorial notch, so we could be able to predict which stroke patient are at risk of neuroworsening based on their morphometric anatomic variation of tentorial notch (preliminary data, not shown).

Having said this, there is a place for the use of ventriculostomy for the treatment of obstructive hydrocephalus complicating cerebellar stroke, subarachnoid hemorrhage, cerebellar hemorrhage, and primary intraparenchymal hemorrhage (ICH). Ventriculostomy use to drain cerebrospinal fluid and relieve hydrocephalus is a treatment option in these conditions. However, posterior fossa brain edema from cerebellar stroke is best treated by surgical craniectomy and decompression rather than by ventriculostomy alone.

Invasive multimodal monitoring, including ptO2 or cerebral microdialysis, have not been sufficiently studied in AIS patients, so the routine use of them cannot be recommended at present. ³

Noninvasive methods of neuromonitoring are becoming a complimentary tool to clinical examination and to predict malignant course of AIS patients:

1. 
   

   Ultrasound: Measure of the optic nerve sheath diameter (ONSD) is an accurate, simple and rapid method for detecting ICP immediate changes and elevated. ¹⁷ Recently, it was described a reliable assessment of cerebral midline shift by transcranial color-coded duplex. It has the advantage of bedside available and favorable safety profile, especially in unstable patients in whom performing serial CT scans expose. ¹⁸

   
   
2. 
   

   Continuous Electroencephalography (cEEG) monitoring. There are limiting data regarding the utility of the cEEG monitoring in predict or manage AIS patients. Resent studies using continuous and quantitative EEG monitoring in ischemic stroke patients, showed a good correlation between loss of fast EEG activity and low CPP, and between brain symmetry index and NIHSS. This could be a promising non-invasive monitoring technique to estimate or predict the outcome after AIS. ^{19 ,​ 20}

10.4 Airway Management and Mechanical Ventilation

Management of the airway is the first step of the medical support of any severe brain injured patient. It is part of the initial triad A, B, C algorithm: airway, breathing and circulation.

Endotracheal intubation is indicated in patients with: a) Decreased level of consciousness; b) Inability to airway protection due to brainstem dysfunction; c) Signs of intracranial hypertension or very large infarcts; d) Any respiratory failure that requires mechanical ventilation.

There are no evidence-based guidelines for specific indication or timing for intubation in AIS. Although, GCS score < 9 is taken as a limit to intubate the decision must be guided by global clinical judgment. In this sense, patients that cannot follow commands due to decreased level of consciousness are candidates to endotracheal intubation, independently of the GCS score. ^{3 ,​ 6 ,​ 10} Intubation within the first 48 hs in large territory ischemic strokes and with low GCS would be reasonable, since that is the window of neuroworsening.

The different guidelines recommend providing supplemental oxygen to keep the oxygen saturation level above 94%. Routine supplemental oxygen for all stroke patients or hyperoxia have not demonstrated a significant benefit in randomized clinical studies. ^{21 ,​ 22 ,​ 23}

Hyperventilation should only be used for short periods of time as a rescue therapeutic intervention to reduce ICP in AIS patients with significant cerebral edema and clinical signs of brain herniation. Due to worsening ischemic injury from vasoconstriction, which elicits further ischemia, prophylactic or routine hyperventilation did not demonstrate benefit of outcome in these patients and is not recommended. ^{3 ,​ 24}

Respiratory insufficiency occurs in approximately 10 – 20% of AIS patients. There are several causes for it: central hypoventilation due to brainstem infarctions, aspiration pneumonitis or pneumonia, acute lung injury or acute respiratory distress syndrome, cardiogenic or neurogenic pulmonary edema, pulmonary embolism, or respiratory muscle weakness. ^{2 ,​ 4}

Noninvasive ventilation should not be used in these patients, because it does not correct underlying problems that frequently present in AIS patients such as central hypoventilation and inability of airway protection due to brainstem dysfunction.

Weaning mechanical ventilation in a severe stroke patient is fundamentally different than the typical non-neurologic ICU patient with respiratory failure. The typical weaning parameters of vital capacity, negative inspiratory force, maximal expiratory force, spontaneous breathing trials, etc do not measure airway control. Airway control and airway protective reflexes are often impaired in severe acute ischemic stroke. While spontaneous breathing trials (SBTs) can be safely done in some patients, SBTs should be avoided in stroke patients with critical brain edema. Gradual weaning of the ventilator rate may be a more cautious approach.

