Abstract
Among medical therapies, osmotherapy with colloidal agents (mannitol, glycerol) or crystalloid agents (hypertonic saline) is reasonable for patients whose condition is deteriorating due to mass effect and herniation from large anterior circulation hemispheric or cerebellar infarcts, especially as bridging therapies to definitive surgical intervention. Hyperventilation and hypothermia may also be reasonable, though are of less certain benefit. Corticosteroids have not been found helpful. Agents that block sulfonylurea receptor-mediated cellular swelling, such as intravenous glyburide, have shown promise but require pivotal trial testing. Ventriculostomy is useful to treat non-communicating hydrocephalus arising from obstruction of CSF flow pathways by swollen brain infarcts. But in select large cerebellar infarcts, ventricular drainage alone may exacerbate upward herniation of swollen cerebellar tissues, and suboccipital craniectomy along with ventricular drainage is preferred. Suboccipital craniectomy, often with resection of infarcted tissue, is recommended for massive cerebellar infarcts that may herniate directly into the brainstem, on the basis of observational evidence. Hemicraniectomy for brain oedema associated with large anterior circulation hemispheric infarction is life-saving, but survivors often have severe disability, especially when over age 60. If pursued, hemicraniectomy is generally best performed within 48 h of onset and before development of advanced herniation and neurological deterioration.
Rationale
Brain oedema in ischaemic stroke sufficient to cause mass effect and herniation is a complication of large infarcts, which account for up to 15% of patients with anterior circulation infarcts and approximately 20% of patients with cerebellar infarcts (Wijdicks et al., 2014). Cerebral oedema in large infarcts produces additional injury, as regionally expanding oedematous tissues compress and compromise adjacent intact brain tissues and cranial nerves, occlude cerebral arteries at points of herniation leading to cerebral infarcts in other vascular territories, or occlude cerebrospinal fluid (CSF) channels causing hydrocephalus. In ischaemic strokes, oedema in the brain arises through two pathophysiological processes: (1) cytotoxic oedema, increased intracellular fluid due to neuronal energetic failure, and (2) vasogenic oedema, increased extracellular fluid due to blood–brain barrier disruption, with cytotoxic oedema providing the greater contribution.
Microscopic cytotoxic oedema begins to develop within the first minutes after ischaemic stroke onset, when it can be detected on magnetic resonance imaging (MRI) as a decrease in the apparent diffusion coefficient of water. However, macroscopically evident mass effect arising from more advanced cytotoxic oedema and vasogenic oedema develops over a longer time period, typically peaking 3–5 days after ischaemic stroke onset.
The management of substantial brain oedema aims to:
reduce mass effect;
prevent secondary injury from herniation;
resolve hydrocephalus, if present; and
maintain adequate cerebral perfusion pressure, sufficient to overcome raised intracranial pressure.
Therapies that have been used in practice to treat brain oedema and its complications include:
elevating head of bed to 20–30 degrees to assist cerebral venous drainage;
mild fluid restriction to deter brain fluid accumulation (but with potential adverse effect on cerebral perfusion pressure);
avoiding hypo-osmolar fluids, such as 5% dextrose in water, to prevent osmotic pull of intravascular water into brain tissue;
administering intravenous crystalloids (hypertonic saline) or colloids (such as mannitol or glycerol) to osmotically draw water out of brain parenchyma;
avoiding cerebral veno-dilating drugs as they expand the intracranial blood volume within the intravascular compartment;
hyperventilation, to induce vasoconstriction and reduce the intracranial blood volume within the intravascular compartment;
hypothermia to reduce intravascular blood volume and stabilize the blood–brain barrier;
corticosteroids to stabilize the blood–brain barrier and reduce vasogenic oedema;
glyburide to decrease water flux across neuronal and glial cell membranes and reduce cytotoxic oedema;
drainage of CSF fluid; and
surgical decompression.
Some of these diverse therapeutic options have been formally tested in randomized clinical trials, while several have been evaluated only in uncontrolled observational series.
