Decompressive Craniectomy for Infarction and Intracranial Hemorrhages

Key Points

Patients with subtotal or complete infarction of the middle cerebral artery (MCA) territory are at high risk for space-occupying edema formation and increased intracranial pressure (ICP). The diagnosis of malignant MCA infarction is based on clinical and radiologic criteria. Patients who present with a severe hemispheric syndrome, deterioration of consciousness with reduced ventilator drive, and definite infarction on neuroimaging of at least two-thirds of the MCA territory including the basal ganglia are at high risk for malignant brain edema.
Medical therapies have not been proven effective in clinical trials, but they may be considered in addition to decompressive hemicraniectomy for patients who deteriorate due to increased ICP.
Decompressive hemicraniectomy for malignant MCA infarction increases the odds of survival, but is associated with increased rates of severe disability, especially in patients older than 60 years. Therefore, the decision toward surgery should be taken on an individual basis after advising patients and families about potential outcome states.
In large space-occupying cerebellar infarction with clinical deterioration, ventriculostomy (extraventricular drainage [EVD]) should be considered if hydrocephalus is present. Otherwise, suboccipital decompressive craniectomy (SDC) with or without strokectomy can be considered, depending on the specific features of the individual patient.
In cerebellar hemorrhage, surgical hematoma evacuation and/or EVD placement are frequently performed in patients with large (>3 cm) hematoma, basal cistern, and brainstem compression, and clinical deterioration due to hydrocephalus.
Decompressive craniectomy is not the treatment of first choice for spontaneous cerebral hemorrhage or subarachnoid hemorrhage. It may, however, be considered for the treatment of refractory increased ICP in these disorders. If decompressive hemicraniectomy is considered, the procedure should be performed early to prevent secondary neurologic damage and herniation. If decompressive hemicraniectomy is considered in spontaneous cerebral hemorrhage, it is reasonable to consider additional hematoma evacuation.

Introduction

Cerebral ischemia and cerebral infarction are often complicated by subacute cerebral edema formation, which can cause life-threatening mass effect. Intracranial hemorrhage can also result in mass effect due to the space-occupying hematoma itself, which may be further impacted by brain edema formation. Regardless of its cause, intracranial mass effect can lead to transtentorial or transforaminal herniation and death or severe neurologic morbidity. Therefore, clinicians should be alert to this possibility and react promptly and effectively. This chapter deals with the surgical decompression of acute, space-occupying malignant middle cerebral artery (MCA) infarction, large cerebellar infarction, intracerebral and intracerebellar hemorrhage, and subarachnoid hemorrhage (SAH).

The pathophysiology of mass effect in ischemic stroke is based on localized cerebral edema resulting in space-occupying brain swelling. Cerebral edema is defined as abnormal parenchymal accumulation of fluid and is classified as cytotoxic, vasogenic, and interstitial. Cytotoxic edema is the result of cellular damage and swelling, associated with a decrease in oxygen supply, failure of energy production, and breakdown of the cell membrane ion pumps. Vasogenic edema is associated with an abnormal increase in vascular and blood-brain barrier (BBB) permeability. Interstitial edema is caused by decreases in cerebrospinal fluid (CSF) reabsorption resulting in acute hydrocephalus. Cytotoxic edema is the primary component of cerebral edema in severe ischemic stroke, but the complete spectrum of mechanisms involved is not yet completely understood. Regardless of its cause, cerebral edema results in intracranial mass effect, causing a shift of brain tissue, compression of adjacent brain structures, steep increases in intracranial pressure (ICP), and potential compromise of cerebral blood flow (CBF), with oxygen and energy deprivation.

Malignant Middle Cerebral Artery Infarction

Large hemispheric stroke patterns resulting from complete or near-complete occlusion of the MCA or the internal carotid artery (ICA) account for 1%–10% of all supratentorial infarctions ( Fig. 78.1 ). ^, These cases are characterized by subsequent development of brain edema and space-occupying mass effect, usually manifesting between the second and fifth day after symptom onset. Massive cerebral edema formation can lead to rapid clinical deterioration, decline in consciousness and, eventually, transtentorial or transforaminal brain herniation and death in about 80% of patients, despite maximal medical treatment. To indicate its severity and importance, the term malignant middle cerebral artery infarction (mMCAI) has been coined for this condition.