There are some criteria to consider in AIS patients prior to extubation: follow more than one command, successful of spontaneous breathing test, absence of oro-pharyngeal saliva collections, and adequate cough effort measured by the white card test. ^{25 ,​ 26}

Tracheostomy should be considered in AIS patients who failed extubation or in whom this is not possible by 7–10 days of mechanical ventilation, typically patients with large MCA or posterior fossa infarcts. The optimal timing for tracheostomy is still matter of debate and is controversial whether early tracheostomy impacts upon AIS patient outcome. ^{27 ,​ 28}

In this sense, Schonenberger et al. developed the SET score, which evaluates three areas of assessment: neurological function (dysphagia, observed aspiration and GCS on admission < 10), type and localization of neurological lesion (brainstem, cerebellar or MCA territory), and general organ function (acute lung injury, APACHE II score > 20, sepsis or neurosurgical intervention). They demonstrated in a prospectively study that a SETscore cut-off value of > 8 points predicts prolonged mechanical ventilation and tracheostomy need with a sensitivity of 64% and a specificity of 86%. ²⁹

Bosel et al. found in a randomized pilot trial of ischemic and hemorrhagic stroke patients, that early tracheostomy group (performed 1–3 days after intubation) had lower ICU mortality (10 vs. 47%) than standard tracheostomy group patients (7–14 days from intubation). ³⁰ Following this line, the SETPOINT2 is an ongoing multicenter randomized trial in which patients with ischemic or hemorrhagic stroke on mechanical ventilation are randomized to early percutaneous tracheostomy (first five days after intubation) or prolonged orotracheal intubation. This trial might clarify the value of early tracheostomy in these patients. ³¹

10.5 Hemodynamic Management

The main objectives of the hemodynamic management of AIS patients are:

1. 
   

   Avoid hypovolemia and hypotension

   
   
2. 
   

   Hypertension control in certain situations

   
   
3. 
   

   Early recognition and treat cardiac complications.

Basic cardiovascular monitoring includes continuous electrocardiography, noninvasive blood pressure (BP), at least once echocardiography and repeated troponin measures, especially in patients with electrocardiographic or echocardiographic abnormalities.

Patients who are hemodynamic unstable or develop significant cardiac complications, benefit of invasive BP monitoring and might require minimally invasive hemodynamic monitoring such as PiCCO or VIGILANCE ® dispositive, which measure continuous cardiac output, systemic vascular resistances and SvO2.

There is not a specific goal for blood pressure range for AIS patients. Hypovolemia and hypotension must be avoided because they could exacerbate brain ischemia, especially in chronic hypertensive patients in which the autoregulation curve shifts to the right. A mean arterial pressure (MAP) > 85 mmHg appears a reasonable goal in ischemic stroke without hemorrhagic transformation.

Fluid balance should be carefully monitored and managed to achieve euvolemia. Fluid resuscitation with isotonic saline solutions (NaCl 0.9%) is the first step of this management, following by vasoactive agents such as norepinephrine if hypotension is unresponsive to volume replacement. Cautious use of balanced crystalloid solutions may be option but avoidance of hyponatremia is paramount in considering fluid selection.

Almost 80% of total AIS patients are hypertensive upon arrival at the hospital emergency and generally normalize over the first 48 hours. This may be related to several reasons: chronic hypertension, stress, pain or agitation or intracranial hypertension. ³² It was observed a U-shaped relationship between BP and mortality in stroke, with adverse effects on outcome associated to systolic blood pressure (SBP) bellow and above 150 mmHg. ³³ BP lowering may exacerbate neurological deterioration by decreasing cerebral perfusion pressure in penumbra tissue when cerebral autoregulation is impaired. On the other hand, severe hypertension can lead to increased brain edema formation and intracerebral hemorrhage due to breakdown of blood-brain barrier, and contribute to cardio-respiratory and renal complications.