Hyperventilation
Reductions in the partial pressure of carbon dioxide (PCO2) produce cerebral vasoconstriction and reduced cerebral blood volume. Hyperventilation reducing PCO2 by 5–10 mm Hg can lower intracranial pressure by 10–40% (Gujjar et al., 1998; Coles et al., 2007). However, this degree of hyperventilation will also reduce cerebral blood flow by 15–30% (Coles et al., 2007), which could aggravate ischaemic brain injury. No randomized trials of hyperventilation for brain oedema in ischaemic stroke have been conducted. Based on its physiological effect profile, hyperventilation, if used at all, likely should be deployed primarily as a temporary measure while preparing surgical interventions to definitively control brain oedema and mass effect.
Corticosteroids
Evidence
Death
A systematic review of eight randomized controlled trial (RCTs) involving 466 patients with early ischaemic stroke revealed that, compared with control, random assignment to corticosteroid treatment within 48 hours of stroke onset did not alter mortality within 1 month (odds ratio [OR]: 0.97, 95% confidence interval [CI]: 0.63–1.47) (Figure 11.1) or within 1 year (OR: 0.87, 95% CI: 0.57–1.34) (Sandercock and Soane, 2011).
Figure 11.1 Forest plot showing the effects of corticosteroids vs control in early ischaemic stroke on death from all causes during study follow-up.
Neurological Impairment
Six trials reported neurological impairment outcomes, but used different neurological deficit assessment scales, precluding formal meta-analytic pooling across studies. At the individual study level, four trials reported no difference, one benefit, and one worsening, in neurological impairment with steroid therapy (Sandercock and Soane, 2011).
Adverse Effects
Adverse effects were not systematically reported in most randomized trials of corticosteroids in ischaemic stroke. However, diabetes or hyperglycaemia was reported to be non-significantly nominally more frequent with steroid therapy in four trials.
Comment
Interpretation of the Evidence
The accumulated randomized trial evidence does not indicate a beneficial effect of corticosteroids in patients with early ischaemic stroke on either death or neurological impairment.
Implications for Practice
Corticosteroid should not be used in the routine management of brain oedema after acute ischaemic stroke.
Implications for Research
The amount of evidence across all trials is modest, and the great preponderance of studies pre-date the modern era of reperfusion therapy for acute ischaemic stroke. While sufficient data have been accumulated to suggest that corticosteroids are not likely to be a useful broad therapy for all patients with ischaemic stroke, the possibility remains that steroid therapy may be effective for targeted subtypes of patients. Studies may be warranted that are confined to patients with large infarcts and substantial cerebral oedema and that deploy higher doses of corticosteroids (e.g. methylprednisolone 500–1000 mg/day), which may be more effective for vasogenic oedema, and for a shorter duration to avoid potential adverse effects (Davis and Donnan, 2004; Norris, 2004). Also, studies would be of interest in patients with imaging evidence of early blood–brain barrier disruption (e.g. abnormal permeability imaging; Nael et al., 2017), or of early cerebral endothelial injury (e.g. vessel wall enhancement; Renú et al., 2017), identifying patients at increased risk for both vasogenic oedema and haemorrhagic transformation.
Osmotic Agents
Osmotic agents are molecules that do not readily cross the blood–brain barrier, so that their systemic administration yields a higher solute concentration in the blood than in the brain. The resulting osmotic gradient draws free water from brain intracellular and interstitial compartments into the systemic intravascular space. The two main categories of osmatic agents are (1) colloids: large, insoluble molecules, and (2) crystalloids: small, water-soluble molecules (Lewis et al., 2018). For brain oedema, the most well-studied agents are the colloids mannitol and glycerol and the crystalloid hypertonic saline (Fink, 2014).