Fig. 78.1, Noncontrast computed tomography scan of a patient with right middle cerebral artery (MCA) territory infarction seen to include nearly the entirety of the right MCA territory. There is extensive edema, sulcal effacement, and new 5-mm contralateral midline shift. The patient’s clinical picture included dense left hemiplegia and progressive loss of consciousness, indicating malignant left middle cerebral artery infarction.

Pathophysiology

Cerebral edema in mMCAI is a self-precipitating condition, which sequentially causes mass effect, an increase in ICP, and reduced cerebral perfusion pressure (CPP), resulting in further brain tissue damage, which in turn leads to additional edema formation. Therapeutic goals are always centered at disrupting this vicious cycle before herniation and irreparable brain damage occurs. Medical management includes various measures aimed at minimizing the metabolic demands of the brain, while optimizing blood supply and oxygenation. ^, While multiple conservative treatment modalities are instituted routinely as soon as cerebral edema is suspected, they have not been shown to provide a significant benefit in mMCAI outcomes, and are therefore primarily used as bridging stabilizing modalities before more permanent therapeutic measures ( Table 78.1 ).

TABLE 78.1

Treatment Recommendations and Level of Evidence for Cerebral and Cerebellar Edema in Stroke.

From Powers WJ, Rabinstein AA, Ackerson T, et al. 2018 Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke . 2018;49:e46–e110.

Treatment Recommendation	Classification of Recommendation, Level of Evidence
Patients with large territorial supratentorial infarctions are at high risk for complicating brain edema and increased intracranial pressure. Discussion of care options and possible outcomes should take place quickly with patients (if possible) and caregivers. Medical professionals and caregivers should ascertain and include patient-centered preferences in shared decision making, especially during prognosis formation and considering interventions or limitations in care.	I, C
Patients with major infarctions are at high risk for complicating brain edema. Measures to lessen the risk of edema and close monitoring of the patient for signs of neurologic worsening during the first days after stroke are recommended. Early transfer of patients at risk for malignant brain edema to an institution with neurosurgical expertise should be considered.	I, C
In patients ≤60 years of age with unilateral MCA infarctions who deteriorate neurologically within 48 h despite medical therapy, decompressive craniectomy with dural expansion is reasonable because it reduces mortality by close to 50%, with 55% of the surgical survivors achieving moderate disability (able to walk) or better (mRS score 2 or 3) and 18% achieving independence (mRS score 2) at 12 months.	II, A
In patients >60 years of age with unilateral MCA infarctions who deteriorate neurologically within 48 h despite medical therapy, decompressive craniectomy with dural expansion may be considered because it reduces mortality by close to 50%, with 11% of the surgical survivors achieving moderate disability (able to walk [mRS score 3]) and none achieving independence (mRS score ≤2) at 12 months.	II, B
Although the optimal trigger for decompressive craniectomy is unknown, it is reasonable to use a decrease in level of consciousness attributed to brain swelling as selection criteria.	II, A
Ventriculostomy is recommended in the treatment of obstructive hydrocephalus after a cerebellar infarct. Concomitant or subsequent decompressive craniectomy may or may not be necessary on the basis of factors such as infarct size, neurologic condition, degree of brainstem compression, and effectiveness of medical management.	I, C
Decompressive suboccipital craniectomy with dural expansion should be performed in patients with cerebellar infarction causing neurologic deterioration from brainstem compression despite maximal medical therapy. When deemed safe and indicated, obstructive hydrocephalus should be treated concurrently with ventriculostomy.	I, B
When considering decompressive suboccipital craniectomy for cerebellar infarction, it may be reasonable to inform family members that the outcome after cerebellar infarct can be good after suboccipital craniectomy.	II, C

MCA , Middle cerebral artery; mRS , modified Rankin score.