This fact, added to the inconsistent evidence of an optimal BP level, makes the management of BP in AIS patients a controversial issue. Clinical guidelines recommend a strategy of permissive hypertension, and BP is allowed to rise as high as 220/120 mm Hg before treatment is started, unless the patient develops cardiac or renal dysfunction due to hypertension (myocardial ischemia, congestive heart or renal failure). In patients who receive intravenous thrombolytic therapy, a strict BP control is needed to keep BP bellow 180/105 mmHg. ^{3 ,​ 10 ,​ 22} Intravenous labetalol is the agent of choice to lower BP, and nicardipine could be a reasonable option when beta blockers are contraindicated. ³⁴ It must be noted that BP variability in the early phase of ischemic stroke is associated with infarct expansion and worse outcome, so it should be avoided. ³⁵

There are at least five randomized clinical trials that studied acute BP lowering in nonthrombolyzed ischemic stroke. Two of them (the CHHIPS and COSSACS trial) where small and underpowered, and did not find substantial differences in outcome between treat or placebo groups. ^{36 ,​ 37} After, three large randomized clinical trials were developed. The SCAST studied enrolled 2029 patients in nine European countries and found that reducing BP with candesartan in the first 7 days of stroke had no beneficial effects on outcome. ³⁸ The CATIS trial was a randomized studied that enrolled 4071 AIS patients in China, and found no reduction in death or major disability. ³⁹ Finally, the ENOS trial, a partial-factorial randomized study of the effect of transdermal glyceryl trinitrate to lowering BP in the first 7 days of ischemic or hemorrhagic stroke, did not improve outcome in these patients. ⁴⁰

Finally, the ENCHANTED trial is an ongoing randomized study to establish the effect of low-dose rtPA and early intensive blood pressure lowering in AIS patients, that might clear out this problem in the next few years ⁴¹ .

There is no strong evidence for application of induced hypertension in AIS patients.

However, in highly selected patients (with fluctuating or deteriorating neurological status, severely stenotic or occluded major vessel and SBP < 150 mmHg), a strategy of induced hypertension could be carefully applied. ^{10 ,​ 42} In this situation, measuring dynamic cerebral autoregulation by transcranial Doppler, like the autoregulation index (ARI) or Mx index, could help to guide hemodynamic management decisions. It is generally accepted that cerebral autoregulation is altered in ischemic and hemorrhagic stroke, and also may change with time and pathological progression. The presence or absence of cerebral autoregulation in acute stroke is critical to maintenance of stable blood flow in the ischemic penumbra and for avoidance of excessive hyperperfusion. It is possible that the subpopulation of AIS patients with relative intact cerebral autoregulation may benefit from aggressive blood pressure treatment to improve clinical outcome. ⁴³ Prospective studies are needed, so today; the practice should be made at individualizing care with the discretion of neurointensivist and neurologist.

There are no specific studies about management of blood pressure in the peri-procedural endovascular setting, such as post mechanical thrombectomy. ⁴⁴

In general, the management should be made at individualizing care based on avoiding hypovolemia and hypotension and evaluating the degree of revascularization, the collateral blood flow of ischemic brain tissue and the extent of infarction. For patients with unsuccessful revascularization, the authors recommend permissive hypertension to augment collateralization or even induced hypertension if there is no neurological improvement.

Cardiac complications are common in AIS and explain up to 20% of the mortality after ischemic stroke. Arrhythmias are present in 57% of patients; elevated serum troponin levels in up to 10 to 18%, and 12% have ventricular wall motion abnormalities on echocardiography. Neurogenic stunned myocardium is a cardiac dysfunction seen in various severe intracranial disorders apart from ischemic stroke, cause by catecholamine release, which lead to contraction band necrosis, an described in specific situations as Takotsubo cardiomyopathy. This name is due to its echocardiographic appearance similar to a Japanese vase. ^{6 ,​ 45 ,​ 46 ,​ 47}

There is no consensus to guide the management of these complications, and the best clinical practice must counterbalance the effects of cardiopulmonary support with the effect of potential neurological worsening. For rapid ventricular rate arrhythmias, beta-blockers and calcium channel blockers are generally used. For hemodynamic instability due to myocardial dysfunction, the therapeutic strategy includes avoidance of fluid overload, supportive care including inotropic drugs and treatment of causal neurologic issues. ^{10 ,​ 48 ,​ 49}