Mannitol
Mannitol is a colloidal osmotic agent and a free radical scavenger. Mannitol (0.25–0.6 g/kg, e.g. 20–40 g) intravenously over 30 minutes lowers intracranial pressure (ICP). It can be given every 6 hours to a maximum daily dose of 2 g/kg (e.g. about 140 g in a 70 kg person). Mannitol decreases brain oedema by exerting osmotic pressure that draws water out of the brain interstitium.
Evidence
Physiological Effects
In a series of ischaemic stroke patients, mannitol was associated with a shrinking of the volume of the contralateral hemisphere by 8% (Videen et al., 2001), and in another study mannitol was associated with positron emission tomography (PET) scan findings of an increase in cerebral blood flow in peri-infarct regions by 18% (p = 0.05) and a trend toward increase in cerebral blood flow in the contralateral hemisphere by 27% (p = 0.11) (Diringer et al., 2011). However, mannitol becomes less effective with repeated doses, and delayed ‘rebound’ increases in ICP can occur as mannitol eventually crosses the blood–brain barrier and enters the cerebral compartment.
Clinical Outcomes
A systematic review identified only one unconfounded RCT in which 77 patients with ischaemic stroke were allocated to mannitol or control (Santambrogio et al., 1978; Bereczki et al., 2007). The mannitol dose was 0.8–0.9 g/kg for 10 days, and the study enrolled both patients with small and those with large ischaemic strokes. The analysed outcomes were the proportion of patients with clinical improvement, clinical stability, and clinical worsening. The outcomes of dependency, death, or adverse effects were not reported. In this small trial, mannitol was not associated with alteration in frequency of improved, unchanged, or worsened patient clinical course as compared with controls. Rates of clinical improvement with mannitol vs control were 33.3% versus 34.1%, risk ratio (RR) 1.02 (95% CI: 0.55–1.91) (Figure 11.2); and rates of clinical worsening were 44.4% versus 43.9%, RR 0.99 (95% CI: 0.60–1.63) (Figure 11.3).
Figure 11.2 Forest plot showing the effects of mannitol vs control in early ischaemic stroke on clinical improvement.
Figure 11.3 Forest plot showing the effects of mannitol vs control in early ischaemic stroke on clinical worsening.
Adverse Effects
The most common complications of mannitol therapy are fluid and electrolyte imbalances, cardiopulmonary oedema, and rebound cerebral oedema. Mannitol also uncommonly causes kidney failure, and hypersensitivity reactions may occur.
Comment
Interpretation of the Evidence
There is not enough reliable evidence to determine whether mannitol is effective or ineffective in patients with cerebral oedema due to ischaemic stroke.
Implications for Practice
In the absence of adequate randomized, clinical outcome data to provide definitive guidance, firm recommendations regarding the use of mannitol to treat brain oedema from ischaemic strokes cannot be given. However, based on findings of potentially beneficial physiological effects in ischaemic stroke patients, and larger experience with use of mannitol in neurological conditions, it is reasonable to use mannitol for patients with large hemispheric or cerebellar infarcts, especially as a temporizing measure until a definitive procedural/surgical intervention is delivered, or until patients have crossed the imminent timepoint of expected peak brain swelling.
Implications for Research
The clinical efficacy of mannitol in acute ischaemic stroke has not yet been properly evaluated. Large, placebo-controlled, unconfounded randomized trials of mannitol as bridging, temporizing therapy in patients with sizeable, space-occupying infarcts are desirable.
Glycerol
Glycerol is a polyol compound that acts as a colloidal osmotic agent. In observational series among patients with diverse neurological conditions causing raised ICP, including ischaemic stroke, glycerol reduced ICP and increased indices of cerebral perfusion pressure (Biestro et al., 1997; Treib et al., 1998). Compared with mannitol, glycerol had not only a longer time until achievement of peak effect, but also a longer duration of effect without rebound.
Evidence
A systematic review identified 7 randomized trials, enrolling 601 patients, comparing intravenous glycerol with control, initiated within the first days after onset, in patients with definite or presumed ischaemic stroke (Righetti et al., 2004).