Decompressive craniectomy (DC) has been touted as the only treatment option that can reliably lead to improved survival rates in mMCAI, albeit with high rates of neurologic deficits and functional dependence for survivors. The rationale behind DC is to remove part of the neurocranium to create space for the expanding brain, minimize brain shift, and prevent ICP elevation and secondary brain tissue damage. ^, However, removing the intracranial hypertension component results in a modest increase in total cerebral tissue swelling, as well as a loss of reflexive arterial hypertension, leaving cerebral perfusion largely unchanged.

Diagnosis

Although the term “malignant brain infarction” was introduced in 1996, there is still no consensus on the diagnostic criteria for the condition. Early prediction of a malignant course can inform clinical decisions so that surgical intervention can be taken in a timely manner. However, most patients with MCA infarctions do not follow a malignant course and will recover with conservative treatment only. Currently, mMCAI diagnosis is based on (1) early and severe neurologic deficits, (2) progressive loss of consciousness, and (3) brain imaging findings suggestive of a large territorial infarct with small penumbra ( Box 78.1 ). ^,

1.
Patients with mMCAI typically present early in their course with severe neurologic symptoms, including dense hemiparesis, gaze deviation, higher cortical deficits, such as multimodal hemineglect, visuo-spatial defects, and global aphasia if the dominant hemisphere is involved. Notably, infarctions of the nondominant hemisphere may be underestimated by the National Institutes of Health stroke scale (NIHSS). The NIHSS score is typically higher than 16–20 when the dominant hemisphere is involved, and higher than 15–18 when the nondominant hemisphere is involved. Older patients may have a more protracted course, due to the presence of cerebral atrophy and compensatory space inside the cranium.
2.
A distinguishing characteristic of mMCAI is the formation of cerebral edema and intracranial hypertension. Early signs include headache, vomiting, and papilledema. A deterioration in consciousness follows, usually over the first 24–48 hours, with a Glasgow Coma Scale of less than 14 or with a score of at least 1 on item 1a of the NIHSS. ^, ^, ^, This is frequently associated with loss of respiratory drive, requiring intubation. Other causes of early deterioration after stroke include failure of collaterals, recurrent stroke, hemorrhagic transformation, seizures, metabolic disturbances, infections, and drugs, all of which should be ruled out.
3.
Imaging supports the diagnosis of mMCAI if there are hypodense changes on computed tomography (CT) imaging extending beyond 50% of the MCA territory, including the basal ganglia. ^, ^, Infarction of the ipsilateral anterior cerebral artery (ACA) or posterior cerebral artery (PCA) further supports the diagnosis but is not required. Larger infarct size on magnetic resonance imaging (MRI), specifically using diffusion-weighted imaging (DWI) or apparent diffusion coefficient (ADC) mapping, is highly predictive of a malignant course. ^,

BOX 78.1

Patients With Acute Middle Cerebral Artery Infarction, With a High Risk of a Malignant Course, and Candidates for Early (<48 Hours) Hemicraniectomy

Age 18–60 years
Severe middle cerebral artery (MCA) syndrome: dense hemiplegia, head and eye deviation, multimodal neglect, global aphasia (in dominant-hemisphere infarction)
National Institutes of Health stroke scale (NIHSS) score >15 (in nondominant-hemisphere infarction) or >20 (in dominant-hemisphere infarction)
Level of consciousness: score of ≥1 on item 1a of the NIHSS or <14 on Glasgow Coma Scale (GCS)
Deterioration of consciousness within the first 48 hours after symptom onset and/or reduced ventilatory drive
Neuroimaging: definite infarction of ≥2/3 of the MCA territory, at least partially including the basal ganglia; additional infarction of the anterior or posterior cerebral artery optional (CT or MRI using diffusion-weighted imaging [DWI] and perfusion imaging); and/or DWI lesion volume >145 mL on DWI and/or >82 mL on apparent diffusion coefficient maps (MRI)

Computed Tomography

Noncontrast CT is widely available in most institutions, and it is the first diagnostic procedure to be performed in acute stroke. Within the first few hours after symptom onset, subtle attenuation changes appear within the gray matter, which appears darker and isointense to the surrounding white matter. These early changes manifest as loss of cortical gray-white matter differentiation, and loss of distinction of the lentiform nucleus and of the insular ribbon, while early edema results in sulcal effacement. Hypoattenuation of the white matter follows, which results in a well-distinguished infarct area. Subsequent cerebral edema, due to extensive infarction, can result in intracranial mass-effect, midline shift, compression of adjacent brain structures, and eventually herniation.