10.6 Cerebral Edema and Intracranial Hypertension Management

The developing of space-occupying edema, which determines progressive neurological deterioration, called malignant cerebral edema, is a life threatening condition in AIS. It occurs in approximately 10% of all ischemic strokes and is associated with an elevated mortality. The main consequences of malignant brain edema are cerebral distortion due to intracranial pressure gradient differences and intracranial hypertension. The common final route of these two phenomena is central or lateral brain herniation. The main cause of intracranial hypertension in AIS is cerebral edema. Other causes are less frequent and required specific treatment such as ventricular drainage of hydrocephalus.

Edema following ischemia and infarction is mediated by various cellular mechanisms: 1) Cytotoxic or ionic edema: the reduced availability of oxygen compromise the energy dependent transport channels of neuron cellular membranes causing loss of ionic gradient and water inflow into neurons, 2) Energy failure of the blood-brain barrier that lead to move fluids into the interstitial space, producing delayed vasogenic edema, 3) Other mechanisms of malignant edema in stroke includes vascular endothelial factors, thrombin and matrix metalloproteinase. ⁵

In recent years, it was found the up-regulation of sulfonylurea receptor (Sur-1) regulated channels as another key molecular event involved in the microvascular dysfunction that generate second injury and edema formation, with novel therapeutic implications. ⁵⁰

Treatment of malignant cerebral edema includes 2 steps: 1) medical transient management and 2) surgical treatment.

1) Medical transient management of cerebral edema and intracranial hypertension:

Stabilization of the airway, breathing and circulation are the first measures for treating intracranial hypertension. Following ABC stabilization, there are general measures to control raised ICP that should be done in all patients: head elevation to 30°, maintenance of head and neck in midline position, avoid tight ties around the neck to improve jugular venous drainage, avoid the so called “lethal H” (hypoxia, hypotension, hypercapnia, hyponatremia, hypo and hyperglycemia), and finally early detection and treating of seizures.

Several specific measures have been employed to transient control of cerebral edema and intracranial hypertension. None of them is supported by strong clinical evidence and its role in the treatment of intracranial hypertension is as much as a transient option while surgical treatment is implemented.

Hyperventilation is employed to reduce ICP due to inducing hypocarbia and cerebral vasoconstriction. Because this effect may worse ischemic injury as it was showed in severe head trauma traumatic brain lesions, prophylactic hyperventilation is not recommending and most authors suggest using it for short period of time as rescue maneuver in patients with clinical signs of brain herniation. ⁵¹

Osmotic therapy reduces brain edema by shifting fluids from interstitial and intracellular spaces to intravascular space due to relative osmotic gradient, in areas of brain with intact blood-brain barrier. The main osmotic agents used in clinical setting are mannitol (in a 20% concentration) and hypertonic saline (in concentrations ranging from 3 to 23.4%). A Cochrane review found that the use of mannitol in AIS patients is not supported by evidence from randomized controlled trials. ⁵² For hypertonic saline the evidence is similar, although a meta-analysis comparing both osmotic agents found greater ICP reductions favoring hypertonic saline despite the small number and size of eligible trials included in the analysis. ⁵³

Specifically in stroke patients, there is some evidence favor hypertonic saline in reducing raised ICP in patients in which mannitol failed to control it. ⁵⁴ It is important to note that osmotic therapy should be guided by ICP monitoring and ICP-blinded therapy is not recommended.

Barbiturates are a therapeutic option for treating cerebral edema refractory to other medical measures. There is no evidence of its benefit in the management of increased ICP in stroke patients, added to its association with significant hypotension. For these reasons barbiturate therapy is not recommended to treat cerebral edema in strokes patients. ⁵⁵

While these therapies aim to treat edema once it has developed, recent preclinical and phase-2 clinical studies have shown that intravenous glyburide, a Sur-1 receptor inhibitor, blocks edema formation in AIS patients. A randomized clinical trial is ongoing and might contribute to evidence in its clinical benefit. ⁵⁶

2) Surgical treatment of malignant brain edema: When significant brain swelling occurs in the cerebral hemisphere or cerebellum, surgical decompression should be considered to relieve the mass effect on the middle brain structures and brainstem, and so, avoid major neurological deterioration and death.