Death
Data regarding death within the scheduled treatment period was available from all 7 trials enrolling 601 patients. Random allocation to glycerol versus control was associated with a reduction in early death, 19.6% versus 25.6% (OR: 0.65, 95% CI: 0.44–0.97; p = 0.03). Information on death by end of scheduled follow-up was available from 6 RCTs enrolling 492 patients. Allocation to glycerol versus control was not associated with alteration in death at final follow-up, 36.7% versus 39.5% (OR: 0.85, 95% CI: 0.59–1.24) (Figure 11.4).
Figure 11.4 Forest plot showing the effects of glycerol vs control in early ischaemic stroke on death by the end of follow-up.
Dependency or Death
Data regarding dependency or death (modified Rankin Scale [mRS] 3–6 or nearest equivalent) at end of follow-up were available from 2 randomized trials enrolling 234 patients with ischaemic stroke. Glycerol compared with control did not show a statistically significant association with dependency or death, 79.5% versus 84.1% (OR: 0.73, 95% CI: 0.37–1.42) (Figure 11.5).
Figure 11.5 Forest plot showing the effects of glycerol vs control in early ischaemic stroke on dependency or death at end of follow-up.
Adverse Effects
Haemolysis was the only relevant adverse effect of glycerol treatment.
Comment
Interpretation of the Evidence
These data indicate a favourable effect of glycerol treatment on short-term survival in patients with definite or presumed ischaemic stroke, though mortality differences did not remain statistically significant at long-term follow-up. Due to the relatively modest total sample size, and the fact that some of the trials were performed in the pre-computed tomography (CT) era and so may have included a few patients with primary haemorrhagic stroke misdiagnosed as ischaemic, the results must be interpreted cautiously. Furthermore, many of the studies included not only patients with large infarcts, but also those with mild-to-moderate infarcts.
Implications for Practice
Given the paucity of evidence available regarding effects on functional outcome, and the mixed evidence of potential benefit on mortality, definitive recommendations regarding use of glycerol to treat brain oedema from ischaemic stroke cannot be formulated. Based on findings of potentially beneficial physiological effects and benefit on short-term mortality, it is reasonable to use glycerol in patients with large hemispheric or cerebellar infarcts, especially as a temporizing measure until a definitive procedural/surgical intervention is employed, or until patients cross the imminent timepoint of expected peak brain swelling.
Implications for Research
As glycerol treatment is inexpensive, may be effective, and appears to be safe, it should continue to be tested as a treatment for brain oedema in ischaemic stroke patients, but in much larger RCTs focused upon patients who have clinical and imaging evidence of mass effect, with assessment of long-term functional outcomes as well as survival.
Hypertonic Saline
Hypertonic saline is a crystalloid osmotic agent, producing elevated serum sodium levels that create osmotic pressure drawing water out of brain tissues. Hypertonic saline doses most often studied are a bolus of 23.4% saline for substantial, short-acting effect and a continuous infusion of 3% saline for more modulated, long-lasting effect (Wijdicks et al., 2014; Ong et al., 2015).
Evidence
Physiological Effects
There have been no unconfounded, randomized studies of hypertonic saline confined to patients with ischaemic stroke. In an observational series of 76 transtentorial herniation events among 68 patients with varied neurological conditions (including 8 with ischaemic stroke), bolus therapy with 23.4% hypertonic saline, in tandem with additional medical interventions, was associated with reversal of clinical signs of herniation in 75% of episodes (Koenig et al., 2008). Independent predictors of successful reversal of herniation were an increase in sodium concentration equal to or greater than 5 mmol/L (OR 12.0, 95% CI: 1.6–90.5) and an attained absolute sodium concentration equal to or greater than 145 mmol/L (OR 26.7, 95% CI: 3.6–200.0). In additional studies in patients with diverse neurological causes of raised ICP, including traumatic brain injury, conditions requiring craniotomy, and subarachnoid haemorrhage, hypertonic saline has been shown to reduce ICP, with comparable or slightly greater physiological efficacy compared with mannitol (Li et al., 2015; Pasarikovski et al., 2017; Fang et al., 2018).