A CT scan can be of substantial help in predicting a malignant course early in the patient’s hospitalization (see Fig. 78.1 ). Early (within 5 hours) hypodensity extension over more than 50% of the MCA territory is a strong prognostic factor for subsequent edema formation and development of mMCAI, with 61% sensitivity and 94% specificity (85% positive predictive value [PPV], 83% negative predictive value [NPV]). ^, ^, Patients with an ASPECTS score of 7 or more are highly likely to develop a malignant course (50% sensitivity, 86% specificity). Additional infarction of the ACA, the PCA, or the ICA reliably predicts worse prognosis but is not required for the diagnosis of mMCAI. ^, ^, Furthermore, the presence of midline shift greater than 3.9 mm the first day after symptom onset, as well as any horizontal displacement of the pineal gland have been shown to be strongly associated with high mortality rates. ^,

Notably, the overall predictive values of routine CT imaging are too low to predict a malignant course with any reliability. Advanced CT technologies, using multimodal imaging such as CT angiography (CTA) and perfusion CT (PCT) mapping may be of additional value to diagnosis ( Fig. 78.2 ). Early involvement of more than two-thirds of the MCA territory on PCT mapping can predict mMCAI with a sensitivity of 92% and specificity of 94%. PCT also assesses the permeability of the BBB, which may indicate the vasogenic component of edema formation. The collateral circulation can be assessed on CTA, with a score less than two being an independent predictor of malignant cerebral edema.

Fig. 78.2, Computed tomography perfusion imaging in the setting of known right M1 occlusion. On the perfusion color maps, there is decreased cerebral blood volume ( CBV <30% volume: 125 mL) and flow (rCBF) involving the entire right middle cerebral artery territory with nearly matching increased mean transit time (MTT) and T-max ( T-max >6.0 seconds volume: 137 mL), consistent with an almost nonexistent penumbra (mismatch volume: 12 mL). This constellation is highly predictive for a malignant course.

Magnetic Resonance Imaging

MRI sequences, including DWI and ADC mapping, may allow for earlier and more accurate prediction of life-threatening brain edema in acute MCA infarction. Lesion volume greater than 145 mL on DWI within 14 hours of symptom onset predicted malignant course with 100% sensitivity and 94% specificity on a retrospective study. Similarly, on 6-hour ADC mapping, lesion volume greater than 82 mL had 87% sensitivity and 91% specificity to predict this course. A subsequent prospective multicentric study demonstrated high PPV but low NPV for these methods (88% and 52%, respectively). To account for brain atrophy in older patients, Goto et al. demonstrated infarct-to-brain volume of more than 7.8% to have sensitivity 86% and specificity 87% in predicting mMCAI.

Detection of a large infarct volume on DWI is fairly specific and should prompt early measures to deal with impending edema formation. However, this fails to account for the imperfect sensitivity of this modality to predict mMCAI. Therefore, patients with lower infarct volumes in the range of 72–82 mL should undergo a repeat CT scan within 6–12 hours after symptom onset, especially if they suffered a severe stroke (NIHSS > 18). ^, ^,

Nuclear Imaging

Single-photon emission computed tomography (SPECT) and positron emission tomography (PET) can provide quantitative measurements of CBF to determine the actual size of the infarcted tissue (ischemic core) and the borders of the hypoperfused, viable penumbra. Characteristically, mMCAI displays a small penumbral area in comparison to the total infarction size. 11C-flumazenil (FMZ) PET can be used to define the ischemic core, where irreversible neuronal damage is characterized by reduced tracer accumulation. In FMZ-PET performed within 24 hours after symptom onset, patients with malignant edema formation demonstrate a larger ischemic core, lower CBF within that core, and a smaller ischemic penumbra volume than patients with a benign course. ^, Pronounced activity deficits on early (6-hour) SPECT using technetium-99m ethyl cysteinate dimer (Tc-99m ECD) predicted aggressive edema formation with 82% sensitivity and 99% specificity, which was more accurate than CT changes or clinical features. Another study, using Tc-99m diethylenetriaminepentaacetic acid (DTPA) SPECT to measure BBB disruption 36 hours after symptom onset, demonstrated the extent of DTPA distribution to be significantly associated with brain herniation later in the course of the patient. These findings indicate that these modalities could predict the development of malignant infarction fairly accurately. However, PET and SPECT studies are only available in specialized research facilities and may not be as well-suited for the selection of patients for invasive treatment in the clinical setting.