There are two scenarios in which surgical treatment has a place in AIS setting: malignant middle cerebral artery (MCA) infarction and cerebellar infarction.

2.1) Malignant MCA infarction.

The term refers to a large supratentorial and hemispheric infarct that occurs as a result of a proximal MCA or internal carotid artery occlusion with significant neurological deterioration. There are variable definition criteria for this denomination: NIHSS score > 15–20 points, brain computed tomography ischemic signs involving > 50% of the MCA territory or magnetic resonance diffusion-weighted imaging infarct volume > 145 cc. With best medical treatment, mortality ranges up to 70–80%. The most effective treatment strategy in this context, is decompressive hemicraniectomy (DHC). ^{15 ,​ 57 ,​ 58}

Three randomized clinical trials (DECIMAL, DESTINY and HAMLET) have demonstrated the benefits of early DHC in patients younger than 60 years with malignant cerebral infarct.

A pooled analysis of them showed reduced mortality with DHC compared with medical management (22% versus 71% mortality) and improvement in the percentage of survivors

with good outcomes (mRS score, 0–3: 43% versus 21%). ^{59 ,​ 60}

Despite these unequivocal effects on survival and functional outcome, some issues remain uncertain about decompressive craniectomy in ischemic stroke patients:

1. 
   

   The selection criteria or optimal trigger to perform the DHC. There is an ongoing controversy whether to wait for signs of neurological deterioration, a major midline shift (probably > 5 mm on computed tomography), or whether to operate as soon as the diagnosis of MCA infarction is made. A strategy of ‘‘prophylactic’’ DHC leads to overutilization, whereas policy of waiting for signs of deterioration may aggravate functional outcome. The authors recommend an individualized clinical decision prioritizing early decompression before signs of herniation occur. ³

   
   
2. 
   

   Surgical technical aspects have not been clarified at all. The size of the DHC is a very important variable that needs to be addressed, since a size smaller than 12 cm has been linked to poor results and increased cerebral complications, so it must be considered suboptimal. Another aspect is the inclusion of the temporal bone resection at the base, to maximize the decompressive effect of the brainstem. Other surgical decisions, such as the storage of the bone flap have not been prospectively studied. ⁶¹

   
   
3. 
   

   The decision of perform a DHC in patients older than 60 years must take in consideration patients and family wishes, since in this age group DHC can reduce mortality but with a higher likelihood of being severely disabled, as it was showed in the DESTINY II trial. ^{62 ,​ 63}

   
   
4. 
   

   Strokectomy or temporal lobectomy could be a potential therapeutic strategy in selected patients who develop clinical failure of DHC. That group of patients characterizes by a persistent and dramatic midline shift after DHC with neurologic deterioration. ⁶⁴

2.2) Cerebellar infarction.

Cerebellar edema complicates patients with cerebellar infarction in 17 – 54% of cases and is generally related to the posterior inferior cerebellar artery (PICA) territory infarction. A rapid neurological deterioration is typically associated with this situation because the posterior fossa provides a little space for compensation mass effect.

Although, the best surgical approach for malignant cerebellar infarction is a matter of debate, a sub-occipital craniectomy with or without resection of necrotic tissue is recommended after significant cerebellar infarction, in which posterior fossa mass effect produce brainstem distortion, ascending or descending herniation or forth ventricle compression with obstructive acute hydrocephalus. ^{65 ,​ 66}

Ventriculostomy should be considered in addition of craniectomy when an obstructive hydrocephalus occurs in this context.

Although the efficacy of surgical treatment of malignant cerebellar infarction has been reported in observational studies, prospective randomized clinical trials will probably never be performed due to depriving affected patients of a life-saving intervention may be considered unethical. In this context, the Statement for Healthcare Professionals from the American Heart Association and American Stroke Association (AHA/ASA) recommends a suboccipital craniectomy with dural expansion in patients with cerebellar infarctions who deteriorate neurologically despite maximal medical therapy (Class I; Level of Evidence B) ^{67 ,​ 68} .