Clinical Outcomes
There have been no unconfounded, randomized trials with neurological deficit, disability, or mortality outcomes comparing hypertonic saline with control or with active colloidal osmotic agent therapy in patients with ischaemic stroke.
Adverse Effects
Adverse effects of hypertonic saline are uncommon, but include hyperchloraemic metabolic acidosis, congestive heart failure, acute renal impairment, and seizures (Jeon et al., 2014).
Comment
Interpretation of the Evidence
The absence of any unconfounded randomized trials precludes reliable assessment of whether hypertonic saline is effective or ineffective in patients with cerebral oedema due to ischaemic stroke.
Implications for Practice
In the absence of randomized, clinical outcome data, definite recommendations regarding the use of hypertonic saline to treat brain oedema from ischaemic strokes cannot be advanced. However, based on findings of potentially beneficial physiological effects in ischaemic stroke patients, and larger experience with use of hypertonic saline in other neurological conditions, it is reasonable to use hypertonic saline therapy for patients with large hemispheric or cerebellar infarcts. Using 3% hypertonic saline to counter cerebral salt wasting and maintain euvolaemia while achieving mild hypernatremia with osmotic effects is a reasonable initial strategy in large infarcts. Administration of 23.4% saline bolus in response to signs of progressive herniation or severely elevated ICP also is reasonable as a temporizing measure until a definitive procedural/surgical intervention is delivered.
Glyburide
In preclinical models, blockade by the sulfonylurea agent glyburide of the inducible sulfonylurea receptor 1 (SUR1)-transient receptor potential melastatin 4 (TRPM4) channel in neurons, astrocytes, and endothelium substantially lessens brain oedema (Simard et al., 2010). Oral glyburide has long been used as an antidiabetic agent, but may not yield predictable blood levels in the critically ill patient, so an intravenous formulation has been developed for potential use as an agent to avert development of substantial brain oedema after large cerebral infarcts (King et al., 2018).
Evidence
One randomized, placebo-controlled, phase 2 trial of intravenous glyburide has been conducted, enrolling patients with large anterior circulation hemispheric infarcts (diffusion MRI infarct volumes 82–300 mL) within 10 hours of onset (Sheth et al., 2016). A total of 86 patients were enrolled and analysed for safety, among whom 77 treated per protocol were analysed for efficacy.
Physiological Outcomes
Allocation to glyburide versus control was associated with reduced growth in midline shift from entry to 72–96 hours, 4.6 versus 8.8 mm, mean difference –4.3 mm (95% CI: –6.3 to –2.4). Differences in growth of ipsilateral hemispheric volume were not statistically significant, 68 versus 78 mL, mean difference –13.4 mL (95% CI: –43.4 to 16.6).
Functional Outcomes
The trial’s primary efficacy outcome composite, of avoiding extreme disability (need for continuous care) or death (mRS 5–6) and avoiding decompressive craniectomy, did not differ between glyburide versus control patients (41% versus 39%; OR: 0.87, 95% CI: 0.32–2.32). However, there was a trend for a favourable shift to reduced disability at 3 months across the 7-level mRS, with mean mRS values 4.1 versus 4.6, non-parametric p = 0.12.
Related posts:
Chapter 5 – Supportive Care during Acute Cerebral Ischaemia
Chapter 9 – Acute Antiplatelet Therapy for the Treatment of Ischaemic Stroke and Transient Ischaemic Attack
Chapter 13 – Acute Treatment of Intracerebral Haemorrhage
Chapter 19 – Long-term Antithrombotic Therapy for Large and Small Artery Occlusive Disease
Chapter 25 – Using Pharmacotherapy to Enhance Stroke Recovery
Chapter 26 – Electrical and Magnetic Brain Stimulation to Enhance Post-stroke Recovery
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