Treatment Modalities

Nonoperative Treatment

The aim of conservative treatment measures in mMCAI is to improve oxygen delivery by increasing CPP and minimizing the metabolic needs of the brain. General measures such as adequate oxygen supply, maintenance of adequate blood pressure, and optimal body and head positioning are necessary. Specific measures used in cerebral edema include deep sedation, barbiturates, buffers, hypothermia, osmotic therapy, steroids, and controlled hyperventilation. However, use of these therapies in mMCAI is controversial; there is inadequate evidence to support these measures, ^, ^, and some reports suggest that they could even be detrimental. ^,

Osmotic therapy is reasonable in patients with clinical deterioration from mMCAI-related cerebral edema. There is, however, limited data on the benefit of this treatment (recommendation class IIa). Osmotic therapy requires an intact BBB to work, which may not be the case in the vicinity of the infarcted territory. Caution is advised, as osmotic therapy may shrink normal cerebral tissue to a greater extent than it does the infarct, potentially aggravating brain shift. Hyperventilation is a useful measure to rapidly reduce cerebral edema in mMCAI. Its mechanism of action, however, is through cerebral vasoconstriction. If hypocapnia is too prolonged or extreme, this could lead to worsened ischemia. Therefore, hyperventilation should be rapidly induced with a target P co ₂ of 30–34 mm Hg and only maintained as briefly as possible.

Conversely, hypothermia or barbiturates in the setting of infarct-induced cerebral edema are currently not recommended. While previous data from smaller studies had shown promise, a more recent meta-analysis of six randomized controlled trials (RCTs) demonstrated that these measures have no significant impact on stroke outcomes. Importantly, a recent RCT comparing post-craniectomy moderate hypothermia (33°C for 72 hours) versus standard treatment found no difference in long-term functional outcomes, but significantly higher adverse event rates in the hypothermia group, and was terminated early for safety concerns. Corticosteroids have also shown no evidence of efficacy in the cerebral edema of mMCAI. Due to their additional potential for infectious complications, they are contraindicated for the management of stroke (see Table 78.1 ).

Currently, glyburide, a sulfonylurea historically used for diabetes mellitus, has shown promise in laboratory and clinical trials in reducing brain edema in stroke. ^, The drug targets sulfonylurea receptor 1–transient receptor potential melastatin 4 (SUR1–TRPM4) channels in infarct-associated cerebral edema. SUR1–TRPM4 co-assembles with aquaporin-4 to mediate astrocyte cellular swelling and is associated with matrix metalloproteinase-9 secretion due to recombinant tissue plasminogen activation in brain endothelial cells and hemorrhagic conversion. Glyburide seems to impede these effects and has shown promise in the Glyburide Advantage in Malignant Edema and Stroke (GAMES) clinical trials, regarding midline shift, functional outcomes, and mortality.

Extraventricular Drain

The role of ventricular drainage in the setting of acute ischemic stroke or temporal hematoma has not been well established. There are few studies that address this specific issue. While ventricular drainage has been described on an anecdotal basis, we do not recommend this as a temporizing measure or treatment modality for malignant cerebral edema. In fact, measurement of ICP can be misleading in the setting of unilateral malignant edema. The mass effect and herniation caused by ischemic stroke are due to a focal increase in pressure and, typically, localized temporal lobe edema. There have been instances of patients who have developed cerebral herniation and expired despite the fact that their ICPs were only slightly elevated. In such situations, ICP monitoring through a ventricular drain may be misleading in the management of patients with unilateral malignant cerebral edema.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here